The Inventions, Researches and Writings - Martin, Thomas Commerford 1856-1924

8/2/2019 The Inventions, Researches and Writings - Martin, Thomas Commerford 1856-1924

http://slidepdf.com/reader/full/the-inventions-researches-and-writings-martin-thomas-commerford-1856-1924 1/802



his book made available by the Internet Archive.



























REFACE.

pHE electrical problems of the present day lie largely inhe ** economical transmission of power and in the radicmprovement of the means and methods of illumination.o many workers and thinkers in the domain of electrica

nvention, the apparatus and devices that are familiar,ppear cumbrous and wasteful, and subject to severemitations. They believe that the principles of currenteneration must be changed, the area of current supply

e enlarged, and the appliances used by the consumer bet once cheapened and simplified. The brilliant successesf the past justify them in every expectancy of still moreenerous fruition.

he present volume is a simple record of the pioneer

work done in such departments up to date, by Mr. Nikolesla, in whom the world has already recognized one of he foremost of modern electrical investigators andnventors. No attempt whatever has been made here tomphasize the importance of his researches andiscoveries. Great ideas and real inventions win their ow

way, determining their own place by intrinsic merit. Butwith the conviction that Mr. Tesla is blazing a patli that

lectrical development must follow for many years toome, the compiler has endeavored to bring together allhat bears the impress of Mr. Tesla's genius, and is

worthy of preservation. Aside from its value as showing

he scope of his inventions, this volume may be of services indicatin the ran e of his thou ht. There is intellectu



rofit in studying the push and play of a vigorous andriginal mind.

Althqugh the lively interest of the public in Mr. Tesla'swork is perhaps of recent growth, this volume covers the

esults of full ten years. It includes his lectures,miscellaneous articles

nd discussions, and makes note of all his inventions thusar known, particularly those bearing on polyphase

motors and the effects obtained with currents of high

otential and high frequency. It will be seen that Mr.esla has ever pressed forward, barely pausing for annstant to work out in detail the utilizations that have atnce been obvious to him of the new principles he haslucidated. Wherever possible his own language has beenmployed.

t may be added that this volume is issued with Mr.esla's sanction and approval, and that permission haseen obtained for the re-publication in it of such papers ave been read before various technical societies of thisountry and Europe. Mr. Tesla has kindly favored the

uthor by looking over the proof sheets of the sectionsmbodying his latest researches. The Work has alsonjoyed the careful revision of the author's friend andditorial associate, Mr. Joseph Wetzler, through whoseands all the proofs have passed.

DECEMBER, 1893. T. C. M.







HAPTER VII.

REGULATOR FOR ROTARY CURRENT MOTORS 45

HAPTER VIII. SINGLE CIRCUIT, SELF-STARTINGYNCHRONIZING MOTORS. .. 50

HAPTER IX.

HANGE FROM DOUBLE CURRENT TO SINGLEURRENT

MO.TORS 56

HAPTER X.

MOTOR WITH " CURRENT LAG " ARTIFICIALLY

ECURED 58

HAPTER XL

ANOTHER METHOD OF TRANSFORMATION FROM ORQUE TO

A SYNCHRONIZING MOTOR ... 62

HAPTER XII. " MAGNETIC LAG " MOTOR 67

HAPTER XIII.

METHOD OF OBTAINING DIFFERENCE OF PHASE B



MAGNETIC SHIELDING 71

HAPTER XIV. TYPE OF TESLA SINGLE-PHASEMOTOR 76

HAPTER XV.

MOTORS WITH CIRCUITS OF DIFFERENTRESISTANCE 79

HAPTER XVI.

MOTOR WITH EQUAL MAGNETIC ENERGIES INIELD AND

ARMATURE g^

HAPTER XVII.

MOTORS WITH COINCIDING MAXIMA OK MAGNETIC EFFECT

N ARMATURE AND FIELD 83

HAPTER XVIII.

MOTOR BASED ON THE DIFFERENCE OF PHASE INHE MAGNETIZATION OF THE INNER AND OUTERARTS OF AN IRON CORE 88

HAPTER XIX. ANOTHER TYPE OF TESLA NDUCTION MOTOR 2



HAPTER XX.

OMBINATIONS OF SYNCHRONIZING MOTOR ANDORQUE

MOTOR 95

HAPTER XXI.

MOTOR WITH A CONDENSER IN THE ARMATUREIRCUIT ... 101

HAPTER XXII. MOTOR WITH CONDENSER IN ONEOF THE FIELD CIRCUITS. 106

HAPTER XXIII.

ESLA POLYPHASE TRANSFORMER 1J9

HAPTER XXIV.

A CONSTANT CURRENT TRANSFORMER WITHMAGNETIC

HIELD BETWEEN COILS OF PRIMARY ANDECONDARY 113

ART II.

HE TESLA EFFECTS WITH HIGH FREQUENCY ANDHIGH POTENTIAL CURRENTS.





REQUENCY 374

HAPTER XXX. ALTERNATE CURRENTLECTROSTATIC INDUCTION APPARATUS 392

HAPTER XXXI.

MASSAGE " WITH CURRENTS OF HIGHREQUENCY 394

HAPTER XXXII.

LECTRIC DISCHARGE IN VACUUM TUBES 396

ART III.

MISCELLANEOUS INVENTIONS AND WEITINGS.

HAPTER XXXIII.

METHOD OF OBTAINING DIRECT FROMALTERNATING CURRENTS 409

HAPTER XXXIV. CONDENSERS WITH PLATES IN

OIL 418

HAPTER XXXV.

LECTROLYTIC REGISTERING ]!UETER . 420

HAPTER XXXVI.



HERMO-MAGNETIC MOTORS AND PYRO-MAGNETIC GENERATORS , 424

HAPTER XXXVII. ANTI-SPARKING DYNAMORUSH AND COMMUTATOR 432

HAPTER XXXVIII.

AUXILIARY BRUSH REGULATION OF DIRECTURRENT DYNAMOS 438

HAPTER XXXIX.

MPROVEMENT IN DYNAMO AND MOTOR ONSTRUCTION 448

HAPTER XL. TESLA DIRECT CURRENT ARCIGHTING SYSTEM 451

HAPTER XLI.

MPROVEMENT IN UNIPOLAR GENERATORS . . . 46

ART IV.

APPENDIX : EARLY PHASE MOTORS AND THE TESLOSCILLATORS.

HAPTER XLIL

MR. TESLA'S PERSONAL EXHIBIT AT THE WORLD'SAIR ... .



HAPTER XLIII.

HE TESLA MECHANICAL AND ELECTRICALOSCILLATORS... 486

ART I.

OLYPHASE CURRENTS.

HAPTER I.

IOGRAPHICAL AND INTRODUCTORY.

As AN introduction to the record contained in this volumf Mr. Tesla' s investigations and discoveries, a few wordf "a biographical nature will, it is deemed, not be out of lace, nor other than welcome.

Nikola Tesla was born in 1857 at Smiljan, Lika, aorderland region of Austro-Hungary, of the Serbian rac

w r hich has maintained against Turkey and all comers snceasing a struggle for freedom. His family is an old andepresentative one among these Switzers of Easternurope, and his father was an eloquent clergyman in the

Greek Church. An uncle is to-day Metropolitan in BosniaHis mother was a woman of inherited ingenuity, and

elighted not only in skilful work of the ordinary ousehold character, but in the construction of such

mechanical appliances as looms and churns and othermachiner re uired in a rural communit . Nikola was



ducated at Gospich in the public school for four years,nd then spent three years in the Real Scliule. He washen sent to Carstatt, Croatia, where he continued histudies for three years in the Higher Real Scliule. Thereor the first time he saw a steam locomotive. He

raduated in 1873, and, surviving an attack of cholera,evoted himself to experimentation, especially inlectricity and magnetism. His father would have had him

maintain the family tradition by Altering the Church, butative genius was too strong, and he was allowed to ente

he Polytechnic School at Gratz, to finish his studies, and

with the object of becoming a professor of mathematicsnd physics. One of the machines there experimented

with was a Gramme dynamo, used as a motor. Despite hnstructor's perfect demonstration of the fact that it \vasmpossible to operate a dynamo without commutator orrushes, Mr. Tesla could not be convinced that such

ccessories \\eiv K-cessary or desirable. He had alreadyeen with quick intuition that a way could be found toispense with them ; and from that time he may

e said to have begun work on the ideas that fructifiedltimately in his rotating field motors.

n the second year of his Gratz course, Mr. Tesla gave uphe notion of becoming a teacher, and took up thengineering curriculum. His studies ended, he returnedome in time to see his father die, and then went torague and Buda-Pesth to study languages, with the

bject of qualifying himself broadly for the practice of the



ngineering profession. For a short time he served as anssistant in the Government Telegraph Engineering

Department, and then became associated with M. Puskapersonal and family friend, and other exploiters of the

elephone in Hungary. He made a number of telephonic

nyentions, but found his opportunities of benefiting by hem limited in various ways. To gain a wider field* of ction, he pushed on to Paris and there securedmployment as an electrical engineer with one of the largompanies in the new industry of electric lighting.

t was during this period, and as early as 1882, that heegan serious and continued efforts to embody theotating field principle' in operative apparatus. He wasnthusiastic about it; believed it to mark a new departurn the electrical arts, and could think of nothing else. Inact, but for the solicitations of a few friends in commerci

ircles who urged him to form a company to exploit thenvention, Mr. Tesla, then a youth of little worldly xperience, would have sought an immediate opportunito publish his ideas, believing them to be worthy of notes a novel and radical advance in electrical theory as wels destined to have a profound influence on all dynamo

lectric machinery.

At last he determined that it would be best to try hisortunes in America. In France he had met many

Americans, and in contact with them learned theesirability of turning every new idea in electricity to

ractical use. He learned also of the ready encouragemen



iven in the United States to any inventor who couldttain some new and valuable result. The resolution \v. .ormed with characteristic quickness, and abandoning allis prospects in Europe, he at once set his face westward

Arrived in the United States, Mr. Tesla took off his coatn-day he arrived, in the Edison Works. That place hadeen a-goal of his ambition, and one can readily imaginehe benefit and stimulus derived from association with

Mr. Edison, for whom Mr. Tesla 'aas always had thetrongest admiration. It was impossible, however, that,

with -his own ideas to carry out, and his

wn inventions to develop, Mr. Tesla could long remain iven the most delightful employ ; and, his work now ttracting attention, he left the Edison ranks to join aompany intended to make and sell an arc lighting system

ased on some of his inventions in that branch of the artWith unceasing diligence he brought the system to

erfection, and saw it placed on the market. But the thinwhich most occupied his time and thoughts, however, allhrough this period, was his old discovery of the rotatingeld principle for alternating current work, and the

pplication of it in motors that have now become knownhe world over.

trong as his convictions on the subject then were, it is aact, that he stood very much alone, for the alternatingurrent had no well recognized place. Few electrical

ngineers had ever used it, and the majority were entirenfamiliar with its value or even its essential features.



ven Mr. Tesla himself did not, until after protractedffort and experimentation, learn how to constructlternating current apparatus of fair efficiency. But thate had accomplished his purpose was shown by the testsf Prof. Anthony, made in the of winter 188T-8, when

esla motors in the hands of that distinguished expertave an efficiency equal to that of direct current motors.Nothing now stood in the way of the commercial

evelopment and introduction of such motors, except thahey had to be constructed with a view to operating onhe circuits then existing, which in this country were all o

igh frequency.

he first full publication of his work in this direction—utside his patents — was a paper read before the

American Institute of Electrical Engineers in New York, May, 1888 (read at the suggestion of Prof. Anthony and

he present writer), when he exhibited motors that hadeen in operation long previous, and with which his beliehat brushes and commutators could be dispensed with,

was triumphantly proved to be correct. The section of tholume devoted to Mr. Tesla's inventions in the utilizatiof polyphase currents will show how thoroughly from the

utset he had mastered the fundamental idea and applie in the greatest variety of ways.

Having noted for years the many advantages obtainablewith alternating currents, Mr. Tesla was naturally led ono experiment with them at higher potentials and higher

re uencies than were common or a roved of. Ever



ressing forward to determine in even the slightestegree the outlines of the unknown, he

was rewarded very quickly in this field with results of thmost-surprising nature. A slight acquaintance with some

f these experiments led the compiler of this volume torge Mr. Tesla to repeat them before the Americannstitute of Electrical Engineers. This was done in May,891, in a lecture that marked, beyond question, a distineparture in electrical theory and practice, and all theesults of which have not yet made themselves fully

pparent. The New York lecture, and its successors, twon number, are also included in this volume, with a few upplementary notes.

Mr. Tesla's work ranges far beyond the vast departmentf polyphase currents and high potential lighting. The "

Miscellaneous " section of this volume includes a greatmany other inventions in arc lighting, transformers, pyromagnetic generators, thermo-magnetic motors, third-

rush regulation, improvements in dynamos, new formsf incandescent lamps, electrical meters, condensers,nipolar dynamos, the conversion of alternating into

irect currents, etc. It is needless to say that at thismoment Mr. Tesla is engaged on a number of interestingdeas and inventions, to be made public in due course. Thresent volume deals simply with his work accomplishedo date.

HAPTER II.



A NEW SYSTEM OF ALTERNATING CURRENTMOTORS AND TRANSFORMERS.

HE present section of this volume deals with polyphaseurrents, and the inventions by Mr. Tesla, made known

hus far, in which he has embodied one feature or anothef the broad principle of rotating field poles or resultantttraction exerted on the armature. It is needless toemind electricians of the great interest aroused by therst enunciation of the rotating field principle, or to dwelpon the importance of the advance from a single

lternating current, to methods and apparatus which deawith more than one. Simply prefacing the considerationere attempted of the subject, with the remark that inowise is the object of this volume of a polemic orontroversial nature, it may be pointed out that Mr.esla's work has not at all been fully understood or

ealized up to date. To many readers, it is believed, thenalysis of what he has done in this department will be aevelation, while it will at the same time illustrate theeautiful flexibility and range of the principles involved.

will be seen that, as just suggested, Mr. Tesla did not stohort at a mere rotating field, but dealt broadly with the

hifting of the resultant attraction of the magnets. It wille seen that he went on to evolve the " multiphase "ystem with many ramifications and turns; that hehowed the broad idea of motors employing currents of iffering phase in the armature with direct currents in theld; that he first described and worked out the idea of armature with a body of iron and coils closed upon



hemselves ; that he worked out both synchronizing andorque motors; that he explained and illustrated how

machines of ordinary construction might be adapted to hystem; that he employed condensers in field andrmature circuits, and went to the bottom of the

undamental principles, testing, approving or rejecting, itwould appear, every detail that inventive ingenuity couldit upon.

Now that opinion is turning so emphatically in favor of ower frequencies, it deserves special note that Mr. Tesla

arly recognized the importance of the low frequency eature in motor work. In fact his first motors exhibitedublicly—and which, as Prof. Anthony showed in his test

n the winter of 1887-8, were the equal of direct currentmotors in efficiency, output and starting torque—were ofhe low frequency type. The necessity arising, however,

o utilize these motors in connection with the existing higrequency circuits, our survey reveals in an interesting

manner Mr. Tesla's fertility of resource in this direction.ut that, after exhausting all the possibilities of this field

Mr. Tesla returns to low frequencies, and insists on theuperiority of his polyphase system in alternating curren

istribution, need not at all surprise us, in view of thetrength of his convictions, so often expressed, on thisubject. This is, indeed, significant, and may be regardeds indicative of the probable development next to be

witnessed.

ncidental reference has been made to the efficiency of



otating field motors, a matter of much importance,hough it is not the intention to dwell upon it here. Prof.

Anthony in his remarks before the American Institute oflectrical Engineers, in May, 1888, on the two small Tesl

motors then shown, which he had tested, stated that one

ave an efficiency of about 50 per cent, and the other attle over sixty per cent. In 1889, some tests wereeported from Pittsburgh, made by Mr. Tesla and Mr.

Albert Schmid, on motors up to 10 H. p. and weighingbout 850 pounds. These machines showed an efficiencyf nearly 90 per cent. With some larger motors it was

hen found practicable to obtain an efficiency, with thehree wire system, up to as high as 94 and 95 per cent.hese interesting figures, which, of course, might beupplemented by others more elaborate and of later datre cited to show that the efficiency of the system has noad to wait until the present late day for any

emonstration of its commercial usefulness. An inventions none the less beautiful because it may lack utility, but

must be a pleasure to any inventor to know that the ideae is advancing are fraught with substantial benefits tohe public.

HAPTEE III.

HE TESLA KOTATING MAGNETIC FIELD. —MOTORS WITH CLOSED CONDUCTORS. —

YNCHRONIZING MOTORS. — KOTATING FIELDRANSFORMERS.

HE best descri tion that can be iven of what he



ttempted, and succeeded in doing, with the rotatingmagnetic field, is to be found in Mr. Tesla's brief paper

xplanatory of his rotary current, polyphase system, reaefore the American Institute of Electrical Engineers, in

New York, in May, 1888, under the title " A New System

f Alternate Current Motors and Transformers." As amatter of fact, which a perusal of the paper will establishMr. Tesla made no attempt in that paper to describe all

is work. It dealt in reality with the few topicsnumerated in the caption of this chapter. Mr. Tesla'seticence was no doubt due largely to the fact that his

ction was governed by the wishes of others with whome was associated, but it may be worth mention that theompiler of this volume — who had seen the motorsunning, and who was then chairman of the Instituteommittee on Papers and Meetings— had great difficulty

n inducing Mr. Tesla to give the Institute any paper at

ll. Mr. Tesla was overworked and ill, and manifested threatest reluctance to an exhibition of his motors, but hisbjections were at last overcome. The paper was writtenhe night previous to the meeting, in pencil, very hastily,nd under the pressure just mentioned.

n this paper casual reference was made to two specialorms of motors not within the group to be considered.hese two forms were : 1. A motor with one of its circuit

n series with a transformer, and the other in theecondary of the transformer. 2. A motor having itsrmature circuit connected to the generator, and the fieloils closed u on themselves. The a er in its essence is



s follows, dealing witli a few leading features of the Teslystem, namely, the rotating magnetic field, motors

with closed conductors, synchronizing motors, andotating field transformers:—

he subject which I now have the pleasure of bringing toour notice is a novel system of electric distribution andransmission of power by means of alternate currents,ffording peculiar advantages, particularly in the way of

motors, which I am confident will at once establish the

uperior adaptability of these currents to theransmission of power and will show that many resultseretofore unattainable can be reached by their use ;esults which are very much desired in the practicalperation of such systems, and which cannot beccomplished by means of continuous currents.

efore going into a detailed description of this system, Ihink it necessary to make a few remarks with referenceo certain conditions existing in continuous currentenerators and motors, which, although generally knownre frequently disregarded.

n our dynamo machines, it is well known, we generatelternate currents which we direct by means of aommutator, a complicated device and, it may be justly aid, the source of most of the troubles experienced in thperation of the machines. Now, the currents so directed

annot be utilized in the motor, but they must—again bymeans of a similar unreliable device— be reconverted int



heir original state of alternate currents.^ The function ohe commutator is entirely external, and in no way does ffect the internal working of the machines. In reality,herefore, all machines are alternate current machines,he currents appearing as continuous only in the externa

ircuit during their transit from generator to motor. Iniew simply of this fact, alternate currents wouldommend themselves as a more direct application of lectrical energy, and the employment of continuousurrents would only be justified if we had dynamos which

would primarily generate, and motors which would be

irectly actuated by, such currents.

ut the operation of the commutator on a motor iswofold ; first, it reverses the currents through the motond secondly, it effects automatically, a progressivehifting of the poles of one of its magnetic constituents.

Assuming, therefore, that both of the useless operationsn the systems, that is to say, the directing of thelternate currents on the generator and reversing theirect currents on the motor, be eliminated, it would stille necessary, in order to cause a rotation of the motor, troduce a progressive

OLYPHASE CURRENTS.

1

hifting of the poles of one of its elements, and the

uestion presented itself—How to perform this operationy the direct action of alternate currents ? I will now



roceed to show how this result was accomplished.

n the first experiment a drum-armature was providedwith

ie. l.

IG. la.

wo coils at right angles to each other, and the ends of hese coils were connected to two pairs of insulatedontact-rings as usual. A ring was then made of thinnsulated plates of sheet-iron and wound with four coils,ach two opposite coils being connected together so as toroduce free poles on diametrically opposite sides of theing. The remaining free ends of the coils were then



onnected to the contact-rings of the generator armatureo as to form two independent circuits, as indicated in Fi.' It may now be seen what results were secured in thisombination, and witli this view I would refer to theiagrams, Figs. 1 to 8#. The field of the generator being

ndependently excited, the rotation of the armature setsp currents in the coils c c l5 varying in

IG

IG. 2a.

trength and direction in the well-known manner. In theosition shown in Fig. 1, the current in coil c is nil, whileoil c { is traversed by its maximum current, and theonnections may be such that the ring is magnetized by

he coils c t <?j, as indicated by the letters N s in Fig. 1#,he magnetizing effect of the coils

c being nil, since these coils are included in the circuit ofoil c.

n Fig. 2, the armature coils are shown in a moredvanced position, one-eighth of one revolution being



ompleted. Fig. la illustrates the corresponding magneticondition of the ring. At this moment the coil c, generatescurrent of the same di-

IG. 3.

IG. 3a.

ection as previously, but weaker, producing the poles wVj upon the ring; the coil c also generates a current of thame direction, and the connections may be such that th

oils c c produce the poles n *, as shown in Fig. 'la. Theesulting polarity is indicated by the letters x s, and it wie observed that the poles of the ring have been shiftedne-eighth of the periphery of the same.

n Fig. 3 the armature has completed one quarter of one

evolution. In this phase the current in coil c is amaximum, and of such direction as to produce the poles

in Fig. 3a, whereas the current in coil c v is nil, this coileing at its neutral position.





leted, and the resulting magnetic condition of the ring isndicated in Fig. 5«. Now the current in coil c is nil, whilehe coil c t yields its maximum current, which is of theame direction as previously ; the magnetizing effect is,herefore, due to the coils, 6 ( ! e n alone, and, referring tig. 5«, it will be observed that the poles N s are shiftedne half of the circumference of the ring. During the nextalf revolution the operations are repeated, asepresented in the Figs, f> to 8a.

A reference to the diagrams will make it clear that durinne

IG. Q.



IG. 6a.

evolution of the armature the poles of the ring arehifted once around its periphery, and, each revolutionroducing like effects, a rapid whirling of the poles in

armony with the rotation of the armature is the result. he connections of either one of the circuits in the ring areversed, the shifting of the poles is made to progress inhe opposite direction, but the operation is identi-

ally the same. Instead of using four wires, with like

esult) three wires may be used, one forming a commoneturn for both circuits.

his rotation or whirling of the poles manifests itself in aeries of curious phenomena. If a delicately pivoted disc teel or other magnetic metal is approached to the ring its set in rapid rotation, the direction of rotation varying

with the position of

IG. 7. FIG. Ta.

he disc. For instance, noting the direction outside of thein it will he found that inside the rin it turns in an



pposite direction, while it is unaffected if placed in aosition symmetrical to the ring. This is easily explainedach time that a pole approaches, it induces an oppositeole in the nearest point on the disc, and an attraction isroduced upon that point; owing to this, as the pole is

hifted further away from the disc a tangential pull isxerted upon the same, and the action being constantly epeated, a more or less rapid rotation of the disc is theesult. As the pull is exerted mainly upon that part whichs nearest to the ring, the rotation outside and inside, oright and left, respectively, is in opposite directions, Fig.

When placed symmetrically to the ring, the pull on thepposite sides of the disc being equal, no rotation resultshe action is based on the magnetic inertia of iron; for theason a disc of hard steel is much more affected than aisc of soft iron, the latter being capable of very rapidariations of magnetism. Such a disc has proved to be a

ery useful instrument in all these investigations, as it hanabled me to detect any irregularity in the action. A urious effect is also produced upon iron tilings. By placinome upon a paper and holding them externally quitelose to the ring, they are set in a vibrating motion,emaining in the same place, although the paper may be

moved back and forth ; but in lifting the paper to a certaeight which seems to be dependent on the intensity of he poles and the speed of rotation, they are thrown awan

direction always opposite to the supposed movement ohe poles. If a paper with filings is put flat upon the ring



nd the current turned on suddenly, the existence of amagnetic whirl may easily be observed.

o demonstrate the complete analogy between the ringnd a revolving magnet, a strongly energized electro-

magnet was rotated by mechanical power, andhenomena identical in every particular to thosementioned above were observed.

Obviously, the rotation of the poles producesorresponding inductive effects and may be utilized to

enerate currents in a closed conductor placed within thenfluence of the poles. For this purpose it is convenient towind a ring with two sets of superimposed coils formingespectively the primary and secondary circuits, as shown Fig. 10. In order to secure the most economical resultshe magnetic circuit should be completely closed, and wit

his object in view the construction may be modified atwill.

he inductive effect exerted upon the secondary coils wie mainly due to the shifting or movement of the

magnetic action ; but there may also be currents set up i

he circuits in consequence of the variations in thentensity of the poles. However, by properly designing thenerator and determining the magnetizing effect of therimary coils, the latter element may be made toisappear. The intensity of the poles being maintainedon-



IG.

IG. 8a.

tant, the action of the apparatus will be perfect, and the

ame result will be secured as though the shifting wereffected by means of a commutator with an infiniteumber of bars. In such case the theoretical relationetween the energizing effect of each set of primary coilsnd their resultant magnetizing effect may be expressedy the equation of a circle having its centre coinciding

with that of an orthogonal system of axes, and in whichhe radius represents the resultant and the co-ordinatesoth

f its components. These are then respectively the sinend cosine of the angle a between*the radius and one of

he axes (O X\ Referring to Fig. 11, we have ,* = x? + f ;where ./• = r cos a, and y = r sin a.

Assuming the magnetizing effect of each set of coils in thransformer to be proportional to the current—which mae admitted for weak degrees of magnetization—then x =

KG and y _ K C ^ w here îs a constant and c and c 1 theurrent in both sets of coils res ectivel . Su osin



urther, the field of the generator to be uniform, we haveor constant speed c 1 = A" 1 sin a and c = K l sin (90° + a

K l cos a, where K l is a constant. See Fig. 12.

herefore, a? = K c — K K^ cos a;

= Kc l = KK l sin a; and

IG. 9.

hat is, for a uniform field the disposition of the two coilst right angles will secure the theoretical result, and thentensity of the shifting poles will be constant. But from ^

x? -J- >f it follows that for y = 0, r = x; it follows that

he joint magnetizing effect of both sets of coils should bequal to the effect of one set when at its maximum actionn transformers and in a certain class of motors theuctuation of the poles is not of great importance, but innother class of these motors it is desirable to obtain theheoretical result.

n a l in this rinci le to the construction of motors



wo typical forms of motor have been developed. First, aorm having a comparatively small rotary effort at thetart but maintaining a perfectly uniform speed at alloads, which motor has been termed synchronous.econd, a form possessing a great rotary effort at the

tart, the speed being dependent on the load.

OL YMIAKK

7

hese motors may be operated in three different ways : By the alternate currents of the source only. 2. By aombined action of these and of induced currents. 3. By he joint action of alternate and continuous currents.

he simplest form of a synchronous motor is obtained by

winding a laminated ring provided with pole projectionswith four coils, and connecting the same in the mannerefore indicated. An iron disc having a segment cut awayn each side may be used



it* 10.

s an armature. Such a motor is shown in Fig. 9. The disceing arranged to rotate freely within the ring in closeroximity to the projections, it is evident that as the polere shifted it will, owing to its tendency to place itself inuch a position as to embrace the greatest number of thenes of force, closely follow the movement of the poles,

nd its motion will be synchronous with that of thermature of the generator; that is, in the peculiarisposition shown in Fig. 9, in which the armatureroduces by one revolution two current impulses in eachf the circuits. It is evident that if, by one revolution of he armature, a greater number of impulses is produced

he speed of the motor will be correspondingly increasedonsidering that the attraction exerted upon the disc isreatest when the same is in close proximity to the polesfollows that such a motor will maintain exactly the sam

peed at all loads within the limits of its capacity.

o facilitate the starting, the disc may be provided with oil closed upon itself. The advantage secured by such aoil is evident. On the start the currents set up in the coitrongly ener-

NVENTIONS OF NIKOLA TKKLA.

ize the disc and increase the attraction exerted u on th



ame by the ring, and currents being generated in the cos long as the speed of the armature is inferior to that of he poles, considerable work may be performed by such

motor even if the speed be below normal. The intensity ohe poles being constant, no currents will be generated in

he coil when the motor is turning at its normal speed.

nstead of closing the coil upon itself, its ends may beonnected to two insulated sliding rings, and a continuouurrent supplied to these from a suitable generator. Theroper way to start such a motor is to close the coil upon

self until the normal speed is reached, or nearly so, andhen turn on the continuous current. If the disc be very trongly energized by a continuous current the motor

may not be able to start, but if it be weakly energized, orenerally so that the magnetizing eifect of the ring



s preponderating, it will start and reach the normal

peed. Such a motor will maintain absolutely the samepeed at all loads. It has also been found that if the motivower of the generator is not excessive, by checking the

motor the speed of the generator is diminished inynchronism with that of the motor. It is characteristic ohis form of motor that it cannot be reversed by reversin

he continuous current through the coil.

he synchronism of these motors may be demonstratedxperimentally in a variety of ways. For this purpose it iest to employ a motor consisting of a stationary field

magnet and an armature arranged to rotate within the

ame, as indicated in Fig. 13. In this case the shifting of he poles of the armature produces a rotation of the latten the opposite direction. It results therefrom that whenhe normal speed is readied, the poles of the armaturessume fixed positions relatively to the

OLYPHASE CURRENTS.

t

eld magnet, and the same is magnetized by induction,xhibiting a distinct pole on each of the pole-pieces. If a

iece of soft iron is approached to the field magnet, it wilt the start be attracted with a ra id vibratin motion



roduced by the reversals of polarity of the magnet, buts the speed of the armature increases, the vibrationsecome less and less frequent and finally entirely cease.hen the iron is weakly but permanently attracted,howing that synchronism is reached and the field magn

nergized by induction.

he disc may also be used for the experiment. If helduite close to the armature it will turn as long as thepeed of rotation of the poles exceeds that of thermature ; but when the normal

IG. 13.

peed is reached, or very nearly so, it ceases to rotate ans permanently attracted.

A crude but illustrative ex eriment is made with an



ncandescent lamp. Placing the lamp in circuit with theontinuous current generator and in series with the

magnet coil, rapid fluctuations are observed in the light ionsequence of the induced currents set up in the coil athe start; the speed increasing, the fluctuations occur at

onger intervals, until they entirely disappear, showinghat the motor has attained its normal speed. A telephoneceiver affords a most sensitive instrument ; whenonnected to any circuit in the motor the synchronism

may be easily detected on the disappearance of thenduced currents.

NVENTIONS OF NIKOLA TKSLA.

he quantity of the shifting magnetism constant,specially if the magnets are not properly subdivided.

o obtain a rotary effort in these motors was the subjectf long thought. In order to secure this result it wasecessary to make such a disposition that while the polesf one element of the motor are shifted by the alternateurrents of the source, the poles produced upon the othelements should always be maintained in the proper

elation to the former, irrespective of the speed of themotor. Such a condition exists in a continuous currentmotor; but in a synchronous motor, such as described,his condition is fulfilled only when the speed is normal.

he object has been attained by placing within the ring a

roperly subdivided cylindrical iron core wound witheveral independent coils closed upon themselves. Two



oils at right angles as

IG. 14.

n Fig. 14, are sufficient, but a greater number may bedvantageously employed. It results from this dispositiohat when the poles of the ring are shifted, currents areenerated in the closed armature coils. These currentsre the most intense at or near the points of the greatestensity of the lines of force, and their effect is to produce

oles upon the armature at right angles to those of theing, at least theoretically so; and since this action isntirely independent of the speed—that is, as far as theocation of the poles is concerned—a continuous pull isxerted upon the periphery of the armature. In many espects these motors are similar to the continuousurrent motors. If load is put on, the speed, and also the



esistance of the motor, is diminished and more current made to pass through the energizing coils, thus

ncreasing the effort. Upon the load being taken off, theounter-electromotive force increases and less current

asses through the primary or energizing coils. Withoutny load the speed is very nearly equal to that of thehifting poles of the iield magnet.

t will be found that the rotary effort in these motors full

IG. 15. FIG. 16. FIG. 17.

quals that of the continuous current motors. The efforteems to be greatest when both armature and field

magnet are without any projections ; but as in suchispositions the field cannot be concentrated, probably he best results will be obtained by leaving pole

rojections on one of the elements only. Generally, it mae stated the projections diminish the torque and productendency to synchronism.

A characteristic feature of motors of this kind is theirroperty of being very rapidly reversed. This follows

rom the peculiar action of the motor. Suppose the



rmature to be rotating and the direction of rotation of he poles to be reversed. The apparatus then representsynamo machine, the power to drive this machine beinghe momentum stored up in the armature and its speedeing the sum of the speeds of the armature and the

oles.

f we now consider that the power to drive such a dynam

\AAA/

IG. 18. FIG. 19. FIG. 20. FIG. 21.

would be very nearly proportional to the third power of he speed, for that reason alone the armature should beuickly reversed. But simultaneously with the reversalnother element is brought into action, namely, as the

movement of the poles with respect to the armature iseversed, the motor acts like a transformer in which the

esistance of the secondarv circuit would be

g INVENTIONS OF NIKOLA TE8LA.

bnormally diminished by producing in this circuit andditional electromotive force. Owing to these causes the

eversal is instantaneous.



f it is desirable to secure a constant speed, and at theame time a certain effort at the start, this result may beasily attained in a variety of ways. For instance, twormatures, one for torque and the other for synchronism

may be fastened on the same shaft and any desired

reponderance may be given to either one, or anrmature may be wound for rotary effort, but a more oress pronounced tendency to synchronism may be giveno it by properly constructing the iron core; and in manyther ways.

As a means of obtaining the required phase of theurrents in both the circuits, the disposition of the twooils at right angles is the simplest, securing the mostniform action ; but the phase may be obtained in many ther ways, varying with the machine employed. Any of he dynamos at present in use may be easily adapted for

his purpose by making connections to proper points of he generating coils. In closed circuit armatures, such assed in the continuous current systems, it is best to mak

our derivations from equi-distant points or bars of theommutator, and to connect the same to four insulatedliding rings on the shaft. In this case each of the motor

ircuits is connected to two diametrically opposite bars ohe commutator. In such a disposition the motor may alse operated at half the potential and on the three-wirelan, by connecting the motor circuits in the proper ordeo three of the contact rings.

n multipolar dynamo machines, such as used in the



onverter systems, the phase is conveniently obtained bwinding upon the armature two series of coils in such amanner that while the coils of one set or series are at themaximum production of current, the coils of the other wi

e at their neutral position, or nearly so, whereby both

ets of coils may be subjected simultaneously oruccessively to the inducing action of the field magnets.

Generally the circuits in the motor will be similarly isposed, and various arrangements may be made to

ulfill the requirements; but the simplest and most

racticable is to arrange primary circuits on stationary arts of the motor, thereby obviating, at least in certainorms, the employment of sliding contacts. In such a casehe magnet coils are connected alternately in both theircuits ; that is, 1, 3, 5 in one, and 2, 4, 6 in the other, an

he coils of each set of series may be connected all in theame

manner, or alternately in opposition ; in the latter case amotor with half the number of poles will result, and its

ction will be correspondingly modified. The Figs. 15, 16,

nd 17, show three different phases, the magnet coils inach circuit being connected alternately in opposition. Inhis case there will be always four poles, as in Figs. 15 an7; four pole projections will be neutral ; and in Fig. 16wo adjacent pole projections will have the same polarityf the coils are connected in the same manner there will

e eight alternating poles, as indicated by the letters n' sn Fi . 1 .



he employment of multipolar motors secures in thisystem an advantage much desired and unattainable inhe continuous current system, and that is, that a motor

may be made to run exactly at a predetermined speed

rrespective of imperfections in construction, of the load,nd, within certain limits, of electromotive force andurrent strength.

n a general distribution system of this kind the followinglan should be adopted. At the central station of supply a

enerator should be provided having a considerableumber of poles. The motors operated from thisenerator should be of the synchronous type, butossessing sufficient rotary effort to insure their starting

With the observance of proper rules of construction itmay be admitted that the speed of each motor will be in

ome inverse proportion to its size, and the number of oles should be chosen accordingly. Still, exceptionalemands may modify this rule. In view of this, it will bedvantageous to provide each motor with a greaterumber of pole projections or coils, the number beingreferably a multiple of two and three. By this means, by

imply changing the connections of the coils, the motormay be adapted to any probable demands.

f the number of the poles in the motor is even, the actiowill be harmonious and the proper result will be obtained

if this is not the case, the best plan to be followed is to

make a motor with a double number of poles and conneche same in the manner before indicated so that half the



umber of poles result. Suppose, for instance, that theenerator has twelve poles, and it would be desired tobtain a speed equal to ^ of the speed of the generator.his would require a motor with seven pole projections o

magnets, and such a motor could not be properly

onnected in the circuits unless fourteen armature coilswould be provided, which would necessitate themployment of sliding

4 INVENTIONS OF NIKOLA TESLA.

ontacts. To avoid this, the motor should be provided wiourteen magnets and seven connected in each circuit, thmagnets in each circuit alternating among themselves.

he armature should have fourteen closed coils. Thection of the motor will not be quite as perfect as in thease of an even number of poles, but the drawback will

ot be of a serious nature.

However, the disadvantages resulting from thisnsymmetrical form will be reduced in the sameroportion as the number of the poles is augmented.

f the generator has, say, n, and the motor % poles, thepeed of the motor will be equal to that of the generator

multiplied by

he speed of the motor will generally be dependent onhe number of the poles, but there may be exceptions to

his rule. The speed may be modified by the phase of theurrents in the circuit or b the character of the current





econdary coil will be proportional to the numerical sumf the variations in the strength of the exciting currenter unit of time; whence it follows that for a givenariation any prolongation of the primary current willesult in a proportional loss. In order to obtain rapid

ariations in the strength of the current, essential tofficient induction, a great number of undulations aremployed ; from this practice various disadvantagesesult. These are : Increased cost and diminishedfficiency of the generator; more waste of energy ineating the cores, and also diminished output of the

ransformer, since the core is not properly utilized, theeversals being too rapid. The inductive effect is also vermall in certain phases, as will be apparent from a graphiepresentation, and there may be periods of inaction, if here are intervals between the succeeding currentmpulses or waves. In producing a shifting of the poles in

ransformer, and thereby inducing currents, the inductios of the ideal character, being always maintained at its

maximum action. It is also reasonable to assume that byhifting of the poles less energy will be wasted than by eversals.

HAPTER IV.

MODIFICATIONS AND EXPANSIONS or THE TESLA OLYPHASE

YSTEMS.

N his earlier papers and patents relative to polyphase



urrents, Mr. Tesla devoted himself chiefly to annunciation of the broad lines and ideas lying at the basisf this new work; but he supplemented this immediatelyy a series of other striking inventions which may beegarded as modifications and expansions of certain

eatures of the Tesla systems. These we shall now roceed to deal with.

n the preceding chapters we have thus shown andescribed the Tesla electrical systems for theransmission of power and the conversion and distributio

f electrical energy, in which the motors and theransformers contain two or more coils or sets of coils,which were connected up in independent circuits withorresponding coils of an alternating current generator,he operation of the system being brought about by theo-operation of the alternating currents in the

ndependent circuits in progressively moving or shiftinghe poles or points of maximum magnetic effect of the

motors or converters. In these systems two independentonductors are employed for each of the independentircuits connecting the generator with the devices foronverting the transmitted currents into mechanical

nergy or into electric currents of another character. Thowever, is not always necessary. The two or moreircuits may have a single return path or wire in common

with a loss, if any, which is so extremely slight that it mae disregarded entirely. For the sake of illustration, if thenerator have two independent coils and the motor twooils or two sets of coils in corres ondin Vela-tions to its



perative elements one terminal of each generator coil isonnected to the corresponding terminals of the motoroils through two independent conductors, while thepposite terminals of the respective coils are bothonnected to one return wire. The following description

eals with the modifica-

OLYPHASE CURRENTS.

7

on. Fig. 22 is a diagrammatic illustration of a generatornd single motor constructed and electrically connected iccordance with the invention. Fig. 23 is a diagram of theystem as it is nsed in operating motors or converters, oroth, in parallel, while Fig. 24 illustrates diagrammaticallhe manner of operating two or more motors or

onverters, or both, in series. Referring to Fig. 22, A A esignate the poles of the field magnets of an alternatingurrent generator, the armature of which, being in thisase cylindrical in form and mounted on a shaft, c, is

wound



IG. 24.

ongitudinally with coils B B'. The shaft c carries threensulated contact-rings, a b c, to two of which, as 5 c, oneerminal of each coil, as e d, is connected. The remaining

erminals, f g, are both connected to the third ring, a.

A motor in this case is shown as composed of a ring, H,wound with four coils, i i j j, electrically connected, so as to-operate in pairs, with a tendency to fix the poles of thing at four points ninety degrees apart. Within the

magnetic ring H is a disc or cylindrical core wound withwo coils, G a', which may be con-

NVENTIONS OF NIKOLA TESLA.

ected to form two closed circuits. The terminals j k of th

wo sets or pairs of coils are connected, respectively, tohe bindin - osts E' F' and the other terminals h i are



onnected to a single binding-post, D'. To operate themotor, three line-wires are used to connect the terminal

f the generator with those of the motor.

o far as the apparent action or mode of operation of this

rrangement is concerned, the single wire D, which is, soo speak,



IG. 23.

common return-wire for both circuits, may be regardes two independent wires. In the illustration, with therder of connection shown, coil B' of the generator is

roducing its maximum current and coil B its minimum ;ence the current which passes through wire e, ring 5,rush b' ', line-wire E, terminal E', wire,;', coils i i, wire orerminal D', line-wire D, brush a', ring a, and wire/, fixeshe polar line of the motor midway between the

wo coils i i; but as the coil B' moves from the positionndicated it generates less current, while coil B, movingnto the field, generates more. The current from coil Basses through the devices and wires designated by the

etters d, c, c' F, F' &, j j, i, D', D, #', «, and g, and theosition of the poles of the motor will be due to theesultant effect of the currents in the two sets of coils—hat is, it will be advanced in proportion to the advance oorward movement of the armature coils. The movemenf the generator-armature through one-quarter of aevolution will obviously bring coil B' into its neutral

osition and coil B into its position of maximum effect, anhis shifts the poles ninety degrees, as they are fixedolely by coils B. This action is repeated for each quarterf a complete revolution.

When more than one motor or other device is employed,

hey may be run either in parallel or series. In Fig. 23 thormer arran ement is shown. The electrical device is



hown as a converter, L, of which the two sets of primaryoils p r are connected, respectively, to the mains F E,

which are electrically connected with the two coils of theenerator. The cross-circuit wires I m, making theseonnections, are then connected to the common return-

wire D. The secondary coils p' p" are in circuits n <>,ncluding, for example, incandescent lamps. Only oneonverter is shown entire in this figure, the others beinglustrated diagrammatically.

When motors or converters are to be run in series, the

wo wires E F are led from the generator to the coils of he first motor or converter, then continued on to theext, and so on through the whole series, and are then

oined to the single wire D, which completes both circuitshrough the generator. This is shown in Fig. 24, in which represent the two coils or sets of coils of the motors.

here are, of course, other conditions under which theame idea may be carried out. For example, in case the

motor and generator each has three independent circuitsne terminal of each circuit is connected to a line-wire,nd the other three terminals to a common return-

onductor. This arrangement will secure similar results those attained with a generator and motor having but twndependent circuits, as above described.-

When applied to such machines and motors as have threr more induced circuits with a common electrical joint,

he three or more terminals of the generator would beim l connected



o those of the motor. Mr. Tesla states, however, that thesults obtained in this manner show a lower efficiency han do the forms dwelt upon more fully above.

HAPTER V.UTILIZING FAMILIAR TYPES OF GENERATOR OF

HE CONTINUOUS CURRENT TYPE.

HE preceding descriptions have assumed the use of lternating current generators in which, in order toroduce the progressive movement of the magnetic poler of the resultant attraction of independent field

magnets, the current generating coils are independent oreparate. The ordinary forms of continuous currentynamos may, however, be employed for the same work

n accordance with a method of adaptation devised by Mesla. As will be seen, the modification involves but slighhanges in their construction, and presents otherlements of economy.

On the shaft of a given generator, either in place of or in

ddition to the regular commutator, are secured as manyairs of insulated collecting-rings as there are circuits toe operated. Now, it will be understood that in theperation of any dynamo electric generator the currentsn the coils in their movement through the field of forcendergo different phases— that is to say, at different

ositions of the coils the currents have certain directionsnd certain stren ths— and that in the Tesla motors or



ransformers it is necessary that the currents in thenergizing coils should undergo a certain order of ariations in strength and direction. Hence, the furthertep — viz., the connection between the induced orenerating coils of the machine and the contact-rings fro

which the currents are to be taken off — will beetermined solely by what order of variations of strengthnd direction in the currents is desired for producing aiven result in the electrical translating device. This maye accomplished in various ways; but in the drawings weive typical instances only of the best and most

racticable ways of applying the invention to three of theeading types of machines in widespread use, in order tolustrate the principle.

ig. 25 is a diagram illustrative of the mode of applyinghe invention to the well-known type of " closed " or

ontinuous cir-

uit machines. Fig. 26 is a similar diagram embodying anrmature with separate coils connected diametrically, or

what is generally called an "open-circuit" machine. Fig. 2s a diagram showing the application of the invention to a

machine the armature-coils of which have -a commonoint.

Keferring to Fig. 25, let A represent a Tesla motor orransformer which, for convenience, we will designate asconverter." It consists of an annular core, B, wound with

our independent coils, c and D, those diametrically osite bein con-



IG. 25.

ected together so as to co-operate in pairs in establishin

ree poles in the ring, the tendency of each pair being tox the poles at ninety degrees from the other. There mae an armature, E, within the ring, which is wound withoils closed upon themselves. The object is to passhrough coils c D currents of such relative strength andirection as to produce a progressive shifting or

movement of the oints of maximum ma netic effect



round the ring, and to thereby maintain a rotary movement of the armature. There are therefore securedo the shaft F of the generator, four insulated contact-ings, abed, upon which bear

he collecting-brushes a' b' c' d', connected by wires G G H, respectively, with the terminals of coils c and D.

Assume, for sake of illustration, that the coils D D are toeceive the maximum and coils c c at the same instant th

minimum current, so that the polar line may be midway

etween the coils D D. The rings a 5 would therefore beonnected to the continuous armature-coil at its neutraloints with respect to the field, or the point correspondin

with that of the ordinary commutator brushes, andetween which exists the greatest difference of potential

while rings c d would be connected to two points in the

oil, between which exists no difference of potential. Theest results will be obtained by making these connectiont points equidistant from one another, as shown. Theseonnections are easiest made by using wires L betweenhe rings and the loops or wires j, connecting the coil i tohe segments of the commutator K. When the converter

re made in this manner, it is evident that the phases of he currents in the sections of the generator coil will beeproduced in the converter coils. For example, afterurning through an arc of ninety degrees the conductors , which before conveyed the maximum current, willeceive the minimum current by reason of the change in

he position of their coils, and it is evident that for the



ame reason the current in these coils lias gradually fallerom the maximum to the minimum in passing throughhe arc of ninety degrees. In this special plan of onnections, the rotation of the magnetic poles of theonverter will be synchronous with that of the armature

oils of the generator, and the result will be the same,whether the energizing circuits are derivations from aontinuous armature coil or from independent coils, as in

Mr. Tesla's other devices.

n Fig. 25, the brushes M M are shown in dotted lines in

heir proper normal position. In practice these brushesmay be removed from the commutator and the field of he generator excited by an external source of current; ohe brushes may be allowed to remain on the commutatond to take off a converted current to excite the field, oro be used for other purposes.

n a certain well-known class of machines known as theopen circuit," the armature contains a number of coilshe terminals of which connect to commutator segmentshe coils being connected across the armature in pairs.his type of machine is represented in Fig. 2fi. In this

machine each pair of coils goes

hrough the same phases as the coils in some of theenerators already shown, and it is obviously only ecessary to utilize them in pairs or sets to operate aesla converter by extending the segments of the

ommutators belonging to each pair of coils and causing aollectin brush to bear on the continuous ortion of each



egment. In this way two or more circuits may be takenff from the generator, each including one or more pairsr sets of coils as may be desired.

n Fig. 2H i i represent the armature coils, T T the poles

f the field magnet, and F the shaft carrying theommutators, which are extended to form continuousortions a I c d. The brushes

IG. 26.

IG. 27.

earing on the continuous portions for taking off the

lternating currents are represented by a' V c' d'. Theollecting brushes, or those which may be used to take of



he direct current, are designated by M M. Two pairs of he armature coils and their commutators are shown inhe figure as being utilized; but all may be utilized in aimilar manner.

here is another well-known type of machine in whichhree or more coils, A' «' c', on the armature have aommon joint, the free ends being connected to theegments of a commutator. This form of generator islustrated in Fig. 27. In this case each terminal of theenerator is connected directly or in derivation to a

ontinuous ring, a 1) <?, and collecting brushes, a' V c',earing

hereon, take oft' the alternating currents that operatehe motor. It is preferable in this case to employ a motorr transformer with three energizing coils, A" B" c",

laced symmetrically with those of the generator, and thircuits from the latter are connected to the terminals ofuch coils either directly—as when they are stationary —r by means of brushes e' and contact rings e. In this, asn the other cases, the ordinary commutator may be usen the generator, and the current taken from it utilized

or exciting the generator iielcl-magnets or for otherurposes.

HAPTER VI.

METHOD OF OBTAINING DESIRED SPEED OF

MOTOR OR GENERATOR.



WITH the object of obtaining the desired speed in motorperated by means of alternating currents of differinghase, Mr. Tesla has devised various plans intended to

meet the practical requirements of the case, in adaptingis system to types of multipolar alternating current

machines yielding a large number of current reversals foach revolution.

or example, Mr. Tesla has pointed out that to adapt aiven type of alternating current generator, you may ouple rigidly two complete machines, securing them

ogether in such a way that the requisite difference inhase will be produced; or you may fasten two armatureo the same shaft within the influence of the same fieldnd with the requisite angular displacement to yield theroper difference in phase between the two currents; orwo armatures may be attached to the same shaft with

heir coils symmetrically disposed, but subject to thenfluence of two sets of field magnets duly displaced; orhe two sets of coils may be wound on the same armaturlternately or in such manner that they will developurrents the phases of which differ in time sufficiently toroduce the rotation of the motor.

Another method included in the scope of the same idea,whereby a single generator may run a number of motorsither at its own rate of speed or all at different speeds, io construct the motors with fewer poles than theenerator, in which case their speed will be greater than

hat of the generator, the rate of speed being higher as



he number of their poles is relatively less. This may benderstood from an example, taking a generator that hawo independent generating coils which revolve betweenwo pole pieces oppositely magnetized ; and a motor withnergizing coils that produce at any given time two

magnetic poles in one element that tend to set up aotation of the motor. A generator thus constructed yieldour reversals, or impulses, in each

evolution, two in each of its independent circuits; and thffect upon the motor is to shift the magnetic poles

hrough three hundred and sixty degrees. It is obvioushat if the four reversals in the same order could beroduced by each half-revolution of the generator the

motor would make two revolutions to the generator's onhis would be readily accomplished by adding two

ntermediate poles to the generator or altering it in any o

he other equivalent ways above indicated. The same rupplies to generators and motors with multiple poles. Fonstance, if a generator be constructed with two circuits,ach of which produces twelve reversals of current to aevolution, and these currents be directed through thendependent energizing-coils of a motor, the coils of whic

re so applied as to produce twelve



IG. 28, FIG. 29.

magnetic poles at all times, the rotation of the two will beynchronous ; but if the motor-coils produce but six polehe movable element will be rotated twice while theenerator rotates once; or if the motor have four poles, iotation will be three times as fast as that of theenerator.

hese features, so far as necessary to an understanding he principle, are here illustrated. Fig. 28 is aiagrammatic illustration of a generator constructed inccordance with the invention. Fig. 29 is a similar view ocorrespondingly constructed motor. Fig. 30 is a diagramf a generator of modified construction. Fig. 31 is aiagram of a motor of corresponding character. Fig. 32 is

diagram of a system containing a generator and severamotors adapted to run at various speeds.

n Fig. 28, let c represent a cylindrical armature corewound longitudinally with insulated coils A A, which areonnected up in series, the terminals of the series being

onnected to collecting-rings a a on the shaft G. By meanf this shaft the armature is mounted to rotate between



he poles of an annular field-magnet D, formed with polarojections wound with coils E, that magnetize the saidrojections. The coils E are included in the circuit of aenerator F, by means of which the field-magnet isnergized. If thus constucted, the machine is a well-

nown form of alternating-current generator. To adapt io his system, however, Mr. Tesla winds on armature c aecond set of coils B B intermediate to the first, or, inther words, in such positions that while the coils of oneet are in the relative positions to the poles of the field-

magnet to produce the maximum current, those of the

ther set will be in the position in which they produce thminimum current. The coils B are connected, also, in

IG. 30.

IG. 81.

eries and to two connecting-rings, secured generally tohe shaft at the opposite end of the armature.

he motor shown in Fig. 29 has an annular field-magnet

H, with four pole-pieces wound with coils i. The armaturs constructed similarl to the enerator but with two



ets of two coils in closed circuits to correspond with theeduced number of magnetic poles in the field. From theoregoing it is evident that one revolution of the armaturf the generator producing eight current impulses in eachircuit will produce two revolutions of the motor-

rmature.

he application of the principle of this invention is not,owever, confined to any particular form of machine. Inigs. 30 and 31 a generator and motor of another well-nown type are shown. In Fig. 30, j j are magnets

isposed in a circle and wound with coils K, which are inircuit with a generator which

upplies the current that maintains the field of force. Inhe usual construction of these machines the armature-onductor L is carried by a suitable frame, so as to be

otated in face of the magnets j .1, or between thesemagnets and another similar set in front of them. Themagnets are energized so as to be of alternately opposite

olarity throughout the series, so that as the conductor cs rotated the current impulses combine or are added tone another, those produced by the conductor in any

iven position being all in the same direction. To adaptuch a machine to his system, Mr. Tesla adds a second sef induced conductors M, in all respects similar to therst, but so placed in reference to it that the currentsroduced in each will differ by a quarter-phase. With sucelations it is evident that as the current decreases in

onductor L it increases in conductor M, and conversel ,



nd that any of the forms of Tesla motor invented for usn this system may be operated by such a generator.

ig. 31 is intended to show a motor corresponding to themachine in Fig. 30. The construction of the motor is

dentical with that of the generator, and if coupled theretwill run synchronously therewith, j' j' are the field-magnets, and K' the coils thereon, i/ is one of the

rmature-conductors and M' the other.

ig. 32 shows in diagram other forms of machine. The

enerator N in this case is shown as consisting of atationary ring o, wound with twenty-four coils p p',lternate coils being connected in series in two circuits.

Within this ring is a disc or drum Q, with projections Q'wound with energizing-coils included in circuit with a

enerator K. By driving this disc or cylinder alternating

urrents are produced in the coils p and p', which arearried off to run the several motors.

he motors are composed of a ring or annular field-magnet s, wound with two sets of energizing-coils T T',

nd armatures u, having projections L T/ wound with

oils v, all connected in series in a closed circuit or eachlosed independently on itself.

uppose the twelve generator-coils p are woundlternately in opposite directions, so that any twodjacent coils of the same set tend to produce a free pole

n the ring o between them and the twelve coils p' to beimilarly wound. A single revolution of the disc or cylinde



Q, the twelve polar projections of which are of oppositeolarity, will therefore produce twelve current impulses

n each of the circuits w w'. Hence the motor x, which

as sixteen coils or eight free poles, will make one and a

alf turns to the generator's one. The motor Y, withwelve coils or six poles, will rotate with twice the speed he generator, and the motor z, with eight coils or fouroles, will revolve three times as fast as the generator.hese multipolar motors have a peculiarity which may bften utilized to great advantage. For ex-



TG. 32.

mple, in the motor x, Fig. 32, the eight poles may beither alternately opposite or there may be at any given

me alternately two like and two opposite poles. This iseadily attained by making the proper electricalonnections. The effect of such a change, however, woulde the same as reducing the number of

oles one-half, and thereby doubling the speed of any

iven motor.

t is obvious that the Tesla electrical transformers whichave independent primary currents may be used with thenerators described. It may also be stated with respecto the devices we now describe that the most perfect and

armonious action of the generators and motors isbtained when the numbers of the poles of each are evennd not odd. If this is not the case, there will be a certainnevenness of action which is the less appreciable as theumber of poles is greater; although this may be in a

measure corrected by special provisions which it is not

ere necessary to explain. It also follows, as a matter of ourse that if the number of the oles of the motor be



reater than that of the generator the motor will revolvet a slower speed than the generator.

n this chapter, we may include a method devised by Mresla for avoiding the very high speeds which would be

ecessary with large generators. In lieu of revolving theenerator armature at a high rate of speed, he secures thesired result by a rotation of the magnetic poles of onelement of the generator, while driving the other at aifferent speed. The effect is the same as that yielded byvery high rate of rotation.

n this instance, the generator which supplies the currenor operating the motors or transformers consists of aubdivided ring or annular core wound with fouriametrically-opposite coils, E F/, Fig. 33. Within the rin

s mounted a cylindrical armature-core wound

ongitudinally with two independent coils, F F', the ends owhich lead, respectively, to two pairs of insulated contact

r collecting rings, D D' G G', on the armature shaft.ollecting brushes d d' g g' bear upon these rings,espectively, and convey the currents through the twondependent line-circuits M M'. In the main line there

may be included one or more motors or transformers, oroth. If motors be used, they are of the usual form of esla construction with independent coils or sets of coils , included, respectively, in the circuits M M'. Thesenergizing-coils are wound on a ring or annular field or oole pieces thereon, and produce by the action of the

lternatin currents assin throu h them a ro ressive



hifting of the magnetism from pole to pole. Theylindrical armature H of the motor is wound with twooils at right angles, which form independent closedircuits.

f transformers be employed, one set of the primary coils N N, wound on a ring or annular core is connected tone circuit, as M', and the other primary coils, N N', to thircuit M. The secondary coils K K' may then be utilizedor running groups of incandescent lamps p p'.

With this generator an exciter is employed. This consistsf



IG. 33.

wo poles, A A, of steel permanently magnetized, or of ron excited by a battery or other generator of continuouurrents, and a cylindrical armature core mounted on ahaft, B, and wound with two longitudinal coils, c c'. Onend of each of these coils is connected to the collecting-ings I c, respectively, while the

ther ends are both connected to a ring, a. Collecting-rushes b' e' bear on the rings b c, respectively, andonductors L L convey tlie currents therefrom throughhe coils E and E of the generator, i/ is a common return

wire to brush a'. Two independent circuits are thus

ormed, one including coils c of the exciter and E E of theenerator, the other coils c' of the exciter and E' E' of theenerator. It results from this that the operation of thexciter produces a progressive movement of the magnetoles of the annular field-core of the generator, thehifting or rotary movement of the poles being

ynchronous with the rotation of the exciter armature.onsidering the operative conditions of a system thus



stablished, it will be found that when the exciter isriven so as to energize the field of the generator, thermature of the latter, if left free to turn, would rotate atspeed practically the same as that of the exciter. If nder such conditions the coils F F' of the generator

rmature be closed upon themselves or short-circuited,o currents, at least theoretically, will be generated inhese armature coils. In practice the presence of slighturrents is observed, the existence of which isttributable to more or less pronounced fluctuations inhe intensity of the magnetic poles of the generator ring.

o, if the armature-coils F F' be closed through the motohe latter will not be turned as long as the movement of he generator armature is synchronous with that of thexciter or of the magnetic poles of its lield. If, on theontrary, the speed of the generator armature be in any

way checked, so that the shifting or rotation of the poles

f the field becomes relatively more rapid, currents wille induced in the armature coils. This obviously followsrom the passing of the lines of force across the armatureonductors. The greater the speed of rotation of the

magnetic poles relatively to that of the armature themore rapidly the currents developed in the coils of the

atter will follow one another, and the more rapidly themotor will revolve in response thereto, and this continue

ntil the armature generator is stopped entirely, as by arake, when the motor, if properly constructed, runs athe speed with which the magnetic poles of the generatorotate.



he effective strength of the currents developed in thermature coils of the generator is dependent upon thetrength of the currents energizing the generator andpon the number of rotations per unit of time of the

magnetic poles of the generator; hence the speed of the

motor armature will depend in all cases

pon the relative speeds of the armature of the generatond of its magnetic poles. For example, if the poles areurned two thousand times per unit of time and thermature is turned eight hundred, the motor will turn

welve hundred times, or nearly so. Very slightiiferences of speed may be indicated by a delicately alanced motor.

et it now be assumed that power is applied to theenerator armature to turn it in a direction opposite to

hat in which its magnetic poles rotate. In such case theesult would be similar to that produced by a generatorhe armature and field magnets of which are rotated inpposite directions, and by reason of these conditions the

motor armature will turn at a rate of speed equal to theum of the speeds of the armature and magnetic poles of

he generator, so that a comparatively low speed of theenerator armature will produce a high speed in the

motor.

t will be observed in connection with this system that oniminishing the resistance of the external circuit of the

enerator armature by checking the speed of the motorr b addin translatin devices in multi le arc in the



econdary circuit or circuits of the transformer thetrength of the current in the armature circuit is greatly ncreased. This is due to two causes: first, to the greatifferences in the speeds of the motor and generator, andecondly, to the fact that the apparatus follows the

nalogy of a transformer, for, in proportion as theesistance of the armature or secondary circuits iseduced, the strength of the currents in the field orrimary circuits of the generator is increased and theurrents in the armature are augmented correspondinglyor similar reasons the currents in the armature-coils of

he generator increase very rapidly when the speed of thrmature is reduced when running in the same directions the magnetic poles or conversely.

t will be understood from the above description that theenerator-armature may be run in the direction of the

hifting of the magnetic poles, but more rapidly, and thatn such case the speed of the motor will be equal to theifference between the two rates.

HAPTER VII.

OR RoTARY CURRENT MoTORS.

AN interesting device for regulating and reversing haseen devised by Mr. Tesla for the purpose of varying thepeed of polyphase motors. It consists of a form of onverter or transformer with one element capable of

movement with respect to the other, whereby thenductive relations may be altered, either manually or



utomatically, for the purpose of varying the strength ofhe induced current. Mr. Tesla prefers to construct thisevice in such manner that the induced or secondary lement may be movable with respect to the other ; andhe invention, so far as relates merely to the construction

f the device itself, consists, essentially, in theombination, with two opposite magnetic poles, of anrmature wound with an insulated coil and mounted on ahaft, whereby it may be turned to the desired extent

within the field produced by the poles. The normalosition of the core of the secondary element is that in

which it most completely closes the magnetic circuitetween the poles of the primary element, and in thisosition its coil is in its most effective position for the

nductive action upon it of the primary coils; but by urning the movable core to either side, the inducedurrents delivered by its coil become weaker until, by a

movement of the said core and coil through 90°, there we no current delivered.

ig. 34 is a view in side elevation of the regulator. Fig. 35s a broken section on line a 1 a? of Fig. 34. Fig. 36 is aiagram illustrating the most convenient manner of

pplying the regulator to ordinary forms of motors, andig. 37 is a similar diagram illustrating the application of he device to the Tesla alternating-current motors. Theegulator may be constructed in many ways to secure thesired result; but that which is, perhaps, its best form ishown in Figs. 34 and 35.



NVENTIONS OF NIKOLA TEKLA.

ng or primary coils c c. i> is a shaft mounted on the sidears, D', and on which is secured a sectional iron core, E,

wound with an' induced or secondary coil, F, the

onvolutions of which are parallel with the axis of thehaft. The ends of the core are rounded off so as to fitlosely in the space between the two poles and permit thore E to be turned to and held at any desiivd point. A andle, G, secured to the projecting end of the shaft D, isrovided for this purpose.

n Fig. 36 let n represent an ordinary alternating currenenerator, the field-magnets of which are excited by auitable source of current, i. Let j designate an ordinary orm of electromagnetic motor provided with anrmature, K, commutator L, and field-magnets M. It is

well known that such a motor, if its



IG. 34.

eld-magnet cores be divided up into insulated sections,may be practically operated by an alternating current;

ut in using this regulator with such a motor, Mr. Tesla

ncludes one element of the motor only—say thermature-coils—in the main circuit of the generator,

making the connections through the brushes and theommutator in the usual way. He also includes one of thelements of the regulator—say the stationary coils—in thame circuit, and in the circuit with the secondary or

movable coil of the regulator he connects up the field-coif the motor. He also prefers to use flexible conductors t

make the connections from the secondary coil of theegulator, as he thereby avoids the use of sliding contactr rings without interfering with the requisite movementf the core E.

OLYPHASE CURRENTS.

7

f the regulator be in its normal position, or that in which

s magnetic circuit is most nearly closed, it delivers itsmaximum induced current, the phases of which soorrespond with those of the primary current that the

motor will run as though both lield and armature werexcited by the main current.

o var the s eed of the motor to an rate between the



minimum and maximum rates, the core E and coils F areurned in either direction to an extent which produces thesired result, for in its normal position the convolutionsf coil F embrace the maximum number of lines of force,ll of which act with the same effect upon the coil; hence

will deliver its maximum current; but by turning the coilout of its position of maximum effect the number of nes of force embraced by it is diminished. The inductiveffect is therefore impaired, and the current delivered byoil F will continue to diminish in proportion to the anglet which the coil F is turned until, after passing through

IG. 36.

n angle of ninety degrees, the convolutions of the coil wie at right angles to those of coils c c, and the inductive

ffect reduced to a minimum.

ncidentally to certain constructions, other causes may nfluence the variation in the strength of the inducedurrents. For example, in the present case it will bebserved that by the first movement of coil F a certain

ortion of its convolutions are carried beyond the line of he direct influence of the lines of force, and that the



magnetic path or circuit for the lines is impaired ; hencehe inductive effect would be reduced. Next, that after

moving through a certain angle, which is obviously etermined by the relative dimensions of the bobbin oroil F, diagonally opposite portions of the coil will be

imultaneously included in the field, but in such positionshat the lines which produce a current-impulse in oneortion of the coil in a certain direction will pro-


uce in the diagonally opposite portion a correspondingmpulse in the opposite direction; hence portions of theurrent will neutralize one another.

As before stated, the mechanical construction of theevice may be greatly varied; but the essential condition

f the principle will be fulfilled in any apparatus in whichhe movement of the elements with respect to onenother effects the same results by varying the inductivelations of the two elements in a manner similar to thatescribed.

t may also be stated that the core E is not indispensableo the operation of the regulator; but its presence isbviously beneficial. This regulator, however, has anothealuable property in its capability of reversing the motor

or if the coil F be turned





ncludes one set, R' R', of the energizing coils of the motowhile the other circuit, as s s, includes the primary coils ohe regulator. The secondary coil of the regulator includehe other coils, R R, of the motor.

While the secondary coil of the regulator is in its normalosition, it produces its maximum current, and themaximum rotary effect is imparted to the motor; but thiffect will be diminished in proportion to the angle at

which the coil F of the regulator is turned. The motor willso be reversed by reversing the position of the coil with

eference to the coils c c, and thereby reversing thehases of the current produced by the generator. Thishanges the direction of the movement of the shiftingoles which the armature follows.

One of the main advantages of this plan of regulation is it

conomy of power. When the induced coil is generating itmaximum current, the maximum amount of energy in th

rimary coils is absorbed ; but as the induced coil isurned from its normal position the self-induction of therimary-coils reduces the expenditure of energy andaves power.

t is obvious that in practice either coils c <: or coil v maye used as primary or secondary, and it is wellnderstood that their relative proportions may be variedo produce any desired difference or similarity in thenducing and induced currents.

HAPTER VIII.



INGLE CIRCUIT, SELF-STARTINGYNCHRONIZING MOTORS.

n the first chapters of this section we have, bearing inmind the broad underlying principle, considered a distinclass of motors, namely, such as require for theirperation a special generator capable of yielding currentf differing phase. As a matter of course, Mr. Teslaecognizing the desirability of utilizing his motors inonnection with ordinary systems of distribution,

ddressed himself to the task of inventing variousmethods and ways of achieving this object. In theucceeding chapters, therefore, we witness the evolutionf a number of ideas bearing upon this important branchf work. It must be obvious to a careful reader, from aumber of hints encountered here and there, that even

he inventions described in these chapters to follow do noepresent the full scope of the work done in these lines.hey might, indeed, be regarded as exemplifications.

We will present these various inventions in the orderwhich to us appears the most helpful to an understandin

f the subject by the majority of readers. It will beaturally perceived that in offering a series of ideas of thature, wherein some of the steps or links are missing,

he descriptions are not altogether sequential; but any ne who follows carefully the main drift of the thoughtsow brought together will find that a satisfactory

omprehension of the principles can be gained.



As is well known, certain forms of alternating-currentmachines have the property, when connected in circuitwith an alternating current generator, of running as amotor in synchronism therewith ; but, while the

lternating current will run the motor after it has attaine

rate of speed synchronous with that of the generator, itwill not start it. Hence, in all instances heretofore wherehese " synchronizing motors," as they are termed, haveeen run, some means have been adopted to bring the

motors up to synchronism with the generator, orpproximately so, before the alternating current of the

enerator is applied to drive them.

n some instances mechanical appliances have beentilized for this purpose. In others special and complicateorms of motor have been constructed. Mr. Tesla hasiscovered a much more simple method or plan of

perating synchronizing motors, which requiresractically no other apparatus than the motor itself. Inther words, by a certain change in the circuit connectionf the motor he converts it at will from a double circuit

motor, or such as have been already described, and whicwill start under the action of an alternating current, into

ynchronizing motor, or one which will be run by theenerator only when it has reached a certain speed of otation synchronous with that of the generator. In this

manner he is enabled to extend very greatly thepplications of his system and to secure all the advantagf both forms of alternating current motor.



he expression " synchronous with that of the generators used here in its ordinary acceptation — that is to say, a

motor is said to synchronize with the generator when itreserves a certain relative speed determined by itsumber of poles and the number of alternations produce

er revolution of the generator. Its actual speed,herefore, may be faster or slower than that of theenerator; but it is said to be synchronous so long as itreserves the same relative speed.

n carrying out this invention Mr. Tesla constructs a

motor which has a strong tendency to synchronism withhe generator. The construction preferred is that in whiche armature is provided with polar projections. Theeld-magnets are wound with two sets of coils, the

erminals of which are connected to a switch mechanismy means of which the line-current may be carried

irectly through these coils or indirectly through paths bwhich its phases are modified. To start such a motor, thewitch is turned on to a set of contacts which includes inne motor circuit a dead resistance, in the other an

nductive resistance, and, the two circuits being inerivation, it is obvious that the difference in phase of th

urrent in such circuits will set up a rotation of the motorWhen the speed of. the motor has thus been brought tohe desired rate the switch is shifted to throw the mainurrent directly through the motor-circuits, and althoughhe currents in both circuits will now be of the same phashe motor will continue to revolve, becoming a trueynchronous motor. To secure greater efficiency, the



rmature or its polar projections are wound with coilslosed on themselves.

n the accompanying diagrams, Fig. 38 illustrates theetails of the plan above set forth, and Figs. 39 and 40

modifications of the same.

Referring to Fig. 38, let A designate the neld-magnets of



K;S. :

motor, the polar projections of which are wound with coils c included in independent circuits, and D the armature

with polar projections wound with coils E closed upon

hemselves, the motor in these respects being similar inonstruction to those

escribed already, but having OH account of the polarrojections on the armature core, or other similar and

well-known features, the properties of a synchronizing-

motor. L i/ represents the conductors of a line from anlternating current generator <j.

Near the motor is placed a switch the action of which ishat of the one shown in the diagrams, which isonstructed as follows : F F' are two conducting plates or

rms, pivoted at their ends and connected by annsulating cross-bar, H, so as to be shifted in parallelism.n the path of the bars F F 7 is the contact •2, which formne terminal of the circuit through coils c, and the contac, which is one terminal of the circuit through coils B. Thpposite end of the wire of coils c is connected to the wire

or bar F' , and the corresponding end of coils B isonnected to wire i and bar F; hence if the bars be



hifted so as to bear on contacts 2 and 4 both sets of coilsc: will be included in the circuit L i/ in multiple arc or

erivation. In the path of the levers F F' are two otherontact terminals, L and 3. The contact 1 is connected toontact 2 through an artificial resistance, i, and contact 3

with contact 4 through a self-induction coil, j, so that whehe switch levers are shifted upon the points ] and 3 theircuits of coils B and c will be connected in multiple arc oerivation to the circuit L i/, and will include theesistance and self-induction coil respectively. A thirdosition of the switch is that in which the levers F and F'

re shifted out of contact with both sets of points. In thisase the motor is entirely out of circuit.

he purpose and manner of operating the motor by thesevices are as follows: The normal position of the switchhe motor being out of circuit, is off the contact points.

Assuming the generator to be running, and that it isesired to start the motor, the switch is shifted until its

evers rest upon points 1 and 3. The two motor-circuitsre thus connected with the generator circuit ; but by eason of the presence of the resistance i in one and theelf-induction coil j in the other the coincidence of the

hases of the current is disturbed sufficiently to producerogression of the poles, which starts the motor inotation. When tl.'e speed of the motor has run up toynchronism with the generator, or approximately so, thwitch is shifted over upon the points 2 and 4, thusutting out the coils i and j, so that the currents in bothircuits have the same phase; but the motor now runs as



synchronous motor.

t will be understood that when brought up to speed themo

>4 INVENTIONS OF NIKOLA TESLA.

or will run with only one of the circuits B or c connectedwith the main or generator circuit, or the two circuits ma

e connected in series. This latter plan is preferable whecurrent having a high number of alternations per unit o

me is employed to drive the motor. In such case thetarting of the motor is more difficult, and the dead andnductive resistances must take up a considerableroportion of the electromotive force of the circuits.

Generally the conditions are so adjusted that thelectromotive force used in each of the motor circuits is

hat which is required to operate the motor when itsircuits are in series. The plan followed in this case islustrated in Fig. 39. In this instance the motor haswelve poles and the armature has polar projections D

wound with closed coils E. The switch used is of ubstantially the same construction as that shown in the

revious figure. There are, however, five contacts,esignated as 5, 6, 7, 8, and 9. The motor-circuits B c,

which include alternate field-coils, are connected to theerminals in the following order: One end of circuit c isonnected to contact 9 and to contact 5 through a deadesistance, i. One terminal of circuit B is connected to

ontact 7 and to contact 6 through a self-induction coil, Jhe o osite terminals of both circuits are connected to



ontact 8.

One of the levers, as F, of the switch is made with anxtension, /, or otherwise, so as to cover both contacts 5nd 6 when shifted into the position to start the motor. I

will be observed that when in this position and with lever' on contact 8 the current divides between the twoircuits B c, which from their difference in electricalharacter produce a progression of the poles that startshe motor in rotation. When the motor has attained theroper speed, the switch is shifted so that the levers

over the contacts 7 and 9, thereby connecting circuits Bnd c in series. It is found that by this disposition the

motor is maintained in rotation in synchronism with theenerator. This principle of operation, which consists inonverting by a change of connections or otherwise aouble-circuit motor, or one operating by a progressive

hifting of the poles, into an ordinary synchronizing motomay be carried out in many other ways. For instance,nstead of using the switch shown in the previous figures

we may use a temporary ground circuit between theenerator and motor, in order to start the motor, inubstantially the manner indicated in Fig. 40. Let G in

his figure represent an ordinary

lternating-current generator with, say, two poles, M Mnd an armature wound with two coils, N N', at rightngles and connected in series. The motor has, forxample, four poles wound with coils B c, which are

onnected in series, and an armature with olar



rojections D wound with closed coils E E. From theommon joint or union between the two circuits of bothhe generator and the motor an earth connection isstablished, while the terminals or ends of these circuitsre connected to the line. Assuming that the motor is a

ynchronizing motor or one that has the capability of unning in synchronism with the generator, but not of tarting, it may be started by the above-describedpparatus by closing the ground connection from bothenerator and motor. The system thus becomes one withtwo-circuit generator and motor, the ground forming a

ommon return for the currents in the two circuits L and/. When by this arrangement of circuits the motor isrought to speed, the ground connection is brokenetween the motor or generator, or both, ground-witches PP' being employed for this purpose. The motorhen runs as a synchronizing motor.

n describing the main features which constitute thisnvention illustrations have necessarily been omitted of he appliances used in conjunction with the electricalevices of similar systems— such, for instance, as drivingelts, fixed and loose pulleys for the motor, and the like;

ut these are matters well understood.

Mr. Tesla believes he is the first to operate electro-magnetic motors by alternating currents in any of theways herein described —that is to say, by producing a

rogressive movement or rotation of their poles or point

f greatest magnetic attraction by the alternating



urrents until they have reached a given speed, and theny the same currents producing a simple alternation of heir poles, or, in other words, by a change in the order oharacter of the circuit connections to convert a motorperating on one principle to one operating on another.

HAPTER IX.

HANGE FROM DOUBLE CURRENT TO SINGLEURRENT MOTOR.

A DESCRIPTION is given elsewhere of a method of perating alternating current motors by first rotatingheir magnetic poles until they have attained synchronoupeed, and then alternating the poles. The motor is thusransformed, by a simple change of circuit connectionsrom one operated by the action of two or more

ndependent energizing currents to one operated eithery a single current or by several currents acting as one.Another way of doing this will now be described.

At the start the magnetic poles of one element or field ofhe motor are progressively shifted by alternating

urrents differing in phase and passed throughndependent energizing circuits, and short circuit the coilf the other element. When the motor thus startedeaches or passes the limit of speed synchronous with thenerator, Mr. Tesla connects up the coils previously hort-circuited with a source of direct current and by a

hange of the circuit connections produces a simplelternation of the poles. The motor then continues to run



n synchronism with the generator. The motor herehown in Fig. 41 is one of the ordinary forms, with field-ores either laminated or solid and with a cylindricalaminated armature wound, for example, with the coils A at right angles. The shaft of the armature carries three

ollecting or contact rings c D E. (Shown, for betterlustration, as of different diameters.)

One end of coil A connects to one ring, as c, and one end ooil B connects with ring D. The remaining ends areonnected to ring E. Collecting springs or brushes F G H

ear upon the rings and lead to the contacts of a switch, te presently described. The field-coils have theirerminals in binding-posts K K, and may be either closedpon themselves or connected w r ith a source of directurrent L, by means of a switch M. The main orontrolling switch has five contacts a b c d e and two

evers/ g, pivoted and connected by an insulating cross-ar A, so as to move in parallelism. These levers areonnected to the line

OLYPHASE CURRENTS.

?

wires from a source of alternating currents N. Contact a onnected to brush o and coil B through a dead resistanc

R and wire P. Contact b is connected with brush F and coA through a self-induction coil s and wire o. Contacts c an

are connected to brushes <; F, respectively, through thwires P o, and contact <l is directly connected with brush



H. The lever/has a widened end, which may span theontacts a 1>. When in such position and with lever g onontact d, the alternating currents divide between the tw

motor-coils, and by reason of their different self-

nduction a difference of current-phase is obtained thattarts the motor in rotation. In starting, the field-coils arhort cir cuited.

When the motor has attained the desired speed, the



witch is shifted to the position shown in dotted lines—hat is to say, with the levers fg resting on points c e. Thionnects up the two armature coils in series, and the

motor will then run as a synchronous motor. The field-oils are thrown into circuit with the direct current sourc

when the main switch is shifted.

HAPTER X.

MOTOR WITH " CURRENT LAG" ARTIFICIALLY ECURED.

ONE of the general ways followed by Mr. Tesla ineveloping his rotary phase motors is to produceractically independent currents differing primarily inhase and to pass these through the motor-circuits.

Another way is to produce a single alternating current, to

ivide it between the motor-circuits, and to effectrtificially a lag in one of these circuits or branches, as byiving to the circuits different self-inductive capacity, an

n other ways. In the former case, in which the necessaryifference of phase is primarily effected in the generationf currents, in some instances, the currents are passed

hrough the energizing coils of both elements of the moto—the field and armature ; but a further result ormodification may be obtained by doing this under theonditions hereinafter specified in the case of motors in

which the lag, as above stated, is artificially secured.

igs. 42 to 4T, inclusive, are diagrams of different ways iwhich the invention is carried out ; and Fig. 48, a side



iew of a foam of motor used by Mr. Tesla for thisurpose.

A B in Fig. 42 indicate the two energizing circuits of amotor, and c D two circuits on the armature. Circuit or co

A is connected in series with circuit or coil c, and the twoircuits B D are similarly connected. Between coils A ands a contact-ring £, forming one terminal of the latter, anbrush «, forming one terminal of the former. A ring d

nd brush c similarly connect coils B and D. The oppositeerminals of the field-coils connect to one binding post h

he motor, and those of the armature coils are similarly onnected to the opposite binding post i through aontact-ring f and brush g. Thus each motor-circuit whiln derivation to the other includes one armature and oneeld coil. These circuits are of different self-induction, an

may be made so in various ways. For the sake of

learness, an artificial resistance R is shown in one of hese circuits, and in the other a self-induction coil s.

When an alternating current is passed

OLYPHASE CURRENTS.

hrough this motor it divides between its two energizing-ircuits. The higher self-induction of one circuit producesgreater retardation or lag in the current therein than in

he other. The difference of phase between the twourrents effects the rotation or shifting of the points of

maximum magnetic effect that secures

—www—HM5&RJT— *& nffiMT—|



l&t-*--*

IGS. 42, 43 and 44.

he rotation of the armature. In certain respects this plaf including both armature and field coils in circuit is a

marked improvement. Such a motor has a good torque atarting ; yet it has also considerable tendency toynchronism, owing to the fact

hat when ro erl constructed the maximum ma netic



ffects in both armature and field coincide — a conditionwhich in the usual construction of these motors withlosed armature coils is not readily attained. The motorhus constructed exhibits too, a better regulation of urrent from no load to load, and there is less difference

etween the apparent and real energy expended inunning it. The true synchronous speed of this form of motor is that of the generator when both are alike — thas to say, if the number of the coils on the armature andn the field is a?, the motor will run normally at the sampeed as a generator driving

Uv^-^Mfa^ Lum'

ms. 45, 46 and 47.

if the number of field magnets or poles of the same belso or.

ig. 43 shows a somewhat modified arrangement of ircuits. There is in this case but one armature coil E, the

winding of which maintains effects corresponding to theesultant poles produced by the two field-circuits.

ig. 44 represents a disposition in which both armature



nd field are wound with two sets of coils, all in multiplerc to the line or main circuit. The armature coils are

wound to correspond with the field-coils with respect toheir self-induction. A modification of this plan is shown iig. 45— that is to say, the

OLYPHASE CURRENTS

1

wo field coils and two armature coils are in derivation to

hemselves and in series with one another. The armaturoils in this case, as in the previous figure, are wound forifferent self-induction to correspond with the field coils.

Another modification is shown in Fig. 46. In this case onlne armature-coil, as D, is included in the line-circuit,

while the other, as c, is short-circuited.

n such a disposition as that shown in Fig. 43, or wherenly one armature-coil is employed, the torque on thetart is somewhat reduced, while the tendency toynchronism is somewhat



IG. 48.

ncreased. In such a disposition as shown in Fig. 4H, thepposite conditions would exist. In both instances,owever, there is the advantage of dispensing with oneontact-ring.

n Fig. 4(5 the two field-coils and the armature-coil D arn multiple arc. In Fig. 47 this disposition is modified, coil

D being shown in series with the two field-coils.

ig. 48 is an outline of the general form of motor in whichhis invention is embodied. The circuit connectionsetween the armature and field coils are made, as

ndicated in the previous figures, through brushes andings, which are not shown.

HAPTER XI.

ANOTHER METHOD OF TRANSFORMATION FROM ORQUE TO A SYNCHRONIZING MOTOR.

N a preceding chapter we have described a method by which Mr. Tesla accomplishes the change in his type of otating field motor from a torque to a synchronizing

motor. As will be observed, the desired end is thereeached b a chan e in the circuit connections at the



roper moment. We will now proceed to describe anotheway of bringing about the same result. The principlenvolved in this method is as follows: —

f an alternating current be passed through the field coils

nly of a motor having two energizing circuits of differenelf-induction and the armature coils be short-circuited,he motor will have a strong torque, but little or noendency to synchronism with the generator; but if theame current which energizes the field be passed alsohrough the armature coils the tendency to remain in

ynchronism is very considerably increased. This is due he fact that the maximum magnetic effects produced inhe field and armature more nearly coincide. On thisrinciple Mr. Tesla constructs a motor having

ndependent field circuits of different self-induction, whicre joined in derivation to a source of alternating current

he armature is wound with one or more coils, which areonnected with the field coils through contact rings andrushes, and around the armature coils a shunt isrranged with means for opening or closing the same. Intarting this motor the shunt is closed around thermature coils, which will therefore be in closed circuit.

When the current is directed through the motor, it dividetween the two circuits, (it is not necessary to considerny case where there are more than two circuits used),

which, by reason of their different self-induction, secure ifference of phase between the two currents in the tworanches, that produces a shifting or rotation of the of tholes. By the alternations of current, other currents are





KJS. 49 ( 50 and 51.

elf-induction coil s in circuit with B. The same result maf course be secured by the winding of the coils, c is the

rmature circuit, the terminals of which are rings a J.rushes c d bear on these rings and connect with the linend field circuits. D is the shunt or short circuit aroundhe armature. E is the switch in the shunt.

t will be observed that in such a disposition as is

lustrated in

vi


ig. 49, the field circuits A and B being of different self-nduction, there will always be a greater lag of the currenn one than the other, and that, generally, the armaturehases will not correspond with either, but with theesultant of both. It is therefore important to observe throper rule in winding the armature. For instance, if the

motor have eight poles—four in each circuit —there wille four resultant poles, and hence the armature windinghould be such as to produce four poles, in order toonstitute a true synchronizing motor.

he diagram, Fig. 50, differs from the previous one only

n respect to the order of connections. In the present cashe armature-coil instead of bein in series with the field



oils, is in multiple arc therewith. The armature-windingmay be similar to that of the field—that is to say, the

rmature may have two or more coils wound or adaptedor different self-induction and

IG. 52.

dapted, preferably, to produce the same difference of hase as the field-coils. On starting the motor the shunt losed around both coils. This is shown in Fig. 51, in whiche armature coils are K <;. To indicate their differentlectrical, character, there are shown in circuit with themespectively, the resistance R' and the self-induction coil. The two armature coils are in series with the field-coil

nd the same disposition of the shunt or short-circuit u issed. It is of advantage in the operation of motors of thisind to construct or wind the armature in such mannerhat when short-circuited on the start it will have aendency to reach a higher speed than that whichynchronizes with the generator. For example, a given

motor having, say, eight poles should run, with thermature coil short-circuited, at two thousand revolution



er minute to bring it up to synchronism. It will generallappen, however, tha't

his speed is not reached, owing to the fact that thermature and field currents do not properly correspond,

o that when the current is passed through the armaturethe motor not being quite up to synchronism) there is aability that it will not "hold on," as it is termed. It isreferable, therefore, to so wind or construct the motorhat on the start, when the armature coils are short-ircuited, the motor will tend to reach a speed higher tha

he synchronous—as for instance, double the latter. Inuch case the difficulty above alluded to is not felt, for thmotor will always hold up to synchronism if theynchronous speed— in the case supposed of twohousand revolutions —is reached or passed. This may bccomplished in various ways; but for all practical

urposes the following will suffice: On the armature arewound two sets of coils. At the start only one of these is

m. 53.



hort-circuited, thereby producing a number of poles onhe armature, which will tend to run the speed up abovehe synchronous limit. When such limit is reached orassed, the current is directed through the other coil,

which, by increasing the number <>f armature poles,

ends to maintain synchronism.

n Fig. 52, such a disposition is shown. The motor havingay, eight poles contains two field-circuits A and B, of ifferent self-induction. The armature has two coils F an

G. The former is closed upon itself, the latter connected

with the field and line through contact-rings a 5, brushesG d, and a switch K. On the start the coil F alone is active

nd the motor tends to run at a speed above theynchronous; but when the coil G is connected to theircuit the number of armature poles is increased, whilehe motor is made a true synchronous motor. This

isposition

as the advantage that the closed armature-circuitmparts to the motor torque when the speed falls off, butt the same time the conditions are such that the motoromes out of synchronism more readily. To increase the

endency to synchronism, two circuits may be used on thrmature, one of which is short-circuited on the start anoth connected with the external circuit after theynchronous speed is reached or passed. This dispositions shown in Fig. 53. There are three contact-rings a b end three brushes c d f, which connect the armature

ircuits with the external circuit. ()n startin , the switch



s turned to complete the connection between oneinding-post p and the field-coils. This short-circuits onef the armature-coils, as G. The other coil F is out of ircuit and open. When the motor is up to speed, thewitch H is turned back, so that the connection from

inding-post p to the field coils is through the coil G, andwitch K is closed, thereby including coil F in multiple arcwith the field coils. Both armature coils arethus active.

rom the above-described instances it is evident thatmany other dispositions for carrying out the invention ar

ossible.

HAPTER XII.

MAGNETIC LAG " MOTOK.

HE following description deals with another form of motor, namely, depending on " magnetic lag " orysteresis, its peculiarity being that in it the attractiveffects or phases while lagging behind the phases of urrent which produce them, are manifestedimultaneously and not successively. The phenomenon

tilized thus at an early stage by Mr. Tesla, was notenerally believed in by scientific men, and Prof. Ayrtonwas probably iirst to advocate it or to elucidate the reaso

f its supposed existence.

ig. 54- is a side view of the motor, in elevation. Fig. 55 i

part-sectional view at right angles to Fig. 54. Fig. 56 isn end view T in elevation and part section of a



modification, and Fig. 57 is a similar view of anothermodification.

n Figs. 54 and 55, A designates a base or stand, and B Bhe supporting-frame of the motor. Bolted to the

upporting-frame are two magnetic cores or pole-pieces , of iron or soft steel. These may be subdivided oraminated, in which case hard iron or steel plates or barshould be used, or they should be wound with closed coil

D is a circular disc armature, built up of sections or platef iron and mounted in the frame between the pole-piece

c', curved to conform to the circular shape thereof. Thisisc may be wound with a number of closed coils E. v Fre the main energizing coils, supported by theupporting-frame, so as to include within their

magnetizing influence both the pole-pieces c c' and thermature i>. The pole-pieces c c' project out beyond the

oils F F on opposite sides, as indicated in the drawings. In alternating current be passed through the coils F F,otation of the armature will be produced, and thisotation is explained by the following apparent action, or

mode of operation: An impulse of current in the coils F Fstablishes two polarities in the motor. The protruding

nd of pole-piece c, for instance, will be

f one sign, and the corresponding end of pole-piece c wile of the opposite sign. The armature also exhibits twooles.at right angles to the coils r F, like poles to those inhe pole-pieces being 011 the same side of the coils. Whil

he current is flowing there is no appreciable tendency to



otation developed ; but after each current impulse ceasr begins to fall, the magnetism in the armature and inhe ends of the pole-pieces c c' lags or continues to

manifest itself, which produces a rotation of the armatury the repellent force between the more closely

pproximating points of maximum magnetic effect. Thisffect is continued by the reversal of current, theolarities of field and armature being simply reversed.

One or both of the elements—the armature or field—maye wound with

IG. 54

losed induced coils to intensify this effect. Although in thlustrations but one of the fields is shown, each element he motor really constitutes a field, wound with the close

oils, the currents being induced mainly in thoseonvolutions or coils which are parallel to the coils r F.



A modified form of this motor is shown in Fig. 5(5. In thiorm G is one of two standards that support the bearingsor the armature-shaft. H H are uprights or sides of arame, preferably magnetic, the ends c c' of which areent in the manner indicated, to conform to the shape ofhe armature D and form field-magnet poles. Theonstruction of the armature may be the same as in therevious figure, or it may be simply a magnetic disc orylinder, as shown, and a coil or coils F F are se-

OLYPHASE CURRENT*.

9

ured in position to surround both the armature and theoles c c'. The armature is detachable from its shaft, the

atter being passed through the armature after it haseen inserted in position. The operation of this form of

motor is the same in principle as that previously escribed and needs no further explanation.

One of the most important features in alternating currenmotors is, however, that they should be adapted to andapable of running efficiently on the alternating circuits iresent use, in which almost without exception theenerators yield a very high number of alternations. Sucmotor, of the type under consideration, Mr. Tesla hasesigned by a development of the principle of the motor

hown in Fig. 56, making a multipolar motor, which islustrated in Fig. 57. In the construction of



IG. 56.

IG. 57.

his motor he employs an annular magnetic frame j, with

nwardly-extending ribs or projections K, the ends of which all bend or turn in one direction and are generally haped to conform to the curved surface of the armatureoils F F are wound from one part K to the one nextdjacent, the ends or loops of each coil or group of wireseing carried over toward the shaft, so as to form y-

haped groups of convolutions at each end of thermature. The pole-pieces C C', being substantially oncentric with the armature, form ledges, along whichhe coils are laid and should project to some extenteyond the the coils, as shown. The cylindrical or drumrmature D is of the same construction as in the other

motors described, and is mounted to rotate within thennular frame and 1 Jet ween the U-sha ed ends or



ends of

he coils F. The coils F are connected in multiple or ineries with a source of alternating currents, and are so

wound that with a current or current impulse of given

irection they will make the alternate pole-pieces c of onolarity and the other pole-pieces c' of the oppositeolarity. The principle of the operation of this motor is thame as the other above described, for, considering any wo pole-pieces c c', a current impulse passing in the coil

which bridges them or is wound over both tends to

stablish polarities in their ends of opposite sign and to sp in the armature core between them a polarity of theame sign as that of the nearest pole-piece c. Upon the far cessation of the current impulse that established thesolarities the magnetism which lags behind the currenthase, and which continues to manifest itself in the polar

rojections c c' and the armature, produces by repulsion otation of the armature. The effect is continued by eacheversal of the current. What occurs in the case of oneair of pole-pieces occurs simultaneously in all, so that thendency to rotation of the armature is measured by theum of all the forces exerted by the pole-pieces, as above

escribed. In this motor also the magnetic lag or effect isntensified by winding one or both cores with closednduced'coils. The armature core is shown as thus wound

When closed coils are used, the cores should be laminate

t is evident that a pulsatory as well as an alternating

urrent mi ht be used to drive or o erate the motors



bove described.

t will be understood that the degree of subdivision, themass of the iron in the cores, their size and the number o

lternations in the current employed to run the motor,

must be taken into consideration in order to properly onstruct this motor. In other words, in all such motorshe proper relations between the number of alternationsnd the mass, size, or quality of the iron must bereserved in order to secure the best results.

HAPTEE XIII.

METHOD OF OBTAINING DIFFERENCE OF PHASE BMAGNETIC SHIELDING.

N that class of motors in which two or more sets of

nergizing magnets are employed, and in which by rtificial means a certain interval of time is made to elapetween the respective maximum or minimum periods ohases of their magnetic attraction or effect, the intervalr difference in phase between the two sets of magnets ismited in extent. It is desirable, however, for the

conomical working of such motors that the strength orttraction of one set of magnets should be maximum, athe time when that of the other set is minimum, andonversely ; but these conditions have not heretoforeeen realized except in cases where the two currentsave been obtained from independent sources in the

ame or different machines. Mr. Tesla has thereforeevised a motor embodying conditions that approach



more nearly the theoretical requirements of perfectworking, or in other words, he produces artificially a

ifference of magnetic phase by means of a current fromingle primary source sufficient in extent to meet theequirements of practical and economical working. He

mploys a motor with two sets of energizing or fieldmagnets, each wound with coils connected with a sourcef alternating or rapidly-varying currents, but formingwo separate paths or circuits. The magnets of one set arrotected to a certain extent from the energizing action ohe current by means of a magnetic shield or screen

nterposed between the magnet and its energizing coil.his shield is properly adapted to the conditions of articular cases, so as to shield or protect the main core

rom magnetization until it has become itself saturatednd no longer capable of containing all the lines of forceroduced by the current. It will be seen that by this

means the energizing action begins in the protected set omagnets a certain arbitrarily-determined period of timeater than in the other, and that by this means alone or inonjunction with other means or devices

eretofore employed a practical difference of magnetic

hase may readily be secured.

ig. 58 is a view of a motor, partly in section, with aiagram illustrating the invention. Fig. 59 is a similar vief a modification of the same.

n Fig. 58, which exhibits the simplest form of thenvention A A is the field-ma net of a motor havin sa



ight poles or inwardly-projecting cores B and c. Theores B form one set of magnets and are energized by oils D. The cores c, forming the other set are energizedy coils E, and the coils are connected, preferablv, ineries with one another, in two derived or branched

ircuits, r o, respectively, from a suitable source of urrent. Each coil E is surrounded by a magnetic shield nwhich is preferably composed of an annealed, insulated,

IG. 58.

IG. 59.

r oxidized iron wire wrapped or wound on the coils in thmanner indicated so as to form a closed magnetic circuit

round the coils and between the same and the magneticores c. Between the pole pieces or cores B c is mountedhe- armature K, which, as is usual in this type of

machines, is wound with coils L closed upon themselves.he o eration resultin from this dis osition is as follows



f a current impulse be directed through the two circuitsf the motor, it will quickly energize the cores B, but noto the cores c, for the reason that in passing through theoils E there is encountered the influence of the closed

magnetic circuits formed by the shields H. The first effec

s to retard effectively the current impulse in circuit G,while at the same time the proportion of current whichoes pass does not magnetize the cores c, which arehielded or

creened by the shields H. As the increasing

lectromotive force then urges more current through theoils E, the iron wire H becomes magnetically saturatednd incapable of carrying all the lines of force, and henceeases to protect the cores c, which becomes magnetizedeveloping their maximum effect after an interval of timubsequent to the similar manifestation of strength in th

ther set of magnets, the extent of which is arbitrarily etermined by the thickness of the shield H, and other

well-understood conditions.

rom the above it will be seen that the apparatus orevice acts in two ways. First, by retarding the current,

nd, second, by retarding the magnetization of one set ofhe cores, from which its effectiveness will readily appea

Many modifications of the principle of this invention areossible. One useful and efficient application of the

nvention is shown in Fig. 59. In this figure a motor is

hown similar in all respects to that above described,xce t that the iron wire H which is wra ed around th



oils E, is in this case connected in series with the coils D.he iron-wire coils H, are connected and wound, so as toave little or no self-induction, and being added to theesistance of the circuit F, the action of the current in thaircuit will be accelerated, while in the other circuit G it

will be retarded. The shield H may be made in many orms, as will be understood, and used in different ways,s appears from the foregoing description.

As a modification of his type of motor with " shielded "elds^ Mr. Tesla has constructed a motor with a field-

magnet having two sets of poles or inwardly-projectingores and placed side Uy side, so as practically to formwo fields of force and alternately disposed—that is to sa

with the poles of one set or field opposite the spacesetween the other. He then connects the free ends of oneet of poles by means of laminated iron bands or bridge-

ieces of considerably smaller cross-section than theores themselves, whereby the cores will all form parts oomplete magnetic circuits. When the coils on each set of

magnets are connected in multiple circuits or branchesrom a source of alternating currents, electromotiveorces are set up in or impressed upon each circuit

imultaneously; but the coils on the magnetically bri'dger shunted cores will have, by reason of the -closed

magnetic-circuits, a high self-induction, which retards thurrent, permitting at the beginning of each impulse butt-




e current to pass. On the other hand, no such oppositioneing encountered in the other set of coils, the currentasses freely through them, magnetizing the poles on

which they are wound. As soon, however, as theaminated bridges become saturated and incapable of

arrying all the lines of force which the risinglectromotive force, and consequently increased currentroduce, free poles are developed at the ends of the core

which, acting in conjunction with the others, produceotation of the armature.

he construction in detail by which this invention islustrated is shown in the accompanying drawings.

ig. 60 is a view in side elevation of a motor embodyinghe principle. Fig. 61 is a vertical cross-section of the

motor. A is the frame of the motor, which should be built

p of sheets of iron punched out to the desired shape andolted together witli



IG. 60.

IG. 61.

nsulation between the sheets. When complete, the frammakes a field-magnet with inwardly projecting pole-

ieces B and c. To adapt them to the requirements of thiarticular case these pole-pieces are out of line with onenother, those marked B surrounding one end of thermature and the others, as c, the opposite end, and they

re disposed alternately — that is to say, the pole-piecesf one set occur in line with the spaces between those of he other sets.

he armature D is of cylindrical form, and is alsoaminated in the'usual way and is wound longitudinally

with coils closed upon themselves. The pole-pieces c areonnected or shunted by bridge-pieces E. These may bemade independently and attached to the pole-pieces, orhey may be parts of the forms or blanks stamped orunched out of sheet-iron. Their size or mass is de-

ermined by various conditions, such as the strength of he current to be em lo ed the mass or size of the cores



o which they are applied, and other familiar conditions.

oils F surround the pole-pieces B, and other coils G arewound on the pole-pieces c. These coils are connected ineries in two circuits, which are branches of a circuit from

generator of alternating currents, and they may be sowound, or the respective circuits in which they arencluded may be so arranged, that the circuit of coils G

will have, independently of the particular constructionescribed, a higher self-induction than the other circuit oranch.

he function of the shunts or bridges E is that they shallorm with the cores c a closed magnetic circuit for aurrent up to a predetermined strength, so that whenaturated by such current and unable to carry more linef force than such a current produces they will to no

urther appreciable extent interfere with theevelopment, by a stronger current, of free magneticoles at the ends of the cores c.

n such a motor the current is so retarded in the coils G,nd the manifestation of the free magnetism in the poles

s so delayed beyond the period of maximum magneticffect in poles B, that a strong torque is produced and the

motor operates with approximately the power developedn a motor of this kind energized by independently enerated currents differing by a full quarter phase.

HAPTEK XIV. TYPE OF TESLA SINGLE-PHASEMOTOR.





n another Tesla motor, however, the torque, instead of eing solely the result of a tjme difference in the magneteriods or phases of the poles or attractive parts to

whatever cause due, is produced by an angular

isplacement of the parts which, though movable withespect to one another, are magnetized simultaneously, opproximately so, by the same currents. This principle operation has been embodied practically in a motor in

which the necessary angular displacement between theoints of greatest magnetic attraction in the two element

f the motor—the armature and field—is obtained by theirection of the lamination of the magnetic cores of thelements.

ig. 62 is a side view of such a motor with a portion of itsrmature core exposed. Fig. 63 is an end or edge view of

he

OLYPHASE CURRENTS.

7

ame. Fig. 64 is a central cross-section of the same, thermature being shown mainly in elevation.

et A A designate two plates built up of thin sections oraminae of soft iron insulated more or less from onenother and held together by bolts a and secured to a

ase B. The inner faces of these plates contain recesses orooves in which a coil or coils D are secured obliquely to



he direction of the laminations. Within the coils D is a di, preferably composed of a spirally-wound iron wire or

ibbon or a series of concentric-rings and mounted on ahaft r, having bearings in the plates A A. Such a device

when acted upon by an alternating current is capable of

otation and constitutes a motor, the operation of whichmay be explained in the following manner: A current orurrent-impulse traversing the coils n tends to magnetizhe

IG. 62.

IG. 63.

IG. 64.

ores A A and E, all of which are within the influence of he lield of the coils. The poles thus established wouldaturally lie in the same line at right angles to the coils D

ut in the plates A they are deflected by reason of theirection of the laminations, and appear at or near the



xtremities of these plates. In the disc, however, wherehese conditions are not present, the poles or points of reatest attraction are on a line at right angles to thelane of the coils; hence there will be a torque establishey this angular displacement of the poles or magnetic

nes, which starts the disc in rotation, the magnetic linesf the armature and field tending toward a position of arallelism. This rotation is continued and maintained byhe reversals of the current in coils D D, which changelternately the polarity of the field-cores A A. This rotarendency or effect will be greatly

ncreased by winding the disc with conductors G, closedpon themselves and having a radial direction, whereby he magnetic intensity of the poles of the disc will bereatly increased by the energizing effect of the currentsnduced in the coils G by the alternating currents in coils

D.

he cores of the disc and field may or may not be of ifferent magnetic susceptibility — that is to say, they

may both be of the same kind of iron, so as to bemagnetized at approximately the same instant by the

oils D; or one may be of soft iron and the other of hard, irder that a certain time may elapse between the periodf their magnetization. In either case rotation will beroduced ; but unless the disc is provided with the closednergizing coils it is desirable that the above-describedifference of magnetic susceptibility be utilized to assist i

s rotation.



he cores of the field and armature may be made inarious ways, as will be well understood, it being only equisite that the laminations in each be in such directions to secure the necessary angular displacement of theoints of greatest attraction. Moreover, since the disc

may be considered as made up of an infinite number of adial arms, it is obvious that what is true of a disc holdsor many other forms of armature.

HAPTER XV.

MOTORS WITH CIRCUITS OF DIFFERENTRESISTANCE.

As lias been pointed out elsewhere, the lag; or retardatiof the phases of an alternating current is directly roportional to the self-induction and inversely

roportional to the resistance of the circuit through whiche current flows. Hence, in order to secure the properifferences of phase between the two motor-circuits, it isesirable to make the self-induction in one much highernd the resistance much lower than the self-induction anesistance, respectively, in the other. At the same time

he magnetic quantities of the two poles or sets of poleswhich the two circuits produce should be approximately

qual. These requirements have led Mr. Tesla to thenvention of a motor having the following generalharacteristics : The coils which are included in thatnergizing circuit which is to have the higher self-

nduction are made of coarse wire, or a conductor of elativel low resistance and with the reatest ossible



ength or number of turns. In the other set of coils aomparatively few turns of liner wire are used, or a wiref higher resistance. Furthermore, in order topproximate the magnetic quantities of the poles excitedy these coils, Mr. Tesla employs in the self-induction

ircuit cores much longer than those in the other oresistance circuit.

ig. 65 is a part sectional view of the motor at right angleo the shaft. Fig. 66 is a diagram of the tield circuits.

n Fig. 66, let A represent the coils in one motor circuit,nd H those in the other. The circuit A is to have theigher self-induction. There are, therefore, used a long

ength or a large number of turns of coarse wire inorming the coils of this circuit. For the circuit B, a smalleonductor is employed, or a conductor of a higher

esistance than copper, such as German silver or iron, anhe coils are wound with fewer turns. In applying theseoils to a motor, Mr. Tesla builds up a field-magnet of lates c, of iron and steel, secured together in the usual

manner

y bolts D. Each plate is formed with four (more or less)ong cores E, around which is a space to receive the coilnd an equal number of short projections F to receive thoils of the resistance-circuit. The plates are generally nnular in shape, having an open space in the centre foreceiving the armature G, which Mr. Tesla prefers to

wind with closed coils. An alternatin current divided



etween the two circuits is retarded as to its phases in thircuit A to a mucli greater extent than in the circuit B. B

IG. 65.

IG.

eason of the relative sizes and disposition of the coresnd coils the magnetic effect of the poles E and F upon thrmature closely approximate.

An important result secured by the construction shownere is that these coils which are designed to have the

igher self-induction are almost completely surroundedy iron, and that the retardation is thus very materially

ncreased.

HAPTER XVI.

MOTOR WITH EQUAL MAGNETIC ENERGIES INIELD AND ARMATURE.



ET it be assumed that the energy as represented in themagnetism in the field of a given rotating field motor is

inety and thafe of the armature ten. The sum of theseuantities, which represents the total energy expended iriving the motor, is one hundred; but, assuming that th

motor be so constructed that the energy in the field isepresented by fifty, and that in the armature by fifty,he sum is still one hundred ; but while in the firstnstance the product is nine hundred, in the second it is

IG. 67.

wo thousand five hundred, and as the energy developed

s in proportion to these products it is clear that thosemotors are the most efficient — other things being equal— in which the magnetic energies developed in the

rmature and field are equal. These results Mr. Teslabtains by using the same amount of copper or ampereurns in both elements when the cores of both are equal,

r approximately so, and the same current energizes botor in cases where the currents in one element are



nduced to those of the other he uses in the induced coilsn excess of copper over that in the primary element oronductor.


he conventional figure of a motor here introduced, Fig.H7, will give an idea of the solution furnished by Mr. Tesor the specific problem. Referring to the drawing, A is theld-magnet, B the armature, c the field coils, and D thermature-coils of the motor.

Generally speaking, if the mass of the cores of armaturend field be equal, the amount of copper or ampere turnf the energizing coils on both should also be equal ; buthese conditions will be modified in different forms of

machine. It will be understood that these results are mo

dvantageous when existing under the conditionsresented where the motor is running with its normaload, a point to be well borne in mind.

HAPTER XVII.

MOTORS WITH COINCIDING MAXIMA OFMAGNETIC EFFECT IN ARMATURE AND FIELD.

N THIS forin of motor, Mr. Tesla's object is to design anuild machines wherein the maxima of the magneticffects of the armature and field will more nearly coincid

han in some of the types previously under considerationhese types are: First, motors having two or more



nergizing circuits of the same electrical character, and inhe operation of which the currents used differ primarilyn phase; second, motors with a plurality of energizingircuits of different electrical character, in or by means o

which the difference of phase is produced artificially, and

hird, motors with a plurality of energizing circuits, theurrents in one being induced from currents in another.onsidering the structural and operative conditions of anne of them —as, for example, that first named — thermature which is mounted to rotate in obedience to theo-operative influence or action of the energizing circuits

as coils wound upon it which are closed upon themselvend in which currents are induced by the energizing-urrents with the object and result of energizing thermature-core; but under any such conditions as mustxist in these motors, it is obvious that a certain time

must elapse between the manifestations of an energizing

urrent impulse in the field coils, and the correspondingmagnetic state or phase in the armature established by he current induced thereby; consequently a given

magnetic influence or effect in the field which is the direcesult of a primary current impulse will have become

more or less weakened or lost before the corresponding

ffect in the armature indirectly produced has reached itmaximum. This is a condition unfavorable to efficientworking in certain cases—as, for instance, when the

rogress of the resultant poles or points of maximumttraction is verj* great, or when a very high number of lternations is employed—for it is apparent that atronger




endency to rotation will be maintained if the maximummagnetic attractions or conditions in both armature and

eld coincide, the energy developed by a motor being

measured by the product of the magnetic quantities of thrmature and field.

o secure this coincidence of maximum magnetic effectsMr. Tesla has devised various means, as explained below

ig. 68 is a diagrammatic illustration of a Tesla motor

ystem in which the alternating currents proceed fromndependent sources and differ primarily in phase.

A designates the field-magnet or magnetic frame of themotor;



IG. 68.

IG. 69.

B, oppositely located pole-pieces adapted to receive thoils of one energizing circuit; and c c, similar pole-pieces

or the coils of the other energizing circuit. These circuitsre designated, respectively, by D E, the conductor B"orming a common return to the generator G. Betweenhese poles is mounted an armature — for example, a rinr annular armature, wound with a series of coils F,orming a closed circuit or circuits. The action or operatio

f a motor thus constructed is now well understood. It we observed, however, that the magnetism of poles B, fo

xample, established by a current impulse in the coilshereon, precedes the magnetic effect set up in thermature by the induced current in coils F. Consequently

he mutual attraction between the armature and field-oles is considerably reduced. The same conditions will bound to exist if, instead of assuming the poles B or c ascting independently, we regard the ideal resultant of oth acting together, which is the real condition. Toemedy this, the motor field is constructed with

econdary poles B' c', which are situated between thethers. These ole- ieces are wound with coils D' E', the



ormer in derivation to the coils D, the latter to coils E.he main or primary coils D and E are wound for aifferent self-induction from that of the coils*D' and E',he relations being so fixed that if the currents in D and Eiffer, for example, by a quarter-phase, the currents in

ach secondary coil, as D' E', will differ from those in itsppropriate primary D or E by, say, forty-five degrees, one-eighth of a period.

Now, assuming that an impulse or alternation in circuit oranch E is just beginning, while in the branch u it is just

alling from maximum, the conditions are those of auarter-phase difference. The ideal resultant of thettractive forces of the two sets of poles B c therefore mae considered as progressing from poles B to poles c, whihe impulse in E is rising to maximum, and that in D isalling to zero or minimum. The polarity set up in the

rmature, however, lags behind the manifestations of fiemagnetism, and hence the maximum points of attractionn armature and field, instead of coinciding, are angularlyisplaced. This effect is counteracted by the supplementoles B' c'. The magnetic phases of these poles succeedhose of poles B c by the same, or nearly the same, perio

f time as elapses between the effect of the poles B c andhe corresponding induced effect in the armature ; hencehe magnetic conditions of poles B' c' and of the armature

more nearly coincide and a better result is obtained. Asoles B' c' act in conjunction with the poles in thermature established by poles B c, so in turn poles c B acimilarly with the poles set up by B' c', respectively.



Under such conditions the retardation of the magneticffect of the armature and that of the secondary poles wiring the maximum of the two more nearly intooincidence and a correspondingly stronger torque or

magnetic attraction secured.

n such a disposition as is shown in Fig. fiS it will bebserved


hat as the adjacent pole-pieces of either circuit are of likolarity they will have a certain weakening effect uponne another. Mr. Tesla therefore prefers to remove theecondary poles from the direct influence of the others.his may be done by constructing a motor with two

ndependent sets of fields, and with either one or two

rmatures electrically connected, or by using twormatures and one field. These modifications arelustrated further on.

ig. 69 is a diagrammatic illustration of a motor andystem in which the difference of phase is artificially

roduced. There are two coils D i) in one branch and twooils E E in another branch



IG. 71.

f the main circuit from the generator o. These twoircuits or branches are of different self-induction, one, a

D, being higher than the other. This is graphically

ndicated by making coils D much larger than coils E. By eason of the difference in the electrical character of thewo circuits, the phases of current in one are retarded toreater extent than the other. Let this difference behirty degrees. A motor thus constructed will rotatender the action of an alternating current; but as happen

n the case previously described the correspondingmagnetic effects of the armature and field do not coincidwing to the time that elapses between a given magneticffect in the armature and

he condition of the field that produces it. The secondary

r supplemental poles B' c' are therefore availed of. Therein thirt de rees difference of hase between the





oils i' K are in circuit with one another, as also are coils j, and there should be a difference of phase between theurrents in coils K and L and their correspondingrimaries of, say, fifteen degrees. If the poles B c are atight angles, the armature-coils should be con-.nected

irectly across, or a single armature core wound from eno end may be used ; but if the poles B c be in line therehould be an angular displacement of the armature coils,s will be well understood.

he operation will be understood from the foregoing. Th

maximum magnetic condition of a pair of poles, as B' B',oincides closely with the maximum effect in thermature, which lags behind the corresponding conditionn poles H n.

HAPTER XVIII.

MOTOR BASED ON THE DIFFERENCE OF PHASE INHE MAGNETIZATION OF THE INNER AND OUTERARTS OF AN IRON CORE.

T is well known that if a magnetic core, even if laminate

r subdivided, be wound with an insulated coil and aurrent of electricity be directed through the coil, themagnetization of the entire core does not immediately nsue, the magnetizing effect not being exhibited in allarts simultaneously. This may be attributed to the facthat the action of the current is to energize first those

aminae or parts of the core nearest the surface anddjacent to the exciting-coil, and from thence the action



rogresses toward the interior. A certain interval of timeherefore elapses between the manifestation of

magnetism in the external and the internal sections orayers of the core. If the core be thin or of small mass, thffect may be inappreciable ; but in the case of a thick

ore, or even of a comparatively thin one, if the number lternations or rate of change of the current strength beery great, the time interval occurring between the

manifestations of magnetism in the interior of the corend in those parts adjacent to the coil is more marked. Inhe construction of such apparatus as motors which are

esigned to be run by alternating or equivalent currentsuch as pulsating or undulating currents generally—Mr.esla found it desirable and even necessary to give due

onsideration to this phenomenon and to make specialrovisions in order to obviate its consequences. With thepecific object of taking advantage of this action or effect

nd to render it more pronounced, he constructs a fieldmagnet in which the parts of the core or cores that exhib

t different intervals of time the magnetic effect imparteo them by alternating or equivalent currents in annergizing coil or coils, are so placed with relation to aotating armature as to exert thereon their attractive

ffect successively in the order of their magnetization. Byhis means he secures a result similar to that which head previously attained in other forms or types of mo-

or in which by means of one or more alternating currente lias produced the rotation or progression of the

magnetic poles.



his new mode of operation will now be described. Fig. 7s a side elevation of such motor. Fig. 73 is a side elevatiof a more practicable and efficient embodiment of the

nvention. Fig. 74 is a central vertical section of the samen the plane of the axis of rotation.

Referring to Fig. 72, let x represent a large iron core,which may be composed of a number of sheets or lamina

f soft iron or steel. Surrounding this core is a coil Y,which is connected with a source E of rapidly varyingurrents. Let us consider now

IGS. 72 and 73.

he magnetic conditions existing in this core at any points 5, at or near the centre, and an other oint, as #,



earer the surface. When a current impulse is started inhe magnetizing coil Y, the section or part at <z, beinglose to the coil, is immediately energized, while theection or part at J, which, to use a convenient expressios " protected " by the intervening sections or layers

etween a and J, does not at once exhibit its magnetism.However, as the magnetization of a increases, 5 becomeslso affected, reaching finally its maximum strength somme later than a. Upon the weakening of the current the

magnetization of a first diminishes, while J still exhibits imaximum strength;

ut the continued weakening of a is attended by aubsequent weakening of b. Assuming the current to ben alternating one, a will now be reversed, while b stillontinues of the first imparted polarity. This actionontinues the magnetic condition of &, following that of a

n the manner above described. If an armature —fornstance, a simple disc F, mounted to rotate freely on anxis—be brought into proximity to the core, a movementf rotation will be imparted to the disc, the .directionepending upon its position relatively to the core, theendency being to turn the portion of the disc nearest to

he core from a to £>, as indicated in Fig. 72.

his action or principle of operation has been embodied ipracticable form of motor, which is illustrated in Fig. 73et A



IG. 74.

n that figure represent a circular frame of iron, fromiametrically opposite points of the interior of which theores project. Each core is composed of three main parts

, B and c, and they are similarly formed with a straightortion or body <?, around which the energizing coil is

wound, a curved arm or extension e, and an inwardly rojecting pole or end d. Each core is made up of twoarts B B, with their polar extensions reaching in oneirection, and a part c between the other two, and with it

olar extension reaching in the opposite direction. Inrder to lessen in the cores the circulation of currents

nduced therein, the several sections are insulated fromne another in the manner usually

ollowed in such cases. These cores are wound with coils

D which are connected in the same circuit either in



arallel or series, arid supplied with an alternating or aulsating current, preferably the former, by a generator

K, represented diagrammatically. Between the cores orheir polar extensions is mounted a cylindrical or similarrmature F, wound with magnetizing coils G, closed upon

hemselves.

he operation of this motor is as follows : When a currenmpulse or alternation is directed through the coils D, theections B B of the cores, being on the surface and in closroximity to the coils, are immediately energized. The

ections c, on the other hand, are protected from themagnetizing influence of the coil by the interposed layersf iron B B. As the magnetism of B B increases, however,he sections c are also energized; but they do not attainheir maximum strength until a certain time subsequento the exhibition by the sections B B of their maximum.

Upon the weakening of the current the magnetic strengtf B B first diminishes, while the sections c have still thei

maximum strength; but as B B continue to weaken thenterior sections are similarly weakened. B B may thenegin to exhibit an opposite polarity, which is followed

ater by a similar change on c, and this action continues. B

and c may therefore be considered as separate field-magnets, being extended so as to act on the armature inhe most efficient positions, and the effect is similar tohat in the other forms of Tesla motor —viz., a rotation orogression of the maximum points of the field of force.

Any armature — such, for instance, as a disc— mounted his field would rotate from the pole first to exhibit its



magnetism to that which exhibits it later.

t is evident that the principle here described may bearried out in conjunction with other means for securing

more favorable or efficient action of the motor. For

xample, the polar extensions of the sections c may bewound or surrounded by closed coils. The effect of theseoils will be to still more effectively retard the

magnetization of the polar extensions of c.

HAPTER XIX.

ANOTHER TYPE OF TESLA INDUCTION MOTOR.

T WILL have been gathered by all who are interested ihe advance of the electrical arts, and who follow arefully, step by step, the work of pioneers, that Mr.

esla ha£ been foremost to utilize inductive effects inermanently closed circuits, in the operation of lternating motors. In this chapter one simple type of uch a motor is described and illustrated, which will servs an exemplification of the principle.

et it be assumed that an ordinary alternating currentenerator is connected up in a circuit of practically no selnduction, such, for example, as a circuit containingncandescent lamps only. On the operation of the

machine, alternating currents will be developed in theircuit, and the phases of these currents will theoretically

oincide with the phases of the impressed electromotiveorce. Such currents may be regarded and designated as



he "unretarded currents."

t will be understood, of course, that in practice there islways more or less self-induction in the circuit, which

modifies to a corresponding extent these conditions; but

or convenience this may be disregarded in theonsideration of the principle of operation, since the samaws apply. Assume next that a path of currents beormed across any two points of the above circuit,onsisting, for example, of the primary of an inductionevice. The phases of the currents passing through the

rimary, owing to the self-induction of the same, will notoincide with the phases of the impressed electromotiveorce, but will lag-behind, such lag being directly roportional to the self-induction and inversely roportional to the resistance of the said coil. The

nsertion of this coil will also cause a lagging or retardatio

f the currents traversing and delivered by the generatoehind the impressed electromotive force, such lag beinghe mean or resultant of the lag of the current throughhe primary alone and of the " unretarded current" in thntire working circuit. Next

OL YPHAXE CURRENTS.

8

onsider the conditions imposed by the association innductive relation with the primary coil, of a secondary

oil. The current generated in the secondary coil will reapon the primary current, modifying the retardation of



he same, according to the amount of self-induction andesistance in the secondary circuit. If the secondary ircuit has but little self-induction —as, for instance, whecontains incandescent lamps only— it will increase the

ctual difference of phase between its own and the

rimary current, first, by diminishing the lag between thrimary current and the impressed electromotive force,nd, second, by its own lag or retardation behind thempressed electromotive force. On the other hand, if theecondary circuit have a high self-induction, its lag behinhe current in the primary is



IG. 7.-).

irectly increased, while it will be still further increased ihe primary have a very low self-induction. The betteresults are obtained when the primary has a low self-nduction.

ig. 75 is a diagram of a Tesla motor embodying this

rinciple. Fig. 76 is a similar diagram of a modification of he same. In Fig. 75 let A designate the field-magnet of a

motor which, as in all these motors, is built up of sectionsr plates. B c are polar projections upon which the coilsre wound. Upon one pair of these poles, as c, are woundrimary coils i>, which are directly connected to theircuit of an alternating current generator a. On the samoles are also wound secondary coils r, either side by sidr over or under the primary coils, and these areonnected with other coils E, which surround the poles B.

he currents in both primary and secondary coils in suchmotor will be retarded or will lag behind the impressed

lectromotive force ; but to secure a proper difference inhase between the primary and secondary currentshemselves, Mr. Tesla increases the resistance of the

ircuit of the secondary and reduces as much asracticable its self-induction. This is done b usin for th



econdary circuit, particularly in the coils E, wire of omparatively small diameter and having but few turnsround the cores; or by using some conductor of higherpecific resistance, such as German silver; or by ntroducing at some point in the secondary circuit an

rtificial resistance K. Thus the self-induction of theecondary is kept down and its resistance increased, withhe result of decreasing the lag between the impressedlectro-motive force and the current in the primary coilsnd increasing the difference of phase between therimary and secondary currents.

n the disposition shown in Fig. 76, the lag in theecondary is increased by increasing the self-induction ohat circuit, while the increasing tendency of the primaryo lag is counteracted by inserting therein a deadesistance. The primary coils D in this case have a low

elf-induction and high resistance, while the coils E F,ncluded in the secondary circuit, have a high self-nduction and low resistance. This may be done by theroper winding of the coils; or in the circuit including theecondary coils E F, we may introducb a self-induction co while in the primary circuit from the generator o and

ncluding coils D, there may be in serted a dead resistancR. By this means the difference of phase between the

rimary and secondary is increased. It is evident thatoth means of increasing the difference of phase—amely, by the special winding as well as by theupplemental or external inductive and dead resistance—

may be employed conjointly.



n the operation of this motor the current impulses in thrimary coils induce currents in the secondary coils, andy the conjoint action of the two the points of greatest

magnetic attraction are shifted or rotated.

n practice it is found desirable to wind the armature witlosed coils in which currents are induced by the actionhereon of the primaries.

HAPTER XX.

OMBINATIONS OF SYNCHRONIZING MOTOR ANDORQUE MOTOR.

N THE preceding descriptions relative to synchronizingmotors and methods of operating them, reference has

een made to the plan adopted by Mr. Tesla, which

onsists broadly in winding or arranging the motor in sucmanner that by means of suitable switches it could betarted as a multiple-circuit motor, or one operating by arogression of its magnetic poles, and then, when up topeed, or nearly so, converted into an ordinary ynchronizing motor, or one in which the magnetic poles

were simply alternated. In some cases, as when a largemotor is used and when the number of alternations is

ery high, there is more or less difficulty in bringing themotor to speed as a double or multiple-circuit motor, forhe plan of construction which renders the motor bestdapted to run as a synchronizing motor impairs its

fficiency as a torque or double-circuit motor under thessumed conditions on the start. This will be readil





motor; Fig. 80 a diagram of the circuit connectionsmployed ; and Figs. 81, 82, 83, 84 and 85 are diagramsf modified dispositions of the two motors.

nasmuch as neither motor is doing any work while the

urrent is acting upon the other, the two armatures areigidly connected, both being mounted upon the samehaft A, the field-magnets B of the synchronizing and c ohe torque motor being secured to

he same base D. The preferably larger synchronizingmotor has polar projections on its armature, which rotatn very close proximity to the poles of the field, and inther respects it conforms to the conditions that areecessary to secure synchronous action. The pole-pieces

f the armature are, however, wound with closed coils E,s this obviates the employment of sliding contacts. Themaller or torque motor, on the other hand, has,referably, a cylindrical armature F, without polarrojections and wound with closed coils G. The field-coilsf the torque motor are connected up in two series H andand the alternating current from the generator is



irected through or divided between these two circuits inny manner to produce a progression of the poles oroints of maximum magnetic effect. This result is securey connecting the two motor-circuits in derivation witlihe circuit

OLYPHASE CURRENTS.

rom the generator, inserting in one motor circuit a deadesistance and in the other a self-induction coil, by which

means a difference in phase between the two divisions of

he current is secured. If both motors have the sameumber of field poles, the torque motor for a givenumber of alternations will tend to run at double thepeed of the other, for, assuming the connections to beuch as to give the best results, its poles are divided intowo series and the number of poles is virtually reduced

ne-half, which being acted upon by the same number oflternations tend to rotate the armature at twice thepeed. By this means the main armature is more easily rought to or above the required speed. -When the speeecessary for synchronism is imparted to the main moto

he current is shifted from the torque motor into the

ther.

A convenient arrangement for carrying out this inventios



IG. 78.

IG. 79.

hown in Fig. 80, in which j .1 are the field coils of the

ynchronizing, and H i the field coils of the torque motor.L' are the conductors of the main line. One end of, say,

oils H is connected to wire L through a self-induction coiM. One end of the other set of coils i is connected to theame wire through a dead resistance N. The oppositends of these two circuits are connected to the contact m

f a switch, the handle or lever of which is in connectionwith the line-wire L', One end of the field circuit of theynchronizing motor is connected to the wire L. The otheerminates in the switch-contact n. From the diagram it

will be readily seen that if the lever p be turned uponontact m, the torque motor will start by reason of theifference of hase between the currents in its two



nergizing circuits. Then when the desired speed isttained, if the lever p be shifted upon con-

act ;/ the entire current will pass through the field coilsf the synchronizing motor and the other will be doing no

work.

he torque motor may be constructed and operated inarious ways, many of which have already been touchedpon. It is not necessary that one motor be cut out of ircuit while the other is in, for both may be acted upon

y current at the same time, and Mr. Tesla has devisedarious dispositions or arrangements of the two motorsor accomplishing this. Some of these arrangements arelustrated in Figs. 81 to 85.

Referring to Fig. 81, let T designate the torque or multip

ircuit motor and s the synchronizing motor, L i,' beinghe line-wires from a source of alternating current. Thewo circuits of the torque motor of different degrees of elf-induction, and designated by N M, are connected inerivation to the wire L. They are then joined andonnected to the energizing circuit of the



IG.

ynchronizing motor, the opposite terminal of which isonnected to wire L'. The two motors are thus in series.

o start them Mr. Tesla short-circuits the synchronizingmotor by a switch P', throwing the whole current throughe torque motor. Then when the desired speed iseached the switch p' is opened, so that the current passhrough both motors. In such an arrangement as this it ibviously desirable for economical and other reasons thaproper relation between the speeds of the two motors

hould be observed.

n Fig. 82 another disposition is illustrated, s is theynchronizing motor and T the torque motor, the circuitf both being in parallel, w is a circuit also in derivation tohe motor circuits and containing a switch P". s' is a switcn the synchronizing motor circuit. On the start, thewitch s' is opened, cutting out the motor s. Then P" ispened, throwing the entire current

OLYPHASE CURRENTS.

hrough the motor T, giving it a very strong torque. Whehe desired speed is reached, switch s' is closed and theurrent divides





IGS. 81, 82, 83, 84 and 85.

etween both motors. By means of switch p" both motormay be cut out.

n Fig. 83 the arrangement is substantially the same,xcept that a switch T' is placed in the circuit whichncludes the two circuits of the torque motor. Fig. 84hows the two motors in series, with a shunt around bothontaining a switch s T. There is also a shunt around theynchronizing motor s, with a switch p'. In Fig. 85 the

ame disposition is shown ; but each motor is providedwitli a shunt, in which are switches P' and T*, as shown.

HAPTER XXL

MOTOR WITH A CONDENSER IN THE ARMATURE

IRCUIT.

WE NOW come to a new class of motors in which resort ad to condensers for the purpose of developing theequired difference of phase and neutralizing the effects elf-induction. Mr. Tesla early began to apply the

ondenser to alternating apparatus, in just how r many ways can only be learned from a perusal of other portionf this volume, especially those dealing with his highrequency work.

ertain laws govern the action or effects produced by a

ondenser when connected to an electric circuit throughwhich an alternatin or in eneral an undulatin current



made to pass. Some of the most important of such effectsre as follows: First, if the terminals or plates of aondenser be connected with two points of a circuit, theotentials of which are made to rise and fall in rapiduccession, the condenser allows the passage, or more

trictly speaking, the transference of a current, althoughs plates or armatures may be so carefully insulated as trevent almost completely the passage of a current of nvarying strength or direction and of moderatelectromotive force. Second, if a circuit, the terminals of

which are connected with the plates of the condenser,

ossess a certain self-induction, the condenser willvercome or counteract to a greater or less degree,ependent upon well-understood conditions, the effects ouch self-induction. Third, if two points of a closed oromplete circuit through w T hich a rapidly rising andalling current flows be shunted or bridged by a

ondenser, a variation in the strength of the currents inhe branches and also a difference of phase of the currenherein is produced. These effects Mr. Tesla has utilizednd applied in a variety of ways in the construction andperation of his motors, such as by producing a differenc

n phase in the two energizing circuits of an alternating

urrent motor by connecting the two circuits in derivationd connecting up a condenser in series in one of theircuits. A further development,


owever, possesses certain novel features of practical



alue and involves a knowledge of facts less generally nderstood. It comprises the use of a condenser orondensers in connection with the induced or armatureircuit of a motor and certain details of the con-



87.

IG. 90.

truction of such motors. In an alternating current motof the type particularly referred to above, or in any othe

which has an armature coil or circuit closed upon itself,he latter represents not only an inductive resistance, bune which is period-

cally varying in value, both of which facts complicate andender difficult the attainment of the conditions bestuited to the most efficient working conditions ; in other

words, they require, first, that for a given inductive effecpon the armature there should be the greatest possibleurrent through the armature or induced coils, and,econd, that there should always exist between theurrents in the energizing and the induced circuits a giveelation of phase. Hence whatever tends to decrease theelf-induction and increase the current in the induced

ircuits will, other things being equal, increase the outpurid efficiency of the motor, and the same will be true of auses that operate to maintain the mutual attractiveffect between the field magnets and armature at its

maximum. Mr. Tesla secures these results by connectingwith the induced circuit or circuits a condenser, in the

manner described below, and he also, with this purpose iiew, constructs the motor in a special manner.





he action, of this motor and the effect of the planollowed in its construction are as follows: The motoreing started in operation and the coils of the field

magnets being traversed by alternating currents,urrents are induced in the armature coils by one set of

eld coils, as B, and the poles thus established are actedpon by the other set, as c. The armature coils, howeverave necessarily a high self-induction, which opposes theow of the currents thus set up. The condenser F not onlermits the passage or transference of these currents, bulso counteracts the effects of self-induction, and by a

roper adjustment of the capacity of the condenser, theelf-induction of the coils, and the periods of the currentshe condenser may be made to overcome entirely theffect of self-induction.

t is preferable on account of the undesirability of using

liding contacts of any kind, to associate the condenserwith the armature directly, or make it a part of the

rmature. In some cases Mr. Tesla builds up thermature of annular plates K K, held by bolts L betweeneads M, which are secured to the driving shaft, and inhe hollow space thus formed he places a condenser F,

enerally by winding the two insulated plates spirally round the shaft. In other cases he utilizes the plates of he core itself as the plates of the condenser. For exampln Figs. 88 and 89, N is the driving shaft, M M are theeads of the armature-core, and K K' the iron plates of

which the core is built up. These plates are insulated fromhe shaft and from one another, and are held together by



ods or bolts L. The bolts pass through a large hole in onelate and a small hole in the one next adjacent, and so ononnecting electrically all of plates K, as one armature of ondenser, and all of plates K' as the other.

o either of the condensers above described the armatuoils may be connected, as explained by reference to Fig6.

n motors in which the armature coils are closed uponhemselves —as, for example, in any form of alternating

urrent motor in which one armature coil or set of coils isn the position of maximum induction with respect to theeld coils or poles, while the other is in the position of

minimum induction — the coils are best connected in oneeries, and two points of the circuit thus formed areridged by a condenser. This is illustrated in Fig. 90, in

which E represents one set of armature coils and E' thether. Their points of uniou are joined through aondenser F. It will be observed that in this dispositionhe self-

nduction of the two branches E and E' varies with their

osition relatively to the field magnet, and that eachranch is alternately the predominating source of the

nduced current. Hence the effect of the condenser F iswofold. First, it increases the current in each of theranches alternately, and, secondly, it alters the phase ohe currents in the branches, this being the well-known

ffect which results from such a disposition of a condensewith a circuit as above described. This effect is favorable



o the proper working of the motor, because it increaseshe flow of current in the armature circuits due to a givennductive effect, and also because it brings more nearly nto coincidence the maximum magnetic effects of theoacting field and armature poles.

t will be understood, of course, that the causes thatontribute to the efficiency of condensers when applied tuch uses as the above must be given due consideration etermining the practicability and efficiency of the

motors. Chief among these is, as is well known, the

eriodicity of the current, and hence the improvementsescribed are mgre particularly adapted to systems in

which a very high rate of alternation or change ismaintained.

Although this invention has been illustrated in connectio

with a special form of motor, it will be understood that its equally applicable to any other alternating current

motor in which there is a closed armature coil wherein thurrents are induced by the action of the field, and theeature of utilizing the plates or sections of a magneticore for forming the condenser is applicable, generally, to

ther kinds of alternating current apparatus.

HAPTEK XXII.

MOTOR WITH CONDENSER IN ONE OF THE FIELDIRCUITS.

F THE field or ener izin circuits of a rotar hase mot



e both derived from the same source of alternatingurrents and a condenser of proper capacity be includedn one of the same, approximately, the desired differencef phase may be obtained between the currents flowingirectly from the source and those flowing through the

ondenser ; but the great size and expense of condensersor this purpose that would meet the requirements of thrdinary systems of comparatively low potential arearticularly prohibitory to their employment.

Another, now well-known, method or plan of securing a

ifference of phase between the energizing currents of motors of this kind is to induce by the currents in oneircuit those in the other circuit or circuits; but as no

means had been proposed that would secure in this way etween the phases of the primary or inducing and theecondary or induced currents that difference —

heoretically ninety degrees—that is best adapted forractical and economical working, Mr. Tesla devised a

means which renders practicable both the aboveescribed plans or methods, and by which he is enabled tbtain an economical and efficient alternating current

motor. His invention consists in placing a condenser in th

econdary or induced circuit of the motor above describend raising the potential of the secondary currents to sucdegree that the capacity of the condenser, which is inart dependent on the potential, need be quite small. Thalue of this condenser is determined in a well-nderstood manner with reference to the self-inductionnd other conditions of the circuit, so as to cause the



urrents which pass through it to differ from the primaryurrents by a quarter phase.

ig. 91 illustrates the invention as embodied in a motor iwhich the inductive relation of the primary and secondar

ircuits is secured by winding them inside the motorartly upon the same cores; but the invention applies,enerally, to

OL YPHA&K d

07ther forms of motor in which one of the energizingurrents is induced in any way from the other.

et A B represent the poles of an alternating currentmotor, of which c is the armature wound with coils D,losed upon themselves, as is now the general practice in

motors of this kind. The poles A, which alternate witholes B, are wound with coils of ordinary or coarse wire E

n such direction as to make them of alternate north andouth polarity, as indicated in the diagram by the

haracters N s. Over these coils, or in other inductiveelation to the same, are wound long fine-wire coils F F,nd in the



IG. 91.

ame direction throughout as the coils E. These coils are

econdaries, in which currents of very high potential arenduced. All the coils E in one series are connected, and ahe secondaries F in another.

On the intermediate poles B are wound line-wirenergizing coils G, which are connected in series with one

nother, and also with the series of secondary coils F, theirection of winding be-.ing such that a current-impulse

nduced from the primary coils K imparts the samemagnetism to the poles B as that produced

n poles A by the primary impulse. Tins condition is

ndicated by the characters N' s'.



n the circuit formed by the two sets of coils F and G isntroduced a condenser H ; otherwise this circuit is closedpon itself, while the free ends of the circuit of coils E areonnected to a source of alternating currents. As theondenser capacity which is needed in any particular

motor of this kind is dependent upon the rate of lternation or the potential, or both, its size or cost, asefore explained, may be brought within economicalmits for use with the ordinary circuits if the potential ofhe secondary circuit in the motor be sufficiently high. Byiving to the condenser proper values, any desired

ifference of phase between the primary and secondary nergizing circuits may be obtained.

HAPTER XXIII.

ESLA POLYPHASE TRANSFORMER.

APPLYING the polyphase principle to the construction oransformers as well to the motors already noticed, Mr.esla has invented some very interesting forms, which honsiders free from the defects of earlier and, at present

more familiar forms. In these transformers he provides

eries of inducing coils and corresponding induced coils,which are generally wound upon a core closed upon itself

sually a ring of laminated iron.

he two sets of coils are wound side by side oruperposed or otherwise placed in well-known ways to

ring them into the most effective relations to onenother and to the core. The inducing or primary coils



wound on the core are divided into pairs or sets by theroper electrical connections, so that while the coils of onair or set co-operate in fixing the magnetic poles of theore at two given diametrically opposite points, the coils he other pair or set — assuming, for sake of illustration,

hat there are but two — tend to fix the poles ninety egrees from such points. With this induction device issed an alternating current generator with coils or sets ooils to correspond with those of the converter, and theorresponding coils of the generator and converter arehen connected up in independent circuits. It results from

his that the different electrical phases in the generatorre attended by corresponding magnetic changes in theonverter; or, in other words, that as the generator coilsevolve, the points of greatest magnetic intensity in theonverter will be progressively shifted or whirled around

ig. 92 is a diagrammatic illustration of the converter anhe electrical connections of the same. Fig. 93 is aorizontal central cross-section of Fig. 92. Fig. 94 is aiagram of the circuits of the entire system, the generatoeing shown in section.

Mr. Tesla uses a core, A, which is closed upon itself— thas to say, of an annular cylindrical or equivalent form—nd as the efficiency of the apparatus is largely increasedy the subdivision

o

NVENTIONS 0V NIKOLA TKKLA.



f tliis core, he makes it of thin strips, plates, or wires .of oft iron electrically insulated as far as practicable. Uponhis core are wound, say, four coils, BBS' B', used asrimary coils, and 'for which long lengths of comparative

ne wire are employed. Over these coils are then woundhorter coils of coarser wire, c c c' c', to constitute thenduced or secondary coils. The construction of this or anquivalent form of converter may be carried further, asbove pointed out, by inclosing these coils with iron —as,or example, by winding over the coils layers of insulated

ron wire.

he device is provided with suitable binding posts, towhich



IGS. 92 and 93.

he ends of the coils are led. The diametrically opposite

oils B R and B' B' are connected, respectively, in series,nd the four terminals are connected to the binding posthe induced coils are connected together in any desired

manner. For example, as shown in Fig. 94, c c may beonnected in multiple arc when a quantity current isesired—as for running a group of incandescent lamps—

while c' c' may be independently connected in series in aircuit including arc lamps or the like. The generator inhis system will be adapted to the converter in the

OLYPHASE CURRENTS.

ll

manner illustrated. For example, in the present casehere are employed a pair of ordinary permanent orlectro-magnets, E E, between which is mounted aylindrical armature on a shaft, F, and wound with two

oils, G G'. The terminals of these coils are connected,espectively, to four insulated contact or collecting rings,

H H H' H', and the four line circuit wires L connect therushes K, bearing on these rings, to the converter in therder shown. Noting the results of this combination, it wie observed that at a given point of time the coil G is in ieutral osition and is eneratin little or no current,





efore—causes a further shifting of the poles through theecond quarter of the ring. The second half -re volution

will obviously be a repetition of the same action. By thehifting of the poles of the ring A, a power-

ul dynamic inductive effect on the coils c c' is produced.esides the currents generated in the secondary coils byynamo-magnetic induction, other currents will be set u

n the same coils in consequence of many variations in thntensity of the poles in the ring A. This should be avoidey maintaining the intensity of the poles constant, to

ccomplish which care should be taken in designing androportioning the generator and in distributing the coils he ring A, and balancing their effect. When this is done,he currents are produced by dynamo-magnetic inductionly, the same result being obtained as though the poles

were shifted by a commutator with an infinite number of

egments.

he modifications which are applicable to other forms of onverter are in many respects applicable to this, such ahose pertaining more particularly to the form of the corhe relative lengths and resistances of the primary and

econdary coils, and the arrangements for running orperating the same.

HAPTEK XXIV.

A CONSTANT CURRENT TRANSFORMER WITH

MAGNETIC SHIELD BETWEEN COILS OF PRIMARY AND SECONDARY.





words, the current in one circuit should be a maximumwhen that in the other circuit is a minimum. To attain tohis condition more perfectly, an increased retardation ohe secondary current is secured in the following mannernstead of bringing the primary and secondary coils or

ircuits of a transformer into the closest possible relations has hitherto


een done, Mr. Tesla protects in a measure the secondar

rom the inductive action or effect of the primary by urrounding either the primary or the secondary with aomparatively thin magnetic shield or screen. Underhese modified conditions, as long as the primary currentas a small value, the shield protects the secondary; buts soon as the primary current has reached a certain

trength, which is arbitrarily determined, the protectingmagnetic shield becomes saturated and the inductive

ction upon the secondary begins. It results, therefore,hat the secondary current begins to How at a certainraction of a period later than it would without thenterposed shield, and since this retardation may be

btained without necessarily retarding the primary urrent also, an additional lag is secured, and the timenterval between the maximum or minimum periods of he primary and secondary currents is increased. Such arans-



IG. 95.

ormer may, by properly proportioning its severallements and determining the proper relations between

he primary and secondary windings, the thickness of thmagnetic shield, and other conditions, be constructed to

ield a constant current at all loads.

ig. 95 is a cross-section of a transformer embodying thimprovement. Fig. 96 is a similar view of a modified form

f transformer, showing diagrammatically the manner ofsing the same.

A A is the main core of the transformer, composed of aing of soft annealed and insulated or oxidized iron wire.

Upon this core is wound the secondary circuit or coil B B

his latter is then covered with a layer or layers of nnealed and insulated iron wires c c, wound in a directiot right angles to the secondary

OLYPHASE CURRENTS.

15



oil. Over the whole is then wound the primary coil orwire D D. From the nature of this construction it will be

bvious that as long as the shield formed by the wires c ielow magnetic saturation the secondary coil or circuit isffectually protected or shielded from the inductive

nfluence of the primary, although on open circuit it may xhibit some electromotive force. When the strength of he primary reaches a certain value, the shield c,ecoming saturated, ceases to protect the secondary from

nductive action, and current is in consequence developeherein. For similar reasons, when the primary current

weakens, the weakening of the secondary is retarded tohe same or approximately the same extent.

he specific construction of the transformer is largely mma-

IG. 90.



erial. In Fig. 90, for example, the core A is built up of thnsulated iron plates or discs. The primary circuit D is

wound next the core A. Over this is applied the shield c,which in this case is made up of thin strips or plates of ron properly insulated and surrounding the primary,

orming a closed magnetic circuit. The secondary B iswound over the shield c. In Fig. 06, also, K is a source of lternating or rapidly changing currents. The primary ofhe transformer is connected with the circuit of theenerator. F is a two-circuit alternating current motor,ne of the circuits being connected with the main circuit

rom the source E, and the other being supplied withurrents from the secondarv of the transformer.

HE TESLA EFFECTS WITH HIGH FREQUENCY ANDHIGH POTENTIAL CURRENTS.

HAPTER XXV.

NTRODUCTION. — THE SCOPE OF THE TESLA ECTURES.

EFORE proceeding to study the three Tesla lectures

ere presented, the reader may find it of some assistanco have his attention directed to the main points of nterest and significance therein. The lirst of theseectures was delivered in New York, at Columbia Collegeefore the American Institute of Electrical Engineers,

May 20,1891. The urgent desire expressed immediately

rom all parts of Europe for an opportunity to witness thrilliant and unusual experiments with which the lecture





eading ideas and experiments can here be touched uponesides, it is preferable that the lectures should bearefully gone over for their own sake, it being more thakely that each student will discover a new beauty ortimulus in them. Taking up the course of reasoning

ollowed by Mr. Tesla in his first lecture, it will be notedhat he started out with the recognition of the fact, whiche has now experimentally demonstrated, that for theroduction of light waves, primarily, electrostatic effects

must be _^Jjrought into play, and continuedjrtudy hased him tpjtheLOpinion that_all electrical and magnetic

ffects may be referred to ek-c-trostajtic _irLolecular _orces. This opinion finds a singular confirmation in one ohe most striking experiments which he describes,amely, the production of a veritable flame by thegitation of electrostatically charged molecules. It is of thighest interest to observe that this result points out a

way of obtaining a flame which consumes no material ann which no chemical action whatever takes place. It alsohrows a light on the nature of the ordinary flame, which

Mr. Tesla believes to be due to electrostatic molecularctions, which, if true, would lead directly to the idea thaven chemical affinities might be electrostatic in their

ature and that, as has already been suggested, moleculaorces in general may be referable to one and the sameause. This singular phenomenon accounts in a plausible

manner for the unexplained fact that buildings arerequently set on fire during thunder storms with'outaving been at all struck by -\v lightning. It may alsoxplain the total disappearance of ships at sea.



One of the striking proofs of the correctness of the ideasdvanced by Mr. Tesla is the fact that, notwithstandinghe employment of the most powerful electromagneticnductive effects, but .feeble luminosity is obtainable, andhis only in close proximity to the source of disturbance;

whereas, when the electrostatic-effects are intensified,he same initial energy suffices to excite luminosity atonsiderable distances from the source. That there arenly electrostatic effects active seems to be clearly provey Mr. Tesla's experiments with an induction coilperated with alternating currents of very high

requency. He shows how tubes may be made to glow rilliantly at considerable distances from any object whelaced in a powerful, rapidly alternating, electrostaticeld, and he describes many interesting phenomenabserved in such a field. His experiments open up theossibility

f lighting an apartment by simply creating in it sucli anlectrostatic field, and this, in a certain way, would appeao be the ideal method of lighting a room, as it would allowhe illuminating device to be freely moved about. Theower with which these exhausted tubes, devoid of any lectrodes, light up is certainly remarkable.

hat the principle propounded by Mr. Tesla is a broadne is evident from the many ways in which it may beractically applied. We need only refer to the variety of he devices shown or described, all of which are novel inharacter and will without doubt lead to further



mportant results at the hands of Mr. Tesla and othernvestigators. The experiment, for instance, of lighting usingle filament or block of refractory material with a

ingle wire, is in itself sufficient to give Mr. Tesla's work he stamp of originality, and the numerous other

xperiments and effects which may be varied at will, arequally new and interesting. Thus, the incandescentlament spinning in an unexhausted globe, the well-nown Crookes experiment on open circuit, and the manthers suggested, will not fail to interest the reader. Mr.esla has made an exhaustive study of the various forms

f the discharge presented by an induction coil whenperated with these rapidly alternating currents, startinrom the thread-like discharge and passing througharious stages to the true electric flame.

A point of great importance in the introduction of high

ension alternating current which Mr. Tesla brings out ishe necessity of carefully avoiding all gaseous matter inhe high tension apparatus. He shows that, at least withery rapidly alternating currents of high potential, theischarge may work through almost any practicablehickness of the best insulators, if air is present. In such

ases the air included within the apparatus is violently gitated and by molecular bombardment the parts may e so greatly heated as to cause a rupture of the.

nsulation. The practical outcome of this is, that, whereaswith steady currents, any kind of insulation may be usedwith rapidly alternating currents oils will probably be the

est to employ, a fact which has been observed, but not



ntil now satisfactorily explained. The recognition of thebove fact is of special importance in the construction of he costly commercial induction coils which are oftenendered useless in an unaccountable manner. The truthf these views of Mr. Tesla is made evident by the in-

teresting experiments illustrative of the behavior of their between charged surfaces, the luminous streamsormed by the charged molecules appearing even whenreat thicknesses of tin-best insulators are interposedetween the charged surfaces. These luminous streams

fford in themselves a very interesting study for thexperimenter. With these rapidly alternating currentshey become far more powerful and produce beautifulght effects when they issue from a wire, pinwheel orther object attached to a terminal of the coil; and it isnteresting to note that they issue from a ball almost as

reely as from a point, when the frequency is very high.

rom these experiments we also obtain a better idea of he importance of taking into account the capacity andelf-induction in the apparatus employed and theossibilities offered by the use of condensers in

onjunction with alternate currents, the employment of urrents of high frequency, among other things, making ossible to reduce the condenser to practicable dirnen-sions. Another point of interest and practical bearing ishe fact, proved by Mr. Tesla, that for alternate currentsspecially those of high frequency, insulators are require

ossessing a small specific inductive capacity, which at th



ame time have a high insulating power.

Mr. Tesla also makes interesting and valuable suggestionn regard to the economical utilization of iron in machinesnd transformers. He shows how, by maintaining by

ontinuous magnetization a flow of lines through the ironhe latter may be kept near its maximum permeability nd a higher output and economy may be secured in sucpparatus. This principle may prove of considerableommercial importance in the development of alternatinystems. Mr. Tesla's suggestion that the same result can

e secured by heating the iron by hysteresis and eddy urrents, and increasing the permeability in this mannerwhile it may appear less practical, nevertheless opens

nother direction for investigation and improvement.

he demonstration of the fact that with alternating

urrents

f high frequency, sufficient energy may be transmittednder

racticable conditions through the glass of an

ncandescent lamp

y electrostatic or electromagnetic induction may lead tode-

-parture in the construction of such devices. Another

mportant



I experimental result achieved is the operation of lampnd even

1 .motors, with the discharges of condensers, thismethod affording

means of converting direct or alternating currents. Inhis connection Mr. Tesla advocates the perfecting of pparatus capable of generating electricity of high tensiorom heat energy, believing this to be a better way of btaining electrical energy for practical purposes,

articularly for the production of light.

While many were probably prepared to encounter curiouhenomena of impedance in the use of a condenserischarged disruptively, the experiments shown werextremely interesting on account of their paradoxical

haracter. The burning of an incandescent lamp at any andle power when connected across a heavy metal bar,he existence of nodes on the bar and the possibility of xploring the bar by means of an ordinary Garde w oltmeter, are all peculiar developments, but perhaps th

most interesting observation is the phenomenon of

mpedance observed in the lamp with a straight filamentwhich remains dark while the bulb glows.

Mr. Tesla's manner of operating an induction coil by means of the disruptive discharge, and thus obtainingnormous differences of potential from comparatively

mall and inexpensive coils, will be appreciated by xperimenters and will find valuable application in



aboratories. Indeed, his many suggestions and hints inegard to the construction and use of apparatus in thesenvestigations will be highly valued and will aid materialln future-research.

he London lecture was delivered twice. In its first formefore the Institution of Electrical Engineers, it was inome respects an amplification of several j>oints notpecially enlarged upon in the JS T ew York lecture, butrought forward many additional discoveries and new

nvestigations. Its repetition, in-.""-] another form, at th

Royal Institution, was due to Prof. Dewar, who with LordRayleigh, manifested a most lively interest in Mr. 'jesla's work, and whose kindness illustrated once more

he strong V } English love of scientific truth andppreciation of its votaries. } As an indefatigablexperimenter, Mr. Tesla was certainly no-^ where more

t home than in the haunts of Faraday, and as the / guesf Faraday's successor. This Royal Institution lecture W ummed up the leading points of Mr. Tesla's work, in theigh / potential, high frequency field, and we may herevail ourselves J of so valuable a summarization, in aimple form, of a subject by no means easv of

omprehension until it has been thoroughly studied.

n these London lectures, among the many notable pointmade was first, the difficulty of constructing the

lternators to obtain, the very high frequencies needed.o obtain the high frequencies it was necessary to provid

everal hundred polar projections, which were necessaril



mall and offered many drawbacks, and this the more asxceedingly high peripheral speeds had to be resorted ton some of the first machines both armature and field haolar projections. These machines produced a curiousoise, especially when the armature was started from th

tate of rest, the field being charged. The most efficientmachine was found to be one with a drum armature, theron body of which consisted of very thin wire annealed

with special care. It was, of course, desirable to avoid themployment of iron in the armature, and several

machines of this kind, with moving or stationary

onductors were constructed, but the results obtainedwere not quite satisfactory, on account of the greatmechanical and other difficulties encountered.

he study of the properties of the high frequency urrents obtained from these machines is very

nteresting, as nearly every experiment disclosesomething new. Two coils traversed by such a currentttract or repel each other with a force which, owing tohe imperfection of our sense of touch, seems continuous

An interesting observation, already noted under anotherorm, is that a piece of iron, surrounded by a coil through

which the current is passing appears to be continuously magnetized. This apparent continuity might be ascribedo the deficiency of the sense of touch, but there isvidence that in currents of such high frequencies one ofhe impulses preponderates over the other.

As might be expected, conductors traversed by such



urrents are rapidly heated, owing to the increase of theesistance, and the heating effects are relatively muchreater in the iron. The hysteresis losses in iron are soreat that an iron core, even if finely subdivided, is heaten an incredibly short time. To give an idea of this, an

rdinary iron wire -^g- inch in diameter inserted within oil having 250 turns, with a current estimated to be fivemperes passing through the coil, becomes within twoeconds' time so hot as to scorch wood. Beyond a certainrequency, an iron core, no matter how finely subdividedxercises a dampening effect, and it was easy to find a

oint at

whicli tlie impedance <>f a coil was not affected by theresence of a core consisting of a bundle of very thin welnnealed and varnished iron wires.

xperiments with a telephone, a conductor in a strongmagnetic field, or with a condenser or arc, seem to affordertain proof that sounds far above the usually acceptedmit of hearing would be perceived if produced withufficient power. The arc produced by these currentsossesses several interesting features. Usually it emits a

ote the pitch of which corresponds to twice therequency of the current, but if the frequency beufficiently high it becomes noiseless, the limit of auditioneing determined principally by the linear dimensions ofhe arc. A curious feature of the arc is its persistency,

which is due partly to the inability of the gaseous column

o cool and increase considerably in resistance, as is the



ase with low frequencies, and partly to the tendency of uch a high frequency machine to maintain a constanturrent.

n connection with these machines the condenser affords

particularly interesting study. Striking effects areroduced by proper adjustments of capacity and self-nduction. It is easy to raise the electromotive force of th

machine to many times the original value by simply djusting the capacity of a condenser connected in thenduced circuit. If the condenser be at some distance from

he machine, the difference of potential on the terminalsf the latter may be only a small fraction of that on theondenser.

ut the most interesting experiences are gained when thension of the currents from the machine is raised by

means of an induction coil. In consequence of thenormous rate of change obtainable in the primary urrent, much higher potential differences are obtainedhan with coils operated in the usual ways, and, owing tohe high frequency, the secondary discharge possesses

many striking peculiarities. Both the electrodes behave

enerally alike, though it appears from some observationhat one current impulse preponderates over the other,s before mentioned.

he physiological effects of the high tension discharge around to be so small that the shock of the coil can be

upported without any inconvenience, except perhaps amall burn roduced b the dischar e u on a roachin



he hand to one of the terminals. The decidedly smallerhysiological effects of these cur-

ents are thought to be due either to a differentistribution through the body or to the tissues acting as

ondensers. But in the case of an induction coil with areat many turns the harmless-ness is principally due tohe fact that but little energy is available in the externalircuit when the same is closed through thexperimenter's body, on account of the great impedancef the coil.

n varying the frequency and strenth of the currentshrough the primary of the coil, the character of theecondary discharge is greatly varied, and no less thanve distincts forms are observed :—A weak, sensitive

hread discharge, a powerful naming discharge, and thre

orms of brush or streaming discharges. Each of theseossesses certain noteworthy features, but the most

nteresting to study are the latter.

Under certain conditions the streams, which areresumably due to the violent agitation of the air

molecules, issue freely from all points of the coil, evenhrough a thick insulation. If there is the smallest airpace between the primary and secondary, they will formhere and surely injure the coil by slowly warming thensulation. As they form even with ordinary frequencies

when the potential is excessive, the air-space must be

most carefully avoided. These high frequency streamersiffer in as ect and ro erties from those roduced b a



tatic machine. The wind produced by them is small andhould altogether cease if still considerably higherrequencies could be obtained. A peculiarity is that they ssue as freely from surfaces as from points. ( hving tohis, a metallic vane, mounted in one of the terminals of

he coil so as to rotate freely, and having one of its sidesovered with insulation, is spun rapidly around. Such aane would not rotate with a steady potential, but with aigh frequency coil it will spin, even if it be entirely overed with insulation, provided the insulation on oneide be either thicker or of a higher specific inductive

apacity. A Crookes electric radiometer is also spunround when connected to one of the terminals of the coiut only at very high exhaustion or at ordinary pressure

here is still another and more striking peculiarity of suchigh frequency streamer, namely, it is hot. The heat is

asily perceptible with frequencies of about 10,000, evef the potential is not excessively high. The heating effects, of course, due to the molecular impacts and collisions.ould the frequency and potential be pushed far enough,

hen a brush could be pr«»-

uced resembling in every particular a flame and givingght and heat, jet without a chemical process taking plac

he hot brush, when properly produced, resembles a jetf burning gas escaping under great pressure, and it emin extraordinary strong smell of ozone. The great

zonizin action is ascribed to the fact that the a itation o



he molecules of the air is more violent in such a brushhan in the ordinary streamer of a static machine. But th

most powerful brush discharges were produced by mploying currents of much higher frequencies than it

was possible to obtain by means of the alternators. Thes

urrents were obtained by disruptively discharging aondenser and setting up oscillations. In this mannerurrents of a frequency of several hundred thousand werbtained.

urrents of this kind, Mr. Tesla pointed out, produce

triking effects. At these frequencies, the impedance of aopper bar is so great that a potential difference of severundred volts can be maintained between two points of ahort and thick bar, and it is possible to keep an ordinaryncandescent lamp burning at full candle power by ttaching the terminals of the lamp to two points of the

ar no more than a few inches apart, When the frequencs extremely high, nodes are found to exist on such a barnd it is easy to locate them by means of a lamp.

y converting the high tension discharges of a low requency coil in this manner, it was found practicable to

eep a few lamps burning on the ordinary circuit in theaboratory, and by bringing the undulation to a low pitch

was possible to operate small motors.

his plan likewise allows of converting high tensionischarges of one direction into low tension unidirectiona

urrents, by adjusting the circuit so that there are noscillations. In assin the oscillatin dischar es throu h



he primary of a specially constructed coil, it is easy tobtain enormous potential differences with only few turnf the secondary.

Great difficulties were at first experienced in producing a

uccessful coil on this plan. It was found necessary to keell air, or gaseous matter in general, away from theharged surfaces, and oil immersion was resorted to. The

wires used were heavily covered with gutta-percha andwound in oil, or the air was pumped out by means of a

prengel pump. The general arrangement was the

ollowing: — An ordinary induction coil, operated from aow frequency alternator, was used to charge Leyden jarhe

ars were made to discharge over a single or multiple gaphrough the primary of the second coil. To insure the

ction of the gap, the arc was blown out by a magnet orir blast. To adjust the potential in the secondary a smallil condenser was used, or polished brass spheres of ifferent sizes were screwed on the terminals and theiristance adjusted.

When the conditions were carefully determined to suitach experiment, magnificent effects were obtained. Tw

wires, stretched through the room, each being connectedo one of the terminals of the coil, emitted streams soowerful that the light from them allowed distinguishinghe objects in the room ; the wires became luminous eve

hough covered with thick and most excellent insulation.When two strai ht wires or two concentric circles of wir



re connected to the terminals, and set at the properistance, a uniform luminous sheet is produced betweenhem. It was possible in this way to cover an ana of morehan one meter square completely with the streams. By ttaching to one terminal a large circle of wire and to the

ther terminal a small sphere, the streams are focusedpon the sphere, produce a strongly lighted spot upon thame, and present the appearance of a luminous cone. A ery thin wire glued upon a plate of hard rubber of greathickness, on the opposite side of which is fastened anfoil coating, is rendered intensely luminous when the

oating is connected to the other terminal of the coil. Sucn experiment can be performed also with low frequencyurrents, but much less satisfactorily.

When the terminals of such a coil, even of a very smallne, are separated by a rubber or glass plate, the

ischarge spreads over the plate in the form of streams,hreads or brilliant sparks, and affords a magnificentisplay, which cannot be equaled by the largest coilperated in the usual ways. By a simple adjustment it isossible to produce with the coil a succession of brilliantparks, exactly as with a Holtz machine.

Under certain conditions, when the frequency of thescillation is very great, white, phantom-like streams areeen to break forth from the terminals of the coil. Thehief interesting feature about them is, that they streamreely against the outstretched hand or other conductingb ect without roducin an sensation, and the hand



may be approached very near to the terminal without apark being induced to jump. This is due presumably tohe fact that a considerable portion of the energy isarried

way or dissipated in the streamers, and the difference ootential between the terminal and the hand isiminished.

t is found in such experiments that the frequency of theibration and the quickness of succession of the sparks

etween the knobs affect to a marked degree theppearance of the streams. When the frequency is very ow, the air gives way in more or less the same manner ay a steady difference of potential, and the streamsonsist of distinct threads, generally mingled with thinparks, which probably correspond to the successive

ischarges occurring between the knobs. But when therequency is very high, and the arc of the dischargeroduces a sound which is loud and smooth (which

ndicates both that oscillation takes place and that theparks succeed each other with great rapidity), then theuminous streams formed are perfectly uniform. They ar

enerally of a purplish hue, but when the molecularibration is increased by raising the potential, they ssume a white color.

he luminous intensity of the streams increases rapidly when the potential is increased; and with frequencies of

nly a few hundred thousand, could the coil be made towithstand a sufficientl hi h otential difference there is



o doubt that the space around a wire could be made tomit a strong light, merely by the agitation of the

molecules of the air at ordinary pressure.

uch discharges of very high frequency which render

uminous the air at ordinary pressure we have very likelccasion to witness in the aurora borealis. From many ofhese experiments it seems reasonable to infer thatudden cosmic disturbances, such as eruptions on the suet the electrostatic charge of the earth in an extremely apid vibration, and produce the glow by the violent

gitation of the air in the upper and even in the lowertrata. It is thought that if the frequency were low? orven more so if the charge were not at all vibrating, theower dense strata would break down as in a lightningischarge. Indications of such breaking down have beenepeatedly observed, but they can be attributed to the

undamental disturbances, which are few in number, forhe superimposed vibration would be so rapid as not tollow a disruptive break.

he study of these discharge phenomena has led Mr.esla to the recognition of some important facts. It was

ound, as already stated, that uascous matter must bemost carefully excluded from

ny dielectric which is subjected to great, rapidly changinlectrostatic stresses. Since it is difficult to exclude the gerfectly when solid insulators are used, it is necessary t

esort to liquid dielectrics. When a solid dielectric is usedmatters little how thick and how ood it is if air be



resent, streamers form, which gradually heat theielectric and impair its insulating power, and theischarge finally breaks through. Under ordinary onditions the best insulators are those which possess thighest specific inductive capacity , but such insulators ar

ot the best to employ when working with these highrequency currents, for in most cases the higher specificnductive capacity is rather a disadvantage. The primeuality of the insulating medium for these currents isontinuity. For this reason principally it is necessary tomploy liquid insulators, such as oils. If two metal plates

onnected to the terminals of the coil, are immersed in oind set a distance apart, the coil may be kept working fony length of time without a break occurring, or withouthe oil being warmed, but if air bubbles are introduced,hey become luminous; the air molecules, by their impacgainst the oil, heat it, and after some time cause the

nsulation to give way. If, instead of the oil, a solid plate ohe best dielectric, even several times thicker than the ointervening between the metal plates, is inserted betweehe latter, the air having free access to the chargedurfaces, the dielectric ivariably is warmed and breaksown.

he employment of oil is advisable or necessary evenwith low frequencies, if the potentials are such thattreamers form, but only in such cases, as is evident fromhe theory of the action. If the potentials are so low thattreamers do not form, then it is even disadvantageous tm lo oil for it ma rinci all b confinin the heat b



he cause of the breaking down of the insulation.

he exclusion of gaseous matter is not only desirable onccount of the safety of the apparatus, but also on accounf economy, especially in a condenser, in which

onsiderable waste of power may occur merely owing tohe presence of air, if the electric density on the chargedurfaces is great.

n the course of these investigations a phenomenon of pecial scientific interest was observed. It may be ranked

mong the brush phenomena, in fact it is a kind of brushwhich forms at, or near, a single terminal in high vacuumn a bulb with a con-

ucting electrode, even if the latter be of aluminum, therush has only a very short existence, but it can be

reserved for a considerable length of time in a bulbevoid of any conducting electrode. To observe thehenomenon it is found best to employ a large sphericalulb having in its centre a small bulb supported on a tubealed to the neck of the former. The large bulb beingxhausted to a high degree, and the inside of the small

ulb being connected to one of the terminals of the coil,nder certain conditions there appears a misty hazeround the small bulb, which, after passing through somtages, assumes the form of a brush, generally at rightngles to the tube supporting the small bulb. When therush assumes this form it may be brought to a state of

xtreme sensitiveness to electrostatic and magneticnfluence. The bulb han in strai ht down and all ob ect



eing remote from it, the approach of the observer withifew paces will cause the brush to fly to the opposite sid

nd if he walks around the bulb it will always keep on thepposite side. It may begin to spin around the terminal

ong before it reaches that sensitive stage. When it begin

o turn around, principally, but also before, it is affectedy a magnet, and at a certain stage it is susceptible tomagnetic influence to an astonishing degree. A small

ermanent magnet, with its poles at a distance of no morhan two centimetres will affect it visibly at a distance ofwo metres, slowing down or accelerating the rotation

ccording to how it is held relatively to the brush.

When the bulb hangs with the globe down, the rotation islways clockwise. In the southern hemisphere it wouldccur in the opposite direction, and on the (magnetic)quator the brush should not turn at all. The rotation ma

e reversed by a magnet kept at some distance. Therush rotates best, seemingly, when it is at right angles the lines of force of the earth. It, very likely rotates, whet its maximum speed, in synchronism with thelternations, say, 10,000 times a second. The rotation cae slowed down or accelerated by the approach or

ecession of the observer, or any conducting body, but itannot be reversed by putting the bulb in any position.

Very curious experiments may be performed with therush when in its most sensitive state. For instance, therush resting in one position, the experimenter may, by electing a proper position, approach the hand at a certaionsiderable distance to the bulb and he ma c uisi' the



rush to pass oft bv merely stiffening the muscles of

he arm, the mere change of configuration of the arm andhe consequent imperceptible displacement beingufficient to disturb the delicate balance. When it begins

o rotate slowly, and tin-hands are held at a properistance, it is impossible to make even the slightestmotion without producing a visible effect upon the brushA metal plate connected to the other terminal of the coil

ffects it at a great distance, slowing down the rotationften to one turn a second.

Mr. Tesla hopes that this phenomenon will prove aaluable aid in the investigation of the nature of the forcecting in an electrostatic or magnetic field. If there is any

motion which is measurable going on in the space, such arush would be apt to reveal it. It is, so to speak, a beam

f light, frictionless, devoid of inertia. On account of itsmarvellous sensitiveness to electrostatic or magnetic

isturbances it may be the means of sending signalshrough submarine cables with any speed, and even of ransmitting intelligence to a .distance without wires.

n operating an induction coil with these rapidly lternating currents, it is astonishing to note, for the firstme, the great importance of the relation of capacity, sel

nduction, and frequency as bearing upon the generalesult. The combined effect of these elements produces

many curious effects. For instance. two metal plates are

onnected to the terminals and set at a small distance, sohat an arc is formed between them. This arc > v-vent



strong current from flowing through the coil. If the art-e interrupted by the interposition of a glass plate, theapacity of the condenser obtained counteracts the self-nduction, and a stronger current is made to pass. Theffects of capacity are the most striking, for in these

xperiments, since the self-induction and frequency bothre high, the critical capacity is very small, and need beut slightly varied to produce a very considerable changhe experimenter brings his body in contact with the

erminals of the secondary of the coil, or attaches to oner both terminals insulated bodies of very small bulk,

uch as exhausted bulbs, and he produces a considerableise or fall of potential on the secondary, and greatly ffects the flow of the current through the primary coil.

n many of the phenomena observed, the presence of thir, or, generally speaking, of a medium of a gaseous

ature (using this term not to imply specific properties,ut in contradistinction to homogeneity or perfectontinuity) plays an important part.

s it allows energy to be dissipated by molecular impactr bombardment. The action is thus explained:—When a

nsulated body connected to a terminal of the coil isuddenly charged to high potential, it acts inductively pon the surrounding air, or whatever gaseous mediumhere might be. The molecules or atoms which are near ire, of course, more attracted, and move through areater distance than the further ones. When the neares

molecules strike the bod the are re elled, and collision



ccur at all distances within the inductive distance. It isow clear that, if the potential be steady, bat little loss ofnergy can be caused in this way, for the molecules whicre nearest to the body having had an additional chargemparted to them by contact, are not attracted until they

ave parted, if not with all, at least with most of thedditional charge, which can be accomplished only after areat many collisions. This is inferred from the fact that

with a steady potential there is but little loss in dry air.When the potential, instead of being steady, is alternatinhe conditions are entirely different. In this case a

hythmical bombardment occurs, no matter whether themolecules after coming in contact with the body lose themparted charge or not, and, what is more, if the charge iot lost, the impacts are all the more violent. Still, if the

requency of the impulses be very small, the loss causedy the impacts and collisions would not be serious unless

he potential was excessive. But when extremely highrequencies and more or less high potentials are used, thoss may be very great, The total energy lost per unit of me is proportionate to the product of the number of

mpacts per second, or the frequency and the energy lostn each impact. But the energy of an impact must be

roportionate to the square of the electric density of theody, on the assumption that the charge imparted to the

molecule is proportionate to that density. It is concludedrom this that the total energy lost must be proportionato the product of the frequency and the square of thelectric density; but this law needs experimentalonfirmation. Assuming the preceding considerations to



e true, then, by rapidly alternating the potential of aody immersed in an insulating gaseous medium, any mount of energy may be dissipated into space. Most of hat energy, then, is not dissipated in the form of longther waves, propagated to considerable distance, as is

hought most generally, but is consumed in impact andollisional losses— that is, heat vibrations — on theurface and in

he vicinity of the body. To reduce the dissipation it isecessary to work with a small electric density — the

maller, the higher the frequency.

he behavior of a gaseous medium to such rapidlternations of potential makes it appear plausible thatlectrostatic disturbances of the earth, produced by osmic events, may have great influence upon the

meteorological condition^. When such disturbances occuoth the frequency of the vibrations of the charge and thotential are in all probability excessive, and the energy onverted into heat may be considerable. Since theensity must be unevenly distributed, either inonsequence of the irregularity of the earth's surface, or

n account of the condition of the atmosphere in variouslaces, the effect produced would accordingly vary fromlace to place. Considerable variations in the temperaturnd pressure of the atmosphere may in this manner beaused at any point of the surface of the earth. Theariations may be gradual or very sudden, according to

he nature of the original disturbance, and may produce



ain and storms, or locally modify the weather in any wa

rom many experiences gathered in the course of thesenvestigations it appears certain that in lightningischarges the air is an element of importance. For

nstance, during a storm a stream may form on a nail orointed projection of a building. If lightning strikesomewhere in the neighborhood*the harmless staticischarge may, in consequence of the oscillations set up,ssume the character of a high-frequency streamer, andhe nail or projection may be brought to a high

emperature by the violent impact of the air molecules.hus, it is thought, a building may be set on fire withouthe lightning striking it. In like manner small metallicbjects may be fused and volatilized —as frequently ccurs in lightning discharges— merely because they areurrounded by air. Were they immersed in a practically

ontinuous medium, such as oil, they would probably beafe, as the energy would have to spend itself elsewhere.

An instructive experience having a bearing on this subjes the following:— A glass tube of an inch or so in diametend several inches long is taken, and a platnium wire

ealed into it, the wire running through the center of theube from end to end. The tube is exhausted to a

moderate degree. If a steady current is passed throughhe wire it is heated uniformly in all parts and the gas inhe tube is of no consequence. But if high

requency discharges are directed through the wire, it iseated more on the ends than in the middle ortion and



he frequency, or rate of charge, is high enough, the wiremight as well be cut in the middle as not, for most of the

eating on the ends is due to the rarefied gas. Here theas might only act as a conductor of no impedance,iverting the current from the wire as the impedance of

he latter is enormously increased, and merely heatinghe ends of the wire by reason of their resistance to theassage of the discharge. But it is not at all necessary thahe gas in the tube should he conducting ; it might be atn extremely low pressure, still the ends of the wire

would be heated ; however, as is ascertained by

xperience, only the two ends would in such case not belectrically connected through the gaseous medium. Now

what with these frequencies and potentials occurs in anxhausted tube, occurs in the lightning discharge atrdinary pressure.

rom the facility with which any amount of energy may e carried off through a gas, Mr. Tesla infers that the be

w T ay to render harmless a lightning discharge is tofford it in some way a passage through a volume of gas.

he recognition of some of the above facts has a bearing

pon far-reaching scientific investigations in whichxtremely high frequencies and potentials are used. Inuch cases the air is an important factor to be consideredo, for instance, if two wires are attached to the terminalf the coil, and the streamers issue from' them, there isissipation of energy in the form of heat and light, and th

wires behave like a condenser of lar er ca acit . If the



wires be immersed in oil, the dissipation of energy isrevented, or at least reduced, and the apparent capacit

s diminished. The action of the air would seem to make iery difficult to tell, from the measured or computedapacity of a condenser in which the air is acted upon, its

ctual capacity or vibration period, especially if theondenser is of very small surface and is charged to aery high potential. As many important results areependant upon the correctness of the estimation of theibration period, this subject demands the most carefulcrutiny of investigators.

n Leyden jars the loss due to the presence of air isomparatively small, principally on account of the greaturface of the coatings and the small external action, buthere are streamers on the top, the loss may beonsiderable, and the period of vibra-

on is affected. In a resonator, the density is small, buthe frequency is extreme, and may introduce aonsiderable error. It appears certain, at any rate, thathe periods of vibration of a charged body in a gaseousnd in a continuous medium, such as oil, are different, on

ccount of the action of the former, as explained.

Another fact recognized, which is of some consequence, ihat in similar investigations the general considerations otatic screening are not applicable when a gaseous

medium is present. This is evident from the following

xperiment:—A short and wide glass tube is taken andovered with a substantial coatin of bronze owder



arely allowing the light to shine a little through. The tubs highly exhausted and suspended on a metallic clasprom the end of a wire. When the wire is connected withne of the terminals of the coil, the gas inside of the tubes lighted in spite of the metal coating. Here the metal

vidently does not screen the gas inside as it ought to,ven if it be very thin and poorly conducting. Yet, in aondition of rest the metal coating, however thin, screenhe inside perfectly.

One of the most interesting results arrived at in pursuing

hese experiments, is the demonstration of the fact that aseous medium, upon which vibration is impressed by apid changes of electrostatic potential, is rigid. Inlustration of this result an experiment made by Mr.esla may by cited :—A glass tube about one inch iniameter and three feet long, with outside condenser

oatings on the ends, was exhausted to a certain point,when, the tube being suspended freely from a wireonnecting the upper coating to one of the terminals of thoil, the discharge appeared in the form of a luminoushread passing through the axis of the tube. Usually thehread was sharply defined in the upper part of the tube

nd lost itself in the lower part. When a magnet or thenger was quickly passed near the upper part of the

uminous thread, it was brought out of position by magnetic or electrostatic influence, and a transversal

ibration like that of a suspended cord, with one or moreistinct nodes, was set up, which lasted for a few minutend raduall died out. B sus endin from the lower



ondenser coating metal plates of different sizes, thepeed of the vibration was varied. This vibration wouldeem to show beyond doubt that the thread possessedigidity, at least to transversal displacements.

Many experiments were tried to demonstrate thisroperty in

ir at ordinary pressure. Though no positive evidence haeen obtained, it is thought, nevertheless, that a highrequency brush or streamer, if the frequency could be

ushed far enough, would be decidedly rigid. A smallphere might then be moved within it quite freely, but ifn-own against it the sphere would rebound. An ordinarame cannot possess rigidity to a marked degree becaus

he vibration is directionless; but an electric arc, it iselieved, must possess that property more or less. A

uminous band excited in a bulb by repeated discharges oLeyden jar must also possess rigidity, and if deformed

nd suddenly released should vibrate.

rom like considerations other conclusions of interest areadied. The most probable medium filling the space is

ne consisting of independent carriers immersed in annsulating fluid. If through' this medium enormouslectrostatic stresses are assumed to act, which vary apidly in intensity, it would allow the motion of a body hrough it, yet it would be rigid and elastic, although theuid itself might be devoid of these properties.

urthermore, on the assumption that the independentarriers are of an confi uration such that the fluid



esistance to motion in one direction is greater than innother, a stress of that nature would cause the carrierso arrange themselves in groups, since they would turn tach other their sides of the greatest electric density , in

which position the fluid resistance to approach would be

maller than to receding. If in a medium of the aboveharacteristics a brush would be formed by a steady otential, an exchange of the carriers would go onontinually, and there would be less carriers per unit of olume in the brush than in the space at some distance

rom the electrode, this corresponding to rarefaction. If

he potential were rapidly changing, the result would beery different; the higher the freqency of the pulses, thelower would be the exchange of the carriers ; finally, the

motion of translation through measurable space wouldease, and, with a sufficiently high frequency andntensity of the stress, the carriers would be drawn

owards the electrode, and compression would result.

An interesting feature of these high frequency currents ihat they allow of operating all kinds of devices by onnecting the device with only one leading wire to thelectric source. In fact, under certain conditions it may b

more economical to supply the electrical energy witli oneead than with two.

An experiment of special interest shown by Mr. Tesla, ishe running, by the use of only one insulated line, of a

motor operating on the principle of the rotating magnetic

eld enunciated b Mr. Tesla. A sim le form of such a



motor is obtained by winding upon a laminated iron corerimary and close to it a secondary coil, closing the endsf the latter and placing a freely movable metal disc

within the influence of the moving field. The secondary oil may, however, be omitted. When one of the ends of

he primary coil of the motor is connected to one of theerminals of the high frequency coil arid the other end ton insulated metal plate, which, it should be stated, is nobsolutely necessary for the success of the experiment,he disc is set in rotation.

xperiments of this kind seem to bring it withinossibility to operate a motor at any point of the earth'surface from a central source, without any connection tohe same except through the earth. If, by means of owerful machinery, rapid variations of the earth'sotential were produced, a grounded wire reaching up to

ome height would be traversed by a current which coule increased by connecting the free end of the wire to aody of some size. The current might be converted to lowension and used to operate a motor or other device. Thexperiment, which would be one of great scientificnterest, would probably best succeed on a ship at sea. In

his manner, even if it were not possible to operatemachinery, intelligence might be transmitted quiteertainly.

n the course of this experimental study special attentionwas devoted to the heating effects produced by these

urrents, which are not only striking, but open up the



ossibility of producing a more efficient illumiuant. It isufficient to attach to the coil terminal a thin wire orlament, to have the temperature of the lattererceptibly raised. If the wire or filament be enclosed in ulb, the heating effect is increased by preventing the

irculation of the air. If the air in the bulb be strongly ompressed, the displacements are smaller, the impactsess violent, and the heating effect is diminished. On theontrary, if the air in the bulb be exhausted, an inclosedamp filament is brought to incandescence, and any mount of light may thus be produced.

he heating of the inclosed lamp filament depends on somany things of a different nature, that it is difficult to giv

generally applicable rule under which the maximumeating

ccurs. As regards the size of the bull), it is ascertainedhat at ordinary or only slightly differing atmosphericressures, when air is a good insulator, the filament iseated more in a small bulb, because of the betteronfinement of heat in this case. At lower pressures, wheir becomes conducting, the heating effect is greater in a

arge bull), but at excessively high degrees of exhaustionhere seems to be, beyond a certain and rather small sizef the vessel, no perceptible difference in the heating.

he shape of the vessel is also of some importance, and ias been found of advantage for reasons of economy to

mploy a spherical bulb with the electrode mounted in itentre where the reboundin molecules collide.



t is desirable on account of economy that all the energy upplied to the bulb from the source should reach withouoss the body to be heated. The loss in conveying thenergy from the source to the body may be reduced by

mploying thin wires heavily coated with insulation, andy the use of electrostatic screens. It is to be remarked,hat the screen, cannot be connected to the ground asnder ordinary conditions.

n the bulb itself a large portion of the energy' supplied

may be lost by molecular bombardment against the wireonnecting the body to be heated with the source.onsiderable improvement was effected by covering thelass stem containing the wire with a closely fittingonducting tube. This tube is made to project a littlebove the glass, and prevents the cracking of the latter

ear the heated body. The effectiveness of the conductinube is limited to very high degrees of exhaustion. Itiminishes the energy lost in bombardment for twoeasons; first, the charge given up by the atoms spreadsver a greater area, and hence the electric density at anyoint is small, and the atoms are repelled with less energ

han if they would strike against a good insulator;econdly, as the tube is electrified by the atoms whichrst come in contact with it, the progress of the followingtoms against the tube is more or less checked by theepulsion which the electrified tube must exert upon theimilarly electrified atoms. This, it is thought, explains

wh the dischar e throu h a bulb is established with



much greater facility when an insulator, than when aonductor, is present.

During the investigations a great many bulbs of differentonstruction, with electrodes of different material, were

xperimented upon, and a number of observations of nterest were made. Mr.

esla has found tlmt the deterioration of the electrode ishe less, the higher the frequency. This was to bexpected, as then the heating is effected by many small

mpacts, instead by fewer and more violent ones, whichuickly shatter the structure. The deterioration is alsomaller when the vibration is harmonic. Thus anlectrode, maintained at a certain degree of heat, lasts

much longer with currents obtained from an alternator,han with those obtained by means of a disruptive

ischarge. One of the most durable electrodes wasbtained from strongly compressed carborundum, whichs a kind of carbon recently produced by Mr. E. G.

Acheson, of Monongahela City , Pa. From experience, it isnferred, that to be most durable, the electrode should ben the form of a sphere with a highly polished surface.

n some bulbs refractory bodies were mounted in aarbon cup and put under the molecular impact. It wasbserved in such experiments that the carbon cup waseated at first, until a higher temperature was reached;hen most of the bombardment was directed against the

efractory body, and the carbon was relieved. In generalwhen different bodies were mounted in the bulb the



ardest fusible would be relieved, and would remain at aonsiderably lower temperature. This was necessitatedy the fact that most of the energy supplied would find it

way through the body \vhioh was more easily fused orevaporated."

uriously enough it appeared in some of the experimentmade, that a body was fused in a bulb under themolecular impact by evolution" of less light than whenused by the application of heat in ordinary ways. This

may be ascribed to a loosening of the structure of the

ody under the violent impacts and changing stresses.

ome experiments seem to indicate that under certainonditions a body, conducting or nonconducting, may,

when bombarded, emit light, which to all appearances isue to phosphorescence, but may in reality be caused by

he incandescence of an infinitesimal layer, the meanemperature of the body being comparatively small. Suc

might be the case if each single rhythmical impact wereapable of instantaneously exciting the retina, and thehythm were just high enough to cause a continuousmpression in the eye. According to this view, a coil

perated by disruptive discharge would be eminently dapted to produce such a result, and it is found by xperience that its power of

xciting phosphorescence is extraordinarily great. It isapable of exciting phosphorescence at comparatively low

egrees of exhaustion, and also projects shadows atressures far reater than those at which the mean free



ath is comparable to the dimensions of the vessel. Theatter observation is of some importance, inasmuch as it

may modify the generally accepted views in regard to thradiant state" phenomena.

A thought which early and naturally suggested itself toVI r. Tesla, was to utilize the great inductive effects of igh frequency currents to produce light in a sealed glassessel without the use of leading in wires. Accordingly,

many bulbs were constructed in which the energy ecessary to maintain a button or filament at high

ncandescence, was supplied through the glass by eitherlectrostatic or electrodynamic induction. It was easy toegulate the intensity of the light emitted by means of anxternally applied condenser coating connected to annsulated plate, or simply by means of a plate attached tohe bulb which at the same time performed the function

f a shade.

A subject of experiment, which has been exhaustively reated in England by Prof. J. J. Thomson, has beenollowed up independently by Mr. Tesla from theeginning of this study, namely, to excite by

lectrodynamic induction a luminous band in a closed tubr bulb. In observing the behavior of gases, and the

uminous phenomena obtained, the importance of thelectrostatic effects was noted and it appeared desirableo produce enormous potential differences, alternating

with extreme rapidity. Experiments in this direction ledo some of the most interestin results arrived at in the



ourse of these investigations. It was found that by rapidlternations of a high electrostatic potential, exhaustedubes could be lighted at considerable distances from aonductor connected to a properly constructed coil, andhat it was practicable to establish with the coil an

lternating electrostatic field, acting through the wholeoom and lighting a tube wherever it was placed withinhe four walls. Phosphorescent bulbs may be excited inuch a field, and it is easy to regulate the effect by onnecting to the bulb a small insulated metal plate. It

was likewise possible to maintain a filament or button

mounted in a tube at bright incandescence, and, in onexperiment, a mica vane was spun by the incandescencef a platinum wire.

oming now to the lecture delivered in Philadelphia andt.

ouis, it may be remarked that to the superficial reader,Mr. Tesla's introduction, dealing with the importance of he eye, might appear as a digression, but the thoughtfuleader will find therein much food for meditation andpeculation. Throughout his discourse one can trace Mr.

esla's effort to present in a popular way thoughts andiews on the electrical phenomena which have in recentears captivated the scientific world, but of which theeneral public has even yet merely received an inkling.

Mr. Tesla also dwells rather extensively on his well-nown method of high-frequency conversion; and the

arge amount of detail information will be gratefully



eceived by students and experimenters in this virgineld. The employment of apt analogies in explaining the

undamental principles involved makes it easy for all toain a clear idea of their nature. Again, the ease with

which, thanks to Mr. Tesla's efforts, these high-frequenc

urrents may now be obtained from circuits carryinglmost any kind of current, cannot fail to result in anxtensive broadening of this field of research, which offeo many possibilities. M r. Tesla, true philosopher as he ioes not hesitate to point out defects in some of his

methods, and indicates the lines which t<> him seem the

most promising. Particular stress is laid by him upon themployment of a medium in which the dischargelectrodes should be immersed in order that this methodf conversion may be brought to the highest perfection.

He has evidently taken pains to give as much usefulnformation as possible to those who wish to follow in his

ath, as he shows in detail the circuit arrangements to bedopted in all ordinary cases met with in practice, andlthough some of these methods were described by himwo years before, the additional information is still timelynd welcome.

n his experiments he dwells first on some phenomenaroduced by electrostatic force, which he considers in theght of modern theories to be the most important force iature for us to investigate. At the very outset he showstrikingly novel experiment illustrating the effect of aapidly varying electrostatic force in a gaseous medium,y touching with one hand one of the terminals of a



00,000 volt transformer and bringing tin-other hand the opposite terminal. The powerful streamers whichssued from his hand and astonished his audiences formecapital illustration of some of the views advanced, and

fforded Mr. Tesla an opportunity of pointing out the tru

easons why.

with these currents, such an amount of energy can beassed through the body with impunity. He then showedy experiment the difference between a steady and aapidly varying force upon the dielectric. This difference

s most strikingly illustrated in the experiment in which ulb attached to the end of a wire in connection with onef the terminals of the transformer is ruptured, althoughll extraneous bodies are remote from the bulb. He nextlustrates how mechanical motions are produced by aarying electrostatic force acting through a gaseous

medium. The importance of the action of the air isarticularly illustrated by an interesting experiment.

aking up another class of phenomena, namely, those ofynamic electricity, Mr. Tesla produced in a number of xperiments a variety of effects by the employment of

nly a single wire with the evident intent of impressingpon his audience the idea that electric vibration orurrent can be transmitted witli ease, without any returnircuit ; also how currents so transmitted can beonverted and used for many practical purposes. A umber of experiments are then shown, illustrating the

ffects of frequency, self-induction and capacity; then a



umber of ways of operating motive and other devices bhe use of a single lead. A number of novel impedancehenomena are also shown which cannot fail to arouse

nterest.

Mr. Tesla next dwelt upon a subject which he thinks of reat importance, that is, electrical resonance, which hexplained in a popular way. He expressed his firmonviction that by observing proper conditions,ntelligence, and possibly even power, can be transmittedhrough the medium or through the earth; and he

onsiders this problem worthy of serious and immediateonsideration.

oming now to the light phenomena in particular, lielustrated the four distinct kinds of these phenomena inn original way, which to many must have been a

evelation. Mr. Tesla attributes these light effects tomolecular or atomic impacts produced by a varyinglectrostatic stress in a gaseous medium. Fie illustrated iseries of novel experiments the effect of the gas

urrounding the conductor and shows beyond a doubthat with high frequency and high potential currents, the

urrounding gas is of paramount importance in theeating of the conductor. He attributes the heatingartially to a conduction current and partially toombardment, and demonstrates that in manv cases the

eating may be practically due to the bombardment

lone. He pointed out also that the skin effect is largely modi lied b the resence of the as or of an atomic



medium in general. He showed also some interestingxperiments in which the effect of convection islustrated. Probably one of the most curious experiment

n this connection is that in which a thin platinum wiretretched along the axis of an exhausted tube is brought

o incandescence at certain points corresponding to theosition of the striae, while at others it remains dark. Thxperiment throws an interesting light upon the nature ohe strife and may lead to important revelations.

Mr. Tesla also demonstrated the dissipation of energy

hrough an atomic medium and dwelt upon the behaviorf vacuous space in conveying heat, and in this connectiohowed the curious behavior of an electrode stream, from

which he concludes that the molecules of a gas probably annot be acted upon directly at measurable distances.

Mr. Tesla summarized the chief results arrived at inursuing his investigations in a manner which will serves a valuable guide to all who may engage in this work.erhaps most interest will centre on his generaltatements regarding the phenomena of hosphorescence, the most important fact revealed in th

irection being that when exciting a phosphorescent bulbcertain definite potential gives the most economical

esult.

he lectures will now be presented in the order of theirate of delivery.

HAPTER XXVI.



XPEEIMENTS WlTH ALTERNATE CURRENTS OFVERY HlGH FREQUENCY AND THEIR APPLICATION

O METHODS OF ARTIFICIAL ILLUMINATION. J

HERE is no subject more captivating, more worthy of tudy, than nature. To understand this great mechanismo discover the forces which are active, and the laws

which govern them, is the highest aim of the intellect of man.

Nature has stored up in the universe infinite energy. Theternal recipient and transmitter of this infinite energy ihe ether. The recognition of the existence of ether, and he functions it performs, is one of the most importantesults of modern scientific research. The merebandoning of the idea of action at a distance, the

ssumption of a medium pervading all space andonnecting all gross matter, has freed the minds of hinkers of an ever present doubt, and, by opening a neworizon — new and unforeseen possibilities — has given

resh interest to phenomena witli which we are familiar old. It has been a great step towards the understanding

he forces of nature and their multifold manifestations tour senses. It has been for the enlightened student of hysics what the understanding of the mechanism of therearm or of the steam engine is for the barbarian.henomena upon which we used to look as wondersaffling explanation, we now see in a different light. The

park of an induction coil, the glow of an incandescentam the manifestations of the mechanical forces of



urrents and magnets are no longer beyond our grasp;nstead of the incomprehensible, as before, theirbservation suggests now in our minds a simple

mechanism, and although as to its precise nature all is stionjecture, yet we know that the truth cannot be much

onger hidden, and instinctively we feel that thenderstanding is dawning upon us. We still admire theseeautiful phenomena, these

A lecture delivered before the American Institute of lectrical Engineers, at Columbia College, N. Y., May 20,

891.

trange forces, but we are helpless no longer ; we can in aertain measure explain them, account for them, and were hopeful of finally succeeding in unraveling the

mystery which surrounds them.

ri how far we can understand the world around us is theltimate thought of every student of nature. Theoarseness of our senses prevents us from recognizing thlterior construction of matter, and astronomy, thisrandest and most positive of natural sciences, can only

each us something that happens, as it were, in ourmmediate neighborhood; of the remoter portions of theoundless universe, with its numberless stars and suns,

we know nothing. But far beyond the limit of perception ur senses the spirit still can guide us, and so we may ope that even these unknown worlds — infinitely small

nd great — may in a measure become known to us. Stillven if this knowled e should reacli us the searchin



mind will find a barrier, perhaps forever unsurpassable,o the true recognition of that which seems to be, the

mere appearcmce of which is the only and slender basis oll our philosophy.

Of all the forms of nature's immeasurable, all-pervadingnergy, which ever and ever changing and moving, like aoul animates the inert universe, electricity and

magnetism are perhaps the most fascinating. The effectsf gravitation, of heat and light we observe daily, and soo

we get accustomed to them, and soon they lose for us the

haracter of the marvelous and wonderful; but electricitynd magnetism, with their singular relationship, withheir seemingly dual character, unique-among the forcesn nature, with their phenomena of attractions, repulsionnd rotations, strange manifestations of mysteriousgents, stimulate and excite the mind to thought and

esearch. "What is electricity, and what is magnetism ?hese questions have been asked again and again. The

most able intellects have ceaselessly wrestled with theroblem ; still the question has not as yet been fully nswered. But while we cannot even to-day state whathese singular forces are, we have made good headway

owards the solution of the problem. We are now onfident that electric and magnetic phenomena arettributable to ether, and we are perhaps justified inaying that the effects of static electricity are effects of ther under strain, and those of dynamic electricity andlectro-magnetism effects of ether in motion. But this stieaves the question, as to what electricity and magnetism



rc, unanswered.

irst, we naturally inquire, What is electricity, and ishere such a thing as electricity ? In interpreting electrichenomena, we may speak of electricity or of an electric

ondition, state or effect. If we speak of electric effects wmust distinguish two such effects, opposite in characternd neutralizing each other, as observation shows thatwo such opposite effects exist. This is unavoidable, for inmedium of the properties, of ether, we cannot possibly

xert a strain, or produce a displacement or motion of an

ind, without causing in the surrounding medium anquivalent and opposite effect. But if we speak of lectricity, meaning a thing, we must, I think, abandonhe idea of two electricities, as the existence of two suchhings is highly improbable. For how can we imagine thathere should be two things, equivalent in amount, alike in

heir properties, but of opposite character, botli clinging matter, both attracting and completely neutralizing each

ther? Such an assumption, though suggested by many henomena, though most convenient for explaining themas little to commend it. If there is such a thing aslectricity, there can be only one such thing, and, excess

nd want of that one thing, possibly; but more probably s condition determines the positive and negativeharacter. The old theory of Franklin, though falling shorn some respects, is, from a certain point of view, after alhe most plausible one. Still, in spite of this, the theory ofhe two electricities is generally accepted, as it apparentlxplains electric phenomena in a more satisfac-tor



manner. But a theory which better explains the facts isot necessarily true. Ingenious minds will invent theorie

o suit observation, and almost every independenthinker has his own views on the subject.

t is not with the object of advancing an opinion, but withhe desire of acquainting you better with some of theesults, which I will describe, to show you the reasoning ave followed, the departures I have made — that Ienture to express, in a few words, the views andonvictions which have led me to these results.



adhere to the idea that there is a thing which we haveeen in the habit of calling electricity . The question is,

What is that thing? or, What, of all tilings, the existence owhich we know, have we the best reason to call electricit

We know that it acts like an incompressible fluid ; thathere must be a constant quantity of it in nature; that itan be neither produced nor destroyed;

nd, what is more important, the electro-magnetic theorf light and all facts observed teach us that electric and

ther phenomena are identical. The idea at once suggestself, therefore, that electricity might be called ether. Inact, this view has in a certain sense been advanced by Dodge. His interesting work has been read by everyonend many have been convinced by his arguments. Hisreat ability and the interesting nature of the subject,

eep the reader spellbound; but when the impressionsade, one realizes that he has to deal only with ingeniousxplanations. I must confess, that I cannot believe in twolectricities, much less in a doubly-constituted ether. Thuzzling behavior of the ether as a solid to waves of lightnd heat, and as a fluid to the motion of bodies through i

s certainly explained in the most natural and satisfactormanner by assuming it to be in motion, as Sir William

homson has suggested; but regardless of this, there isothing which would enable us to conclude with certainty

hat, while a fluid is not capable of transmittingransverse vibrations of a few hundred or thousand per

econd, it might not be capable of transmitting such



ibrations when they range into hundreds of millionmillions per second. Nor can anyone prove that there areransverse ether waves emitted from an alternateurrent machine, giving a small number of alternationser second; to such slow disturbances, the ether, if at res

may behave as a true fluid.

Returning to the subject, and bearing in mind that thexistence of two electricities is, to say the least, highly mprobable, we must remember, that we have novidence of electricity, nor can we hope to get it, unless

ross matter is present. Electricity, therefore, cannot bealled ether in the broad sense of the term ; but nothingwould seem to stand in the way of calling electricity ethe

ssociated with matter, or bound ether; or, in otherwords, that the so-called static charge of the molecule isther associated in some way with the molecule. Looking

t it in that light, we would be justified in saying, thatlectricity is concerned in all molecular actions.

Now, precisely what the ether surrounding the molecules, wherein it differs from ether in general, can only beonjectured. It cannot differ in density, ether being

ncompressible ; it must, therefore, be under some strainr in motion, and the latter is the most probable. Tonderstand its functions, it would be necessary to have axact idea of the physical con-

' •

truction of matter, of which, of course, we can only form



mental picture.

ut of all the views on nature, the one which assumes onmatter and one force, and a perfect uniformity hroughout, is the most scientific and most likely to be

rue. An infinitesimal world, with the molecules and theirtoms spinning and moving in orbits, in much the samemanner as celestial bodies, carrying with them and

robably spinning with them ether, or in other words,arrying with them static charges, seems to my mind the

most probable view, and one which, in a plausible manne

ccounts for most of the phenomena observed. Thepinning of the molecules and their ether sets up thether tensions or electrostatic strains; the equalization ofther tensions sets up ether motions or electric currentsnd the orbital movements produce the effects of electrond permanent magnetism.

About fifteen years ago, Prof. Rowland demonstrated amost interesting and important fact, namely, that a statiharge carried around produces the effects of an electricurrent. Leaving out of consideration the precise nature he mechanism, which produces the attraction and

epulsion of currents, and conceiving the electrostaticallyharged molecules in motion, this experimental fact gives a fair idea of magnetism. We can conceive lines orubes of force which physically exist, being formed of rowf directed moving molecules ; we can see that these line

must be closed, that they must tend to shorten and

xpand, etc. It likewise explains in a reasonable way, the



most puzzling phenomenon of all, permanent magnetismnd, in general, lias all the beauties of the Ampere theory

without possessing the vital defect of the same, namely,he assumption of molecular currents. Without enlargingurther upon the subject, 1 would say, that I look upon al

lectrostatic, current and magnetic phenomena as beingue to electrostatic molecular forces.

he preceding remarks I have deemed necessary to a funderstanding of the subject as it presents itself to my

mind.

Of all these phenomena the most important to study arehe current phenomena, on account of the already xtensive and evergrowing use of currents for industrialurposes. It is now a century since the first practicalource of current was produced, and, ever since, the

henomena which accompany the flow of currents haveeen diligently studied, and through the untiring efforts cientific men the simple laws which govern them have

een discovered. But these laws are found to hold goodnly when the currents are of a steady character. When

he currents are rapidly varying in strength, quiteifferent phenomena, often unexpected, presenthemselves, and quite different laws hold good, whichven now have not been determined as fully as isesirable, though through the work, principally, of Engliscientists, enough knowledge has been gained on the

ubject to enable us to treat simple cases which now resent themselves in dail ractice.



he phenomena which are peculiar to the changingharacter of the currents are greatly exalted when theate of change is increased, hence the study of theseurrents is considerably facilitated by the employment o

roperly constructed apparatus. It was with this andther objects in view that I constructed alternate currenmachines capable of giving more than two millioneversals of current per minute, and to this circumstanceis principally due, that I am able to bring to your

ttention some of the results thus far reached, which I

ope will prove to be a step in advance on account of theiirect bearing upon one of the most important problemsamely, the production of a practical and efficient sourcef light.

he study of such rapidly alternating currents is very

nteresting. Nearly every experiment discloses somethinew. Many results may, of course, be predicted, but man

more are unforeseen. The experimenter makes many nteresting observations. For instance, we take a piece ofron and hold it against a magnet. Starting from low lternations and running up higher and higher we feel th

mpulses succeed each other faster and faster, get weakend weaker, and finally disappear. We then observe aontinuous pull ; the pull, of course, is not continuous ; itnly appears so to us ; our sense of touch is imperfect.

We may next establish an arc between the electrodes an

bserve, as the alternations rise, that the note whichccom anies alternatin arcs ets shriller and shriller



radually weakens, and finally ceases. The air vibrationsf course, continue, but they are too weak to beerceived; our sense of hearing fails us.

We observe the small physiological effects, the rapid

eating of the iron cores and conductors, curious inductivffects, interesting condenser phenomena, and still morenteresting light phenomena with a high tension inductiooil. All these experiments and observations would be of he greatest interest to the

tudent, but their description would lead me too far fromhe principal subject. Partly for this reason, and partly onccount of their vastly greater importance, I will confine

myself to the description of the light effects produced byhese currents.

n the experiments to this end a high tension inductionoil or equivalent apparatus for converting currents of omparatively low into currents of high tension is used.

f you will be sufficiently interested in the results I shallescribe as to enter into an experimental study of this

ubject ; if you will be convinced of the truth of therguments I shall advance— your aim will be to produceigh frequencies and high potentials j in other words,owerful electrostatic effects. You will then encounter

many difficulties, which, if completely overcome, wouldllow us to produce truly wonderful results.

irst will be met the difficult of obtainin the re uired



requencies by means of mechanical apparatus, and, if hey be ob. tained otherwise, obstacles of a differentature will present themselves. Next it will be foundifficult to provide the requisite insulation withoutonsiderably increasing the size of the apparatus, for the

otentials required are high, and, owing to the rapidity ohe alternations, the insulation presents peculiarifficulties. So, for instance, when a gas is present, theischarge may work, by the molecular bombardment of he gas and consequent heating, through as much as annch of the best solid insulating material, such as glass,

ard rubber, porcelain, sealing wax, etc. ; in fact, throughny known insulating substance. The chief requisite in th

nsulation of the apparatus is, therefore, the exclusion of ny gaseous matter.

n general my experience tends to show that bodies whic

ossess the highest specific inductive capacity, such aslass, afford a rather inferior insulation to others, which,

while they are good insulators, have a much smallerpecific inductive capacity, such as oils, for instance, theielectric losses being no doubt greater in the former. Thifficulty of insulating, of course, only exists when the

otentials are excessively high, for with potentials such afew thousand volts there is no particular difficulty

ncountered in conveying currents from a machineiving^ say, 20,000 alternations per second, to quite aistance. This number of alternations, however, is by faroo small for many purposes, though quite sufficient forome practical applications. This difficulty of insulating is



ortunately not a vital drawback;

affects mostly the size of the apparatus, for, whenxcessively high potentials would be used, the light-givinevices would be located not far from the apparatus, and

ften they would be quite close to it. As the air-ombardment of the insulated wire is dependent onondenser action, the loss may be reduced to a trifle by sing excessively thin wires heavily insulated.

Another difficulty will be encountered in the capacity and

elf-induction necessarily possessed by the coil. If the coie large, that is, if it contain a great length of wire, it wille generally un suited for excessively high frequencies; ibe small, it may be well adapted for such frequencies,

ut the potential might-then not be as high as desired. Aood insulator, and preferably one possessing a small

pecific inductive capacity, would afford a two-folddvantage. First, it would enable us to construct a very mall coil capable of withstanding enormous differences ootential ; and secondly, such a small coil, by reason of itmaller capacity and self-induction, would be capable of uicker and more vigorous vibration. The problem then

onstructing a coil or induction apparatus of any kindossessing the requisite qualities I regard as one of nomall importance, and it has occupied me for aonsiderable time.

he investigator who desires to repeat the experiments

which I will describe, with an alternate current machine,a able of su l in currents of the desired fre uenc





he field magnet of this machine consists of two like partwhich either enclose an exciting coil, or else arendependently wound.

IG. 97.

ach part has 480 pole projections, the projections of one

acing those of the other. The armature consists of awheel of hard bronze, carrying the conductors whichevolve between the projections of the field magnet. To

wind the armature conductors, I have found it mostonvenient to proceed in the following manner. Ionstruct a ring of hard bronze of the required size. This

ing and the rim of the wheel are provided with thero er number of ins and both fastened u on a late.



he armature conductors being wound, the pins are cutff and the ends of the conductors fastened by two rings

which screw to the

NVENTIONS OF NIKOLA TE8LA.

ronze ring and the rim of the wheel, respectively. Thewhole may then be taken off and forms a solid structure.

he conductors in such a type of machine should consistf sheet copper, the thickness of which, of course, dependn the thickness of the pole projections; or else twisted

hin wires should be employed.

ig. 99 is a smaller machine, in many respects similar tohe former, only here the armature conductors and thexciting coil are kept stationary, while only a block of

wrought iron is revolved.

t would be uselessly lengthening this description were Io



well more 011 the details of construction of thesemachines. Besides, they have been described somewhatmore elaborately in The Electrical Engineer, of March 18

891. I deem it well, however, to call the attention of thenvestigator to two things, the importance of which,hough self evident, he is nevertheless apt tonderestimate; namely, to the local action in the

onductors which must be carefully avoided, and to thelearance, which must be small. I may add, that since it iesirable to use very high peripheral speeds, thermature should be of very large diameter in order tovoid impracticable belt speeds. Of

he several types of these machines which have beenonstructed by me, I have found that the type illustratedn Fig. 97 caused me the least trouble in construction, as

well as in maintenance, and on the whole, it has been aood experimental machine.

n operating an induction coil with very rapidly lternating currents, among the first luminoushenomena noticed are naturally those presented by theigh-tension discharge. As the number of alternations peecond is increased, or as— the number being high — theurrent through the primary is varied, the discharge

radually changes in appearance. It would be difficult toescribe the minor chan es which occur, and the



onditions which

IG. 99.

ring them about, but one may note five distinct forms ohe discharge.

irst, one may observe a weak, sensitive discharge in theorm of a thin, feeble-colored thread. (Fig. lOOa.) Itlways occurs when, the number of alternations perecond being high, the current through the primary isery small. In spite of the excessively small current, theate of change is great, and the difference of potential at

he terminals of the secondary is therefore considerable,o that the arc is established at great distances; but the



uantity of " electricity " set in motion is insignificant,arely sufficient to maintain a thin, threadlike arc. It isxcessively sensitive and may be made so to such aegree that the mere act of breathing near the coil willffect it, and unless it is perfectly

well protected from currents of air, it wriggles aroundonstantly. Nevertheless, it is in this form excessively ersistent, and when the terminals are approached to,ay, one-third of the striking distance, it can be blown ounly with difficulty. This exceptional persistency, when

hort, is largely due to the arc being excessively thin;resenting, therefore, a very small surface to the blast.ts great sensitiveness, when very long, is probably due he motion of the particles of dust suspended in the air.

When the current through the primary is increased, the

ischarge gets broader and stronger, and the effect of thapacity of the coil becomes visible until, finally, underroper conditions, a white naming arc, Fig. 100 B, often ahick as one's finger, and striking across the whole coil, isroduced. It develops remarkable heat, and may be

urther characterized by the absence of the high note

which accompanies the less powerful discharges. To takeshock from the coil under these conditions would not



IG. lOOa. FIG. lOOh.

e advisable, although under different conditions, theotential being much higher, a shock from the coil may baken with impunity. To produce this kind of dischargehe number of alternations per second must not be tooreat for the coil used ; and, generally speaking, certain

elations between capacity, self-induction and frequencymust be observed.

he importance of these elements in an alternate currenircuit is now well-known, and under ordinary conditionshe general rules are applicable. But in an induction coil

xceptional conditions prevail. First, the self-induction isf little importance before the arc is established, when itsserts itself, but perhaps never as prominently as inrdinary alternate current circuits, because the capacity s distributed all along the coil, and by reason of the facthat the coil usually discharges through very great

esistances; hence the currents are exceptionally small.econdly,



he capacity goes on increasing continually as theotential rises, in consequence of absorption which takeslace to a considerable extent. Owing to this there existso critical relationship between these quantities, and

rdinary rules would not seem to be applicable. As theotential is increased either in consequence of thencreased frequency or of the increased current throughhe primary, the amount of the energy stored becomesreater and greater, and the capacity gains more and

more in importance. Up to a certain point the capacity is

eneficial, but after that it begins to be an enormousrawback. It follows from this that each coil gives the beesult with a given frequency and primary current. A ery large coil, when operated with currents of very high

requency, may not give as much as •£• incli spark. By dding capacity to the terminals, the condition may be

mproved, but what the coil really wants is a lowerrequency.

When the flaming discharge occurs, the conditions arevidently such that the greatest current is made to flow hrough the circuit. These conditions may be attained by

arying the frequency within wide limits, but the highestrequency at which the flaming arc can still be produced,etermines, for a given primary current, the maximumtriking distance of the coil. In the flaming discharge theclat effect of the capacity is not perceptible ; the rate at

which the energy is being stored then just equals the rat

t which it can be disposed of through the circuit. Thisind of dischar e is the severest test for a coil the break



when it occurs, is of the nature of that in an overchargedey den jar. To give a rough approximation I would state

hat, with an ordinary coil of, say 10,000 ohms resistanche most powerful arc would be produced with about2,000 alternations per second.

When the frequency is increased beyond that rate, theotential, of course, rises, but the striking distance may,evertheless, diminish, paradoxical as it may seem. As

he potential rises the coil attains more and more theroperties of a static machine until, finally, one may

bserve the beautiful phenomenon of the streamingischarge, Fig. 101, which may be produced across the

whole length of the coil. At that stage streams begin tossue freely from all points and projections. These stream

will also be seen to pass in abundance in the spaceetween the primary and the insulating tube. When the

otential is excessively high they will always appear, evef the frequency be low, and even if the primary beurrounded by as much as an inch of wax, hard nib-

er, glass, or any other insulating substance. This limitsreatly the'output of the coil, but I will later show how I

ave been able to overcome to a considerable extent thisisadvantage in the ordinary coil.

esides the potential, the intensity of the streamsepends on the frequency; but if the coil be very largehey show themselves, no matter how low the frequencie

sed. For instance in a ver lar e coil of a resistance of



7,000 ohms, constructed by me some time ago, they ppear with as low as 100 alternations per second andess, the insulation of the secondary being f inch of bonite. When very intense they produce a noise similaro that produced by the charging of a Holtz machine, but

much more powerful, and they emit a strong smell of zone. The lower the frequency, the more apt they are tuddenly injure the coil. With excessively high frequencihey may pass freely

IG. 101.

without producing any other effect than to heat thensulation slowly and uniformly.

he existence of these streams shows the importance of onstructing an expensive coil so as to permit of one'seeing through the tube surrounding the primary, and thatter should be easily exchangeable; or else the spaceetween the primary and secondary should be completelled up with insulating material so as to exclude all air.

he non-observance of this simple rule in theonstruction of commercial coils is responsible for the



estruction of many an expensive coil.

At the stage when the streaming discharge occurs, or witomewhat higher frequencies, one may, by approachinghe terminals quite nearly, and regulating properly the

ffect of capacity, produce a veritable spray of smallilver-white sparks, or a bunch of excessively thin silverhreads (Fig. 102) amidst a powerful brush—each spark r thread possibly corresponding

o one alternation. This, when produced under proper

onditions, is probably the most beautiful discharge, andwhen an air blast is directed against it, it presents aingular appearance. The spray of sparks, when receivedhrough the body, causes some inconvenience, whereas,

when the discharge simply streams, no tiling at all is likeo be felt if large conducting objects are held in the hands

o protect them from receiving small burns.

f the frequency is still more increased, then the coilefuses to give any spark unless at comparatively smallistances, and the fifth typical form of discharge may bebserved (Fig. 103). The tendency to stream out and

issipate is then so great that when the brush is producet one terminal no sparking occurs, even if, as I haveepeatedly tried, the hand, or any conducting object, iseld within the stream; and, what is more singular, the

umi-



IG. 103. FIG. 104.

ous stream is not at all easily deflected by the approachf a conducting body.

At this stage the streams seemingly pass with thereatest freedom through considerable thicknesses of nsulators, and it is particularly interesting to study their

ehavior. For this purpose it is convenient to connect tohe terminals of the coil two metallic spheres which may e placed at any desired distance, Fig. 104. Spheres arereferable to plates, as the discharge can be betterbserved. By inserting dielectric bodies between thespheres, beautiful discharge phenomena may be

bserved. If the spheres be quite close and a spark belaying between them, by interposing a thin plate of



bonite between the spheres the spark instantly ceasesnd the discharge spreads into an intensely luminousircle several inches in diameter, provided the spheresre

ufficiently large. The passage of the streams heats, and,fter a while, softens, the rubber so much that two platemay be made to stick together in this manner. If thepheres are so far apart that no spark occurs, even if there far beyond the striking distance, by inserting a thick late of glass the discharge is instantly induced to pass

rom the spheres to the glass in the form of luminoustreams. It appears almost as though these streams passhrough the dielectric. In reality this is not the case, as thtreams are due to the molecules of the air which areiolently agitated in the space between the oppositely harged .surfaces of the spheres. When no dielectric othe

han air is present, the bombardment goes on, but is tooweak to be visible ; by inserting a dielectric the inductiveffect is much increased, and besides, the projected air

molecules find an obstacle and the bombardmentecomes so intense that the streams become luminous. Iy any mechanical means we could effect such a violent

gitation of the molecules we could produce the samehenomenon. A jet of air escaping through a small holender enormous pressure and striking against an

nsulating substance, such as glass, may be luminous inhe dark, and it might be possible to produce ahosphorescence of the glass or other insulators in this

manner.



he greater the specific inductive capacity of thenterposed dielectric, the more powerful the effectroduced. Owing to this, the streams show themselves

with excessively high potentials even if the glass be asmuch as one and one-half to two inches thick. But beside

he heating due to bombardment, some heating goes onndoubtedly in th^ dielectric, being apparently greater ilass than in ebonite. I attribute this to the greaterpecific inductive capacity of the glass, in consequence of

which, with the same potential difference, a greatermount of energy is taken up in it than in rubber. It is lik

onnecting to a battery a copper and a brass wire of theame dimensions. The copper wire, though a more perfeonductor, would heat more by reason of its taking moreurrent. Thus what is otherwise considered a virtue of thlass is here a defect. Glass usually gives way muchuicker than ebonite ; when it is heated to a certain

egree, the discharge suddenly breaks through at oneoint, assuming then the ordinary form of an arc.

he heating effect produced by molecular bombardmentf the dielectric would, of course, diminish as the pressurf the

ir is increased, and at enormous pressure it would beegligible, unless the frequency would increaseorrespondingly.

t will be often observed in these experiments that when

he s heres are be ond the strikin distance, the



pproach of a glass plate, for instance, may induce thepark to jump between the spheres. This occurs when thapacity of the spheres is somewhat below the criticalalue which gives the greatest difference of potential athe terminals of the coil. By approaching a dielectric, the

pecific inductive capacity of the space between thepheres is increased, producing the same effect as if theapacity of the spheres were increased. The potential athe terminals may then rise so high that the air space isracked. The experiment is best performed with denselass or mica.

Another interesting observation is that a plate of nsulating material, when the discharge is passing throug, is strongly attracted by either of the spheres, that is b

he nearer one, this being obviously due to the smallermechanical effect of the bombardment on that side, and

erhaps also to the greater electrification.

rom the behavior of the dielectrics in these experimentwe may conclude that the best insulator for these rapidly

lternating currents would be the one possessing themallest specific inductive capacity and at the same time

ne capable of withstanding the greatest differences of otential; and thus two diametrically opposite ways of ecuring the required insulation are indicated, namely, tose either a perfect vacuum or a gas under great pressurbut the former would be preferable. Unfortunately either of these two ways is easily carried out in practice

t is es eciall interestin to note the behavior of an



xcessively high vacuum in these experiments. If a testube, provided with external electrodes and exhausted the highest possible degree, be connected to the terminaf the coil, Fig. 105, the electrodes of the tube are

nstantly brought to a high temperature and the glass at

ach end of the tube is rendered intensely hosphorescent, but the middle appears comparatively ark, and for a while remains cool.

When the frequency is so high that the discharge shown ig. 103 is observed, considerable dissipation no doubt

ccurs in the coil. Nevertheless the coil may be worked folong time, as the heating is gradual.

n spite of the fact that the difference of potential may b


normous, little is felt when the discharge is passedhrough the body, provided the hands are armed. This iso some extent due to the higher frequency, butrincipally to the fact that less energy is availablexternally, when the difference of potential reaches an

normous value, owing to the circumstance that, with thise of potential, the energy absorbed in the coil increases the square of the potential. Up to a certain point thenergy available externally increases with the rise of otential, then it begins to fall off rapidly. Thus, with therdinary high tension induction coil, the curious paradox

xists, that, while with a given current through therimary the shock might be fatal, with many times that



urrent it might be perfectly harmless, even if therequency be the same. With high frequencies andxcessively high potentials when the terminals are notonnected to bodies of some size, practically all the energupplied to the primary is

IG. 10").

IG. l<m.

aken up by the coil. There is no breaking through, noocal injury, but all the material, insulating andonducting, is uniformly heated.

o avoid misunderstanding in regard to the physiologicaffect of alternating currents of very high frequency, Ihink it necessary to state that, while it is an undeniableact that they are incomparably less dangerous thanurrents of low frequencies, it should not be thought thathey are altogether harmless. What has just been said

efers only to currents from an ordinary high tensionnduction coil which currents are necessaril ver small



f received directly from a machine or from a secondary oow resistance, they produce more or less powerfulffects, and may cause serious injury, especially whensed in conjunction with condensers.

he streaming discharge of a high tension induction coiliffers in many respects from that of a powerful staticmachine. In color it has neither the violet of the positive,

or the brightness of the negative, static discharge, butes somewhere between, being, of course, alternatively ositive and negative. But since the streaming is more

owerful when the point or terminal is electrifiedositively, than when electrified negatively, it follows thahe point of the brush is more like the positive, and theoot more like the negative, static discharge. In the dark

when the brush is very powerful, the root may appearlmost white. The wind produced by the escaping

treams, though it may be very strong—often indeed touch a degree that it may be felt quite a distance from thoil—is, nevertheless, considering the quantity of theischarge, smaller than that produced by the positive



IG. 107. FIG. 108.

rush of a static machine, and it affects the flame muchess powerfully. From the nature of the phenomenon we

an conclude that the higher the frequency, the smallermust, of course, be the wind produced by the streams,

nd with sufficiently high frequencies no wind at all woule produced at the ordinary atmospheric pressures. Witrequencies obtainable by means of a machine, the

mechanical effect is sufficiently great to revolve, withonsiderable speed, large pin-wheels, which in the dark resent a beautiful appearance owing to the abundance ohe streams (Fig. 106).

n general, most of the experiments usually performedwith a static machine can be performed with an inductionoil when operated witli very rapidly alternating currenthe effects produced, however,'are much more striking,eing of incomparably

reater power. When a small length of ordinary cottonovered wire, Fig. 107, is attached to one terminal of theoil, the streams issuing from all points of the wire may bo intense as to produce a considerable light effect. Whenhe potentials and frequencies are very high, a wirensulated with gutta percha or rubber and attached to onf the terminals, appears to be covered with a luminous

lm. A very thin bare wire when attached to a terminalmits owerful streams and vibrates continuall to and



ro or spins in a circle, producing a singular effect (Fig.08). Some of these experiments have been described by

me in The Electrical W(M, of February 21, 1891.

Another peculiarity of the rapidly alternating discharge o

he induction coil is its radically different behavior withespect to points and rounded surfaces.

f a thick wire, provided with a ball at one end and with aoint at the other, be attached to the positive terminal ofstatic machine, practically all the charge will be lost

hrough the point, on account of the enormously greaterension, dependent on the radius of curvature. But if sucwire is attached to one of the terminals of the induction

oil, it will be observed that with very high frequenciestreams issue from the ball almost as copiously as fromhe point (Fig. 109).

t is hardly conceivable that we could produce such aondition to an equal degree in a static machine, for theimple reason, that the tension increases as the square ohe density, which in turn is proportional to the radius ofurvature ; hence, with a steady potential an enormous

harge would be required to make streams issue from aolished ball while it is connected with a point. But with a

nduction coil the discharge of which alternates with greaapidity it is different, Here we have to deal with twoistinct tendencies. First, there is the tendency to escape

which exists in a condition of rest, and which depends OH

he radius of curvature; second, there is the tendency toissi ate into the surroundin air b condenser action



which depends on the surface. When one of theseendencies is a maximum, the other is at a minimum. Athe point the luminous stream is principally due to the ai

molecules coming bodily in contact with the point; they re attracted and repelled, charged and discharged, and,

heir atomic charges being thus disturbed, vibrate andmit light waves. At the ball, on the contrary, there is nooubt that the effect is to a great extent produced indue

vely, the air molecules not necessarily coining in contacwith the ball, though they undoubtedly do so. To convinc

urselves of this we only need to exalt the condenserction, for instance, by enveloping the ball, at someistance, by a better conductor than the surrounding

medium, the conductor being, of course, insulated ; or elsy surrounding it with a better dielectric and approachinn insulated conductor; in both cases the streams will

reak forth more copiously. Also, the larger the ball withiven frequency, or the higher the frequency, the more

will the ball have the advantage over the point. But, sinccertain intensity of action is required to render the

treams visible, it is obvious that in the experimentescribed the ball should not be taken too large.

n consequence of this two-fold tendency, it is possible toroduce by means of points, effects identical to thoseroduced by



IG. 109. FIG. 110.

apacity. Thus, for instance, by attaching to one terminaf the coil a small length of soiled wire, presenting many oints and offering great facility to escape, the potential

he coil may be raised to the same value as by attachingo the terminal a polished ball of a surface many timesreater than that of the wire.

An interesting experiment, showing the effect of theoints, may be performed in the following manner: Attac

o one of the terminals of the coil a cotton covered wirebout two feet in length, and adjust the conditions so thatreams issue from the wire. In this experiment therimary coil should be preferably placed so that itxtends only about half way into the secondary coil. Nowouch the free terminal of the secondary with a

onducting object held in the hand, or else connect it to ansulated

ody of some size. In this manner the potential on thewire may be enormously raised. The effect of this will beither to increase, or to diminish, the streams. If they

ncrease the wire is too short if the diminish it is too



ong. By adjusting the length of the wire, a point is foundwhere the touching of the other terminal does not at all

ffect the streams. In this case the rise of potential isxactly counteracted by the drop through the coil. It wille observed that small lengths of wire produce

onsiderable difference in the magnitude and luminosity f the streams. The primary coil is placed side wise forwo reasons: First, to increase the potential at the wire;nd, second, to increase the drop through the coil. Theensitiveness is thus augmented.

here is still another and far more striking peculiarity ofhe brush discharge produced by very rapidly alternatinurrents. To observe this it is best to replace the usualerminals of the coil by two metal columns insulated withgood thickness of ebonite. It is also well to close allssures and cracks with wax so that the brushes cannot

orm anywhere except at the tops of the columns. If theonditions are carefully adjusted—which, of course, muste left to the skill of the experimenter —so that theotential rises to an enormous value, one may producewo powerful brushes several inches long, nearly white aheir roots, which in the dark bear a striking resemblanc

o two flames of a gas escaping under pressure (Fig. 110)ut they do not only resemble, they are veritable flames

or they are hot. Certainly they are not as hot as a gasurner, but they would be so if the frequency cmd theotential would be sufficiently high. Produced with, say,wenty thousand alternations per second, the heat isasily perceptible even if the potential is not excessively



igh. The heat developed is, of course, due to the impactf the air molecules against the terminals and against eacther. As, at the ordinary pressures, the mean free path xcessively small, it is possible that in spite of thenormous initial speed imparted to each molecule upon

oming in contact with the terminal, its progress—by ollision with other molecules—is retarded to such anxtent, that it does not get away far from the terminal,ut may strike the same many times in succession. Theigher the frequency, the less the molecule is able to getway, and this the more so, as for a given effect the

otential required is smaller; and a frequency isonceivable—perhaps even obtainable—at

which practically the same molecules would strike theerminal. Under such conditions the exchange of the

molecules would be very slow, and the heat produced at,

nd very near, the terminal would be excessive. But if threquency would go on increasing constantly, the heatroduced would begin to diminish for obvious reasons. Inhe positive brush of a static machine the exchange of th

molecules is very rapid, the stream is constantly of oneirection, and there are fewer collisions ; hence the

eating effect must be very small. Anything that impairshe facility of exchange tends to increase the local heatroduced. Thus, if a bulb be held over the terminal of theoil so as to enclose the brush, the air contained in theulb is very quickly brought to a high temperature. If alass tube be held over the brush so as to allow theraught to carry the brush upwards, scorching hot air





r on a projection of any kind, more or less conducting, oendered so by dampness, a powerful brush may appearf the lightning strikes somewhere in the

eighborhood the enormous potential may be made to

lternate or fluctuate perhaps many million times aecond. The air molecules are violently attracted andepelled, and by their impact produce such a powerfuleating effect that a lire is started. It is conceivable that hip at sea may, in this manner, catch fire at many pointt once. When we consider, that even with the

omparatively low frequencies obtained from a dynamomachine, and with potentials of no more than one or twoundred thousand volts, the heating effects areonsiderable, we may imagine how much more powerfulhey must be with frequencies and potentials many timereater; and the above explanation seems, to say the

east, very probable. Similar explanations may have beenuggested, but I am not aware that, up to the present, theating effects of a brush produced by a rapidly lternating potential



IG. ill.

ave been experimentally demonstrated, at least not touch a remarkable degree.

y preventing completely the exchange of the airmolecules^ the local heating effect may be so exalted as

ring a body to incandescence. Thus, for instance, if amall button, or preferably a very thin wire or filament bnclosed in an unexhausted globe and connected with theerminal of the coil, it may be rendered incandescent. Thhenomenon is made much more interesting by the rapipinning round in a circle of the top of the filament, thusresenting the appearance of a luminous funnel, Fig. Ill,

which widens when the potential is increased. When theotential is small the end of the filament may perf orm

rregular motions, suddenly changing from one to thether, or it may describe an ellipse; but when theotential is very high it always spins in a circle; and sooes generally a thin

traight wire attached freely to the terminal of the coil.

hese motions are, of course, due to the impact of themolecules, and the irregularity in the distribution of theotential, owing to the roughness and dissymmetry of th

wire or filament. With a perfectly symmetrical andolished wire such motions would probably not occur.hat the motion is not likely to be due to others causes is

vident from the fact that it is not of a definite direction,nd that in a ver hi hl exhausted lobe it ceases



ltogether. The possibility of bringing a body toncandescence in an exhausted globe, or even when not all enclosed, would seem to afford a possible way of btaining light eifects, which, in perfecting methods of roducing rapidly alternating potentials, might be

endered available for useful purposes.

n employing a commercial coil, the production of very owerful brush effects is attended with considerableifficulties, for

IG. 112*.

when these high frequencies and enormous potentials ar

sed, the best insulation is apt to give way. Usually theoil is insulated well enough to stand the strain fromonvolution to convolution, since two double silk coveredaraffined wires will withstand a pressure of severalhousand volts; the difficulty lies principally in preventinhe breaking through from the secondary to the primary

which is greatly facilitated by the streams issuing fromhe latter. In the coil, of course, the strain is greatest from



ection to section, but usually in a larger coil there are somany sections that the danger of a sudden giving way is

ot very great. No difficulty will generally be encounteren that direction, and besides, the liability of injuring theoil internally is very much reduced by the fact that the

ffect most likely to be produced is simply a gradualeating, which, when far enough

dvanced, could not fail to be observed. The principalecessity is then to prevent the streams between therimary and the tube, not only on account of the heating

nd possible injury, but also because the streams may iminish very considerably the potential differencevailable at the terminals. A few hints as to how this maye accomplished will probably be found useful in most ofhese experiments with the ordinary induction coil.

One of the ways is to wind a short primary, Fig. 112a, sohat the difference of potential is not at that length greatnough to cause the breaking forth of the streamshrough the insulating tube. The length of the primary hould be determined by experiment. Both the ends of he coil should be brought out on one end through a plug

f insulating material fitting in the tube as illustrated. Inuch a disposition one terminal of the secondary isttached to a body, the surface of which is determined

with the



IG. 112b.

reatest care so as to produce the greatest rise in theotential. At the other terminal a powerful brush appear

which may be experimented upon.

he above plan necessitates the employment of a primarf comparatively small size, and it is apt to heat whenowerful effects are desirable for a certain length of timen such a case it is better to employ a larger coil, Fig.12b, and introduce it from one side of the tube, until the

treams begin to appear. In this case the nearest terminf the secondary may be connected to the primary or tohe ground, which is practically the same thing, if therimary is connected directly to the machine. In the casef ground connections it is well to determinexperimentally the frequency which is best suited under

he conditions of the test. Another way of obviating thetreams, more or less, is to

make the primary in sections and supply it from separatwell insulated sources.

n many of these experiments, when powerful effects arwanted for a short time it is advanta eous to use iron



ores with the primaries. In such case a very largerimary coil may be wound and placed side by side witlihe secondary, and, the nearest terminal of the lattereing connected to the primary, a laminated iron core is

ntroduced through the primary into the secondary as fa

s the streams will permit. Under these conditions anxcessively powerful brush, several inches long, whichmay be appropriately called " St. Elmo's hot fire," may baused to appear at the other terminal of the secondary,roducing striking effects. It is a most powerful ozonizero powerful indeed, that only a few minutes are sufficien

o fill the whole room with the smell of ozone, and itndoubtedly possesses the quality of exciting chemicalffinities.

or the production of ozone, alternating currents of veryigh frequency are eminently suited, not only on account

f the advantages they offer in the way of conversion bulso because of the fact, that the ozonizing action of aischarge is dependent on the frequency as well as on thotential, this being undoubtedly confirmed by bservation.

n these experiments if an iron core is used it should bearefully watched, as it is apt to get excessively hot in anncredibly short time. To give an idea of the rapidity of he heating, I will state, that by passing a powerfulurrent through a coil with many turns, the inserting

within the same of a thin iron wire for no more than one

econd's time is sufficient to heat the wire to somethin



ke 100° C.

ut this rapid heating need not discourage us in the use oron cores in connection with rapidly alternating currentshave for a long time been convinced that in the

ndustrial distribution by means of transformers, someuch plan as the following might be practicable. We may se a comparatively small iron core, subdivided, orerhaps not even subdivided. We may surround this cor

with a considerable thickness of material which is fire-roof and conducts the heat poorly, and on top of that we

may place the primary and secondary windings. By usingither higher frequencies or greater magnetizing forces,we may by hysteresis and eddy currents heat the ironore so far as to bring it nearly to its maximumermeability, which, as Hopkinson has

hown, may be as much as sixteen times greater thanhat at ordinary temperatures. If the iron core wereerfectly enclosed, it would not be deteriorated by theeat, and, if the enclosure of tire-proof material would beufficiently thick, only a limited amount of energy coulde radiated in spite of the high temperature.

ransformers have been constructed by me on that planut for lack of time, no thorough tests have as yet been

made.

Another way of adapting the iron core to rapidlternations, or, generally speaking, reducing the friction

osses, is to produce by continuous magnetization a flow oomethin like seven thousand or ei ht thousand lines e



quare centimetre through the core, and then work withweak magnetizing forces and preferably high frequencies

round the point of greatest permeability. A higherfficiency of conversion and greater output are obtainabln this manner. I have also employed this principle in

onnection with machines in which there is no reversal oolarity. In these types of machines, as long as there arenly few pole projections, there is no great gain, as the

maxima and minima of magnetization are far from theoint of maximum permeability ; but when the number ohe pole projections is very great, the required rate of

hange may be obtained, without the magnetizationarying so far as to depart greatly from the point of

maximum permeability, and the gain is considerable.

he above described arrangements refer only to the usef commercial coils as ordinarily constructed. If it is

esired to construct a coil for the express purpose of erforming with it such experiments as I have describedr, generally, rendering it capable of withstanding thereatest possible difference of potential, then aonstruction as indicated in Fig. 113 will be found of dvantage. The coil in this case is formed of two

ndependent parts which are wound oppositely, theonnection between both being made near the primary.he potential in the middle being zero, there is not much

endency to jump to the primary and not much insulations required. In some cases the middle point may,owever, be connected to the primary or to the ground.n such a coil the laces of reatest difference of otentia



re far apart and the coil is capable of withstanding annormous strain. The two parts may be movable so as tollow a slight adjustment of the capacity effect.

As to the manner of insulating the coil, it will be found

on-

enient to proceed in the following way: First, the wirehould be boiled in paraffine until all the air is out ; thenhe coil is wound by running the wire through meltedaraffine, merely for the purpose of fixing the wire. The

oil is then taken off from the spool, immersed in aylindrical vessel filled with pure melted wax and boiledor a long time until the bubbles cease to appear. The

whole is then left to cool down thoroughly, and then themass is taken out of the vessel and turned up in a lathe. Aoil made in this manner and with care is capable of

withstanding enormous potential differences.

t may be found convenient to immerse the coil inaraffine oil or some other kind of oil; it is a most effectiv

way of insulating, principally on account of the perfectxclusion of air, but it may



IG. 113.

e found that, after all, a vessel filled with oil is not a veronvenient thing to handle in a laboratory.

f an ordinary coil can be dismounted, the primary may e taken out of the tube and the latter plugged up at onend, filled with oil, and the primary reinserted. This

ffords an excellent insulation and prevents the formatiof the streams.

Of all the experiments which may be performed withapidly alternating currents the most interesting arehose which concern the production of a practical

luminant. It cannot be denied that the present methodshough they were brilliant advances, are very wasteful.ome better methods must be invented, some moreerfect apparatus devised. Modern research has openedew possibilities for the production of an efficient sourcef light, and the attention of all has been turned in the

irection indicated

y able pioneers. Many have been carried away by thenthusiasm and passion to discover, but in their zeal toeach results, some have been misled. Starting with thedea of producing electromagnetic waves, they turned

heir attention, perhaps, too much to the study of electroma netic effects, and ne lected the stud of electrostatic



henomena. Naturally, nearly every investigator availedimself of an apparatus similar to that used in earlierxperiments. But in those forms of apparatus, while thelectromagnetic inductive effects are enormous, thelectrostatic effects are excessively small.

n the Hertz experiments, for instance, a high tensionnduction coil is short circuited by an arc, the resistance o

which is very small, the smaller, the more capacity isttached to the terminals ; and the difference of potentiat these is enormously diminished. On the other hand,

when the discharge is not passing between the terminalshe static effects may be considerable, but only ualitatively so, not quantitatively, since their rise and fa

s very sudden, and since their frequency is small. Ineither case, therefore, are powerful electrostatic effectserceivable. Similar conditions exist when, as in some

nteresting experiments of Dr. Lodge, Ley den jars areischarged disruptively. It has been thought — and Ielieve asserted — that in such cases most of the energy

s radiated into space. In the light of the experimentswhich I have described above, it will now not be thoughto. I feel safe in asserting that in such cases 'most of the

nergy is partly taken up and converted into heat in therc of the discharge and in the conducting and insulating

material of the jar, some energy being, of course, given oy electrification of the air; but the amount of the directladiated energy is very small.

When a high tension induction coil, operated by currents



lternating only 20,000 times a second, has its terminalslosed through even a very small jar, practically all thenergy passes through the dielectric of the jar, which iseated, and the electrostatic effects manifest themselvesutwardly only to a very weak degree. Now the external

ircuit of a Leyden jar, that is, the arc and the connectionf the coatings, may be looked upon as a circuit generatinlternating currents of excessively high frequency andairly high potential, which is closed through the coatingsnd the dielectric between them, and from the above it isvident that the external electrostatic effects must be

ery small, even if a

ecoil circuit be used. These conditions make it appearhat with the apparatus usually at hand, the observationf powerful electrostatic effects was impossible, and whaxperience has been gained in that direction is only due t

he great ability of the investigators.

ut powerful electrostatic eifects are a sitw qua -non of ght production on the lines indicated by theory. Electro

magnetic eifects are primarily unavailable, for the reasonhat to produce the required effects we would have to

ass current impulses through a conductor, which, longefore the required frequency of the impulses could beeached, would cease to transmit them. On the otherand, electro-magnetic waves many times longer than

hose of light, and producible by sudden discharge of aondenser, could not be utilized, it would seem, except w

vail ourselves of their effect upon conductors as in the



resent methods, which are wasteful. We could not affecy means of such waves the static molecular or atomicharges of a gas, cause them to vibrate and to emit light.ong transverse waves cannot, apparently, produce suchffects, since excessively small electro-magnetic

isturbances may pass readily through miles of air. Suchark waves, unless they are of the length of true lightwaves, cannot, it would seem, excite luminous radiation

Geissler tube, and the luminous effects, which areroducible by induction in a tube devoid of electrodes, Im inclined to consider as being of an electrostatic natur

o produce such luminous effects, straight electrostatichrusts are required; these, whatever be their frequency

may disturb the molecular charges and produce light.ince current impulses of the required frequency cannotass through a conductor of measurable dimensions, we

must work with a gas, and then the production of owerful electrostatic effects becomes an imperativeecessity.

t has occurred to me, however, that electrostatic effectsre in many ways available for the production of light. Fo

nstance, we may place a body of some refractory material in a closed, and preferably more or lessxhausted, globe, connect it to a source of • high, rapidly lternating potential, causing the molecules of the gas totrike it many times a second at enormous speeds, and inhis manner, with trillions of invisible hammers, pound it

ntil it gets incandescent ; or we may place a body in a



ery highly exhausted globe, in a non-striking vacuum,nd, by employing very

igh frequencies and potentials, transfer sufficient energrom it to other bodies in the vicinity, or in general to the

urroundings, to maintain it at any degree of ncandescence ; or we may, by means of such rapidly lternating high potentials, disturb the ether carried by he molecules of a gas or their static charges, causinghem to viorate and to emit light.

ut, electrostatic eifects being dependent upon theotential and frequency, to produce the most powerfulction it is desirable to increase both as far as practicablet may be possible to obtain quite fair results by keepingither of these factors small, provided the other isufficiently great; but we are limited in both directions.

My experience demonstrates that we cannot go below aertain frequency, for, first, the potential then becomes sreat that it is dangerous ; and, secondly, the lightroduction is less efficient.

have found that, by using the ordinary low frequencies

he physiological effect of the current required tomaintain at a certain degree of brightness a tube four feeong, provided at the ends with outside and insideondenser coatings, is so powerful that, I think, it mightroduce serious injury to those not accustomed to suchhocks; whereas, with twenty thousand alternations per

econd, the tube may be maintained at the same degreef bri htness without an effect bein felt. This is due



rincipally to the fact that a much smaller potential isequired to produce the same light effect, and also to theigher efficiency in the light production. It is evident thahe efficiencv in such cases is the greater, the higher therequency, for the quicker the process of charging and

ischarging the molecules, the less energy will be lost inhe form of dark radiation. But, unfortunately, we cannoto beyond a certain frequency on account of the difficultyf producing and conveying the effects.

have stated above that a body inclosed in an

nexhausted bulb may be intensely heated by simply onnecting it with a source of rapidly alternating potentiahe heating in such a case is, in all probability, due mostl

o the bombardment of the molecules of the gas containen the bulb. When the bulb is exhausted, the heating of he body is much more rapid, and there is no difficulty

whatever in bringing a wire or filament to any degree of ncandescence by simply connecting it to one terminal ofoil of the proper dimensions. Thus, if the well-knownpparatus of Prof. Crookes, consisting of a bent platinum

wire with

anes mounted over it (Fig. 114), be connected to oneerminal of the coil — either one or both ends of thelatinum wire being connected— the wire is renderedlmost instantly incandescent, and the mica vanes areotated as though a current from a battery were used. Ahin carbon filament, or, preferably, a button of some

efractor material (Fi . 115), even if it be a



omparatively poor conductor, inclosed in an exhaustedlobe, may be rendered highly incandescent ; and in this

manner a simple lamp capable of giving any desiredandle power is provided.

he success of lamps of this kind would depend largely ohe selection of the light-giving bodies contained withinhe bulb. Since, under the conditions described, refractorodies— which are very poor conductors and capable of

withstanding for a long time excessively high degrees of emperature — may be used, such illuminating devices

may be rendered successful.

t might be thought at first that if the bulb, containing th

IG. 11 . FIG. 11 .



lament or button of refractory material, be perfectly wexhausted— that is, as far as it can be done by the use ofhe best apparatus—the heating would be much lessntense, and that in a perfect vacuum it could not occur a

ll. This is not confirmed by my experience; quite theontrary, the better the vacuum the more easily theodies are brought to incandescence. This result is

nteresting for many reasons.

At the outset of this work the idea presented itself to me

whether two bodies of refractory material enclosed in aulb exhausted to such a degree that the discharge of aarge induction coil, operated in the usual manner, cannoass through, could be rendered incandescent by mereondenser action. Obviously, to reach this resultnormous potential differences and very high frequencie

re required, as is evident from a simple calculation.

ut such a lamp would possess a vast advantage over anrdinary incandescent lamp in regard to efficiency. It is

well-known that the efficiency of a lamp is to some extenfunction of the degree of incandescence, and that, could

we but work a filament at many times higher degrees of ncandescence, the efficiency would be much greater. Inn ordinary lamp this is impracticable on account of theestruction of the filament, and it has been determined bxperience how far it is advisable to push thencandescence. It is impossible to tell how much higher

fficiency could be obtained if the filament couldwithstand indefinitel as the investi ation to this end



bviously cannot be carried beyond a certain stage ; buthere are reasons for believing that it would be very onsiderably higher. An improvement might be made inhe ordinary lamp by employing a short and thick carbonbut then the leading-in wires would have to be thick,

nd, besides, there are many other considerations whichender such a modification entirely impracticable. But inamp as above described, the leading in wires may beery small, the incandescent refractory material may be

n the shape of blocks offering a very small radiatingurface, so that less energy would be required to keep

hem at the desired incandescence ; and in addition tohis, the refractory material need not be carbon, but maye manufactured from mixtures of oxides, for instance,

with carbon or other material, or may be selected fromodies which are practically non-conductors, and capablef withstanding enormous degrees of temperature.

All this would point to the possibility of obtaining a muchigher efficiency with such a lamp than is obtainable inrdinary lamps. In my experience it has beenemonstrated that the blocks are brought to high degreef incandescence with much lower potentials than those

etermined by calculation, and the blocks may be set atreater distances from each other. We may freely ssume, and it is probable, that the molecularombardment is an important element in the heating,ven if the globe be exhausted with the utmost care, as Iave done; for although the number of the molecules is,om arativel s eakin insi nificant et on account of





IG. 116.

IG. 117.

ormer were the case, then a thin filament enclosed in a

erfectly exhausted globe, and connected to a source of normous, steady potential, would be brought toncandescence.

Various forms of lamps on the above described principlewith the refractory bodies in the form of filaments, Fig.

16, or blocks, Fi . 117, have been constructed and



perated by me, and investigations are being carried on his line. There is no difficulty in reaching such highegrees of incandescence that ordinary carbon is to allppearance melted and volatilized. If the vacuum could b

made absolutely perfect, such a lamp, although

noperative with apparatus ordinarily used, would, if perated with cur-

ents of the required character, afford an illuminant whicwould never be destroyed, and which would be far morefficient than an ordinary incandescent lamp. This

erfection can, of course, never be reached, and a very low destruction and gradual diminution in size alwaysccurs, as in incandescent filaments ; but there is noossibility of a sudden and premature disabling whichccurs in the latter by the breaking of the filament,specially when the incandescent bodies are in the shape

f blocks.

With these rapidly alternating potentials there is,owever, no necessity of enclosing two blocks in a globe,ut a single block, as in Fig. 115, or filament, Fig. 118, mae used. The potential in this case must of course be

igher, but is easily obtainable, and besides it is notecessarily dangerous.

he facility with which the button or filament in such aamp



IG. 118.

s brought to incandescence, other things being equal,epends on the size of the globe. If a perfect vacuumould be obtained, the size of the globe would not be of mportance, for then the heating would be wholly due to

he surging of the charges, and all the energy would beiven off to the surroundings by radiation. But this canever occur in practice. There is always some gas left in

he globe, and although the exhaustion may be carried tohe highest degree, still the space inside of the bulb muste considered as conducting when such high potentials ar

sed, and I assume that, in estimating the energy thatmay be given off from the filament to the surroundings,we may consider

he inside surface of the bulb as one coating of aondenser, the air and other objects surrounding the bulormin the other coatin . When the alternations are ver



ow there is no doubt that a considerable portion of thenergy is given off by the electrification, of theurrounding air.

n order to study this subject better, I carried on some

xperiments with excessively high potentials and low requencies. I then observed that when the hand ispproached to the bulb, — the filament being connected

with one terminal of the coil, — a powerful vibration iselt, being due to the attraction and repulsion of the

molecules of the air which are electrified by induction

hrough the glass. In some cases when the action is very ntense I have been able to hear a sound, which must beue to the same cause.

When the alternations are low, one is agt to get anxcessively



IG. 119. FIG. 120.

owerful shock from the bulb. In general, when onettaches bulbs or objects of some size to the terminals ofhe coil, one should look out for the rise of potential, for it

may happen that by merely connecting a bulb or plate tohe terminal, the potential may rise to many times its

riginal value. When lamps are attached to the terminalss illustrated in Fig. 119, then the capacity of the bulbshould be such as to give the maximum rise of potentialnder the existing conditions. In this manner one may btain the required potential with fewer turns of wire.

he life of such lamps as described above depends, of ourse, largely on the degree of exhaustion, but to somextent also on the shape of the block of refractory

material. Theoretically it


would seem that a small sphere of carbon enclosed in aphere of glass would not suffer deterioration from

molecular bombardment, for, the matter in the globeeing radiant, the molecules would move in straight linesnd would seldom strike the sphere obliquely. An

nterestin thou ht in connection with such a lam is, tha



n it " electricity" and electrical energy apparently mustmove in the same lines.

he use of alternating currents of very high frequency makes it possible to transfer, by electrostatic or

lectromagnetic induction through the glass of a lamp,ufficient energy to keep a fila-



IG. 121a.

IG. 121b.

ment at incandescence and so do away with the leading-iwires. Such lamps have been proposed, but for want of

roper apparatus they have not been successfully perated. Many forms of lamps on this principle withontinuous and broken filaments have been constructedy me and experimented upon. When using a secondary nclosed within the lamp, a condenser is advantageously ombined with the secondary. When the transference isffected by electrostatic induction, the potentials usedre, of course, very high with frequencies obtainable frommachine. For instance, with a condenser surface of fort

quare centimetres,

which is not impracticably large, and with glass of gooduality 1 ram. thick, using currents alternating twenty housand times a second, the potential required ispproximately 9,000 volts. This may seem large, but

ince each lamp may be included in- the secondary of aransformer of very small dimensions, it would not benconvenient, and, moreover, it would not produce fatalnjury. The transformers would all be preferably in seriehe regulation would offer no difficulties, as with currentf such frequencies it is very easy to maintain a constant

urrent.



n the accompanying engravings some of the types of amps of this kind are shown. Fig. 120 is such a lamp withbroken filament, and Figs. 121 A and 121 B one with a

ingle outside and inside coating and a single filament. Iave also made lamps with two outside and inside

oatings and a continuous loop connecting the latter. Sucamps have been operated by me with current impulses he enormous frequencies obtainable by the disruptiveischarge of condensers.

he disruptive discharge of a condenser is especially

uited for operating such lamps — with no outwardlectrical connections — by means of electromagneticnduction, the electromagnetic inductive effects beingxcessively high ; and I have been able to produce theesired incandescence with only a few short turns of wirncandescence may also be produced in this manner in a

imple closed filament.

eaving now out of consideration the practicability of sucamps, I would only say that they possess a beautiful andesirable feature, namely, that they can be rendered, at

will, more or less brilliant simply by altering the relative

osition of the outside and inside condenser coatings, ornducing and induced circuits.

When a lamp is lighted by connecting it to one terminalnly of the source, this may be facilitated by providing thlobe with an outside condenser coating, which serves at

he same time as a reflector, and connecting this to annsulated bod of some size. Lam s of this kind are



lustrated in Fig. 122 and Fig. 123. Fig. 124 shows thelan of connection. The brilliancy of the lamp may, in thisase, be regulated within wide limits by varying the size he insulated metal plate to which the coating isonnected.

t is likewise practicable to light with one leading wireamps such as illustrated in Fig. 116 and Fig. 117, by onnecting one

erminal of the lamp to one terminal of the source, and

he other to an insulated body of the required size. In allases the insulated body serves to give off the energy inthe surrounding space, and is equivalent to a return wire

Obviously, in the two last-named cases, instead of onnecting the wires to an insulated body, connections

may be made to the ground.

he experiments which will prove most suggestive and omost interest to the investigator are probably those

erformed with exhausted tubes. As might be anticipatesource of such rapidly alternating potentials is capable

xciting the tubes at a considerable distance, and the ligh

ffects produced are remarkable.

During my investigations in this line I endeavored toxcite



IG. 122. FIG. 123.

ubes, devoid of any electrodes, by electromagneticnduction, making the tube the secondary of the inductioevice, and passing through the primary the discharges o

Leyden jar. These tubes were made of many shapes,nd I was able to obtain luminous effects which I thenhought were due wholly to electromagnetic induction.ut on carefully investigating the phenomena I found tha

he effects produced were more of an electrostatic naturt may be attributed to this circumstance that this mode

f exciting tubes is very wasteful, namely, the primary ircuit being closed, the potential, and consequently the



lectrostatic inductive effect, is much diminished.

When an induction coil, operated as above described, issed, there is no doubt that the tubes are excited by lectrostatic induction, and that electromagnetic inductio

as little, if anything, to do with the phenomena.

his is evident from many experiments. For instance, if ube be taken in one hand, the observer being near theoil, it is brilliantly lighted and remains so no matter in

what position it is held relatively to the observer's body.

Were the action electromagnetic, the tube could not beghted when the observer's body is interposed between nd the coil, or at least its luminosity should beonsiderably diminished. When the tube is held exactly ver the centre of the coil —the latter being wound inections and the primary placed symmetrically to the

econdary— it may remain completely dark, whereas it iendered intensely luminous by moving it slightly to theight or left from the centre of the coil. It does not lightecause in the

IG. 124.



middle both halves of the coil neutralize each other, andhe electric potential is zero. If the action werelectromagnetic, the tube should light best in the planehrough the centre of the coil, since the electromagneticffect there should be a maximum. When an arc is

stablished between the terminals, the tubes and lamps he vicinity of the coil go out, but light up again when therc is broken, on account of the rise of potential. Yet thelectromagnetic effect should be practically the same inoth cases.

y placing a tube at some distance from the coil, andearer to one terminal — preferably at a point on the axf the coil — one may light it by touching the remoteerminal with an insulated body of some size or with theand, thereby raising the potential at that terminalearer to the tube. If the tube is shifted nearer to the coi

o that it is lighted by the action of the nearer termi-

al, it may be made to go out by holding, on an insulatedupport, the end of a wire connected to the remoteerminal, in the vicinity of the nearer terminal, by this

means counteracting the action of the latter upon the

ube. These effects are evidently electrostatic. Likewise,when a tube is placed at a considerable distance from theoil, the observer may, standing upon an insulatedupport between coil and tube, light the latter by pproaching the hand to it; or he may even render ituminous by simply stepping between it and the coil. Thi

would be impossible with electro-magnetic induction, for



he body of the observer would act as a screen.

When the coil is energized by excessively weak currentshe experimenter may, by touching one terminal of theoil with the tube, extinguish the latter, and may again

ght it by bringing it out of contact with the terminal andllowing a small arc to form. This is clearly due to theespective lowering and raising of the potential at thaterminal. In the above experiment, when the tube isghted through a small arc, it may go out when the arc isroken, because the electrostatic inductive effect alone is

oo weak, though the potential may be much higher ; butwhen the arc is established, the electrification of the endf the tube is much greater, and it consequently lights.

f a tube is lighted by holding it near to the coil, and in thand which is remote, by grasping the tube anywhere

with the other hand, the part between the hands isendered dark, and the singular effect of wiping out theght of the tube may be produced by passing the handuickly along the tube and at the same time withdrawinggently from the coil, judging properly the distance so

hat the tube remains dark afterwards.

f the primary coil is placed sidewise, as in Fig. 112 B fornstance, and an exhausted tube be introduced from thether side in the hollow space, the tube is lighted mostntensely because of the increased condenser action, andn this position the striae are most sharply defined. In all

hese experiments described, and in many others, thection is clearl electrostatic.



he effects of screening also indicate the electrostaticature of the phenomena and show something of theature of electrification through the air. For instance, if a

ube is placed in the direction of the axis of the coil, and a

nsulated metal plate be interposed, the tube willenerally increase in brilliancy, or if it l>e too far from thoil to light, it may even be rendered lumin-

us by interposing an insulated metal plate. Themagnitude of the effects depends to some extent on the

ize of the plate. But if the metal plate be connected'by awire to the ground, its interposition will always make theube go out even if it be very near the coil. In general, thnterposition of a body between the coil and tube,ncreases or diminishes the brilliancy of the tube, or itsacility to light up, according to whether it increases or

iminishes the electrification. When experimenting withn insulated plate, the plate should not be taken too larglse it will generally produce a weakening effect by reasof its great facility for giving off energy to theurroundings.

f a tube be lighted at some distance from the coil, and alate of hard rubber or other insulating substance be

nterposed, the tube may be made to go out. Thenterposition of the dielectric in this case only slightly ncreases the inductive effect, but diminishes considerabhe electrification through the air.

n all cases, then, when we excite luminosity in exhauste



ubes by means of such a coil, the effect is due to theapidly alternating electrostatic potential ; and,urthermore, it must be attributed to the harmoniclternation produced directly by the machine, and not tony superimposed vibration which might be thought to

xist. Such superimposed vibrations are impossible whenwe work with an alternate current machine. If a spring bradually tightened and released, it does not performndependent vibrations; for this a sudden release isecessary. So with the alternate currents from a dynamo

machine ; the medium is harmonically strained and

eleased, this giving rise to only one kind of waves ; audden contact or break, or a sudden giving way of theielectric, as in the disruptive discharge of a Leyden jar,re essential for the production of superimposed waves.

n all the last described experiments, tubes devoid of any

lectrodes may be used, and there is no difficulty inroducing by their means sufficient light to read by. Theght effect is, however, considerably increased by the usf phosphorescent bodies such as yttria, uranium glass,tc. A difficulty will be found when the phosphorescent

material is used, for with these powerful effects, it is

arried gradually away, and it is preferable to usematerial in the form of a solid.

nstead of depending on induction at a distance to lighthe tube, the same may be provided with an external —nd, if desired, also with an internal — condenser coating

nd it may then




e suspended anywhere in the room from a conductoronnected to one terminal of the coil, and in this manner oft illumination may be provided.

he ideal way of lighting a hall or room would, however,e



IG. 125.

o produce such a condition in it that an illuminatingevice could be moved and put anywhere, and that it is

ghted, no matter where it is put and without beinglectrically connected to

nything. I have been able to produce such a condition breating in the room a powerful, rapidly alternatinglectrostatic field. For this purpose I suspend a sheet of

metal a distance from the ceiling on insulating cords andonnect it to one terminal of the induction coil, the othererminal being preferably connected to the ground. Orlse I suspend two sheets as illustrated in Fig. 125, eachheet being connected with one of the terminals of theoil, and their size being carefully determined. Anxhausted tube may then be carried in the handnywhere between the sheets or placed anywhere, evenertain distance beyond them ; it remains alwaysuminous.

n such an electrostatic field interesting phenomena maye observed, especially if the alternations are kept low





f the energy, whereas in an electromagnetic alternatingeld the conductor tends to take up tlie least energy, the

waves being reflected with but little-loss.

his is one reason why it is difficult to excite an exhauste

ube, at a distance, by electromagnetic induction. I havewound coils of very large diameter and of many turns of wire, and connected a Geissler tube to the ends of the cowith the object of exciting the tube at a distance ; butven with the powerful inductive effects producible by ey den jar discharges, the tube could not be excited

nless at a very small distance, although some judgmentwas used as to the dimensions of the coil. I have alsoound that even the most powerful Leyden jar dischargere capable of exciting only feeble luminous effects in alosed exhausted tube, and even these effects uponhorough examination I have been forced to consider of

n electrostatic nature.

How then can we hope to produce the required effects atdistance by means of electromagnetic action, when eve

n the closest proximity to the source of disturbance,nder the most advantageous conditions, we can excite

ut faint luminosity '* It is true that when acting at aistance we have the resonance to help us out. We canonnect an exhausted tube, or whatever the illuminatingevice may be, with an insulated system of the properapacity, and so it may be possible to increase the effectualitatively, and only qualitatively, for we would not get

more energy through the device. So we may, by



esonance effect, obtain the required electromotive forcen an exhausted tube, and excite faint luminous effects,ut we cannot get enough energy to render the lightractically available, and a simple calculation, based onxperimental results, shows that even if all the energy

which a tube would receive at a certain distance from theource should be wholly converted into light, it wouldardly satisfy the practical requirements. Hence theecessity of directing, by means of a conducting circuit,he energy to the place of transformation. But in so doing

we cannot very sensibly depart from present methods,

nd all we could do would be to improve the 1 apparatus

rom these considerations it would seem that if this ideaway of lighting is to be rendered practicable it will be onl

y the use of electrostatic effects. In such a case the mosowerful electrostatic inductive effects are needed ; the

pparatus employed must, therefore, be capable of roducing high electrostatic potentials changing in value

with extreme rapidity. High frequencies are especially wanted, for practical considerations make it desirable to

eep down the potential. By the employment of machine

r, generally speaking, of any mechanical apparatus, butow frequencies can be reached ; recourse must,herefore, be had to some other means. The discharge ofondenser affords us a means of obtaining frequencies byar higher than are obtainable mechanically, and I haveccordingly employed condensers in the experiments to

he above end.



When the terminals of a high tension induction coil, Fig.20, are connected to a Leyden jar, and the latter isischarging dis-ruptively into a circuit, we may look upohe arc playing between the knobs as being a source of lternating, or generally speaking, undulating currents,

nd then we have to deal with the familiar system of aenerator of such currents, a circuit connected to it, and ondenser bridging the circuit. The condenser in such cas a veritable transformer, and since the frequency isxcessive, almost any ratio in the strength of the currentn both the branches may be obtained. In reality the

nalogy is not quite complete, for in the disruptiveischarge we have most generally a fundamental

nstantaneous variation of comparatively low frequency,nd a superimposed harmonic vibration, and the lawsoverning the flow of currents are not the same for both

n converting in this manner, the ratio of conversionhould not be too great, for the loss in the arc between thnobs increases with the square of the current, and if the

ar be discharged through very thick and shortonductors, with the view of obtaining a very rapidscillation, a very considerable portion of the energy

tored is lost. On the other hand, too small ratios are notracticable for many obvious reasons.

As the converted currents flow in a practically closedircuit, the electrostatic effects are necessarily small, andtherefore convert them into currents or effects of the

e uired character. I have effected such conversions in



everal ways. The preferred plan of connections islustrated in Fig. 127. The manner of operating renders asy to obtain by means of a small and inexpensivepparatus enormous differences of potential which haveeen usually obtained by means of large and expensive

oils. For this it is only necessary to take an ordinary mall coil, adjust to it a condenser and discharging circuitorming, the primary of an auxiliary small coil, andonvert upward. As the inductive effect of the primary urrents is excessively great, the second coil need haveomparatively but very few turns. By properly adjusting

he elements, remarkable results may be secured.

n endeavoring to obtain tlie required electrostatic effectn this manner, I have, as might be expected, encountere

many difficulties which I have been gradually overcominut I am not as yet prepared to dwell upon my

xperiences in this direction.

believe that the disruptive discharge of a condenser willay an important part in the future, for it offers vastossibilities, not only in the way of producing light in a

more efficient manner and in the line indicated by theory

ut also in many other respects.

or years the efforts of inventors have been directedowards obtaining electrical energy from heat by means he thermopile. It might seem invidious to remark thatut few know what is the real trouble with the

hermopile. It is not the inefficiency or small output —hou h these are reat drawbacks— but the fact that the



hermopile has its phylloxera, that is, that by constant uis deteriorated, which has thus far prevented its

IG. 126.

ntroduction on an industrial scale. Now that all modernesearch seems to point with certainty to the use of lectricity of excessively high tension, the question mustresent itself to many whether it is not possible to obtain

n a practicable manner this form of energy from heat. Wave been used to look upon an electrostatic machine as

laything, and somehow we couple with it the idea of thenefficient and impractical. But now we must think ifferently, for now w r e know that everywhere we haveo deal with the same forces, and that it is a mereuestion of inventing proper methods or apparatus forendering them available.

n the present systems oij electrical distribution, themployment of the iron with its wonderful magneticroperties allows us to reduce considerably the size of thpparatus ; but, in spite of this, it is still very umbersome. The more we progress in the study of

lectric and magnetic phenomena, the more we be-





ncountered in the arc of the discharge which I have beeble to overcome to a great extent; besides this, and thedjustment necessary for the proper working, no otherifficulties have been met with, and it was easy to operatrdinary lamps, and even motors, in this manner. The lin

eing connected to the ground, all the wires could beandled with perfect impunity, no matter how high theotential at the terminals of the condenser. In thesexperiments a high tension induction coil, operated fromattery or from an alternate current machine, wasmployed to charge the condenser ; but the induction coi

might be replaced by an apparatus of a different kind,apable of giving electricity of such high tension. In this

manner, direct or alternating currents may be convertednd in both cases the current-impulses may be of any esired frequency. "When the currents charging theondenser are of the

ame direction, and it is desired that the convertedurrents should also he of one direction, the resistance ofhe discharging circuit should, of course, be so chosen thahere are no oscillations.

n operating devices on the above plan I have observedurious phenomena of impedance which are of interest.or instance if a thick copper bar be bent, as indicated inig. 128, and shunted by ordinary incandescent lamps,

hen, by passing the discharge between the knobs, theamps may be brought to incandescence although they a

hort-circuited. When a large induction coil



IG. 128.

s employed it is easy to obtain rendered evident by theifferent degree of brilliancy of the lamps, as shownoughly in Fig. 12S. The nodes are never clearly delined,ut they are simply maxima and minima of potentialslong the bar. This is probably due to the irregularity of he arc between the knobs. In general when the above-escribed plan of conversion from high to low tension issed, the behavior of the disruptive discharge may be

losely studied. The nodes may also be investigated by means of an ordinarv Cardew voltmeter



which should be well insulated. Geissler tubes may also bghted across the points of the bent bar; in this case, of ourse, it is better to employ smaller capacities. I haveound it practicable to light up in this manner a lamp, andven a Geissler tube, shunted by a short, heavy block of

metal, and this result seems at first very curious. In facthe thicker the copper bar in Fig. 128, the better it is forhe success of the experiments, as they appear moretriking. When lamps with long slender filaments are usew T ill be often noted that the filaments are from time

o time violently vibrated, the vibration being smallest ahe nodal points. This vibration seems to be due to anlectrostatic action between the filament and the glass ofhe bulb.

n some of the above experiments it is preferable to usepecial lamps having a straight filament as shown in Fig.29. When such a lamp is used a still more curioushenomenon than those

ro. 129.



escribed may be observed. The lamp may be placedcross the copper bar and lighted, and by using somewhaarger capacities, or, in other words, smaller frequenciesr smaller impulsive impedances, the filament may berought to any desired degree of incandescence. But

Avhen the impedance is increased, a point is reachedwhen comparatively little current passes through thearbon, and most of it through the rarefied gas; orerhaps it may be more correct to state that the currentivides nearly evenly through both, in spite of thenormous difference in the resistance, and this would be

rue unless the gas and the filament behave differently. s then noted that the whole bulb is brilliantly illuminatednd the ends of the leading-in wires become incandescennd often throw off sparks in consequence of the violentombardment, but the carbon filament remains dark.his is illustrated in Fig. 129. Instead of the filament a

ingle

wire extending through the whole bulb may be used, andn this case the phenomenon would seem to be still morenteresting.

rom the above experiment it will be evident, that whenrdinary lamps are operated by the converted currents,hose should be preferably taken in which the platinum

wires are far apart, and the frequencies used should note too great, else the discharge will occur at the ends of he filament or in the base of the lamp between the

eading-in wires, and the lamp might then be damaged.



n presenting to you these results of my investigation onhe subject under consideration, I have paid only aassing notice to facts upon which I could have dwelt at

ength, and among many observations I have selectednly those which I thought most likely to interest you.

he field is wide and completely unexplored, and at evertep a new truth is gleaned, a novel fact observed.

How far the results here borne out are capable of practicpplications will be decided in the future. As regards theroduction of light, some results already reached are

ncouraging and make me confident in asserting that theractical solution of the problem lies in the direction Iave endeavored to indicate. Still, whatever may be the

mmediate outcome of these experiments I am hopefulhat they will only prove a step in further developmentowards the ideal and final perfection. The possibilities

which are opened by modern research are so vast thatven the most reserved must feel sanguine of the futureminent scientists consider the problem of utilizing oneind of radiation without the others a rational one. In anpparatus designed for the production of light by onversion from any form of energy into that of light, suc

result can never be reached, for no matter what therocess of producing the required vibrations, be itlectrical, chemical or any other, it will not be possible tobtain the higher light vibrations without going throughhe lower heat vibrations. It is the problem of impartingo a body a certain velocity without passing through allower velocities. But there is a ossibilit of obtainin



nergy not only in the form of light, but motive power,nd energy of any other form, in some more direct way rom the medium. The time will be when this will beccomplished, and the time has come when one may utteuch words before an enlightened audience without being

onsidered a visionary. AVe are whirling through

ndless space with an inconceivable speed, all around usverything is spinning, everything is moving, everywhers energy. There must be some way of availing ourselvesf this energy more directly. Then, with the light obtaine

rom the medium, with the power derived from it, withvery form of energy obtained without effort, from thetore forever inexhaustible, humanity will advance withiant strides. The mere contemplation of these

magnificent possibilities expands our minds, strengthensur hopes and fills our hearts with supreme delight.

HAPTEE XXVII.

XPERIMENTS WITH ALTERNATE CURRENTS OFHIGH POTENTIAL

AND HlGH FREQUENCY. 1

CANNOT find words to express how deeply I feel theonor of addressing some of the foremost thinkers of theresent time, and so many able scientific men, engineersnd electricians, of the country greatest in scientific

chievements.



he results which I have the honor to present before sucgathering I cannot call my own. There are among youot a few who can lay better claim than myself on any

eature of merit which this work may contain. I need notmention many names which are world-known— names o

hose among you who are recognized as the leaders in thnchanting science; but one, at least, I must mention —aame which could not be omitted in a demonstration of his kind. It is a name associated with the most beautifulnvention ever made: it is Crookes!

When I was at college, a good while ago, I read, in aranslation (for then I was not familiar with yourmagnificent language), the description of his experiment

n radiant matter. I read it only once in my life —thatme—yet every detail about that charming work I canemember to this day. Few are the books, let me say,

which can make such an impression upon the mind of atudent.

ut if, on the present occasion, I mention this name as onf many your Institution can boast of, it is because I hav

more than one reason to do so. For what I have to tell yo

nd to show you this evening concerns, in a largemeasure, that same vague world which Professor Crooke

as so ably explored; and, more than this, when I traceack the mental process which led me to these advances

—which even by myself cannot be considered trifling,ince they are so appreciated by you—I believe that thei

eal origin, that which started me to work in this



. Lecture delivered before the Institution of Electricalngineers, London, February, 1892.

irection, and brought me to them, after a long period ofonstant thought, was that fascinating little book which I

ead many years ago.

And now that I have made a feeble effort to express my omage and acknowledge my indebedness to him andthers among you, I will make a second effort, which Iope you will not find so feeble as the first, to entertain

ou.

Give me leave to introduce the subject in a few words.

A short time ago I had the honor to bring before ourAmerican Institute of Electrical Engineers some results

hen arrived at by me in a novel line of work. I need notssure you that the many evidences which I haveeceived that English scientific men and engineers werenterested in this work have been for me a great rewardnd encouragement. I will not dwell upon thexperiments already described, except with the view of

ompleting, or more clearly expressing, some ideas^dvanced by me before, and also with the view of endering the study here presented self-contained, and

my remarks on the subject of this evening's lectureonsistent.

his investigation, then, it goes without saying, deals witlternatin currents and to be more recise with



lternating currents of high potential and high frequencyust in how much a very high frequency is essential forhe production of the results presented is a question

which, even with my present experience, wouldmbarrass me to answer. Some of the experiments may

e performed with low frequencies; but very highrequencies are desirable, not only on account of the manffects secured by their use, but also as a convenient

means of obtaining, in the induction apparatus employedhe high potentials, which in their turn are necessary tohe demonstration of most of the experiments here

ontemplated.

Of the various branches of electrical investigation,erhaps the most interesting and the most immediately romising is that dealing with alternating currents. Therogress in this branch of applied science has been so

reat in recent years that it justifies the most sanguineopes. Hardly have we become familiar with one fact,

when novel .experiences are met and new avenues of esearch are opened. Even at this hour possibilities notreamed of before are, by the use of these currents,artly realized. As in nature all is ebb and tide, all is wav

motion, so it seems that in all branches of industry lternating currents—electric wave motion— will have thway.

oo INVENTIONS OF NIKOLA TESLA.

One reason, perhaps, why this branch of science is beingo ra idl develo ed is to be found in the interest which



ttached to its experimental study. We wind a simple rinf iron with coils ; we establish the connections to theenerator, and with wonder and delight we note theffects of strange forces which we bring into play, whichllow us to transform, to transmit and direct energy at

will. We arrange the circuits properly, and we see themass of iron and wires behave as though it were endowewith life, spinning a heavy armature, through invisibleonnections, with great speed and power— with thenergy possibly conveyed from a great distance. Webserve how the energy of an alternating current

raversing the wire manifests itself— not so much in thewire as in the surrounding space— in the most surprisingmanner, taking the forms of heat, light, mechanicalnergy, and, most surprising of all, even chemical affinity

All these observations fascinate us, and fill us with anntense desire to know more about the nature of these

henomena. Each day we go to our work in the hope of iscovering,— in the hope that some one, no matter who

may find a solution of one of the pending great problems—and each succeeding day we return to our task withenewed ardor ; and even if we are unsuccessful, our

work has not been in vain, for in these strivings, in these

fforts, we have found hours of untold pleasure, and weave directed our energies to the benefit of mankind.

We may take —at random, if you choose— any of themany experiments which may be performed with

lternating currents ; a few of which only, and by nomeans the most strikin form the sub ect of this



vening's demonstration ; they are all equally interestingqually inciting to thought.

Here is a simple glass tube from which the air has beenartially exhausted. I take hold of it ; I bring my body in

ontact with a wire conveying alternating currents of higotential, and the tube in my hand is brilliantly lighted. Iwhatever position I may put it, wherever I move it inpace, as far as I can reach, its soft, pleasing light persist

with undiminished brightness.

Here is an exhausted bulb suspended from a single wire.tanding on an insulated support, I grasp it, and alatinum button mounted in it is brought to vivid

ncandescence.

Here, attached to a leading wire, is another bulb, which, a

touch its metallic socket, is filled with magnificent colorf phosphorescent light.

HIGH FREQUENCY AND HIGH POTENTIALURRENTS. 201

Here still another, which by my fingers' touch casts ahadow —the Crookes shadow—of the stem inside of it.

Here, again, insulated as I stand on this platform, I bringmy body in contact with one of the terminals of theecondary of this induction coil—with the end of a wire

many miles long— and you see streams of light break orth from its distant end, which is set in violent vibratio



Here, once more, I attach these two plates of wire gauzeo the terminals of the coil ; I set them a distance apart,nd I set the coil to work. You may see a small spark pasetween the plates. I insert a thick plate of one of the beielectrics between them, and instead of renderingltogether impossible, as we are used to expect, I aid theassage of the discharge, which, as I insert the plate,

merely changes in appearance and assumes the form of uminous streams.

s there, I ask, can there be, a more interesting study han that of alternating currents ?

n all these investigations, in all these experiments, whicre so very, very interesting, for many years past— everince the greatest experimenter who lectured in this halliscovered its principle—we have had a steady ompanion, an appliance familiar to every one, a playthinnce, a thing of momentous importance now—the

nduction coil. There is no dearer appliance to thelectrician. From the ablest among you, I dare say, downo the inexperienced student, to your lecturer, we all hav

assed many delightful hours in experimenting with thenduction coil. We have watched its play, and thought anondered over the beautiful phenomena which it discloseo our ravished eyes. So well known is this apparatus, soamiliar are these phenomena to every one, that my ourage nearly fails me when I think that I have venture

o address so able an audience, that I have ventured tontertain ou with that same old sub ect. Here in realit



he same apparatus, and here are the same phenomena,nly the apparatus is operated somewhat differently, thehenomena are presented in a different aspect. Some of he results we find as expected, others surprise us, but aaptivate our attention, for in scientific investigation each

ovel result achieved may be the centre of a new eparture, each novel fact learned may lead to importanevelopments.

Usually in operating an induction coil we have set up aibration of moderate frequency in the primary, either b

means of an

nterrupter or break, or by the use of an alternator.arlier English investigators, to mention only pottiswoode and J. E. H. Gordon, have used a rapidreak in connection with the coil. Our knowledge and

xperience of to-day enables us to see clearly why theseoils under the conditions of the test did not disclose any emarkable phenomena, and why able experimentersailed to perceive many of the curious effects which haveince been observed.

n the experiments such as performed this evening, weperate the coil either from a specially constructedlternator capable of giving many thousands of reversalsf current per second, or, by disrupt!vely discharging aondenser through the primary, we set up a vibration inhe secondary circuit of a frequency of many hundred

housand or millions per second, if we so desire; and insin either of these means we enter a field as et



nexplored.

t is impossible to pursue an investigation in any novelne without finally making some interesting observationr learning some useful fact. That this statement is

pplicable to the subject of this lecture the many curiousnd unexpected phenomena which we observe afford aonvincing proof. By way of illustration, take for instancehe most obvious phenomena, those of the discharge of he induction coil.

Here is a coil which is operated by currents vibrating witxtreme rapidity , obtained by disruptively discharging aeyden jar. It would not surprise a student were the

ecturer to say that the secondary of this coil consists of amall length of comparatively stout wire ; it would noturprise him were the lecturer to state that, in spite of

his, the coil is capable of giving any potential which theest insulation of the turns is able to withstand ; butlthough he may be prepared, and even be indifferent aso the anticipated result, yet the aspect of the discharge he coil will surprise and interest him. Every one isamiliar with the discharge of an ordinary coil; it need no

e reproduced here. But, by way of contrast, here is aorm of discharge of a coil, the primary current of which ibrating several hundred thousand times per second.he discharge of an ordinary coil appears as a simple liner band of light. The discharge of this coil appears in theorm of powerful brushes and luminous streams issuing

rom all oints of the two strai ht wires attached to the



erminals of the secondary. (Fig. 130.)

ompare this phenomenon which you have just witnesse

with the discharge of a Holtz or Wimshurst machine —hat other interesting appliance so dear to thexperimenter. What a difference there is between thesehenomena ! And yet, had I made the necessary rrangements — which could have been made easily,

were it not that they would interfere with otherxperiments — I could have produced with this coil

parks which, had I the coil



IG. 131.

idden from your view and only two knobs exposed, evehe keenest observer among you would find it difficult, ifot impossible, to distinguish from those of an influence o

riction machine. This may be done in many ways —fornstance, by operating the induction coil which charges thondenser from an alternatin -current machine of ver



ow frequency, and preferably adjusting the dischargeircuit so that there are no oscillations set up in it. Wehen obtain in the secondary circuit, if the knobs are of he required size and properly set, a more or less

apid succession of sparks of great intensity and smalluantity, which possess the same brilliancy, and areccompanied by the same sharp crackling sound, as thosbtained from a friction or influence machine.

Another way is to pass through two primary circuits,

aving a common secondary, two currents of a slightly ifferent period, which produce in the secondary circuitparks occurring at comparatively long intervals. But,ven with the means at hand this evening, I may succeedn imitating the spark of a Holtz machine. For this purposestablish between the terminals of the coil which

harges the condenser a long, unsteady arc, which iseriodically interrupted by the upward current of airroduced by it. To increase the current of air I place onach side of the arc, and close to it, a large plate of mica.he condenser charged from this coil discharges into therimary circuit of a second coil through a small air gap,

which is necessary to produce a sudden rush of currenthrough the primary. The scheme of connections in theresent experiment is indicated in Fig. 131.

G is an ordinarily constructed alternator, supplying therimary P of an induction coil, the secondary s of which

harges the condensers or jars c c. The terminals of theecondar are connected to the inside coatin s of the ars



he outer coatings being connected to the ends of therimary p p of a second induction coil. This primary p pas a small air gap a b.

he secondary s of this coil is provided with knobs or

pheres K K of the proper size and set at a distanceuitable for the experiment.

A long arc is established between the terminals A B of he first induction coil. M M are the mica plates.

ach time the arc is broken between A and B the jars areuickly charged and discharged through the primary p producing a snapping spark between the knobs K K. Upohe arc forming between A and B the potential falls, andhe jars cannot be charged to such high potential as toreak through the air gap a ~b until the arc is again

roken by the draught.

n this manner sudden impulses, at long intervals, areroduced in the primary p p, which in the secondary sive a corresponding number of impulses of great

ntensity. If the secondary knobs or spheres, K K, are of

he proper size, the sparks show much resemblance tohose of a Holtz machine.

ut these two effects, which to the e,ye appear so very iffer-

ut, are only two of the many discharge phenomena. Wenl need to chan e the conditions of the test and a ain



we make other observations of interest.

When, instead of operating the induction coil as in the laswo experiments, we operate it from a high frequency lternator, as in the next experiment, a systematic study

f the phenomena is rendered much more easy. In suchase, in varying the strength and frequency of theurrents through the primary, we may observe liveistinct forms of discharge, which I have described in my

ormer paper on the subject before the Americannstitute of Electrical Engineers, May 20, 1891.

t would take too much time, and it would lead us too farrom the subject presented this evening, to reproduce allhese forms, but it seems to me desirable to show you onf them. It is a brush discharge, which is interesting in

more than one respect. Viewed from a near position it

esembles much a jet of gas escaping under greatressure. We know that the phenomenon is due to thegitation of the molecules near the terminal, and wenticipate that some heat must be developed by thempact of the molecules against the terminal or againstach other. Indeed, we find that the brush is hot, and onl

little thought leads us to the conclusion that, could weut reach sufficiently high frequencies, we could producebrush which would give intense light and heat, and

which would resemble in every particular an ordinary ame, save, perhaps, that both phenomena might not beue to the same agent—save, perhaps, that chemical

ffinity might not be electrical in its nature.



As the production 'of heat and light is here due to thempact of the molecules, or atoms of air, or something elsesides, and, as we can augment the energy simply by aising the potential, we might, even with frequenciesbtained from a dynamo machine, intensify the action to

uch a degree as to bring the terminal to melting heat. Buwith such low' frequencies we would have to deal alwayswith something of the nature of an electric current. If I

pproach a conducting object to the brush, a thin littlepark passes, yet, even with the frequencies used thisvening, the tendency to spark is not very great. So, for

nstance, if I hold a metallic sphere at some distancebove the terminal, you may see the whole space betweehe terminal and sphere illuminated by the streams

without the spark passing; and with the much higherrequencies obtainable by the disrup-

ve discharge of a condenser, were it not for the suddenmpulses^ which are comparatively few in number,parking would not occur even at very small distances.

However, with incomparably higher frequencies, whichwe may yet lind means to produce efficiently, and

rovided that electric impulses of such high frequencies

ould be transmitted through a conductor, the electricalharacteristics of the brush discharge would completely anish —no spark would pass, no shock would he felt —et we would still have to deal with an electrichenomenon, but in the broad, modern interpretation ofhe word. In my first paper, before referred to, I haveointed out the curious ro erties of the brush and



escribed the best manner of producing it, but I havehought it worth while to endeavor to express myself

more clearly in regard to this phenomenon, because of itbsorbing interest.

When a coil is operated with currents of very highreqency, beautiful brush effects may be produced, evenhe coil be of comparatively small dimensions. Thexperimenter may vary them in many ways, and, if it

were for nothing else, they afford a pleasing sight. Whatdds to their interest is that they may be produced with

ne single terminal as well as with two — in fact, oftenetter with one than with two.

ut of all the discharge phenomena observed, the mostleasing to the eye, and the most instructive, are thosebserved with a coil which is operated by means of the

isruptive discharge of a condenser. The power of therushes, the abundance of the sparks, when theonditions are patiently adjusted, is often amazing. Withven a very small coil, if it be so well insulated as to standdifference of potential of several thousand volts per

urn, the sparks may be so abundant that the whole coil

may appear a complete mass of fire.

uriously enough the sparks, when the terminals of theoil are set at a considerable distance, seem to dart invery possible direction as though the terminals wereerfectly independent of each other. As the sparks would

oon destroy the insulation, it is necessary to preventhem. This is best done b immersin the coil in a ood



quid insulator, such as boiled-out oil. Immersion in aquid may be considered almost an absolute necessity fohe continued and successful working of such a coil.

t is, of course, out of the question, in an experimental

ecture, with only a few minutes at disposal for theerformance of each experiment, to show these discharghenomena to advantage,

s, to produce each phenomenon at its best, a very carefdjustment is required. But even if imperfectly produced

s they are likely to be this evening, they are sufficientlytriking to interest an intelligent audience.

efore showing some of these curious effects I must, forhe sake of completeness, give a short description of theoil and other apparatus used in the experiments with th

isruptive discharge this evening.

t is contained in a box u (Fig. 13:2) of thick boards of ard



wood, covered on the outside with a zinc sheet z, which isarefully soldered all around. It might be advisable, in atrictly scientific investigation, when accuracy is of greatmportance, to do away with the metal cover, as it might

ntroduce many errors, principally on account of itsomplex action upon the coil, as a condenser of very smaapacity and as an electrostatic and electromagneticcreen. When the coil is used for such experiments as areere contemplated, the employment of the metal coverffers some practical advantages, but these are not of

ufficient importance to be dwelt upon.

he coil should be placed symmetrically to the metalover,

nd the space between should, of course, not be too smal

ertainly not less than, say, five centimetres, but muchmore if possible ; especially the two sides of the zinc box,which are at right angles to the axis of the coil, should beufficiently remote from the latter, as otherwise they

might impair its action and be a source of loss.

he coil consists of two s ools of hard rubber R K, held



part at a distance of 10 centimetres by bolts c and nutsw, likewise of hard rubber. Each spool comprises a tube T

f approximately 8 centimetres inside diameter, and 3millimetres thick, upon which are screwed two flanges F

, 24 centimetres square, the space between the flanges

eing about 3 centimetres. The secondary, s s, of the besutta percha-covered wire, has 26 layers, 10 turns inach, giving for each half a total of 260 turns. The twoalves are wound oppositely and connected in series, theonnection between both being made over the primary.his disposition, besides being convenient, has the

dvantage that when the coil is well balanced—that is,when both of its terminals TJ, T,, are connected to bodie

r devices of equal capacity—there is not much danger oreaking through to the primary, and the insulationetween the primary and the secondary need not behick. In using the coil it is advisable to attach to both

erminals devices of nearly equal capacity, as, when theapacity of the terminals is not equal, sparks will be apt tass to the primary. To avoid this, the middle point of thecondary may be connected to the primary, but this isot always practicable.

he primary p p is wound in two parts, and oppositely,pon a wooden spool w, and.the four ends are led out of he oil through hard rubber tubes t t. The ends of theecondary T t T t are also led out of the oil throughubber tubes t± t v of great thickness. The primary andecondary layers are insulated by cotton cloth, thehickness of the .insulation, of course, bearing some



roportion to the difference of potential between theurns of the differ" ent layers. Each half of the primary as four layers, 24 turns in each, this giving a total of 96urns. When both the parts are connected in series, thisives a ratio of conversion of about 1:2.7, and with the

rimaries in multiple, 1 :5.4; but in operating with very apidly alternating currents this ratio does not convey ven an approximate idea of the ratio of the E. M. F'S. inhe primary and secondary circuits. The coil is held inosition in the oil on wooden supports, there being aboutcentimetres

hickness of oil all round. Where the oil is not specially eeded, the space is filled with pieces of wood, and for thurpose principally the wooden box B surrounding the

whole is used.

he construction here shown is, of course, not the best oeneral principles, but I believe it is a good andonvenient one for the production of effects in which anxcessive potential and a very small current are needed.

n connection with the coil I use either the ordinary form

f* discharger or a modified form. In the former I haventroduced two changes which secure some advantages,nd which are obvious. If they are mentioned, it is only inhe hope that some experimenter may find them of use.

One of the changes is that the adjustable knobs A and B

Fig. 183), of the discharger are held in jaws of brass, .1T, by spring pressure, this allowing of turning them



uccessively into different

IG. 133.

ositions, and so doing away with the tedious process of requent polishing up.

he other change consists in the employment of a stronglectromagnet N s, which is placed with its axis at right

ngles to the line joining the knobs A and B, and producestrong magnetic field between them. The pole pieces of

he magnet are movable and properly formed so as torotrude between the brass knobs, in order to make theeld as intense as possible; but to prevent the discharge

rom jumping to the magnet the pole pieces are protecte

y a layer of mica, M M, of sufficient thickness; s t s l and9 2 .«? 2 are screws for fastening the wires. On each sidene of the screws is for large and the other for small

wires. L L are screws for fixing in position the rods R K,which support the knobs.

n another arrangement with the magnet I take the



ischarge between the rounded pole pieces themselves,which in such case are insulated and preferably providedwith polished brass caps.

he employment of an intense magnetic field is of

dvantage principally when the induction coil orransformer which charges the condenser is operated byurrents of very low frequency. In such a case theumber of the fundamental discharges between the knob

may be so small as to render the currents produced in thecondary unsuitable for many experiments. The intens

magnetic field then serves to blow out the arc betweenhe knobs as soon as it is formed, and the fundamentalischarges occur in quicker succession.

nstead of the magnet, a draught or blast of air may bemployed with some advantage. In this case the arc is

referably

IG. 134.

stablished between the knobs A B, in Fig. 181 (the knobl> being generally joined, or entirely done away with), a

n this disposition the arc is long and unsteady, and isasily affected by the draught.



When a magnet is employed to break the arc, it is bettero choose the connection indicated diagrammatically inig. 134, as in this case the currents forming the arc are

much more powerful, and the magnetic field exercises areater influence. The use of the magnet permits,owever, of the arc being replaced by a vacuum tube, buhave encountered great difficulties in working with anxhausted tube.

he other form of discharger used in these and similar

xperiments is indicated in Figs. 135 and 13H. It consistsf a number of brass pieces e c (Fig. 135), each of whichomprises a spherical middle portion /// with anxtension e below—which is merely used to fasten theiece in a lathe when polishing up the discharging

urface — and a column above, which consists of a knurleange f surmounted by a threaded stem I carrying a nutw, by means of which a wire is fastened to the column.

he flange/ conveniently serves for holding the brassiece when fastening the

IG. 135.



wire, and also for turning it in any position when itecomes necessary to present a fresh discharging surfacwo stout strips of hard rubber K K, with planed grooveg (Fig. 136) to fit the middle portion of the pieces c c,

erve to clamp the latter and hold them firmly in position

y means of two bolts c c (of which only one is shown)assing through the ends of the strips.

n the use of this kind of discharger I have found threerincipal advantages over the ordinary form. First, theielectric strength of a given total widtli of air space is

reater when a great many small air gaps are usednstead of one, which permits

IG. 136.

f working with a smaller length of air gap, and thatmeans smaller loss and less deterioration of the metal;econdly, by reason of splitting the arc up into smallerrcs, the polished surfaces are made to last much longer;nd, thirdly, the appa-



atus affords some gauge in the experiments. I usually she pieces by putting between them sheets of uniformhickness at a certain very small distance which is knownrom the experiments of Sir William Thomson to requireertain electromotive force to be bridged by the spark.

t should, of course, be remembered that the sparkingistance is much diminished as the frequency is increasey taking any number of spaces the experimenter has a

ough idea of the electromotive force, and he finds itasier to repeat an experiment, as he has not the trouble

f setting the knobs again and again. With this kind of ischarger I have been able to maintain an oscillating

motion without any spark being visible with the nakedye between the knobs, and they would not show a very ppeciable rise in temperature. This form of dischargelso lends itself to many arrangements of condensers and

ircuits which are often very convenient and time-savinghave used it preferably in a disposition similar to that

ndicated in Fig. 131, when the currents forming the arcre small.

may here mention that I have also used dischargers

with single or multiple air gaps, in which the dischargeurfaces were rotated with great speed. No particulardvantage was, however, gained by this method, exceptn cases where the currents from the condenser werearge and the keeping cool of the surfaces was necessary,nd in cases Avhen, the discharge not being oscillating ofself the arc as soon as established was broken b the a



urrent, thus starting the vibration at intervals in rapiduccession. I have also used mechanical interrupters in

many ways. To avoid the difficulties with frictionalontacts, the preferred plan adopted was to establish therc and rotate through it at great speed a rim of mica

rovided with many holes and fastened to a steel plate. Is understood, of course, that the employment of amagnet, air current, or other interrupter, produces noffect worth noticing, unless the self-induction, capacity nd resistance are so related that there are oscillations sp upon each interruption.

will now endeavor to show you some of the mostoteworthy of these discharge phenomena.

have stretched across the room two ordinary cottonovered wires, each about seven metres in length. They

re supported 011 insulating cords at a distance of abouthirty centimetres. I attach now to each of the terminalsf the coil one of the wires.

nd set the coil in action. Upon turning the lights off in thoom yon see the wires strongly illuminated by the

treams issuing abundantly from their whole surface inpite of the cotton covering, which may even be very hick. When the experiment is performed under goodonditions, the light from the wires is sufficiently intenseo allow distinguishing the objects in a room. To producehe best result it is, of course, necessary to adjust

arefully the capacity of the jars, the arc between thenobs and the len th of the wires. M ex erience is that



alculation of the length of the wires leads, in such case, to result whatever. The experimenter will do best to tak

he wires at the start very long, and then adjust by utting off first long pieces, and then smaller and smallernes as he approaches the right length.



A convenient way is to use an oil condenser of very smalapacity, consisting of two small adjustable metal plates,n connection with this and similar experiments. In suchase I take wires rather short and at the beginning set th

ondenser plates at maximum distance. If the streamsrom the wires increase by approach of the plates, theength of the wires is about right; if they diminish, the

wires are too long for that frequency and potential. Whencondenser is used in connection with experiments with

uch a coil, it should be an oil condenser by all means, as

n using an air condenser considerable energy might bewasted. The wires leading to the plates in the oil should bery thin, heavily coated with some insulating compoundnd provided with a conducting covering — thisreferably extending under the surface of the oil. Theonducting cover should not be too near the terminals, or

nds, of the wire, as a spark would be apt to jump fromhe wire to it. The conducting coating is used to diminishhe air losses, in virtue of its action as an electrostaticcreen. As to the size of the vessel containing the oil, andhe size of the plates, the experimenter gains at once andea from a rough trial. The size of the plates in oil is,

owever, calculable, as the dielectric losses are very mall.

n the preceding experiment it is of considerable interesto know what relation the quantity of the light emittedears to the frequency and potential of the electric

mpulses. My opinion is that the heat as well as light



ffects produced should be proportionate, undertherwise equal conditions of test, to the product of requency and square of potential, but the experimentalerification of the law, whatever it may be, would bexceedingly

ifficult. One thing is certain, at any rate, and that is, than augmenting the potential and frequency we rapidly ntensify the streams; and, though it may be very anguine, it is surely not altogether hopeless to expecthat we may succeed in producing a practical illuminant

n these lines. We would then be simply using burners orames, in which there would be no chemical process, noonsumption of material, but merely a transfer of energynd which would, in all probability, emit more light andess heat than ordinary flames.

he luminous intensity of the streams is, of course,onsiderably



IG. 137.

ncreased when they are focused upon a small surface.his may be shown by the following experiment:

attach to one of the terminals of the coil a wire w (Fig.37), bent in a circle of about 30 centimetres in diameter

nd to the other terminal I fasten a small brass sphere she surface of the wire being preferably equal to theurface of the sphere, and the centre of the latter being iline at right angles to the plane of the wire circle andassing through its centre. When the discharge isstablished under proper conditions, a luminous hollow

one is formed, and in the dark one-half of the brassphere is strongly illuminated, as shown in the cut.

y some artifice or other it is easy to concentrate thetreams

pon small surfaces and to produce very strong lightffects. Two thin wires may thus be rendered intensely uminous.

n order to intensify the streams the wires should be verhin and short ; but as in this case their capacity would benerall too small for the coil — at least for such a one a



he present— it is necessary to augment the capacity tohe required value, while, at the same time, the surface ohe wires remains very small. This may be done in many

ways.

Here, for instance, I have two plates, K K, of hard rubberFig. 188), upon which I have glued two very thin wires ww, so as to form a name. The wires may be bare orovered with the best insulation—it is immaterial for theuccess of the experiment. Well insulated wires, if nything, are preferable. On the back

IG. 138.

f each plate, indicated by the shaded portion, is a tinfoiloating t t. The plates are placed in line at a sufficient

istance to prevent a spark passing from one wire to thether. The two tinfoil coatings I have joined by a



onductor c, and the two wires I presently connect to theerminals of the coil. It is now easy, by varying thetrength and frequency of the currents through therimary, to find a point at which the capacity of theystem is best suited to the conditions, and the wires

ecome so strongly luminous that, when the light in theoom is turned off the name formed by them appears inrilliant letters.

t is perhaps preferable to perform this experiment withcoil operated from an alternator of high frequency, as

hen,

wing to the harmonic rise and fall, the streams are veryniform, though they are less abundant than whenroduced with such a coil as the present one. Thisxperiment, however, may be performed with low

requencies, but much less satisfactorily.

When two wires, attached to the terminals of the coil, areet at the proper distance, the streams between them

may be so intense as to produce a continuous luminousheet. To show this phenomenon I have here two circles,

ndc (Fig. 139), of rather stout wire, one being about 80entimetres and the other 30 centimetres in diameter. Tach of the terminals of the coil I attach one of the circleshe supporting wires are so bent that



IG. 139.

he circles may be placed in the same plane, coinciding as

early as possible. When the light in the room is turnedff and the coil set to work, you see the whole spaceetween the wires uniformly filled with streams, formingluminous disc, which could be seen from a considerableistance, such is the intensity of the streams. The outerircle could have been much larger than the present one

n fact, with this coil I have used much larger circles, andave been able to produce a strongly luminous sheet,overing an area of more than one square metre, which iremarkable effect with this very small coil. To avoidncer-

ainty, the circle has been taken smaller, and the area isow about 0. s uare metre.



he frequency of the vibration, and the quickness of uccession of the sparks between the knobs, affect to a

marked degree the appearance of the streams. When therequency is very low, the air gives way in more or less

he same manner, as by a steady difference of potential,nd the streams consist of distinct threads, generally mingled with thin sparks, which probably correspond tohe successive discharges occurring between the knobs.ut when the frequency is extremely high, and the arc o

he discharge produces a very loud and smooth sound —

howing both that oscillation takes place and that theparks succeed each other with great rapidity — then theuminous streams formed are perfectly uniform. To reachis result very small coils and jars of small capacity hould be used. I take two tubes of thick Bohemian glassbout 5 centimetres in diameter and 20 centimetres long

n each of the tubes I slip a primary of very thick copperwire. On the top of each tube I wind a secondary of muchhinner gutta-percha covered wire. The two secondariesonnect in series, the primaries preferably in multiple arhe tubes are then placed in a large glass vessel, at aistance of 10 to 15 centimetres from each other, on

nsulating supports, and the vessel is filled witli boiled-ouil, the oil reaching about an inch above the tubes. Theree ends of the secondary are lifted out of the coil andlaced parallel to each other at a distance of about tenentimetres. The ends which are scraped should beipped in the oil. Two four-pint jars joined in series may e used to dischar e throu h the rimar . When the



ecessary adjustments in the length and distance of thewires above the oil and in the arc of discharge are made, uminous sheet is produced between the wires which iserfectly smooth and tex-tureless, like the ordinary ischarge through a moderately exhausted tube.

have purposely dwelt upon this apparently insignificanxperiment. In trials of this kind the experimenterrrives at the startling conclusion that, to pass ordinary uminous discharges through gases, no particular degreef exhaustion is needed, but that the gas may be at

rdinary or even greater pressure. To accomplish this, aery high frequency is essential ; a high potential iskewise required, but this is merely an incidentalecessity. These experiments teach us that, inndeavoring to dis-

over novel methods of producing light by the agitation otoms, or molecules, of a gas, we need not limit ouresearch to the vacuum tube, but may look forward quiteriously to the possibility of obtaining the light effects

without the use of any vessel whatever, with air atrdinary pressure.

uch discharges of very high frequency, which renderuminous the air at ordinary pressures, we have probablccasion often to witness in Nature. I have no doubt that

f, as many believe, the aurora borealis is produced by udden cosmic disturbances, such as eruptions at the

un's surface, which set the electrostatic charge of thearth in an extremel ra id vibration the red low



bserved is not confined to the upper rarefied strata of he air, but the discharge traverses, by reason of its veryigh frequency, also the dense atmosphere in the form oglow, such as we ordinarily produce in a slightly

xhausted tube. If the frequency were very low, or even

more so, if the charge were not at all vibrating, the denseir would break down as in a lightning discharge.ndications of such breaking down of the lower densetrata of the air have been repeatedly observed at theccurence of this marvelous phenomenon ; but if it doesccur, it can only be attributed to the fundamental

isturbances, which are few in number, for the vibrationroduced by them would be far too rapid to allow aisruptive break. It is the original and irregular impulses

which affect the instruments; the superimposedibrations probably pass unnoticed.

When an ordinary low frequency discharge is passedhrough moderately rarefied air, the air assumes aurplish hue. If by some means or other we increase the

ntensity of the molecular, or atomic, vibration, the gashanges to a white color. A similar change occurs atrdinary pressures with electric impulses of very high

requency. If the molecules of the air around a wire aremoderately agitated, the brush formed is reddish or

iolet; if the vibration is rendered sufficiently intense, thtreams become white. We may accomplish this in variou

ways. In the experiment before shown with the two wirecross the room, I have endeavored to secure the result

ushin to a hi h value both the fre uenc and



otential; in the experiment with the thin wires glued onhe rubber plate I have concentrated the action upon aery small surface—in other words, I have worked with reat electric density.

A most curious form of discharge is observed with such aoil

when the frequency and potential are pushed to thextreme limit. To perform the experiment, every part ofhe coil should be heavily insulated, and only two small

pheres —or, better still, two sharp-edged metal discs (d, Fig. 140) of no more than a few centimetres iniameter—should be exposed to the air. The coil heresed is immersed in oil, and the ends of the secondary eaching out of the oil are covered with an air-tight coverf hard rubber of great thickness. All cracks, if there are

ny, should be carefully stopped up, so that the brushischarge cannot form anywhere except on the smallpheres or plates which are exposed to the air. In thisase, since there are no large plates or other bodies of apacity attached to the terminals, the coil is capable of axtremely rapid vibration.



IG. 140.

he potential may be raised by increasing, as far as thexperimenter judges proper, the rate of change of therimary current. With a coil not widely differing from th

resent, it is best to connect the two primaries in multiplrc; but if the secondary should have a much greaterumber of turns the primaries should preferably be used

n series, as otherwise the vibration might be too fast forhe secondary. It occurs under these conditions that

misty white streams break forth from the edges of the

iscs and spread out phantom-like into space. With thisoil, when fairly well produced, they are about 25 to 30entimetres long. When the hand is held against them noensation is produced, and a spark, causing a shock, jumprom

he terminal only upon the hand being brought muchearer. If the oscillation of the primary current isendered intermittent by some means or other, there is orresponding throbbing of the streams, and now theand or other conducting object may be brought in stillreater proximity to the terminal without a spark being

aused to jump.



Among the many beautiful phenomena which may beroduced with such a coil, I have here selected only thos

which appear to possess some features of novelty, andead us to some conclusions of interest. One will not tind t all difficult to produce in the laboratory, by means of i

many other phenomena which appeal to the eye evenmore than these here shown, but present no particulareature of novelty.

arly experimenters describe the display of sparksroduced by an ordinary large induction coil upon an

nsulating plate separating the terminals. Quite recently iemens performed some experiments in which fineffects were obtained, which were seen by many withnterest. No doubt large coils, even if operated withurrents of low frequencies, are capable of producingeautiful effects. But the largest coil ever made could not

y far, equal the magnificent display of streams andparks obtained from such a disruptive discharge coil

when properly adjusted. To give an idea, a coil such as thresent one will cover easily a plate of one metre iniameter completely with the streams. The best way toerform such experiments is to take a very thin rubber o

glass plate and glue on one side of it a narrow ring of nfoil of very large diameter, and on the other a circular

washer, the centre of the latter coinciding with that of thing, and the surfaces of both being preferably equal, so ao keep the coil well balanced. The washer and ring shoue connected to the terminals by heavily insulated thin

wires. It is easy in observing the effect of the capacity to



roduce a sheet of uniform streams, or a fine network ofhin silvery threads, or a mass of loud brilliant sparks,

which completely cover the plate.

ince I have advanced the idea of the conversion by

means of the disruptive discharge, in my paper before thAmerican Institute of Electrical Engineers at theeginning of the past year, the interest excited in it haseen considerable. It affords us a means for producingny potentials by the aid of inexpensive coils operatedrom ordinary systems of distribution, and—what is

erhaps more appreciated—it enables us to convertunvnt* <>!'

ny frequency into currents of any other lower or higherrequency. But its chief value will perhaps be found in thelp which it will afford us in the investigations of the

henomena of phosphorescence, which a disruptiveischarge coil is capable of exciting in innumerable cases

where ordinary coils, even the largest, would utterly fail.

onsidering its probable uses for many practicalurposes, and its possible introduction into laboratories

or scientific research, a few additional remarks as to theonstruction of such a coil will perhaps not be founduperfluous.

t is, of course, absolutely necessary to employ in such aoil wires provided with the best insulation.

Good coils may be produced by employing wires covered



with several layers of cotton, boiling the coil a long time iure wax, and cooling under moderate pressure. Thedvantage of such a coil is that it can be easily handled,ut it cannot probably give as satisfactory results as a co

mmersed in pure oil. Besides, it seems that the presence

f a large body of wax affects the coil disadvantageously,whereas this does not seem to be the case with oil.erhaps it is because the dielectric losses in the liquid aremaller.

have tried at iirst silk and cotton covered wires with oil

mmersions, but I have been gradually led to use gutta-ercha covered wires, which proved most satisfactory.Gutta-percha insulation adds, of course, to the capacity ohe coil, and this, especially if the coil be large, is a greatisadvantage when extreme frequencies are desired; bun the other hand, gutta-percha will withstand much

more than an equal thickness of oil, and this advantagehould be secured at any price. Once the coil has beenmmersed, it should never be taken out of the oil for morhan a few hours, else the gutta-percha will crack up andhe coil will not be worth half as much as before. Gutta-ercha is probably slowly attacked by the oil, but after a

mmersion of eight to nine months I have found no illffects.

have obtained two kinds of gutta-percha wire known inommerce : in one the insulation sticks tightly to the

metal, in the other it does not. Unless a special method is

ollowed to expel all air, it is much safer to use the iirst



ind. I wind the coil within an oil tank so that allnterstices are filled up with the oil. Between the layers Ise cloth boiled out thoroughly in oil, calculating thehickness according to the difference of potential

etween the turns. There seems not to be a very greatifference whatever kind of oil is used; I use paraffine ornseed oil.

o exclude more perfectly the air, an excellent way toroceed, and easily practicable with small coils, is the

ollowing: Construct a box of hardwood of very thick oards which have been for a long time boiled in oil. Theoards should be so joined as to safely withstand thexternal air pressure. The coil being placed and fastenedn position within the box, the latter is closed with atrong lid, and covered with closely fitting metal sheets,

he joints of which are soldered very carefully. On the towo small holes are drilled, passing through the metalheet and the wood, and in these holes two small glassubes are inserted and the joints made air-tight. One of he tubes is connected to a vacuum pump, and the other

with a vessel containing a sufficient quantity of boiled-ou

il. The latter tube has a very small hole at the bottom,nd is provided with a stopcock. When a fairly goodacuum has been obtained, the stopcock is opened andhe oil slowly fed in. Proceeding in this manner, it ismpossible that any big bubbles, which are the principalanger, should remain between the turns. The air is mos

ompletely excluded, probably better than by boiling out



which, however, when gutta-percha coated wires aresed, is not practicable.

or the primaries I use ordinary line wire with a thick otton coating. Strands of very thin insulated wires

roperly interlaced would, of course, be the best tomploy for the primaries, but they are not to be had.

n an experimental coil the size of the wires is not of greamportance. In the coil here used the primary is No. 12nd the secondary No. 24 Brown & Sharpe gauge wire;

ut the sections may be varied considerably. It wouldnly imply different adjustments ; the results aimed atwould not be materially affected.

have dwelt at some length upon the various forms of rush discharge because, in studying them, we not only

bserve phenomena which please our eye, but also affords food for thought, and lead us to conclusions of practicamportance. In the use of alternating currents of very igh tension, too much precaution cannot be taken torevent the brush discharge. In a main conveying suchurrents, in an induction coil or transformer, or in a

ondenser, the brush discharge is a source of great dangeo the insulation. In a condenser, especially, the gaseous

matter must

e most carefully expelled, for in it the charged surfacesre near each other, and if the potentials are high, just

ssure as a weight will fall if let go, so the insulation willive way if a single gaseous bubble of some size be



resent, whereas, if all gaseous matter were carefully xcluded, the condenser would safely withstand a muchigher difference of potential. A main conveyinglternating currents of very high tension may be injured

merely by a blow hole or small crack in the insulation, th

more so as a blowhole is apt to contain gas at low pressurand as it appears almost impossible to completely bviate such little imperfections, I am led to believe that

n our future distribution of electrical energy by currentsf very high tension, liquid insulation will be used. Theost is a great drawback, but if we employ an oil as an

nsulator the distribution of electrical energy withomething like 100,000 volts, and even more, becomes,t least with higher frequencies, so easy that it could beardly called an engineering feat. With oil insulation andlternate current motors, transmissions of power can beffected with safety and upon an industrial basis at

istances of as much as a thousand miles.

A peculiar property of oils, and liquid insulation in generawhen subjected to rapidly changing electric stresses, is to

isperse any gaseous bubbles which may be present, andiffuse them through its mass, generally long before any

njurious break can occur. This feature may be easily bserved with an ordinary induction coil by taking therimary out, plugging up the end of the tube upon whichhe secondary is wound, and filling it with some fairly ransparent insulator, such as paraffme oil. A primary ofiameter something like six millimetres smaller than the

nside of the tube may be inserted in the oil. When the co



s set to work one may see, looking from the top throughhe oil, many luminous points—air bubbles which areaught by inserting the primary, and which are rendereduminous in consequence of the violent bombardment.he occluded air, by its impact against the oil, heats it; th

il begins to circulate, carrying some of the air along with, until the bubbles are dispersed and the luminous poinisappear. In this manner, unless large bubbles areccluded in such way that circulation is renderedmpossible, a damaging break is averted, the only effecteing a moderate warming up of the oil. If, instead of the

quid, a solid insulation, no matter how thick, were usedbreaking through and injury of the apparatus would be

nevitable.

he exclusion of gaseous matter from any apparatus inwhich the dielectric is subjected to more or less rapidly

hanging electric forces is, however, not only desirable inrder to avoid a possible injury of the apparatus, but alson account of economy. In a condenser, for instance, as

ong as only a solid or only a liquid dielectric is used, theoss is small; but if a gas under ordinary or small pressure present the loss may be very great. Whatever the

ature of the force acting in the dielectric may be, iteems that in a solid or liquid the molecular displacemenroduced by the force is small: hence the product of forcnd displacement is insignificant, unless the force be veryreat; but in a gas the displacement, and therefore thisroduct, is considerable ; the molecules are free to movehey reach high speeds, and the energy of their impact is



ost in heat or otherwise. If the gas be strongly ompressed, the displacement due to the force is mademaller, and the losses are reduced.

n most of the succeeding experiments I prefer, chiefly o

ccount of the regular and positive action, to employ thelternator before referred to. This is one of the severalmachines constructed by me for the purpose of thesenvestigations. It has 384 pole projections, and is capablef giving currents of a frequency of about 10,000 perecond. This machine has been illustrated and briefly

escribed in my first paper before the American Institutf Electrical Engineers, May 20th, 1891, to which I havelready referred. A more detailed description, sufficient nable any engineer to build a similar machine, will beound in several electrical journals of that period.

he induction coils operated from the machine are rathemall, containing from 5,000 to 15,000 turns in theecondary. They are immersed in boiled-out linseed oil,ontained in wooden boxes covered with zinc sheet.

have found it advantageous to reverse the usual positio

f the wires, and to wind, in these coils, the primaries onhe top; thus allowing the use of a much larger primary,

which, of course, reduces the danger of overheating andncreases the output of the coil. I make the primary onach side at least one centimetre shorter than theecondary, to prevent the breaking through on the ends,

which would surely occur unless the insulation on the topf the secondar be ver thick and this of course would



e disadvantageous.

When the primary is made movable, which is necessary

ome experiments, and many times convenient for the

urposes of adjustment, I cover the secondary with waxnd turn it off in a lathe to a diameter slightly smallerhan the inside of the primary coil. The latter I provide

with a handle reaching out of the oil, which serves to shifin any position along the secondary.

will now venture to make, in regard to the generalmanipulation of induction coils, a few observations bearinpon points which have not been fully appreciated inarlier experiments with such coils, and are even now ften overlooked.

he secondary of the coil possesses usually such a highelf-induction that the current through the wire isnappreciable, and may be so even when the terminals aroined by a conductor of small resistance. If capacity isdded to the terminals, the self-induction is counteractednd a stronger current is made to flow through the

econdary, though its terminals are insulated from eachther. To one entirely unacquainted with the properties lternating currents nothing will look more puzzling. Thieature was illustrated in the experiment performed athe beginning with the top plates of wire gauze attachedo the terminals and the rubber plate. When the plates o

wire gauze were close together, and a small arc passedetween them, the arc prevented a strong current from



assing through the secondary, because it did away withhe capacity on the terminals; when the rubber plate wanserted between, the capacity of the condenser formedounteracted the self-induction of the secondary, atronger current passed now, the coil performed more

work, and the discharge was by far more powerful.

he first thing, then, in operating the induction coil is toombine capacity with the secondary to overcome theelf-induction. If the frequencies and potentials are veryigh, gaseous matter should be carefully kept away from

he charged surfaces. If Leyden jars are used, they shoule immersed in oil, as otherwise considerable dissipationmay occur if the jars are greatly strained. When highrequencies are used, it is of equal importance to combincondenser with the primary. One may use a condenser

onnected to the ends of the primary or to the terminals

f the alternator, but the latter is not to be recommendes the machine might be injured. The best way isndoubtedly to use the condenser in series with therimary and with the alternator, and to adjust its capacito as to annul the


elf-induction of both the latter. The condenser should bdjustable by very small steps, and for a finer adjustmensmall oil condenser with movable plates may be used

onveniently.

think it best at this uncture to brin before ou a



henomenon, observed by me some time ago, which tohe purely scientific investigator may perhaps appear

more interesting than any of the results which I have therivilege to present to you this evening.

t may be quite properly ranked among the brushhenomena — in fact, it is a brush, formed at, or near, aingle terminal in high vacuum.

n bulbs provided with a conducting terminal, though it bf



IG. 141.

IG. 142.

luminum, the brush -has but an ephemeral existence,nd cannot, unfortunately, be indefinitely preserved in it

most sensitive state, even in a bulb devoid of any onducting electrode. In studying the phenomenon, by a

means a bulb having no leading-in wire should be used. I

ave found it best to use bulbs constructed as indicated iigs. 141 and 142.

n Fig. 141 the bulb comprises an incandescent lamp glob, in the neck of which is sealed a barometer tube &, thend of which is blown out to form a small sphere s. This

phere should be sealed as closely as possible in theentre of the large globe. Before sealing, a thin tube t, of luminum sheet, may be slipped in the barometer tube,ut it is not important to employ it.

he small hollow sphere s is filled with some conducting

owder, and a wire w is cemented in the neck for theur ose of connectin the conductin owder with the



enerator.

he construction shown in Fig. 142 was chosen in order temove from the brush any conducting body which mighossibly affect it. The bulb consists in this case of a lamp

lobe Z, which has a neck n, provided with a tube b andmall sphere s, sealed to it, so that two entirely ndependent compartments are formed, as indicated inhe drawing. When the bulb is in use the neck n isrovided with a tinfoil coating, which is connected to theenerator and acts inductively upon the moderately

arefied and highly conducted gas inclosed in the neck.rom there the current passes through the tube b intohe small sphere *, to act by induction upon the gasontained in the globe L.

t is of advantage to make the tube £-very thick, the hol

IG. 143.



hrough it very small, and to blow the sphere * very thint is of the greatest importance that the sphere * belaced in the centre of the globe L.

igs. 143, 144 and 145 indicate different forms, or stages

f the brush. Fig. 143 shows the brush as it first appearsn a bulb provided with a conducting terminal ; but, as inuch a bulb it very soon disappears — often after a few

minutes — I will confine myself to the description of thehenomenon as seen in a bulb without conductinglectrode. It is observed under the following conditions:

When the globe L (Figs. 141 and 142) is exhausted to aery high degree, generally the bulb is not excited upononnecting the wire w (Fig. 141) or the tinfoil coating of he bulb (Fig.


42) to the terminal of the induction coil. To excite it, it isually sufficient to grasp the globe L with the hand. An

ntense phosphorescence then spreads at tirst over thelobe, but soon gives place to a white, misty light. Shortly

fterward one may notice that the luminosity is unevenlyistributed in the globe, and after passing the current forome time the bulb appears as in Fig. 144. From thistage the phenomenon will gradually pass to thatndicated in Fig. 145, after some minutes, hours, days or

weeks, according as the bulb is worked. Warming the bu

r increasing the potential hastens the transit.



When the brush assumes the form indicated in Fig. 145, may be brought to a state of extreme sensitiveness to

lectrostatic



IG. 144.

IG. 145.

nd magnetic influence. The bulb hanging straight downrom a wire, and all objects being remote from it, thepproach of the observer at a few paces from the bulb wause the brush to fly to the opposite side, and if he walkround the bulb it will always keep on the opposite side.

may begin to spin around the terminal long before iteaches that sensitive stage. When it begins to turnround, principally, but also before, it is affected by a

magnet, and at a certain stage it is susceptible to magnetnfluence to an astonishing degree. A small permanent

magnet, with its poles at a distance of no more than twoentimetres, will aft'ect it visibly at a distance of two

metres, slowing down or ac-•elerating the rotationccording to how it is held relatively to

he brush. I think I have observed that at the stage wheis most sensitive to magnetic, it is not most sensitive to

lectrostatic, influence. My explanation is, that thelectrostatic attraction between the brush and the glass he bulb, which retards the rotation, grows much quickehan the magnetic influence when the intensity of thetream is increased.

When the bulb han s with the lobe L down the rotation



s always clockwise. In the southern hemisphere it wouldccur in the opposite direction and on the equator therush should not turn at all. The rotation may beeversed by a magnet kept at some distance. The brushotates best, seemingly, when it is at right angles to the

nes of force of the earth. It very likely rotates, when ats maximum speed, in synchronism with the alternationay, 10,000 times a second, The rotation can be slowedown or accelerated by the approach or receding of thebserver, or any conducting body, but it cannot beeversed by putting the bulb in any position. When it is i

he state of the highest sensitiveness and the potential orrequency be varied, the sensitiveness is rapidly iminished. Changing either of these but little willenerally stop the rotation. The sensitiveness is likewiseffected by the variations of temperature. To attain greaensitiveness it is necessary to have the small sphere s in

he centre of the globe Z, as otherwise the electrostaticction of the glass of the globe will tend to stop theotation. The sphere s should be small and of uniformhickness ; any dissymmetry of course has the effect toiminish the sensitiveness.

he fact that the brush rotates in a delinite direction in aermanent magnetic tield seems to show that inlternating currents of very high frequency the positivend negative impulses are not equal, but that one alwaysreponderates over the other.

Of course, this rotation in one direction may be due to th



ction of the two elements of the same current upon eachther, or to the action of the field produced by one of thelements upon the other, as in a series motor, withoutecessarily one impulse being stronger than the other.he fact that the brush turns, as far as I could observe, i

ny position, would speak for this view. In such case itwould turn at any point of the earth's surface. But, on thther hand, it is then hard to explain why a permanent

magnet should reverse the rotation, and one must assumhe preponderance of impulses of one kind.

As to the causes of the formation of the brush or stream,

hink it is due to the electrostatic action of the globe andhe dissymmetry of the parts. If the small bulb * and thelobe Z were perfect concentric spheres, and the glasshroughout of the same thickness and quality, I think the

rush would not form, as the tendency to pass would bequal on all sides. That the formation of the stream is duo an irregularity is apparent from the fact that it has theendency to remain in one position, and rotation occurs

most generally only when it is brought out of this positiony electrostatic or magnetic influence. When in an

xtremely sensitive state it rests in one position, mosturious experiments may be performed with it. Fornstance, the experimenter may, by selecting a properosition, approach the hand at a certain considerableistance to the bulb, and he may cause the brush to passif by merely stiffening the muscles of the arm. When it

egins to rotate slowly, and the hands are held at a prop



istance, it is impossible to make even the slightestmotion without producing a visible effect upon the brushA metal plate connected to the other terminal of the coil

ffects it at a great distance, slowing down the rotationften to one turn a second.

am firmly convinced that such a brush, when we learnow to produce it properly, will prove a valuable aid in th

nvestigation of the nature of the forces acting in anlectrostatic or magnetic field. If there is any motion

which is measurable going on in the space, such a brush

ught to reveal it. It is, so to speak, a beam of light,rictionless, devoid of inertia.

think that it may find practical applications inelegraphy. With such a brush it would be possible to senispatches across the Atlantic, for instance, with any

peed, since its sensitiveness may be so great that thelightest changes will affect it. If it were possible to makehe stream more intense and very narrow, its deflectionsould be easily photographed.

have been interested to find whether there is a rotation

f the stream itself, or whether there is simply a stressraveling around the bulb. For this purpose I mounted aght mica fan so that its vanes were in the path of therush. If the stream itself was rotating the fan would bepun around. I could produce no distinct rotation of thean, although I tried the experiment repeatedly ; but as

he fan exerted a noticeable influence on the stream, andhe a arent rotation of the latter was in this case neve



uite satisfactory, the experiment did not appear to beonclusive.

have been unable to produce the phenomenon with theisruptive discharge coil, although every other of these

henomena can be well produced by it — many, in fact,much better than with coils operated from an alternator

t may be possible to produce the brush by impulses of ne direction, or even by a steady potential, in which caswould be still more sensitive to magnetic influence.

n operating an induction coil with rapidly alternatingurrents, we realize with astonishment, for the first timehe great importance of the relation of capacity, self-nduction and frequency as regards the general results.he effects of capacity are the most striking, for in these

xperiments, since the self-induction and frequency bothre high, the critical capacity is very small, and need beut slightly varied to produce a very considerable changhe experimenter may bring his body in contact with th

erminals of the secondary of the coil, or attach to one oroth terminals insulated bodies of very small bulk, such

s bulbs, and lie may produce a considerable rise or fall ootential, and greatly affect the now of the currenthrough the primary. In the experiment before shown, in

which a brush appears at a wire attached to one terminand the wire is vibrated when the experimenter bringsis insulated body in contact with the other terminal of

he coil, the sudden rise of potential was made evident,



may show you the behavior of the coil in anothermanner which possesses a feature of some interest. I

ave here a little light fan of aluminum sheet, fastened toneedle and arranged to rotate freely in a metal piece

crewed to one of the terminals of the coil. When the coil

s set to work, the molecules of the air are rhythmically ttracted and repelled. As the force with which they areepelled is greater than that with which they arettracted, it results that there is a repulsion exerted onhe surfaces of the fan. If the fan were made simply of a

metal sheet, the repulsion would be equal on the opposit

ides, and would produce no effect. But if one of thepposing surfaces is screened, or if, generally speaking,he bombardment on this side is weakened in some way r other, there remains the repulsion exerted upon thether, and the fan is set in rotation. The screening is bestffected by fastening upon one of the opposing sides of th

an insulated conducting coatings, or, if the fan is made inhe shape of an ordinary propeller screw, by fastening onne


ide, and close to it, an insulated metal plate. The staticcreen may, however, be omitted, and simply a thicknesf insulating material fastened to one of the sides of thean.

o show the behavior of the coil, the fan may be placed

pon the terminal and it will readily rotate when the coils o erated b currents of ver hi h fre uenc . With a



teady potential, of course, and even with alternatingurrents of very low frequency, it would not turn, becausf the very slow exchange of air and, consequently,maller bombardment; but in the latter case it might turf the potential were excessive. With a pin wheel, quite th

pposite rule holds good; it rotates best with a steady otential, and the eifort is the smaller the higher therequency. Now, it is very easy to adjust the conditions shat the potential is normally not sufficient to turn thean, but that by connecting the other terminal of the coil

with an insulated body it rises to a much greater value, s

s to rotate the fan, and it is likewise possible to stop theotation by connecting to the terminal a body of differentize, thereby diminishing the potential.

nstead of using the fan in this experiment, we may usehe " electric " radiometer with similar effect. But in this

ase it will be found that the vanes will rotate only at higxhaustion or at ordinary pressures; they will not rotatet moderate pressures, when the air is highly conductinghis curious observation was made conjointly by rofessor Crcokes and myself. I attribute the result to thigh conductivity of the air, the molecules of which then

o not act as independent carriers of electric charges, buct all together as a single conducting body. In such casef course, if there is any repulsion at all of the moleculesrom the vanes, it must be very small. It is possible,owever, that the result is in part due to the fact that thereater part of the discharge passes from the leading-in

wire throu h the hi hl conductin as instead of assin



ff from the conducting vanes.

n trying the preceding experiment with the electricadiometer the potential should not exceed a certain limis then the electrostatic attraction between the vanes an

he glass of the bulb may be so great as to stop theotation.

A most curious feature of alternate currents of highrequencies and potentials is that they enable us toerform many experiments by the use of one wire only.

n many respects this feat, ure is of great interest.

n a type of alternate current motor invented by me somears ago I produced rotation by inducing, by means of aingle alternating current passed through a motor circuitn the mass or other circuits of the motor, secondary

urrents, which, jointly with the primary or inducingurrent, created a moving field of force. A simple butrude form of such a motor is obtained by winding uponn iron core a primary, and close to it a secondary coil,

oining the ends of the latter and placing a freely movablmetal disc within the influence of the field produced by

oth. The iron core is employed for obvious reasons, buts not essential to the operation. To improve the motor,he iron core is made to encircle the armature. Again tomprove, the secondary coil is made to partly overlap therimary, so that it cannot free itself from a strong

nductive action of the latter, repel its lines as it may.

Once more to improve, the proper difference of phase isbtained between the rimar and secondar currents b



condenser, self-induction, resistance or equivalentwindings.

had discovered, however, that rotation is produced by means of a single coil and core; my explanation of the

henomenon, and leading thought in trying thexperiment, being that there must be a true time lag inhe magnetization of the core. I remember the pleasure ad when, in the writings of Professor Ayrton, whichame later to my hand, I found the idea of the time lagdvocated. Whether there is a true time lag, or whether

he retardation is due to eddy currents circulating inminute paths, must remain an open question, but the facs that a coil wound upon an iron core and traversed by alternating current creates a moving field of force, capabf setting an armature in rotation. It is of some interest, onjunction with the historical Arago experiment, to

mention that in lag or phase motors I have producedotation in the opposite direction to the moving field,

which means that in that experiment the magnet may nootate, or may even rotate in the opposite direction to th

moving disc. Here, then, is a motor (diagrammatically lustrated in Fig. 146), comprising a coil and iron core,

nd a freely movable copper disc in proximity to theatter.

o demonstrate a novel and interesting feature, I have,or a reason which I will explain, selected this type of

motor. When the ends of the coil are connected to the

erminals of an alternator the disc is set in rotation. But i



s not this experiment, now well known, which I desire toerform. What I wish to

how you is that this motor rotates with one singleonnection between it and the generator; that is to say,

ne terminal of the motor is connected to one terminal ofhe generator — in this case the secondary of a high-ension induction coil —the other terminals of motor andenerator being insulated in space. To produce rotation is generally (but not absolutely) necessary to connect theree end of the motor coil to an insulated body of some

ize. The experimenter's body is more than sufficient. If e touches the free terminal with an object held in theand, a current passes through the coil and the copperisc is set in rotation. If an exhausted tube is put in serie

with the coil, the tube lights brilliantly, showing theassage of a strong current. In-



IG. 146.

tead of the experimenter's body, a small metal sheetuspended on a cord may be used with the same result. I

his case the plate acts as a condenser in series with theoil. It counteracts the self-induction of the latter andllows a strong current to pass. In such a combination, threater the self-induction of the coil the smaller need behe plate, and this means that a lower frequency, orventually a lower potential, is required to operate the

motor. A single coil wound upon a core has a high self-nduction ; for this reason, principally, this type of motor

was chosen to perform the experiment. Were a secondarlosed coil wound upon the core, it would tend to diminishe self-

nduction, and then it would be necessary to employ amuch higher frequency and potential. Neither would be

dvisable, for a higher potential would endanger thensulation of the small primary coil, and a higherrequency would result in a materially diminished torque

t should be remarked that when such a motor with alosed secondary is used, it is not at all easy to obtainotation with excessive frequencies, as the secondary cuff almost completely the lines of the primary—and this,f course, the more, the higher the frequency—and allow

he passage of but a minute current. In such a case, unlehe secondar is closed throu h a condenser it is almost



ssential, in order to produce rotation, to make therimary and secondary coils overlap each other more or

ess.

ut there is an additional feature of interest about this

motor, namely, it is not necessary to have even a singleonnection between the motor and generator, except,erhaps, through the ground; for not only is an insulatedlate capable of giving off energy into space, but it iskewise capable of deriving it from an alternatinglectrostatic field, though in the latter case the available

nergy is much smaller. In this instance one of the motorerminals is connected to the insulated plate or body ocated within the alternating electrostatic field, and thether terminal preferably to the ground.

t is quite possible, however, that such " no wire " motor

s they might be called, could be operated by conductionhrough the rarefied air at considerable distances.

Alternate currents, especially of high frequencies, passwith astonishing freedom through even slightly rarefied

ases. The upper strata of the air are rarefied. To reach umber of miles out into space requires the overcoming

ifficulties of a merely mechanical nature. There is nooubt that with the enormous potentials obtainable by he use of high frequencies and oil insulation, luminousischarges might be passed through many miles of arefied air, and that, by thus directing the energy of

many hundreds or thousands of horsepower, motors or

amps might be operated at considerable distances from



tationary sources. But such schemes are mentionedmerely as possibilities. We shall have no need to transmi

ower in this way. We shall have no need to transmitower at all. Ere many generations pass, our machinery

will be driven by a power obtainable at any point of the

niverse. This idea is

ot novel. Men have been led to it long ago by instinct oreason. It has been expressed in many ways, and in manlaces, in the history of old and new. We find it in theelightful myth of Antheus, who derives power from the

arth; we find it among the subtle speculations of one of our splendid mathematicians, and in many hints andtatements of thinkers of the present time. Throughoutpace there is energy. Is this energy static or kinetic ? Iftatic our hopes are in vain; if kinetic—and this we know s, for certain—then it is a mere question of time when

men will succeed in attaching their machinery to the verwheelwork of nature. Of all, living or dead, Crookes came

earest to doing it. His radiometer will turn in the light oay and in the darkness of the night; it will turnverywhere where there is heat, and heat is everywhereut, unfortunately, this beautiful little machine, while it

oes down to posterity as the most interesting, mustkewise be put on record as the most inefficient machinever invented!

he preceding experiment is only one of many equally nteresting experiments which may be performed by the

se of only one wire with alternations of high potential



nd frequency. We may connect an insulated line to aource of such currents, we may pass an inappreciableurrent over the line, and on any point of the same we arble to obtain a heavy current, capable of fusing a thick opper wire. Or we may, by the help of some artifice,

ecompose a solution in any electrolytic cell by connectinnly one pole of the cell to the line or source of energy. Owe may, by attaching to the line, or only bringing into its

icinity, light up an incandescent lamp, an exhaustedube, or a phosphorescent bulb.

However impracticable this plan of working may appearn many cases, it certainly seems practicable, and evenecommend-able, in the production of light. A perfectedamp would require but little energy, and if wires weresed at all we ought to be able to supply that energy

without a return wire.

t is now a fact that a body may be rendered incandescenr phosphorescent by bringing it either in single contactr merely in the vicinity of a source of electric impulses ohe proper character, and that in this manner a quantityf light sufficient to afford a practical illuminant may be

roduced. It is, therefore, to say the least, worth while tottempt to determine the best conditions and to inventhe best appliances for attaining this object.

ome experiences have already been gained in thisirection, and I will dwell on them briefly, in the hope tha

hey might prove useful.



he heating of a conducting body inclosed in a bulb, andonnected to a source of rapidly alternating electricmpulses, is dependent on so many things of a differentature, that it would be difficult to give a generally pplicable rule under which the maximum heating occur

As regards the size of the vessel, I have lately found thatt ordinary or only slightly differing atmosphericressures, when air is a good insulator, and henceractically the same amount of energy by a certainotential and frequency is given off from the body,

whether the bulb be small or large, the body is brought t

higher temperature if enclosed in a small bulb, becausef the better confinement of heat in this case.

At lower pressures, when air becomes more or lessonducting, or if the air be sufficiently warmed to becomonducting, the body is rendered more intensely

ncandescent in a large bulb, obviously because, undertherwise equal conditions of test, more energy may beiven off from the body when the bulb is large.

At very high degrees of exhaustion, when the matter inhe bulb becomes " radiant," a large bulb has still an

dvantage, but a comparatively slight one, over the smaulb.

inally, at excessively high degrees of exhaustion, whichannot be reached except by the employment of special

means, there seems to be, beyond a certain and rather

mall size of vessel, no perceptible difference in theeatin -.



hese observations were the result of a number of xperiments, of which one, showing the effect of the sizef the bulb at a high degree of exhaustion, may beescribed and shown here, as it presents a feature of

nterest. Three spherical bulbs of 2 inches, 3 inches and 4nches diameter were taken, and in the centre of each wamounted an equal length of an ordinary incandescentamp filament of uniform thickness. In each bulb the piecf filament was fastened to the leading-in wire of latinum, contained in a glass stem sealed in the bulb ;

are being taken, of course, to make everything as nearlylike as possible. On each glass stem in the inside of theulb was slipped a highly polished tube made of aluminuheet, which fitted the'stem and was held on it by springressure. The function of this aluminum tube will boxplained subsequently. In each bulb an equal length of

la-

ment protruded above the metal tube. It is sufficient toay now that under these conditions equal lengths of lament of the same thickness—in other words, bodies oqual bulk — were brought to incandescence. The three

ulbs were sealed to a glass tube, which was connected tSprengel pump. When a high vacuum had been reache

he glass tube carrying the bulbs was sealed off. A currenwas then turned on successively on each bulb, and it wasound that the filaments came to about the samerightness, and, if anything, the smallest bulb, which was

laced midwa between the two lar er ones, ma have



een slightly brighter. This result was expected, for wheither of the bulbs was connected to the coil theuminosity spread through the other two, hence the threulbs constituted really one vessel. When all the threeulbs were connected in multiple arc to the coil, in the

argest of them the filament glowed brightest, in the nexmaller it was a little less bright, and in the smallest itnly came to redness. The bulbs were then sealed off andeparately tried. The brightness of the filaments was nowuch as would have been expected on the supposition thahe energy given off was proportionate to the surface of

he bulb, this surface in each case representing one of thepatings of a condenser. Accordingly, there was lessifference between the largest and the middle sized thanetween the latter and the smallest bulb.

An interesting observation was made in this experiment

he three bulbs were suspended from a straight barewire connected to a terminal of a coil, the largest bulb

eing placed at the end of the wire, at some distance fromthe smallest bulb, and at an equal distance from the

atter the middle-sized one. The carbons glowed then inoth the larger bulbs about as expected, but the smallest

id not get its share by far. This observation led me toxchange the position of the bulbs, and I then observedhat whichever of the bulbs was in the middle was by faress bright than it was in any other position. This

mystifying result was, of course, found to be due to thelectrostatic action between the bulbs. When they werelaced at a considerable distance, or when they were



ttached to the corners of an equilateral triangle of coppewire, they glowed in about the order determined by theiurfaces.

As to the shape of the vessel, it is also of some

mportance, especially at high degrees of exhaustion. Of ll the possible constructions, it seems that a sphericallobe with the refractory body

mounted in its centre is the best to employ. By experienlias been demonstrated that in such a globe a refractor

ody of a given bulk is more easily brought toncandescence than when differently shaped bulbs aresed. There is also an advantage in giving to the

ncandescent body the shape of a sphere, for self-evidenteasons. In any case the body should be mounted in theentre, where the atoms rebounding from the glass

ollide. This object is best attained in the spherical bulb;ut it is also attained in a cylindrical vessel with one orwo straight filaments coinciding with its axis, andossibly also in parabolical or spherical bulbs withefractory body or bodies placed in the focus or foci of thame; though the latter is not probable, as the electrified

toms should in all cases rebound normally from theurface they strike, unless the speed were excessive, in

which case they would probably follow the general law ofeflection. ]S r o matter what shape the vessel may havef the exhaustion be low, a filament mounted in the globes brought to the same degree of incandescence in all

arts; but if the exhaustion be high and the bulb be



pherical or pear-shaped, as usual, focal points form andhe filament is heated to a higher degree at or near suchoints.

o illustrate the effect, I have here two small bulbs whic

re alike, only one is exhausted to a low and the other toery high degree. When connected to the coil, the filamenn the former glows uniformly throughout all its length ;

whereas in the latter, that portion of the filament which n the centre of the bulb glows far more intensely than thest. A curious point is that the phenomenon occurs even

f two filaments are mounted in a bulb, each beingonnected to one terminal of the coil, and, what is stillmore curious, if they be very near together, provided the

acuum be very high. I noted in experiments with suchulbs that the filaments would give way usually at aertain point, and in the first trials I attributed it to a

efect in the carbon. But when the phenomenon occurremany times in succession I recognized its real cause.

n order to bring a refractory body inclosed in a bulb toncandescence, it is desirable, on account of economy, thall the energy supplied to the bulb from the source shoul

each without loss the body to be heated; from there, anrom nowhere else, it should be radiated. It is, of course,ut of the question to reach this theoretical result, but it ossible by a proper construction of the illuminatingevice to approximate it more or less.

or many reasons, the refractory body is placed in theentre of the bulb and it is usuall su orted on a lass



tem containing the leading-in wire. As the potential of his wire is alternated, the rarefied gas surrounding thetem is acted upon inductively, and the glass stem isiolently bombarded and heated. In this manner by farhe greater portion of the energy supplied to the bulb—

specially when exceedingly high frequencies are used—may be lost for the purpose contemplated. To obviate thoss, or at least to reduce it to a minimum, I usually creen the rarefied gas surrounding the stem from thenductive action of the leading-in wire by providing thetem with a tube or coating of conducting material. It

eems beyond doubt that the best among metals tomploy for this purpose is aluminum, on account of its

many remarkable properties. Its only fault is that it isasily fusible, and, therefore, its distance from thencandescing body should be properly estimated. Usuallythin tube, of a diameter somewhat smaller than that of

he glass stem, is made of the finest aluminum sheet, andlipped on the stem. The tube is conveniently preparedy wrapping around a rod fastened in a lathe a piece of luminum sheet of proper size, grasping the sheet firmly

with clean chamois leather or blotting paper, and spinninhe rod very fast. The sheet is wound tightly around the

od, and a highly polished tube of one or three layers of he sheet is obtained. When slipped on the stem, theressure is generally sufficient to prevent it from slippinff, but, for safety, the lower edge of the sheet may beurned inside. The upper inside corner of the sheet—thats, the one which is nearest to the refractory incandescenod —should be cut out dia onall as it often ha ens



hat, in consequence of the intense heat, this corner turnoward the inside and comes very near to, or in contact

with, the wire, or filament, supporting the refractory ody. The greater part of the energy supplied to the bulb

s then used up in heating the metal tube, and the bulb is

endered useless for the purpose. The aluminum sheethould project above the glass stem more or less—onench or so—or else, if the glass be too close to thencandescing body, it may be strongly heated and becom

more or less conducting, whereupon it may be ruptured,r may, by its conductivity, establish a good electrical

onnection between the metal tube and the leading-inwire, in which case, again, most of the energy will be lostn heating the former. Perhaps the best way is to makehe top of the glass tube, for about an inch, of a

much smaller diameter. To still further reduce the dange

rising from the heating of the glass stem, and also withhe view of preventing an electrical connection betweenhe metal tube and the electrode, I preferably wrap thetem with several layers of thin mica, which extends ateast as far as the metal tube. In some bulbs I have alsosed an outside insulating cover.

he preceding remarks are only made to aid thexperimenter in the first trials, for the difficulties whiche encounters he may soon find means to overcome in hwn way.

o illustrate the effect of the screen, and the advantage osin it I have here two bulbs of the same size with the



tems, leading-in wires and incandescent lamp filamentsed to the latter, as nearly alike as possible. The stem ofne bulb is provided with an aluminum tube, the stem ofhe other has none. Originally the two bulbs were joinedy a tube which was connected to a Sprengel pump. Whe

high vacuum had been reached, first the connectingube, and then the bulbs, were sealed off; they areherefore of the same degree of exhaustion. When they re separately connected to the coil giving a certainotential, the carbon filament in the bulb provided withhe aluminum screen is rendered highly incandescent,

while the filament in the other bulb may, with the sameotential, not even come to redness, although in reality he latter bulb takes generally more energy than theormer. When they are both connected together to theerminal, the difference is even more apparent, showinghe importance of the screening. The metal tube placed o

he stem containing the leading-in wire performs really wo distinct functions: First, it acts more or less as anlectrostatic screen, thus economizing the energy supplieo the bulb; and, second, to whatever extent it may fail tct electrostatically, it acts mechanically, preventing theombardment, and consequently intense heating and

ossible deterioration of the slender support of theefractory incandescent body, or of the glass stemontaining the leading-in wire. I say slender support, fors evident that in order to confine the heat moreompletely to the incandescing body its support should bery thin, so as to carry away the smallest possiblemount of heat b conduction. Of all the su orts used I



ave found an ordinary incandescent lamp filament to behe best, principally because among conductors it can

withstand the highest degree of heat.

he effectiveness of the metal tube as an electrostatic

creen depend? largely on the degree of exhaustion.

At excessively high degrees of exhaustion—which areeached by using great care and special means inonnection with the Sprengel pump—when the matter inhe globe is in the ultra-radiant state, it acts most

erfectly. The shadow of the upper edge of the tube ishen sharply defined upon the bulb.

At a somewhat lower degree of exhaustion, which is abouhe ordinary "non-striking" vacuum, and generally as lons the matter moves predominantly in straight Hues, the

creen still does well. In elucidation of the precedingemark it is necessary to state that what is a "non-triking" vacuum for a coil operated as ordinarily, by mpulses, or currents, of low frequency, is not so, by far,

when the coil is operated by currents of very highrequency. In such case the discharge may pass with

reat freedom through the rarefied gas through which aow frequency discharge may not pass, even though theotential be much higher. At ordinary atmosphericressures just the reverse rule holds good: the higher th

requency, the less the spark discharge is able to jumpetween the terminals, especially if they are knobs or

pheres of some size.



inally, at very low degrees of exhaustion, when the gas well conducting, the metal tube not only does not act as alectrostatic screen, but even is a drawback, aiding to aonsiderable extent the dissipation of the energy laterallrom the leading-in wire. This, of course, is to be

xpected. In this case, namely, the metal tube is in goodlectrical connection with the leading-in wire, and most ohe bombardment is directed upon the tube. As long ashe electrical connection is not good, the conducting tubes always of some advantage, for although it may notreatly economize energy, still it protects the support of

he refractory button, and is the means of concentratingmore energy upon the same.

o whatever extent the aluminum tube performs theunction of a screen, its usefulness is therefore limited toery high degrees of exhaustion when it is insulated from

he electrode— that is, when the gas as a whole is non-onducting, and the molecules, or atoms, act asndependent carriers of electric charges.

n addition to acting as a more or less effective screen, inhe true meaning of the word, the conducting tube or

oating may also act, by reason of its conductivity, as aort of equalizer or dampener of the bombardmentgainst the stem. To be explicit, I assume the action to bs follows: Suppose a rhythmical bom-

ardment to occur against the conducting tube by reason

f its imperfect action as a screen, it certainly musta en that some molecules or atoms strike the tube



ooner than others. Those which come first in contactwith it give up their superfluous charge, and the tube islectrified, the electrification instantly spreading over itsurface. But this must diminish the energy lost in theombardment, for two reasons : first, the charge given u

y the atoms spreads over a great area, and hence thelectric density at any point is small, and the atoms areepelled with less energy than they would be if they truck against a good insulator ; secondly, as the tube islectrified by the atoms which first come in contact with, the progress of the following atoms against the tube is

more or less checked by the repulsion which



IG. 147. FIG. 148.

he electrified tube must exert upon the similarly

lectrified atoms. This repulsion may perhaps beufficient to prevent a large portion of the atoms fromtriking the tube, but at any rate it must diminish thenergy of their impact. It is clear that when thexhaustion is very low, and the rarefied gas wellonducting, neither of the above effects can occur, and, o

he other hand, the fewer the atoms, with the greaterreedom they move ; in other words, the higher theegree of exhaustion, up to a limit, the more telling will both the effects.

What I have just said may afford an explanation of the

henomenon observed by Prof. Crookes, namely, that aischarge through a bulb is established \vith muchreater facility when an

nsulator than when a conductor is present in the same. Imy opinion, the conductor acts as a dampener of the

motion of the atoms in the two wa s ointed out; hence,



o cause a visible discharge to pass through the bulb, amuch higher potential is needed if a conductor, especially

f much surface, be present.

or the sake of elucidating of some of the remarks before

made, I must now refer to Figs. 147, 148 and 149, whichlustrate various arrangements with a type of bulb mostenerally used.

ig. 147 is a section through a spherical bulb L, with thelass stem *, contains the leading-in wire ?r, which has a

amp filament I fastened to it, serving to support theefractory button m in the centre. M is a sheet of thinmica wound in several layer* around the stem s, and a ishe aluminum tube.

ig. 148 illustrates such a bulb in a somewhat more

dvanced stage of perfection. A metallic tube s is fasteney means of some cement to the neck of the tube. In theube is screwed a plug P, of insulating material, in theentre of which is fastened a metallic terminal t, for theonnection to the leading-in wire w. This terminal muste well insulated from the metal tube s; therefore, if the

ement used is conducting—and most generally it isufficiently so—the space between the plug P and theeck of the bulb should be filled with some good insulatin

material, such as mica powder.

ig. 149 shows a bulb made for experimental purposes. I

his bulb the aluminum tube is provided with an externaonnection, which serves to investigate the eifect of the



ube under various conditions. It is referred to chiefly touggest a line of e.xprri-ment followed.

ince the bombardment against the stem containing theeading-in wire is due to the inductive action of the latter

pon the rarefied gas, it is of advantage to reduce thisction as far as practicable by employing a very thin wireurrounded by a verv thick insulation of glass or other

material, and by making the wire passing through thearefied gas as short as practicable. To combine theseeatures I employ a large tube T (Fig. 150), which

rotrudes into the bulb to some distance, and carries onhe top a very short glass stem «, into which is sealed theeading-in wire w, and I protect the top of the glass stemgainst the heat by a small aluminum tube a and a layerf mica underneath the same, as usual. The wire u\assing through the large tube to the outside of the bulb

hould be well insulated—with a s;lass tube,

or instance— and the space between ought to be filled ouwith some excellent insulator. Among many insulating

owders I have found that mica powder is the best tomploy. If this precaution is not taken, the tube T,

rotruding into the bulb, will surely be cracked inonsequence of the heating by the brushes which are apto form in the upper part of the tube, near the exhaustedlobe, especially if the vacuum be excellent, and thereforhe potential necessary to operate the lamp be very high

ig. 151 illustrates a similar arrangement, with a largeube T rotrudin into the art of the bulb containin th



efractory button -ni. In this case the wire leading fromhe outside into the bulb is omitted, the energy requiredeing supplied through



IG. 149.

IG. 150.

ondenser coatings c o. The insulating packing p should in

his construction be tightly litting to the glass, and ratherwide, or otherwise the discharge might avoid passinghrough the wire ie, which connects the inside condenseroating to the incandescent button ///.

he molecular bombardment against the glass stem in th

ulb is a source of great trouble. As an illustration I willite a phenomenon only too frequently and unwillingly bserved. A bulb, preferably a large one, may be taken,nd a good conducting body, such as a piece of carbon,

may be mounted in it upon a platinum wire sealed in thelass stem. The bulb may be exhausted to a fairly high

egree, nearly to the point when phosphorescence

egins to appear. When the bulb is connected with theoil, the piece of carbon, if small, may become highly ncandescent at first, but its brightness immediately iminishes, and then the discharge may break through

he glass somewhere in the middle of the stem, in theorm of bright sparks, in spite of the fact that thelatinum wire is in good electrical connection with thearefied gas through the piece of carbon or metal at theop. The first sparks are singularly bright, recalling thoserawn from a clear surface of mercury. But, as they heat

he lass ra idl the of course lose their bri htness



nd cease when the glass at the ruptured place becomesncandescent, or generally sufficiently hot to conduct.

When observed for the first time the phenomenon mustppear very curious, and shows in a striking manner howadically different alternate currents, or impulses, of high

requency behave, as compared with steady currents, orurrents of low frequency. With such currents — namelyhe latter — the phenomenon would of course not occur.

When frequencies such as are obtained by mechanicalmeans are used, I think that the rupture of the glass ismore or less the consequence of the bombard, ment,

which warms it up and impairs its insulating power ; butwith frequencies obtainable with condensers I have no

oubt that the glass may give way without previouseating. Although this appears most singular at first, it is

n reality what we might expect to occur. The energy upplied to the wire leading into the bulb is given off

artly by direct action through the carbon button, andartly by inductive action through the glass surroundinghe wire. The case is thus analogous to that in which aondenser shunted by a conductor of low resistance isonnected to a source of alternating current. As long ashe frequencies are low, the conductor gets the most and

he condenser is perfectly safe ; but when the frequency ecomes excessive, the role of the conductor may becomuite insignificant. In the latter case the difference of otential at the terminals of the condenser may becomeo great as to rupture the dielectric, notwithstanding theact that the terminals are joined by a conductor of low esis tance.



t is, of course, not necessary, when it is desired toroduce the incandescence of a body inclosed in a bulb by

means of these currents, that the body should be aonductor, for even a perfect non-conductor may be quits readily heated. For this purpose it is sufficient to

urround a conducting electrode with a non-con-

material, as, for instance, in the bulb described before inig. 150, in which a thin incandescent lamp filament isoated with a non-conductor, and supports a button of thame material on the top. At the start the bombardment

oes on by inductive action through the non-conductor,ntil the same is sufficiently heated to become conductin

when the bombardment continues in the ordinary way.

A different arrangement used in some of the bulbsonstructed is illustrated in Fig. 152. In this instance a

on-conductor ra is mounted in a piece of common arcght carbon so as to project some small distance abovehe latter. The carbon piece is connected to the leading-i

wire passing through a glass stem, which



IG. 151.

IG. 152.

s wrapped with several layers of mica. An aluminum tubis employed as usual for screening. It is so arranged tha reaches very nearly as high as the carbon and only theon-conductor m projects a little above it. The

ombardment goes at first against the upper surface of arbon the lower arts bein rotected b the aluminum



ube. As soon, however, as the non-conductor m is heateis rendered good conducting, and then it becomes the

entre of the bombardment, being most exposed to theame.

have also constructed during these experiments many uch single-wire bulbs with or without internal electroden which the radiant matter was projected against, orocused upon, the body

o be rendered incandescent. Fig. 153 (page 263)

lustrates one of the bulbs used. It consists of a sphericalobe L, provided with a long neck n, on top, for increasinhe action in some cases by the application of an externaonducting coating. The globe L is blown out on theottom into a very small bulb Z>, which serves to hold itrmly in a socket s of insulating material into which it is

emented. A fine lamp filament f, supported on a wire w,asses through the centre of the globe L. The filament isendered incandescent in the middle portion, where theombardment proceeding from the lower inside surface he globe is most intense. The lower portion of the globe,s far as the socket s reaches, is rendered conducting,

ither by a tinfoil coating or otherwise, and the externallectrode is connected to a terminal of the coil.

he arrangement diagrammatically indicated in Fig. 153was found to be an inferior one when it was desired toender incandescent a filament or button supported in th

entre of the lobe but it was convenient when the ob ec



was to excite phosphorescence.

n many experiments in which bodies of different kindwere mounted in the bulb as, for instance, indicated in

ig. 152, some observations of interest were made.

t was found, among other things, that in such cases, nomatter where the bombardment began, just as soon as a

igh temperature was reached there was generally one ohe bodies which seemed to take most of theombardment upon itself, the other, or others, being

hereby relieved. The quality appeared to dependrincipally on the point of fusion, and on the facility withwhich the body was " evaporated," or, generally speakin

isintegrated—meaning by the latter term not only thehrowing off of atoms, but likewise of large lumps. Thebservation made was in accordance with generally

ccepted notions. In a highly exhausted bulb, electricity arried off from the electrode by independent carriers,

which are partly the atoms, or molecules, of the residualtmosphere, and partly the atoms, molecules, or lumpshrown off from the electrode. If the electrode isomposed of bodies of different character, and if one of

hese is more easily disentegrated than the other, most ohe electricity supplied is carried off from that body,

which is then brought to a higher temperature than thethers, and this the more, as upon an increase of theemperature the body is still more easily dis-intregrated

t seems to me quite probable that a similar process takelace in the bulb even with a homo eneous electrode an



think it to be the principal cause of the disintegration.here is bound to be some irregularity, even if the surfa

s highly polished, which, of course, is impossible withmost of the refractory bodies employed as electrodes.Assume that a point of the electrode gets hotter ;

nstantly most of the discharge passes through that pointnd a minute patch it probably fused and evaporated. It ow possible that in consequence of the violentisintegration the spot attacked sinks in temperature, orhat a counter force is created, as in an arc ; at any rate,he local tearing off meets with the limitations incident to

he experiment, whereupon the same process occurs onnother place. To the eye the electrode appears uniformrilliant, but there are upon it points constantly shiftingnd wandering around, of a temperature far above the

mean, and this materially hastens the process of eterioration. That some such thing occurs, at least when

he electrode is at a lower temperature, sufficientxperimental evidence can be obtained in the following

manner : Exhaust a bulb to a very high degree, so thatwith a fairly high potential the discharge cannot pass—hat is, not a luminous one, for a weak invisible dischargeccurs always, in all probability. Now raise slowly and

arefully the potential, leaving the primary current on nomore than for an instant. At a certain point, two, three, o

alf a dozen phosphorescent spots will appear on thelobe. These places of the glass are evidently moreiolently bombarded than others, this being due to thenevenly distributed electric density, necessitated, of ourse b shar ro ections or enerall s eakin



rregularities of the • electrode. But the luminous patchesre constantly changing in position, which is especially

well observable if one manages to produce very few, andhis indicates that the configuration of the electrode isapidly changing.

rom experiences of this kind I am led to infer that, inrder to be most durable, the refractory button in theulb should be in the form of a sphere with a highly olished surface. Such a small sphere could be

manufactured from a diamond or some other crystal, bu

better way would be to fuse, by the employment of xtreme degrees of temperature, some oxide—as, fonstance, zirconia — into a small drop, and then keep it inhe bulb at a temperature somewhat below its point of usion.

nteresting and useful results can, no doubt, be reached ihe

irection of extreme degrees of heat. How can such highemperatures he arrived at ? How are the highest degref heat readied in nature ? By the impact of stars, by hig

peeds and collisions. In a collision any rate of heateneration may be attained. In a chemical process we armited. When oxygen and hydrogen combine, they fall,

metaphorically speaking, from a definite height. Weannot go very far with a blast, nor by confining heat in aurnace, but in an exhausted bulb we can concentrate an

mount of energy upon a minute button. Leavingracticabilit out of consideration this then would be th



means which, in my opinion, would enable us to reach thighest temperature. But a great difficulty whenroceeding in this way is encountered, namely, in mostases the body is carried off before it can fuse and form arop. This difficulty exists principally with an oxide, such

s zirconia, because it cannot be compressed in so hard aake that it would not be carried off quickly. I havendeavored repeatedly to fuse zirconia, placing it in a cupf arc light carbon, as indicated in Fig. 152. It glowed withmost intense light, and the stream of the particlesrojected out of the carbon cup was of a vivid white ; but

whether it was compressed in a cake or made into a pastwith carbon, it was carried off before it could be fused.

he carbon cup, containing zirconia, had to be mountedery low in the neck of a large bulb, as the heating of thelass by the projected particles of the oxide was so rapidhat in the first trial the bulb was cracked almost in an

nstant, when the current was turned on. The heating of he glass by the projected particles was found to belways greater when the carbon cup contained a body

which was rapidly carried off—I presume, because in sucases, with the same potential, higher speeds wereeached, and also because, per unit of time, more matter

was projected— that is, more particles would strike thelass.

he before-mentioned difficulty did not exist, however,when the body mounted in the carbon cup offered greatesistance to deterioration. For instance, when an oxide

was first fused in an ox en blast and then mounted in



he bulb, it melted very readily into a drop.

Generally, during the process of fusion, magnificent lightffects were noted, of which it would be difficult to give adequate idea. Fig. 152 is intended to illustrate the effect

bserved with a ruby drop. At first one may see a narrowunnel of

white light projected against the top of the globe, where roduces an irregularly outlined phosphorescent patch.

When the point of the ruby fuses, the phosphorescence

ecomes very powerful ; but as the atoms are projectedwith much greater speed from the surface of the drop,oon the glass gets hot and "tired," and now only theuter edge of the patch glows. In this manner an intensehosphorescent, sharply defined line, £, corresponding tohe outline of the drop, is produced, which spreads slowly

ver the globe as the drop gets larger. When the massegins to boil, small bubbles and cavities are formed,

which cause dark colored spots to sweep across the globehe bulb may be turned downward without fear of therop falling off, as the mass possesses considerableiscosity .

may mention here another feature of some interest,which I believe to have noted in the course of thesexperiments, though the observations do not amount to ertitude. It appeared that under the molecular impactaused by the rapidly alternating potential, the body was

used and maintained in that state at a lower temperaturn a hi hl exhausted bulb than was the case at normal



ressure and application of heat in the ordinary way —hat is, at least, judging from the quantity of the lightmitted. One of the experiments performed may be

mentioned here by way of illustration. A small piece of umice stone was stuck on a platinum wire, and first

melted to it in a gas burner. The wire was next placedetween two pieces of charcoal, and a burner applied, sos to produce an intense heat, sufficient to melt down theumice stone into a small glass-like button. The platinum

wire had to be taken of sufficient thickness, to prevent itmelting in the fire. While in the charcoal fire, or when hel

n a burner to get a better idea of the degree of heat, theutton glowed with great brilliancy. The wire with theutton was then mounted in a bulb, and upon exhaustinghe same to a high degree, the current was turned onlowly, so as to prevent the cracking of the button. Theutton was heated to the point of fusion, and when it

melted, it did not, apparently, glow with the samerilliancy as before, and this would indicate a loweremperature. Leaving out of consideration the observer'ossible, and even probable, error, the question is, can aody under these conditions be brought from a solid to aquid state with the evolution of less light ?

When the potential of a body is rapidly alternated, it isertain

hat the structure is jarred. When the potential is very igh, although the vibrations may be few—say 20,000

er second—the effect u on the structure ma be



onsiderable. Suppose, for example, that a ruby is meltednto a drop by a steady application" of energy. When itorms a drop, it will emit visible and invisible waves,

which will be in a definite ratio, and to the eye the dropwill appear to be of a certain brilliancy. Next, suppose we

iminish to any degree we choose the energy steadily upplied, and, instead, supply energy which rises and fallccording to a certain law. Now, when the drop is formedhere will be emitted from it three different kinds of ibrations — the ordinary visible, and two kinds of

nvisible waves : that is, the ordinary dark waves of all

engths, and, in addition, waves of a well definedharacter. The latter would not exist by a steady supply f the energy ; still they help to jar and loosen thetructure. If this really be the case, then the ruby drop

will emit relatively less visible and more invisible waveshan before. Thus it would seem that when a platinum

wire, for instance, is fused by currents alternating withxtreme rapidity, it emits at the point of fusion less lightnd more ..visible radiation than it does when melted byteady current, though the total energy used up in therocess of fusion is the same in both cases. Or, to citenother example, a lamp filament is not capable of

withstanding as long with currents of extreme frequencys it does with steady currents, assuming that it be

worked at the same luminous intensity. This means thator rapidly alternating currents the filament should behorter and thicker. The higher the frequency— that is,he greater the departure from the steady flow — the

worse it would be for the filament. But if the truth of this



emark were demonstrated, it would be erroneous toonclude that such a refractory button as used in theseulbs would be deteriorated quicker by currents of xtremely high frequency than by steady or low requency currents. From experience I may say that jus

he opposite holds good : the button withstands theombardment better with currents of very highrequency. But this is due to the fact that a high frequencischarge passes through a rarefied gas with muchreater freedom than a steady or low frequency ischarge, and this will mean that with the former we can

work with a lower potential or with a less violent impact.As long, then, as the gas is of no consequence, a steady oow frequency current is better; but as soon as the actionf the gas is desired and important, high frequencies arereferable.

n the course of these experiments a great many trialswere made with all kinds of carbon buttons. Electrodesmade of ordinary carbon buttons were decidedly more

urable when the buttons were obtained by thepplication of enormous pressure. Electrodes prepared bepositing carbon in well known ways did not show up

well; they blackened the globe very quickly. From manyxperiences I conclude that lamp filaments obtained inhis manner can be advantageously used only with low otentials and low frequency currents. Some kinds of arbon withstand so well that, in order to bring them tohe point of fusion, it is necessary to employ very smalluttons. In this case the observation is rendered very



ifficult on account of the intense heat produced.Nevertheless there can be no doubt that all kinds of arbon are fused under the molecular bombardment, buhe liquid state must be one of great instability. Of all theodies tried there were two which withstood best—

iamond and carborundum. These two showed up aboutqually, but the latter was preferable for many reasons.As it is more than likely that this body is not yet generall

nown, I Avill venture to call your attention to it.

t has been recently produced by Mr. E. G. Acheson, of

Monongahela City, Pa., II. S. A. It is intended to replacerdinary diamond powder for polishing precious stones,tc., and I have been informed that it accomplishes thisbject quite successfully. I do not know why the name "arborundum" has been given to it, unless there isomething in the process of its manufacture which

ustifies this selection. Through the kindness of thenventor, I obtained a short while ago some samples

which I desired to test in regard to their qualities of hosphorescence and capability of withstanding highegrees of heat.

arborundum can be obtained in two forms—in the formf "crystals" and of powder. The former appear to theaked eye dark colored, but are very brilliant; the latter

s of nearly the same color as ordinary diamond powder,ut very much finer. When viewed under a microscopehe samples of crystals given to me did not appear to hav

ny definite form, but rather resembled pieces of broken



p egg coal of fine quality. The majority were opaque, buhere were some which were transparent and colored.he crystals are a kind of carbon containing some

mpurities; they are extremely hard, and withstand for aong time even an oxygen blast. When the blast is directe

gainst them they at first form a cake of someompactness, probably in consequence of the fusion of mpurities they contain. The mass withstands for a very ong time the blast without further fusion ; but a slow arrying off, or burning, occurs, and, finally, a small

uantity of a glass-like residue is left, w r hich, I supposes melted alumina. When compressed strongly they onduct very well, but not as well as ordinary carbon. Thowder, which is obtained from the crystals in some way

s practically non-conducting. It affords a magnificentolishing material for stones.

he time has been too short to make a satisfactory studyf the properties of this product, but enough experienceas been gained in a few weeks I have experimentedpon it to say that it does possess some remarkableroperties in many respects. It withstands excessively

igh degrees of heat, it is little deteriorated by molecularombardment, and it does not blacken the globe asrdinary carbon does. The only difficulty which I havexperienced in its use in connection with thesexperiments was to find some binding material which

would resist the heat and the effect of the bombardment

s successfully as carborundum itself does.



have here a number of bulbs which I have provided wiuttons of carborundum. To make such a button of arborundum crystals I proceed in the following mannertake an ordinary lamp filament and dip its point in tar,r some other thick substance or paint which may be

eadily carbonized. I next pass the point of the filamenthrough the crystals, and then hold it vertically over a holate. The tar softens and forms a drop on the point of thlament, the crystals adhering to the surface of the dropy regulating the distance from the plate the tar is slowlyried out and the button becomes solid. I then once mor

ip the button in tar and hold it again over a plate untilhe tar is evaporated, leaving only a hard mass whichrmly binds the crystals. When a larger button isequired I repeat the process several times, and Ienerally also cover the filament a certain distance belowhe button with crystals. The button being mounted in a

ulb, when a good vacuum has been reached, first a weaknd then a strong discharge is passed through the bulb toarbonize the tar and expel all gases, and later it isrought to a very intense incandescence.

When the powder is used I have found it best to proceed

s follows : I make a thick paint of carborundum and tar,nd pass a lamp filament through the paint. Taking then

most of the

aint off by rubbing the filament against a piece of hamois leather, I hold it over a hot plate until the tar

va orates and the coatin becomes firm. I re eat this



rocess as many times as it is necessary to obtain aertain thickness of coating. On the point of the coatedlament I form a button in the same manner.

here is no doubt that such a button—properly prepared

nder great pressure— of carborundum, especially of owder of the best quality, will withstand the effect of thombardment fully as well as anything we know. Theifficulty is that the binding material gives way, and thearborundum is slowly thrown off after some time. As itoes not seem to blacken the globe in the least, it might

e found useful for coating the filaments of ordinary ncandescent lamps, and I think that it is even possible troduce thin threads or sticks of carborundum which wileplace the ordinary filaments in an incandescent lamp. Aarborundum coating seems to be more durable thanther coatings, not only because the carborundum can

withstand high degrees of heat, but also because it seemo unite with the carbon better than any other material Iave tried. A coating of zirconia or any other oxide, for

nstance, is far more quickly destroyed. I prepareduttons of diamond dust in the same manner as of arborundum, and these came in durability nearest to

hose prepared of carborundum, but the binding pasteave way much more quickly in the diamond buttons;his, however, I attributed to the size and irregularity of he grains of the diamond.

t was of interest to find whether carborundum possesse

he quality of phosphorescence. One is, of course,



repared to encounter two difficulties: first, as regardshe rough product, the "crystals," they are goodonducting, and it is a fact that conductors do nothosphoresce; second, the powder, being exceedingly fin

would not be apt to exhibit very prominently this quality

ince we know that when crystals, even such as diamondr ruby, are finely powdered, they lose the property of hosphorescence to a considerable degree.

he question presents itself here, can a conductorhosphoresce ? What is there in such a body as a metal,

or instance, that would deprive it of the quality of hosphoresence, unless it is that property whichharacterizes it as a conductor ? For it is a fact that mostf the phosphorescent bodies lose that quality when theyre sufficiently heated to become more or less conductin

hen, if a metal be in a large measure, or perhapsntirely, deprived of that property, it should be capable hosphoresence. Therefore it is quite possible that atome extremely high frequency, when behavingractically as a non-conductor, a metal or any otheronductor might exhibit the quality of phosphoresence,

ven though it be entirely incapable of phosphorescingnder the impact of a low-frequency discharge. There isowever, another possible way how a conductor might at

east appear to phosphoresce.

onsiderable doubt still exists as to what really is

hosphorescence, and as to whether the varioushenomena com rised under this head are due to the



ame causes. Suppose that in an exhausted bulb, underhe molecular impact, the surface of a piece of metal orther conductor is rendered strongly luminous, but at thame time it is found that it remains comparatively cool,

would not this luminosity be called phosphorescence?

Now such a result, theoretically at least, is possible, for its a mere question of potential or speed. Assume theotential of the electrode, and consequently the speed ofhe projected atoms, to be sufficiently high, the surface ohe metal piece, against which the atoms are projected,

would be rendered highly incandescent, since the proces

f heat generation would be incomparably faster than thf radiating or conducting away from the surface of theollision. In the eye of the observer a single impact of thetoms would cause an instantaneous flash, but if the

mpacts were repeated with sufficient rapidity, they would produce a continuous impression upon his retina.

o him then the surface of the metal would appearontinuously incandescent and of constant luminousntensity, while in reality the light would be eitherntermittent, or at least changing periodically in intensityhe metal piece would rise in temperature untilquilibrium was attained—that is, until the energy

ontinuously radiated would equal that intermittently upplied. But the supplied energy might under suchonditions not be sufficient to bring the body to any morehan a very moderate mean temperature, especially if threquency of the atomic impacts be very low—just enoughat the fluctuation of the intensity of the light emittedould not be detected b the e e. The bod would now



wing to the manner in which the energy is supplied, emstrong light, and yet be at a comparatively very low

mean temperature. How should the observer name theuminosity thus produced ? Even if

he analysis of the light would teach him somethingefinite, still he would probably rank it under thehenomena of phosphorescence. It is conceivable that inuch a way both conducting and non-conducting bodies

may be maintained at a certain luminous intensity, buthe energy required would very greatly vary with the

ature and properties of the bodies.

hese and some foregoing remarks of a speculativeature were made merely to bring out curious features olternate currents or electric impulses. By their help we

may cause a body to emit more light, while at a certain

mean temperature, than it would emit if brought to thatemperature by a steady supply ; and, again, we may ring a body to the point of fusion, and cause it to emit

ess light than when fused by the application of energy inrdinary ways. It all depends on how we supply thenergy, and what kind of vibrations we set up; in one cas

he vibrations are more, in the other less, adapted toffect our sense of vision.

ome effects, which I had not observed before, obtainedwith carborundum in the first trials, I attributed to

hosphorescence, but in subsequent experiments it

ppeared that it was devoid of that quality. The crystalsossess a noteworth feature. In a bulb rovided with a



ingle electrode in the shape of a small circular metal discor instance, at a certain degree of exhaustion thelectrode is covered with a milky, film, which is separatey a dark space from the glow filling the bulb. When the

metal disc is covered with carborundum crystals, the film

s far more intense, and snow-white. This I found later toe merely an effect of the bright surface of the crystals,or when an aluminum electrode was highly polished, itxhibited more or less the same phenomenon. I made aumber of experiments with the samples of crystalsbtained, principally because it would have been of speci

nterest to find that they are capable of phosphorescencen account of their being conducting. I could not producehosphorescence distinctly, but I must remark that aecisive opinion cannot be formed until otherxperimenters have gone over the same ground.

he powder behaved in some experiments as though itontained alumina, but it did not exhibit with sufficientistinctness the red of the latter. Its dead color brightenonsiderably under the molecular impact, but I am now onvinced it does not phosphoresce. Still, the tests withhe powder are not conclusive, because powdered

arborundum probably does not behave like a

hosphorescent sulphide, for example, which could benely powdered without impairing the phosphorescenceut rather like powdered ruby or diamond, and therefor would be necessary, in order to make a decisive test, t

btain it in a large lump and polish up the surface.



f the carborundum proves useful in connection •withhese and similar experiments, its chief value will beound in the production of coatings, thin conductors,uttons, or other electrodes capable of withstandingxtremely high degrees of heat.

he production of a sniall electrode, capable of withstan<lino-enormous temperatures, I regard as of th

reatest importance in the manufacture of light. It wouldnable us to obtain, by means of currents of very highrequencies, certainly 20 times, if not more, the quantity

f light which is obtained in the present incandescentamp by the same expenditure of energy. This estimate

may appear to many exaggerated, but in reality I think is far from being so. As this statement might be

misunderstood, I think it is necessary to expose clearly he problem with which, in this line of work, we are

onfronted, and the manner in which, in my opinion, aolution will be arrived at.

Any one who begins a study of the problem will be apt tohink that what is wanted in a lamp with an electrode is aery high degree of incandescence of the electrode. Ther

e will be mistaken. The high incandescence of the buttos a necessary evil, but what is really wanted is the highncandescence of the gas surrounding the button. In othe

words, the problem in such a lamp is to bring a mass of as to the highest, possible incandescence. The higher thncandescence, the quicker the mean vibration, thereater is the econom of the li ht roduction. But to



maintain a mass of gas at a high degree of incandescencen a glass vessel, it will always be necessary to keep thencandescent mass away from the glass; that is, to confin

as much as possible to the central portion of the globe.

n one of the experiments this evening a brush wasroduced at the end of a wire. The brush was a flame, aource of heat and light. It did not emit much perceptibleeat, nor did it glow with an intense light; but is it the lesflame because it does not scorch my hand { Is it the lesflame because it does not hurt my eyes by its brilliancy

The problem is precisely to produce in the bulb such aame, much smaller in size, but incomparably moreowerful. Were there means at hand for

roducing electric impulses of a sufficiently highrequency, and for transmitting them, the bulb could be

one away with, unless it were used to protect thelectrode, or to economize the energy by confining theeat. But as such means are not at disposal, it becomesecessary to place the terminal in the bulb and rarefy thir in the same. This is done merely to enable thepparatus to perform the work which it is not capable of

erforming at ordinary air pressure. In the bulb we areble to intensify the action to any degree—so far that therush emits a powerful light.

he intensity of the light emitted depends principally onhe frequency and potential of the impulses, and on the

lectric density on the surface of the electrode. It is of threatest im ortance to em lo the smallest ossible



utton, in order to push the density very far. Under theiolent impact of the molecules of the gas surrounding it,he small electrode is of course brought to an extremely igh temperature, but around it is a mass of highly

ncandescent gas, a flame photosphere, many hundred

mes the volume of the electrode. With a diamond,arborundum or zirconia button the photosphere can bes much as one thousand times the volume of the button

Without much reflection one would tllink that in pushingo far the incandescence of the electrode it would benstantly volatilized. But after a careful consideration one

would find that, theoretically, it should not occur, and inhis fact —which, moreover, is experimentally emonstrated — lies principally the future value of such

amp.

At first, when the bombardment begins, most of the wor

s performed on the surface of the button, but when aighly conducting photosphere is formed the button isomparatively relieved. The higher the incandescence ofhe photosphere, the more it approaches in conductivity o that of the electrode, and the more, therefore, the solind the gas form one conducting body. The consequence

s that the further the incandescence is forced the morework, comparatively, is performed on the gas, and theess on the electrode. The formation of a powerfulhotosphere is consequently the very means forrotecting the electrode. This protection, of course, is aelative one, and it should not be thought that by pushinhe incandescence hi her the electrode is actuall less



eteriorated. Still, theoretically, with extremerequencies, this result must be reached, but probably attemperature too high for most of the refractory bodies

nown. Given, then, an electrode which can withstand to

ery high limit the effect of the bombardment andutward strain, it would be safe, no matter how much itwas forced beyond that limit. In an incandescent lamp

uite different considerations apply. There the gas is nott all concerned ; the whole of the work is performed onhe filament ; and the the life of the lamp diminishes so

apidly with the increase of the degree of incandescencehat economical reasons compel us to work it at a low ncandescence. But if an incandescent lamp is operated

with currents of very high frequency, the action of the gaannot be neglected, and the rules for the mostconomical working must be considerably modified.

n order to bring such a lamp with one or two electrodeso a great perfection, it is necessary to employ impulses oery high frequency. The high frequency secures, amongthers, two chief advantages, which have a most

mportant bearing upon the economy of the light

roduction. First, the deterioration of the electrode iseduced by reason of the fact that we employ a great

many small impacts, instead of a few violent ones, whichuickly shatter the structure ; secondly, the formation oflarge photo-shere is facilitated.

n order to reduce the deterioration of the electrode tohe minimum it is desirable that the vibration be



armonic, for any suddenness hastens the process of estruction. An electrode lasts much longer when kept a

ncandescence by currents, or impulses, obtained from aigh frequency alternator, which rise and fall more or lesarmonically, than by impulses obtained from a

isruptive discharge coil. In the latter case there is nooubt that most of the damage is done by theundamental sudden discharges.

One of the elements of loss in such a lamp is theombardment of the globe. As the potential is very high,

he molecules are pro jected with great speed; they strikhe glass, and usually excite a strong phosphorescence.he effect produced is very pretty, but for economicaleasons it would be perhaps preferable to prevent, or ateast reduce to a minimum, the bombardment against thlobe, as in such case it is, as a rule, not the object to

xcite phosphorescence, and as some loss of energy esults from the bombardment. This loss in the bulb isrincipally dependent on the potential of the impulses ann the electric density on the surface of the electrode. Inmploying \cry high frecjuen-

HIGH FREQUENCY AND HIGH POTENTIALURRENTS. 261

ies the loss of energy by the bombardment is greatly educed, for, first, the potential needed to perform a givemount of work is much smaller ; and, secondly, by

roducing a highly conduct-ting photosphere around thelectrode the same result is obtained as thou h the



lectrode were much larger, which is equivalent to amaller electric density. But be it by the diminution of th

maximum potential or of the density, the gain is effectedn the same manner, namely, by avoiding violent shocks,

which strain the glass much beyond its limit of elasticity.

f the frequency could be brought high enough, the lossue to the imperfect elasticity of the glass would bentirely negligible. The loss due to bombardment of thelobe may, however, be reduced by using two electrodesnstead of one. In such case each of the electrodes may bonnected to one of the terminals; or else, if it is

referable to use only one wire, one electrode may beonnected to one terminal and the other to the ground oro an insulated body of some surface, as, for instance, ahade on the lamp. In the latter case, unless someudgment is used, one of the electrodes might glow morentensely than the other.

ut on the whole I find it preferable, when using such higrequencies, to employ only one electrode and oneonnecting wire. I am convinced that the illuminatingevice of the near future will not require for its operation

more than one lead, and, at any rate, it will have no

eading-in wire, since the energy required can be as wellransmitted through the glass. In experimental bulbs theeading-in wire is not generally used on account of onvenience, as in employing condenser coatings in the

manner indicated in Fig. 151, for example, there is someifficulty in titting the parts, but these difficulties wouldot exist if a reat man bulbs were manufactured



therwise the energy can be conveyed through the glasss well as through a wire, and with these high frequenciehe losses are very small. Such illustrating devices willecessarilly involve the use of very high potentials, and

his, in the eyes of practical men, might be an

bjectionable feature. Yet, in reality, high potentials areot objectionable—certainly not in the least so far as theafety of the devices is concerned.

here are two ways of rendering an electric applianceafe. One is to use low potentials, the other is to

etermine the dimensions of the apparatus so that it isafe, no matter how high a potential is used. Of the two,he latter seems to me the better

way, for then the safety is absolute, unaffected by any ossible combination of circumstances which might

ender even alow-potential appliance dangerous to lifend property. But the practical conditions require notnly the judicious determination of the dimensions of thepparatus; they likewise necessitate the employment of nergy of the proper kind. It is easy, for instance, toonstruct a transformer capable of giving, when operated

rom an ordinary alternate current machine of low ension, say 50,000 volts, which might be required toght a highly exhausted phosphorescent tube, so that, inpite of the high potential, it is perfectly safe, the shock rom it producing no inconvenience. Still such aransformer would be expensive, and in itself inefficient;

nd, besides, what energy was obtained from it would no



e economically used for the production of light. Theconomy demands the employment of energy in the formf extremely rapid vibrations. The problem of producingght has been likened to that of maintaining a certainigh-pitcli note by means of a bell. It should be said a

arely audible note ; and even these words xvould notxpress it, so wonderful is the sensitiveness of the eye.We may deliver powerful blows at long intervals, waste a

ood deal of energy, and still not get what we want; or wmay keep up the note by delivering frequent taps, and g

earer to the object sought by the expenditure of much

ess energy. In the production of light, as far as theluminating device is concerned, there can be only oneule—that is, to use as high frequencies as can be obtainebut the means for the production and conveyance of

mpulses of such character impose, at present at least,reat limitations. Once it is decided to use very high

requencies, the return wire becomes unnecessary, and ahe appliances are simplified. By the use of obvious meanhe same result is obtained as though the return wire

were used. It is sufficient for this purpose to bring inontact with the bulb, or merely in the vicinity of theame, an insulated body of some surface. The surface

eed, of course, be the smaller, the higher the frequencynd potential used, and necessarily, also, the higher theconomy of the lamp or other device.

his plan of working has been resorted to on severalccasions this evening. So, for instance, when thencandescence of a button was produced by grasping the



ulb with the hand, the body of the experimenter merelyerved to intensify the action. The bulb used was similaro that illustrated in Fig. 148, and

ie coil was excited to a small potential, not sufficient to

ring the button to incandescence when the bull) wasanging from the Avire ; and incidentally, in order toerform the experiment in a more suitable manner, theutton was taken so large that a perceptible time had tolapse before, upon grasping the bulb, it could beendered incandescent. The contact with the bulb was, o

ourse, quite unnecessary. It is easy, by using a ratherarge bulb with an exceedingly small electrode, to adjusthe conditions so that the latter is brought to brightncandescence by the mere approach of the experimente

within a few feet of the bulb, and that the incandescenceubsides upon his receding.



IG. 153.

IG. 154.

n another experiment, when phosphorescence wasxcited, a similar bulb was used. Here again, originally,he potential was not sufficient to excite phosphorescenc

ntil the action was intensified—in this case, however, to



resent a different feature, by touching the socket with ametallic object held in the hand. The electrode in the bulbwas a carbon button so large that it could not be broughto incandescence, and thereby spoil the effect producedy phosphorescence.

Again, in another of the early experiments, a bulb wassed,

s illustrated in Fig. 141. In this instance, by touching theulb with one or two fingers, one or two shadows of the

tem inside were projected against the glass, the touch ohe finger producing the same results as the application on external negative electrode under ordinary ircumstances.

n all these experiments the action was intensified by

ugmenting the capacity at the end of the lead connectedo the terminal. As a rule, it is not necessary to resort touch means, and would be quite unnecessary with stilligher frequencies; but when it Is desired, the bulb, orube, can be easily adapted to the purpose.

n Fig. 153, for example, an experimental bull), i,, ishown, which is provided with a neck, n, on the top, forhe application of an external tinfoil coating, which may bonnected to a body of larger surface. Such a lamp aslustrated in Fig. 154 may also be lighted by connectinghe tinfoil coating on the neck n, to the terminal, and the

eading-in wire, w, to an insulated plate. If the bulb standn a socket upright, as shown in the cut, a shade of



onducting material may be slipped in the neck, n, and thction thus magnified.

A more perfected arrangement used in some of theseulbs is illustrated in Fig. 155. In this case the

onstruction of the bulb is as shown and described beforewhen reference was made to Fig. 148. A zinc sheet, z, wittubular extension, T, is applied over the metallic socke The bulb hangs downward from the terminal, t, the zin

heet, z, performing the double office of in-tensifier andeflector. The reflector is separated from the terminal, t,

y an extension of the insulating plug, P.

A similar disposition with a phosphorescent tube islustrated in Fig. 156. The tube, T, is prepared from twohort tubes of different diameter, which are sealed on thnds. Oil the lower end is placed an inside conducting

oating, c, which connects to the wire w. The wire has aook on the upper end for suspension, and passes throughe centre of the inside tube, which is filled witli someood and tightly packed insulator. On the outside of thepper end of the tube, T, is another conducting coating, o

} upon which is slipped a metallic reflector z, which

hould be separated by a thick insulation from the end ofwire u\

he economical use of such a reflector or intensifier woulequire that all energy supplied to an air condenserhould be recoverable, or, in other words, that there

hould not be any losses,



either in the gaseous medium nor through its actionlsewhere. This is far from being so, but, fortunately, theosses may be reduced to anything desired. A few emarks are necessary on this subject, in order to makehe experiences gathered in the course of these

nvestigations perfectly clear.

uppose a small helix with many well insulated turns, asn experiment Fig. 146, has one of its ends connected tone of the terminals of the induction coil, and the other tometal plate, or, for the sake of simplicity, a sphere,

nsulated in space. When the coil is set to work, theotential of the sphere is alternated, and a small helix noehaves as though its free end were connected to thether terminal of the induction coil. If an iron rod be held

within a small helix, it is quickly brought to a high

IG. 155.



emperature, indicating the passage of a strong currenthrough the helix. How does the insulated sphere act inhis case ? It can be a condenser, storing and returninghe energy supplied to it, or it can be a mere sink of nergy, and the conditions of the experiment determine

whether it is rather one than the other. The sphere beinharged to a high potential, it acts inductively upon theurrounding air, or whatever gaseous medium there

might be. The molecules, or atoms, which are near thephere, are of course more attracted, and move throughreater distance than the farther ones. When the neares

molecules strike the sphere, they are repelled, andollisions occur at all distances within the inductive actionf the sphere. It is now clear that, if the poten-

al be steady, but little loss of energy can be caused inhis way, for the molecules which are nearest to the

phere, having had an additional charge imparted to themy contact, are not attracted until they have parted, if no

with all, at least with most of the additional charge, whichan be accomplished only after a great many collisions.rom the fact, that with a steady potential there is butttle loss in dry air, one must come to such a conclusion.

When the potential of a sphere, instead of being steady, ilternating, the conditions are entirely different. In thisase a rhythmical bombardment occurs, no matter

whether the molecules, after coming in contact with thephere, lose the imparted



IG. 156.

harge or not; what is more, if the charge is not lost, the

mpacts are only the more violent. Still, if the frequency he impulses be very small, the loss caused by thempacts and collisions would not be serious, unless theotential \vere excessive. But when extremely high

requencies and more or less high potentials are used, thoss may very great. The total energy lost per unit of tims proportionate to the product of the number of impactser second, or the frequency and the energy lost in each

mpact. But the energy of an impact must beroportionate to the square of the electric density of thephere, since the charge imparted

o the molecule is ro ortionate to that densit . I



onclude from this that the total energy lost must beroportionate to the product of the frequency and thequare of the electric density; but this law needsxperimental confirmation. Assuming the preceding-onsiderations to be true, then, by rapidly alternating th

otential of a body immersed in an insulating gaseousmedium, any amount of energy may be dissipated intopace. Most of that energy then, I believe, is notissipated in the form of long ether waves, propagated toonsiderable distance, as is thought most generally, but ionsumed—in the case of an insulated sphere, for

xample—in impact and collisional losses— that is, heatibrations — on the surface and in the vicinity of thephere. To reduce the dissipation, it is necessary to work

with a small electric density—the smaller, the higher therequency.

ut since, on the assumption before made, the loss isiminished with the square of the density , and sinceurrents of very high frequencies involve considerable

waste when transmitted through conductors, it followshat, on the whole, it is better to employ one wire thanwo. Therefore, if motors, lamps, or devices of any kind

re perfected, capable of being advantageously operatedy currents of extremely high frequency, economicaleasons will make it advisable to use only one wire,specially if the distances are great.

When energy is absorbed in a condenser, the same

ehaves as though its capacity were increased.



Absorption always exists more or less, but generally it ismall and of no consequence us long as the frequenciesre not very great, In using extremely high frequencies,nd, necessarily in such case, also high potentials, thebsorption—or, what is here meant more particularly by

his term, the loss of energy due to the presence of "aaseous medium —is an important factor to beonsidered, as the energy absorbed in the air condenser

may be any fraction of the supplied energy. This wouldeem to make it very difficult to tell from the measured omputed capacity of an air condenser its actual capacity

r vibration period, especially if the condenser is of very mall surface and is charged to a very high potential. As

many important results are dependent upon theorrectness of the estimation of the vibration period, thisubject demands the most careful scrutiny of othernvestigators. To reduce the probable error as much as

ossible in experiments of the kind alluded to, it isdvisable to use spheres or plates of large surface, so as t

make the density exceedingly small. Otherwise, when it racticable, an oil condenser should be used in preferencn oil or other liquid dielectrics there are seemingly no

uch losses as in gaseous media. It being impossible toxclude entirely the gas in condensers with solidielectrics, such condensers should be immersed in oil, foconomical reasons, if nothing else ; they can then betrained to the utmost, and will remain cool. In Leydenars the loss due to air is comparatively small, as thenfoil coatings are large, close together, and the charged



urfaces not directly exposed; but when the potentials arery high, the loss may be more or less considerable at, oear, the upper edge of the foil, where the air isrincipally acted upon. If the jar be immersed in boiled-ut oil, it will be capable of performing four times the

mount of work which it can for any length of time whensed in the ordinary way, and the loss will benappreciable.

t should not be thought that the loss in heat in an airondenser is necessarily associated with the formation of

-/VJ/r streams or brushes. If a small electrode, inclosen an unexhausted bulb, is connected to one of theerminals of the coil, streams can be seen to issue fromhe electrode, and the air in the bulb is heated; if insteadf a small electrode a large sphere is inclosed in the bulb,o streams are observed, still the air is heated.

ôr should it be thought that the temperature of an airondenser would give even an approximate idea of theoss in heat incurred, as in such case heat must be givenff much more quickly, since there is, in addition to therdinary radiation, a very active carrying away of heat b

ndependent carriers going on, and since not only thepparatus, but the air at some distance from it is heatedn consequence of the collisions which must occur.

Owing to this, in experiments with such a coil, a rise of emperature can be distinctly observed only when the

ody connected to the coil is very small. But witharatus on a lar er scale even a bod of considerable



ulk would be heated, as, for instance, the body of aerson; and I think that skilled physicians might makebservations of utility in such experiments, which, if thepparatus were judiciously designed, would not presenthe slightest danger.

A question of some interest, principally to meteorologists

resents itself here. How does the earth behave ? Thearth is an air condenser, but is it a perfect or a very mperfect one—a mere sink of energy ? There can be litt

oubt that to such small disturbance as might be causedn an experiment, the earth behaves as an almost perfecondenser. But it might be different when its charge is sen yibration by some sudden disturbance occurring in theeavens. In such case, as before stated, probably only ttle of the energy of the vibrations set up would be lost

nto space in the form of long ether radiations, but most ohe energy, I think, would spend itself in molecularmpacts and collisions, and pass off into space in the formf short heat, and possibly light, waves. As both therequency of the vibrations of the charge and the potentire in all probability excessive, the energy converted int

eat may be considerable. Since the density must benevenly distributed, either in consequence of the

rregularity of the earth's surface, or on account of theondition of the atmosphere in various places, the effectroduced would accordingly vary from place to place.onsiderable variations in the temperature and pressure

f the atmos here ma in this manner be caused at an



oint of the surface of the earth. The variations may beradual or very sudden, according to the nature of theeneral disturbance, and may produce rain and storms, o

ocally modify the weather in any way.

rom the remarks before made, one may see what anmportant factor of loss the air in the neighborhood of aharged surface becomes when the electric density isreat and the frequency of the impulses excessive. Buthe action, as explained, implies that the air is insulating

— that is, that it is composed of independent carriers

mmersed in an insulating medium. This is the case only when the air is at something like ordinary or greater, or xtremely small, pressure. When the air is slightly areiied and conducting, then true conduction losses occulso. In such case, of course, considerable energy may beissipated into space even with a steady potential, or wit

mpulses of low frequency, if the density is very great.

When the gas is -at very low pressure, an electrode iseated more because higher speeds can be reached. If thas around the electrode is strongly compressed, theisplacements, and consequently the speeds, are very

mall, and the heating is in-signincant. But if in such casehe frequency could be sufficiently increased, thelectrode would be brought to a high tern-


erature as well as if the gas were at very low pressure ;n fact, exhausting the bulb is only necessary because we



annot produce, (and possibly not convey) currents of thequired frequency.

Returning to the subject of electrode lamps, it is obviouslf advantage in such a lamp to confine as much as possib

he heat to the electrode by preventing the circulation ofhe gas in the bulb. If a very small bulb be taken, it woulonfine the heat better than a large one, but it might note of sufficient capacity to be operated from the coil, or, io, the glass might get too hot. A simple way to improve his direction is to employ a globe of the required size, bu

o place a small bulb, the diameter of which is properly stimated, over the refractory button con-



IG. 157.

ained in the globe. This arrangement is illustrated in Fig57. The globe L has in this case a large neck n, allowinghe small bulb b to slip through. Otherwise the

onstruction is the same as shown in Fig. 147, forxample. The small bulb is conveniently supported uponhe stem s, carrying the refractory button in. It iseparated from the aluminum tube a by several layers o

mica M, in order to prevent the cracking of the neck by he rapid heating of the aluminum tube upon a suddenurning on of the current. The inside bulb should be asmall as possible when it is desired to obtain light only byncandescence of the electrode. If it is desired to producehosphorescence, the bulb

hould be larger, else it would be apt to get too hot, andhe hos horescence would cease. In this arran ement



sually only the small bulb shows phosphorescence, ashere is practically no bombardment against the outerlobe. In some of these bulbs constructed as illustrated iig. 157, the small tube was coated with phosphorescentaint, and beautiful effects were obtained. Instead of

making the inside bulb large, in order to avoid undueeating, it answers the purpose to make the electrode marger. In this case the bombardment is weakened by eason of the smaller electric density.

Many bulbs were constructed on the plan illustrated in

ig. 158. Here a small bulb Z>, containing the refractory utton >//, upon being exhausted to a very high degreeAvas sealed in a large globe L, which w T as thenmoderately exhausted and sealed off. The principal

dvantage of this construction was that it allowed of eaching extremely high vacua, and, at the same time of

sing a large bulb. It was found, in the course of xperiments with bulbs such as illustrated in Fig. 158,hat it was well to make the stem *, near the seal at <*,ery thick, and the leading-in wire //• thin, as it occurreometimes that the stem at e was heated and the bulb

was cracked. Often the outer globe L was exhausted only

ust enough to allow the discharge to pass through, andhe space between the bulbs appeared crimson, producincurious effect. In some cases, when the exhaustion inlobe L was very low, and the air good conducting, it wasound necessary, in order to bring the button in to highncandescence, to place, preferably on the upper part of he neck of the globe, a tinfoil coating which was



onnected to an insulated body, to the ground, or to thether terminal of the coil, as the highly conducting air

weakened the effect somewhat, probably by being actedpon inductively from the wire w, where it entered theulb at e. Another difficulty— which, however, is always

resent when the refractory button is mounted in a verymall bulb —existed in the construction illustrated in Fig58, namely, the vacuum in, the bulb 1> would bempaired in a comparatively short time.

he chief idea in the two last described constructions wa

o confine the heat to the central portion of the globe by reventing the exchange of air. An advantage is securedut owing to the heating of the inside bulb and slow vaporation of the glass, the vacuum is hard to maintainven if the construction illustrated in Fig. 157 be chosen,n which both bulbs communicate.

ut by far the better way— the ideal way — would be toeach sufficiently high frequencies. The higher therequency, the slower would be the exchange of the air,nd I think that a frequency may be reached, at whichhere would be no exchange whatever of the air molecule

round the terminal. We would then produce a flame inwhich there would be no carrying away of material, and

ueer flame it would be, for it would be rigid ! With suchigh frequencies the inertia of the particles would come

nto play. As the brush, or flame, would gain rigidity inirtue of the inertia of the particles, the exchange of the

atter would be prevented. This would necessarily occur,



or, the number of impulses being augmented, theotential energy of each would diminish, so that finally nly atomic vibrations could be set up, and the motion ofranslation through measurable space would cease. Thusn ordinary gas burner connected to a source of rapidly

lternating potential might have its efficiency augmentedo a certain limit, and this for two reasons—because of thdditional vibration imparted, and because of a slowingown of the process of carrying off. But the renewal beinendered difficult, a renewal being necessary to maintainhe burner, a continued increase of the frequency of the

mpulses, assuming they could be transmitted to andmpressed upon the flame, would result in the "xtinction" of the latter, meaning by this term only theessation of the chemical process.

think, however, that in the case of an electrode

mmersed in a fluid insulating medium, and surroundedy independent carriers of electric charges, which can bected upon inductively, a sufficient high frequency of thempulses would probably result in a gravitation of the gall around toward the electrode. For this it would be onlyecessary to assume that the independent bodies are

rregularly shaped ; they would then turn toward thelectrode their side of the greatest electric density, andhis would be a position in which the fluid resistance topproach .would be smaller than that offered to theeceding.

he general opinion, I do not doubt, is that it is out of the



uestion to reach any such frequencies as might —ssuming some of the views before expressed to be true

— produce any of the re. suits which I have pointed out amere possibilities. This may be so, but in the course of hese investigations, from the observation of many

henomena, I have gained the conviction that theserequencies would be much lower than one is apt tostimate at

rst. In a flame we set up light vibrations by causingmolecules, or atoms, to collide. But what is the ratio of th

requency of the collisions and that of the vibrations setp? Certainly it must be incomparably smaller than thatf the strokes of the bell and the sound vibrations, or thaf the discharges and the oscillations of the condenser. W

may cause the molecules of the gas to collide by the use olternate electric impulses of high frequency, and so we

may imitate the process in a flame; and from experimenwith frequencies which we are now able to obtain, I thinkhat the result is producible with impulses which areransmissible through a conductor.



n connection with thoughts of a similar nature, itppeared to me of great interest to demonstrate theigidity of a vibrating gaseous column. Although with sucow frequencies as, say 10,000 per second, which I was

ble to obtain without difficulty from a specially onstructed alternator, the task looked discouraging atrst, I made a series of experiments. The trials with air ardinary pressure led to no result, but with air

moderately rarefied I obtain what I think to be annmistakable experimental evidence of the property

ought for. As a result of this kind might lead ablenvestigators to conclusions of importance, I will describene of the experiments performed.

t is well known that when a tube is slightly exhausted,he discharge may be passed through it in the form of a

hin luminous thread. When produced with currents of ow frequency, obtained from a coil operated as usual, thhread is inert. If a magnet be approached to it, the partear the same is attracted or repelled, according to theirection of the lines of force of the magnet. It occurred t

me that if such a thread would be produced with current

f very high frequency, it should be more or less rigid, ans it was visible it could be easily studied. Accordingly Irepared a tube about one inch in diameter and one

metre long, with outside coating at each end. The tubewas exhausted to a point at which, by a little working, thhread discharge could be obtained. It must be remarked

ere that the general aspect of the tube, and the degree



xhaustion, are quite other than when ordinary low requency currents are used. As it was found preferableo work with one terminal, the tube prepared wasuspended from the end of a wire connected to theerminal, the tinfoil coating being connected to the wire,

nd to the lower coating sometimes a small insulated pla

was attached. When the thread was formed, it extendedhrough the upper part of the tube and lost itself in theower end. If it possessed rigidity it resembled, notxactly an elastic cord stretched tight between two

upports, but a cord suspended from a height with a smaweight attached at the end. When the finger or a smallmagnet was approached to the upper end of the luminouhread, it could be brought locally out of position by lectrostatic or magnetic action ; and when the disturbinbject was very quickly removed, an analogous result wa

roduced, as though a suspended cord would be displacend quickly released near the point of suspension. Inoing this the luminous thread was set in vibration, andwo very sharply marked nodes, and a third indistinctne, were formed. The vibration, once set up, continuedor fully eight minutes, dying gradually out. The speed of

he vibration often varied perceptibly, and it could bebserved that the electrostatic attraction of the glassffected the vibrating thread; but it was clear that thelectrostatic action was not the cause of the vibration, forhe thread was most generally stationary, and couldlways be set in vibration by passing the finger quickly ear the upper part of the tube. With a magnet the threa



ould be split in two and both parts vibrated. By pproaching the hand to the lower coating of the tube, or

nsulation plate if attached, the vibration was quickened lso, as far as I could see, by raising the potential orrequency. Thus, either increasing the frequency or

assing a stronger discharge of the same frequency orresponded to a tightening of the cord. I did not obtainny experimental evidence with condenser discharges. Auminous band excited in the bulb by repeated dischargef a Leyden jar must possess rigidity, and if deformed anuddenly released, should vibrate. But probably the

mount of vibrating matter is so small that in spite of thextreme speed, the inertia cannot prominently assertself. Besides, the observation in such a case is renderedxtremely difficult on account of the fundamentalibration.

he demonstration of the fact—which still needs betterxperimental confirmation—that a vibrating gaseousolumn possesses rigidity, might greatly modify the viewf thinkers. When with low frequencies and insignificantotentials indications of that property may be noted, how

must a gaseous medium be-liave under the influence of

normous electrostatic stresses which may be active inhe interstellar space, and which may alternate

with inconceivable rapidity ? The existence of such anlectrostatic, rhythmically throbbing force—of a vibratinlectrostatic field — would show a possible way how solid

might have formed from the ultra-gaseous uterus, and



ow transverse and all kinds of vibrations may beransmitted through a gaseous medium filling all space.hen, ether might be a true fluid, devoid of rigidity, andt rest, it being merely necessary as a connecting link tonable interaction. What determines the rigidity of a bod

It must be the speed and the amount of motive mattern a gas the speed may be considerable, but the density xceedingly small; in a liquid the speed would be likely toe small, though the density may be considerable ; and ioth cases the inertia resistance offered to displacement

s practically nil. But place a gaseous (or liquid) column in

n intense,rapidly alternating electrostatic field, set thearticles vibrating with enormous speeds, then the inertesistance asserts itself. A body might move with more oess freedom through the vibrating mass, but as a whole

would be rigid.

here is a subject which I must mention in connectionwith these experiments: it is that of high vacua. This is aubject, the study of which is not only interesting, butseful, for it may lead to results of great practical

mportance. In commercial ap-' paratus, such asncandescent lamps, operated from ordinary systems of

istribution, a much higher vacuum than is obtained atresent would not secure a very great advantage. In succase the work is performed on the filament, and the ga

s little concerned ; the improvement, therefore, would but trifling. But when we begin to use very highrequencies and potentials, the action of the gas becomesll important, and the degree of exhaustion materially



modifies the results. As long as ordinary coils, even very arge ones, were used, the study of the subject wasmited, because just at a point when it became most

nteresting it had to be interrupted on account of the "on-striking " vacuum being reached. But at present we

re able to obtain from a small disruptive discharge coilotentials much higher than even the largest coil wasapable of giving, and, what is more, we can make theotential alternate with great rapidity. Both of theseesults enable us now to pass a luminous dischargehrough almost any vacua obtainable, and the field of our

nvestigations is greatly extended. Think we as we may,f all the possible directions to develop a practicallnminant, the line of

igh vacua seems to be the most promising at present.ut to reach extreme vacua the appliances must be muc

more improved, and ultimate perfection will not bettained until we shall have discharged the mechanicalnd perfected an electrical vacuum pump. Molecules andtoms can be thrown out of a bulb under the action of annormous potential : this will be the principle of theacuum pump of the future. For the present, we must

ecure the best results we can with mechanicalppliances. In this respect, it might not be out of the wayo say a few words about the method of, and apparatusor, producing excessively



IG. 159.

igh degrees of exhaustion of which I have availed mysen the course of these investigations. It is very probablehat other experimenters have used similarrrangements; but as it is possible that there may be anem of interest in their description, a few remarks, whic

will render this investigation more complete, might be

ermitted.

he apparatus is illustrated in a drawing shown in Fig.59. s represents a Sprengel pump, which has beenpecially constructed to better suit the work required.he stop-cock which

s usually employed has been omitted, and instead of it a



ollow stopper s has been fitted in the neck of theeservoir K. This stopper has a small hole A, through

which the mercury descends ; the size of the outlet oeing properly determined with respect to the section ofhe fall tube £, which is sealed to the reservoir instead of

eing connected to it in the usual manner. Thisrrangement overcomes the imperfections and troubleswhich often arise from the use of the stopcock on theeservoir and the connections of the latter with the fallube.

he pump is connected through a (Jsna ped tube t to aery large reservoir & lm Especial care was taken intting the grinding surfaces of the stoppers p and p lt anoth of these and the mercury caps above them were

made exceptionally long. After the U-shaped tube wastted and put in place, it was heated, so as to soften and

ake off the strain resulting from imperfect fitting. The (Jshaped tube was provided with a stopcock c. and tworound connections y and y l —one for a small bulb b,sually containing caustic potash, and the other for theeceiver /•, to be exhausted.

he reservoir E b was connected by means of a rubberube to a slightly larger reservoir R^, each of the twoeservoirs being provided with a stopcock c t and c 2 ,espectively. The reservoir RJ could be raised andowered by a wheel and rack, and the range of its motion

was so determined that when it was filled with mercury

nd the sto cock c 2 closed so as to form a Torricellian



acuum in it when raised, it could be lifted so high thathe reservoir EJ would stand a little above stopcock c^ ;nd when this stopcock was closed and the reservoir Egescended, so as to form a Torricellian vacuum ineservoir R,, it could be lowered so far as to completely

mpty the latter, the mercury filling the reservoir RJ upo a little above stopcock c 2 .

he capacity of the pump and of the connections wasaken as small as possible relatively to the volume of eservoir, E l5 since, of course, the degree of exhaustion

epended upon the ratio of these quantities.

With this apparatus I combined the usual means indicatey former experiments for the production of very highacua. In most of the experiments it was most convenieno use caustic potash. I may venture to say, in regard to

s use, that much time is saved and a more perfect actiof the pump insured by fusing and boiling the potash asoon as, or even before, the

ump settles down. If this course is not followed, theticks, as ordinarily employed, may give off moisture at a

ertain very slow rate, and the pump may work for manyours without reaching a very high vacuum. The potash

was heated either hy a spirit lamp or by passing aischarge through it, or by passing a current through a

wire contained in it. The advantage in the latter case washat the heating couldjbe more rapidly repeated.

Generally the process of exhaustion was the following :—At the start the sto -cocks c and c t bein o en and all



ther connections closed, the reservoir n> was raised soar that the mercury filled the reservoir R t and a part ofhe narrow connecting U-shaped tube. When theump^was set to work, the mercury would, of course,uickly rise in the tube, and reservoir RJ was lowered,

he experimenter keeping ^the mercury at about theame level. The reservoir RJ wasj balanced by a longpring which facilitated the operation, and the friction of he parts was generally sufficient to keep it in almost anyosition. When the Sprengel pump had done its work,he'reservoir R% was further lowered and the mercury

escended in R t and tilled R>, whereupon stopcock 02was closed. The air adhering to the walls of R t and that

bsorbed by the mercury was carried off, and to free themercury of all air the reservoir E 2 was for a long timeworked up and down. During this process some air, whicwould gather below stopcock c 2 , was expelled from R 2

y lowering it far enough and opening the stopcock,losing the latter again before raising the reservoir. Whell the air had been expelled from the mercury, and no a

would gather in Rg when it was lowered, the causticotash was resorted to. The reservoir RJ was now againaised until the mercury in RJ stood above stopcock Ci.

he caustic potash was fused and boiled, andmoisture'partly carried off by the pump and partly re-

bsorbed ; and this process of heating and cooling wasepeated many times, and each time, upon the moistureeing absorbed or carried off, the reservoir B^ was for a

ong time raised and lowered. In this manner all themoisture was carried off from the mercur and both the



eservoirs were in proper condition to be used. Theeservoir R 2 was then again raised to the top, and theump was kept working for a long time. When the highesacuum obtainable with the pump had been reached, theotash bulb was usually wrapped with cotton which was

prinkled with ether so as to keep the potash at a very ow temperature, then the reservoir R 2 was lowered, anpon reservoir R t being emptied the receiver

was]quickly sealed up.

When a new bulb was put on, the mercury was always

aised above stopcock c,, which was closed, so as to alwayeep the mercury and both the reservoirs in fineondition, and the mercury was never withdrawn from Rexcept when the pump had reached the highest degreef exhaustion. It is necessary to observe this rule if it isesired to use the apparatus to advantage.

y means of this arrangement I was able to proceed veryuickly, and when the apparatus was in perfect order it

was possible to reach the phosphorescent stage in a smaulb in less than fifteen minutes, which is certainly very uick work for a small laboratory arrangement requiring

ll in all about 100 pounds of mercury. With ordinary mall bulbs the ratio of the capacity of the pump, receivend connections, and that of reservoir R was about 1 to0, and the degrees of exhaustion reached wereecessarily very high, though I am unable to make arecise and reliable statement how far the exhaustion wa

arried.



What impresses the investigator most in the course of hese experiences is the behavior of gases when subjecteo great^rap-idly alternating^ electrostatic stresses. Bute must remain in doubt as to whether the effectsbserved are due wholly to the molecules, or atoms, of th

as which chemical analysis discloses to us, or whetherhere enters into play another medium of a gaseousature, comprising atoms, or molecules, immersed in auid pervading the space. Such a medium surely mustxist, and I am convinced that, for instance, even if air

were absent, the surface and neighborhood of a body in

pace would be heated by rapidly alternating the potentif the body; but no such heating of the surface oreighborhood could occur if all free atoms were removednd only a homogeneous, incompressible, and elastic flui

—such as ether is supposed to be — would remain, forhen there would be no impacts, no collisions. In such a

ase, as far as the body itself is concerned, only f rictionaosses in the inside could occur.

t is a striking fact that the discharge through a gas isstablished with ever-increasing freedom as therequency of the impulses is augmented. It behaves in

his respect quite contrarily to a metallic conductor. Inhe latter the impedance enters prominently into play ashe frequency is increased, but the gas acts much as aeries of condensers would ; the facility with which theischarge passes through, seems to depend on the rate ohange of potential. If it acts so, then in a vacuum tubeven



f great length, and no matter how strong the current,elf-induction could not assert itself to any appreciableegree. We have, then, as far as we can now see, in theas a conductor which is capable of transmitting electricmpulses of any frequency which we may be able to

roduce. Could the frequency be brought high enough,hen a queer system of electric distribution, which woulde likely to interest gas companies, might be realized :

metal pipes filled with gas— the metal being the insulatohe gas the conductor—supplying phosphorescent bulbs,r perhaps devices as yet uninvented. It is certainly

ossible to take a hollow core of copper, rarefy the gas inhe same, and by passing impulses of sufficiently highrequency through a circuit around it, bring the gas insido a high degree of incandescence ; but as to the nature ohe forces there would be considerable uncertainty, for it

would be doubtful whether with such impulses the coppe

ore would act as a static screen. Such paradoxes andpparent impossibilities we encounter at every step inhis line of work, and therein lies, to a great extent, theharm of the study. I have here a short and wide tube

which is exhausted to a high degree and covered with aubstantial coating of bronze, the coating barely allowinghe light to shine through. A metallic cap, with a hook foruspending the tube, is fastened around the middleortion of the latter, the clasp being in contact with theronze coating. I now want to light the gas inside by uspending the tube on a wire connected to the coil. Any

ne who would try the experiment for the first time, notavin an revious ex erience, would robabl take



are to be quite alone when making the trial, for fear thae might become the joke of his assistants. Still, the bulbghts in spite of the metal coating, and the light can beistinctly perceived through the latter. A long tubeovered with aluminum bronze lights when held in one

and— the other touching the terminal of the coil— quiteowerfully. It might be objected that the coatings are noufficiently conducting ; still, even if they were highly esistant, they ought to screen the gas. They certainly creen it perfectly in a condition of rest, but far fromerfectly when the charge is surging in the coating. But

he loss of energy which occurs within the tube,otwithstanding the screen, is occasioned principally by he presence of the gas. Were we to take a large hollow

metallic sphere and fill it with a perfect, incompressible,uid dielectric, there would be no loss inside of the sphernd

onsequently the inside might be considered as perfectlycreened, though the potential be very rapidly lternating. Even were the sphere filled with oil, the loss

would be incomparably smaller than when the fluid iseplaced by a gas, for in the latter case the force produce

isplacements; that means impact and collisions in thenside.

No matter what the pressure of the gas may be, itecomes an important factor in the heating of a conducto

when the electric density is great and the frequency very

igh. That in the heating of conductors by lightning



ischarges, air is an element of great importance, islmost as certain as an experimental fact. I may illustrathe action of the air by the following experiment: I take ahort tube which is exhausted to a moderate degree andas a platinum wire running through the middle from on

nd to the other. I pass a steady or low frequency currenhrough the wire, and it is heated uniformly in all parts.he heating here is due to conduction, or frictional lossesnd the gas around the wire has—as far as we can see—nunction to perform. But now let me pass suddenischarges, or high frequency currents, through the wire

Again the wire is heated, this time principally on the endnd least in the middle portion ; and if the frequency of he impulses, or the rate of change, is high enough, the

wire might as well be cut in the middle as not, forractically all heating is due to the rarefied gas. Here theas might only act as a conductor of no impedance

iverting the current from the wire as the impedance of he latter is enormously increased, and merely heatinghe ends of the wire by reason of their resistance to theassage of the discharge. But it is not at all necessary thahe gas in the tube should be conducting ; it might be atn extremely low pressure, still the ends of the wire

would be heated— as, however, is ascertained by xperience— only the two ends would in such case not belectrically connected through the gaseous medium. Now

what with these frequencies and potentials occurs in anxhausted tube, occurs in the lightning discharges atrdinary pressure. We only need remember one of theacts arrived at in the course of these investigations,



amely, that to impulses of very high frequency the gas rdinary pressure behaves much in the same manner ashough it were at moderately low pressure. I think that ightning discharges frequently wires or conductingbjects are volatilized merely because air is present, and

hat, were the conductor im-

merged in an insulating liquid, it would be safe, for thenhe energy would have to spend itself somewhere else.rom the behavior of gases under sudden impulses of igh potential, I am led to conclude that there can be no

urer way of diverting a lightning discharge than by ffording it a passage through a volume of gas, if such ahing can be done in a practical manner.

here are two more features upon which I think itecessary to dwell in connection with these experiments

— the " radiant state " and the " non-striking vacuum."

Any one who has studied Crookes' work must haveeceived the impression that the " radiant state " is aroperty of the gas inseparably connected with anxtremely high degree of exhaustion. But it should be

emembered that the phenomena observed in anxhausted vessel are limited to the character and capacitf the apparatus which is made use of. 1 think that in aulb a molecule, or atom, does not precisely move in atraight line because it meets no obstacle, but because thelocity imparted to it is sufficient to propel it in a

ensibly straight line. The mean free path is one thing, buhe velocit — the ener associated with the movin



ody— is another, and under ordinary circumstances Ielieve that it is a mere question of potential or speed. Aisruptive discharge coil, when the potential is pushedery far, excites phosphorescence and projects shadows,t comparatively low degrees of exhaustion. In a lightnin

ischarge, matter moves in straight lines at ordinary ressure when the mean free path is exceedingly small,nd frequently images of wires or other metallic objectsave been produced by the particles thrown off in straighnes.

have prepared a bulb to illustrate by an experiment thorrectness of these assertions. In a globe L, Fig. 160, Iave mounted upon a lamp filament f a piece of lime /.he lamp filament is connected with a wire which leads

nto the bulb, and the general construction of the latter iss indicated in Fig. 148, before described. The bulb being

uspended from a wire connected to the terminal of theoil, and the latter being set to work, the lime piece I andhe projecting parts of the filament f are bombarded. Theegree of exhaustion is just such that with the potentialhe coil is capable of giving, phosphorescence of the glasss produced, but disappears as soon as the vacuum is

mpaired. The lime containing moisture, and moistureeing given off as soon as heating occurs, thehosphorescence lasts only for

few moments. When the lime has been sufficiently eated, enough moisture has been given oft' to impair

materiall the vacuum of the bulb. As the bombardment



oes on, one point of the lime piece is more heated thanther points, and the result is that finally practically allhe discharge passes through that point which is intenseleated, and a white stream of lime particles (Fig. 160)hen breaks forth from that point. This stream is

omposed of " radiant" matter, yet the degree of xhaustion is low. But the particles move in straight linesecause the velocity imparted to them is great, and this ue to three causes — to the great electric density, theigh temperature of the small point, and the fact that thearticles of the lime are easily

IG. 160.

orn and thrown off—far more easily than those of carbo

With frequencies such as we are able to obtain, thearticles are bodily thrown off and projected to a



onsiderable distance ; but with sufficiently highrequencies no such thing would occur ; in such case onlytress would spread or a vibration would be propagatedhrough the bulb. It would be out of the question to reaclny such frequency on the assumption that the atoms

move with the speed of light; but I believe that such ahing is impossible ; for this an enormous potential woulde required. With potentials which we are able to obtain,ven with a disruptive discharge coil, the speed must beuite insignificant.

As to the " non-striking vacuum," the point to be noted ihat it can occur only with low frequency impulses, and its


ecessitated by the impossibility of carrying off enoughnergy with such impulses in high vacuum, since the fewtoms which are around the terminal upon coining inontact with the same, are repelled and kept at a distancor a comparatively long period of time, and not enough

work can be performed to render the effect perceptible t

he eye. If the difference of potential between theerminals is raised, the dielectric breaks down. But withery high frequency impulses there is no necessity foruch breaking down, since any amount of work can beerformed by continually agitating the atoms in thexhausted vessel, provided the frequency is higli enough

t is easy to reach—even with



IG. 161.



IG. 162.

requencies obtained from an alternator as here used —atage at which the discharge does not pass between twolectrodes in a narrow tube, each of these being connecte

o one of the terminals of the coil, but it is difficult to reacpoint at which a luminous discharge would not occur

round each electrode.

A thought which naturally presents itself in connectionwith high frequency currents, is to make use of their

owerful electro-dynamic inductive action to produceght effects in a sealed glass globe. The leading-in wire isne of the defects of the present incandescent lamp, and o other improvement were made, that imperfection at

east should be done away with. Following

his thought, I have carried on experiments in variousirections, of which some were indicated in my formeraper. I may here mention one or two more lines of xperiment which have been followed up.

Many bulbs were constructed as shown in Fig. 161 and

ig. 162.

n Fig. 161, a wide tube, T, was sealed to a smaller W haped tube u, of phosphorescent glass. In the tube T,

was placed a coil c, of aluminum wire, the ends of whichwere provided with small spheres, t and ^, of aluminum,

nd reached into the u tube. The tube T was sli ed into



ocket containing a primary coil, through which usually he discharges of Leyden jars were directed, and thearefied gas in the small u tube was excited to stronguminosity by the high-tension current induced in the co. When Leyden jar discharges were used to induce

urrents in the coil c, it was found necessary to pack theube T tightly with insulating powder, as a dischargewould occur frequently between the turns of the coil,

specially when the primary was thick and the air gap,hrough which the jars discharged, large, and no littlerouble was experienced in this way.

n Fig. 162 is illustrated another form of the bulbonstructed. In this case a tube T is sealed to a globe L.he tube contains a coil c, the ends of which pass through

wo small glass tubes t and ti, which are sealed to the tub. Two refractory buttons m and m t are mounted on

amp filaments which are fastened to the ends of the wirassing through the glass tubes t and £,. Generally inulbs made on this plan the globe L communicated withhe tube T. For this purpose the ends of the small tubes nd ti were heated just a trifle in the burner, merely toold the wires, but not to interfere with the

ommunication. The tube T, with the small tubes, wireshrough the same, and the refractory buttons in and m l9

were first prepared, and then sealed to globe L,whereupon the coil c was slipped in and the connectionsmade to its ends. The tube was then packed withnsulating powder, jamming the latter as tight as possiblep to very nearly the end ; then it was closed and only a



mall hole left through which the remainder of the powdwas introduced, and finally the end of the tube was closeUsually in bulbs constructed as shown in Fig. 162 an

luminum tube a was fastened to the upper end s of eachf the tubes t and £ b in order to protect that end agains

he heat. The buttons m and m^ could be brought to anyegree

f incandescence by passing the discharges of Leyden jarround the coil c. In such bulbs with two buttons a very urious effect is produced by the formation of the

hadows of each of the two buttons.

Another line of experiment, which has been assiduously ollowed, was to induce by electro-dynamic induction aurrent or luminous discharge in an exhausted tube orulb. This matter has received such able treatment at th

ands of Prof. J. J. Thomson, that I could add but little towhat he has made known, even had I made it the specialubject of this lecture. Still, since experiments in this lineave gradually led me to the present views and results, a

ew words must be devoted here to this subject.

t has occured, no doubt, to many that as a vacuum tubes made longer, the electromotive force per unit length ofhe tube, necessary to pass a luminous discharge throughhe latter, becomes continually smaller ; therefore, if thexhausted tube be made long enough, even with low requencies a luminous discharge could be induced in suc

tube closed upon itself. Such a tube might be placedround a hall or on a ceilin and at once a sim le



ppliance capable of giving considerable light would bebtained. But this would be an appliance hard to

manufacture and extremely unmanageable. It would noto to make the tube up of small lengths, because there

would be with ordinary frequencies considerable loss in

he coatings, and besides, if coatings were used, it woulde better to supply the current directly to the tube by onnecting the coatings to a transformer. But even if allbjections of such nature were removed, with low requencies the light conversion itself would be inefficiens I have before stated. In using extremely high

requencies the length of the secondary — in other wordshe size of the vessel— can be reduced as much asesired, and the efficiency of the light conversion is

ncreased, provided that means are invented forfficiently obtaining such high frequencies. Thus one ised, from theoretical and practical considerations, to the

se of high frequencies, and this means highlectromotive forces and small currents in the primary.

When one works with condenser charges—and they arehe only means up to the present known for reachinghese extreme frequencies—one gets to electromotiveorces of several thousands of volts per turn of the

rimary. "We cannot multiply the electro-dynamicnductive effect by taking

more turns in the primary, for we arrive at the conclusiohat the best way is to work with one single turn —hough we must sometimes depart from this rule — and

we must et alon with whatever inductive effect we can



btain with one turn. But before one has longxperimented with the extreme frequencies required toet up in a small bulb an electromotive force of severalhousands of volts, one realizes the great importance of lectrostatic effects, and these effects grow relatively to

he electro-dynamic in significance as the frequency isncreased.

Kow, if anything is desirable in this case, it is to increasehe frequency, and this would make it still worse for thelectro-dynamic effects. On the other hand, it is easy to

xalt the electrostatic action as far as one likes by takingmore turns on the secondary, or combining self-inductiond capacity to raise the potential. It should also beemembered that, in reducing the the current to themallest value and increasing the potential, the electricmpulses of high frequency can be more easily

ransmitted through a conductor.

hese and similar thoughts determined me to devotemore at tention to the electrostatic phenomena, and tondeavor to produce potentials as high as possible, andlternating as fast as they could be made to alternate. I

hen found that I could excite vacuum tubes atonsiderable distance from a conductor connected to aroperly constructed coil, and that I could, by convertinghe oscillatory current of a conductor to a higher potentiastablish electrostatic alternating fields which actedhrough the whole extent of the room, lighting up a tube

o matter where it was held in space. I thought I



ecognized that I had made a step in advance, and I haveersevered in this line ; but I wish to say that I share witll lovers of science and progress the one and only desire

— to reach a result of utility to men in any direction towhich thought or experiment may lead me. I think that

his departure is the right one, for I cannot see, from thebservation of the phenomena which manifest themselves the frequency is increased, what there would remain tct between two circuits conveying, for instance, impulsef several hundred millions per second, exceptlectrostatic forces. Even with such trifling frequencies

he energy would be practically all potential, and my onviction has grown strong that, to whatever kind of

motion light may be due, it is produced by tremendouslectrostatic stresses vibrating with extreme rapidity.

Of all these phenomena observed with currents, or

lectric impulses, of high frequency, the most fascinatingor an audience are certainly those which are noted in anlectrostatic field acting through considerable distance;nd the best an unskilled lecturer can do is to begin andnish with the exhibition of these singular effects. I take

ube in my hand and move it about, and it is lighted

wherever I may hold it; throughout space the invisibleorces act. But I may take another tube and it might notght, the vacuum being very high. I excite it by means odisruptive discharge coil, and now it will light in the

lectrostatic

IG. 163. FIG. 164.



eld. I may put it away for a few weeks or months, still itetains the faculty of being excited. What change have Iroduced in the tube in the act of exciting it? If a motion

mparted to atoms, it is difficult to perceive how it canersist so long without being arrested by f rictional losse

and if a strain exerted in the dielectric, such as a simplelectrification would produce, it is easy to see how it mayersist indefinitely, but very difficult to understand why uch a condition should aid the excitation when we haveo deal with potentials which are rapidly alternating.

ince I have exhibited these phenomena for the first timhave obtained some other interesting effects. For

nstance, I have produced the incandescence of a button,lament, or wire enclosed in a tube. To get to this result

was necessary to economize the energy which is obtainedrom the field, and direct most of it on the small body to

e rendered incandescent. At the beginning the task ppeared difficult, but the experiences gatheredermitted me to reach the result easily. In Fig. 163 andig. 164, two such tubes are illustrated, which arerepared for the occasion. In Fig. 163 a short tube TJ,ealed to another long tube T, is provided with a stem .$

with a platinum wire sealed in the latter. A very thin lamlament I, is fastened to this wire and connection to theutside is made through a thin copper wire w. The tube irovided with outside and inside coatings, c and GJ,espectively, and is filled as far as the coatings reach withonducting, and the space above with insulating, powderhese coatin s are merel used to enable me to erform



wo experiments with the tube—namely, to produce theffect desired either by direct connection of the body of he experimenter or of another body to the wire w, or bycting inductively through the glass. The stem s isrovided with an aluminum tube «, for purposes before

xplained, and only a small part of the filament reachesut of this tube. By holding the tube T : anywhere in thelectrostatic field, the filament is rendered incandescent.

A more interesting piece of apparatus is illustrated in Fig64. The construction is the same as before, only instead

f the lamp filament a small platinum wire ^>, sealed in tem s, and bent above it in a circle, is connected to theopper wire w, which is joined to an inside coating c. A mall stem * M is provided with a needle, on the point of

which is arranged, to rotate very freely, a very light fan mica v. To prevent the fan from falling out, a thin stem o

lass </, is bent properly and fastened to the aluminumube. When the glass tube is held anywhere in thelectrostatic field the platinum wire becomesncandescent, and the mica vanes are rotated very fast.

ntense phosphorescence may be excited in a bulb by

merely connecting it to a plate within the field, and thelate need not be any larger than an ordinary lamp shadhe phosphorescence excited with these currents is

ncomparably more powerful than with ordinary pparatus. A small phosphorescent bulb, when attachedo a wire connected to a coil, emits sufficient light

o allow readin ordinar rint at a distance of five to six



aces. It was of interest to see how some of thehosphorescent bulbs of Professor Crookes would behav

with these currents, and he has had the kindness to lendme a few for the occasion. The effects produced aremagnificent, especially by the sulphide of calcium and

ulphide of zinc. With the disruptive discharge coil they low intensely merely by holding them in the hand andonnecting the body to the terminal of the coil.

o whatever results investigations of this kind may leadhe chief interest lies, for the present, in the possibilities

hey offer for the production of an efficient illuminatingevice. In no branch of electric industry is an advance

more desired than in the manufacture of light. Every hinker, when considering the barbarous methodsmployed, the deplorable losses incurred in our bestystems of light production, must have asked himself,

What is likely to be the light of the future ? Is it to be anncandescent solid, as in the present lamp, or anncandescent gas, or a phosphorescent body, or somethinke a burner, but incomparably more efficient ?

here is little chance to perfect a gas burner; not,

erhaps, because human ingenuity has been bent uponhat problem for centuries without a radical departureaving been made— though the argument is not devoid o

orce—but because in a burner the highest vibrations canever be reached, except by passing through all the low nes. For how is a flame to proceed unless by a fall of

fted wei hts ? Such rocess cannot be maintained



without renewal, and renewal is repeated passing fromow to high vibrations. One way only seems to be open tomprove a burner, and that is by trying to reach higheregrees of incandescence. Higher incandescence isquivalent to a quicker vibration : that means more light

rom the same material, and that again, means nioreconomy. In this direction some improvements have beemade, but the progress is hampered by many limitationsDiscarding, then, the burner, there remains the threeways first mentioned, which are essentially electrical.

uppose the light of the immediate future to be a solid,endered incandescent by electricity. Would it not seemhat it is better to employ a small button than a fraillament ? From many considerations it certainly must boncluded that a button is capable of a higher economy,ssuming, of course, the difficulties connected with the

peration of such a lamp to be effec-

vely overcome. But to light such a lamp we require aigh potential ; and to get this economically, we must useigh frequencies.

uch considerations apply even more to the production oght by the incandescence of a gas, or by hosphorescence. In all cases we require high frequenciend high potentials. These thoughts occurred to me a lonme ago.

ncidentally we gain, by the use of high frequencies, mandvantages, such as higher economy in the light



roduction, the possibility of working with one lead, theossibility of doing away with the leading-in wire, etc.

he question is, how far can we go with frequencies ?Ordinary conductors rapidly lose the facility of

ransmitting electric impulses when the frequency isreatly increased. Assume the means for the productionf impulses of very great frequency brought to the utmoerfection, every one will naturally ask how to transmithem when the necessity arises. In transmitting suchmpulses through conductors we must remember that w

ave to deal with pressure and flow, in the ordinary nterpretation of these terms. Let the pressure increaseo an enormous value, and let the flow correspondingly iminish, then such impulses— variations merely of ressure, as it were — can no doubt be transmittedhrough a wire even if their frequency be many hundred

f millions per second. It would, of course, be out of uestion to transmit such impulses through a wire

mmersed in a gaseous medium, even if the wire wererovided with a thick and excellent insulation, for most ohe energy would be lost in molecular bombardment andonsequent heating. The end of the wire connected to th

ource would be heated, and the remote end wouldeceive but a trifling part of the energy supplied. Therime necessity, then, if such electric impulses are to besed, is to find means to reduce as much as possible theissipation.

he first thought is, to employ the thinnest possible wire



urrounded by the thickest practicable insulation. Theext thought is to employ electrostatic screens. The

nsulation of the wire may be covered with a thinonducting coating and the latter connected to the grounut this would not do, as then all the energy would pass

hrough the conducting coating to the ground and nothinwould get to the end of the wire. If a ground connection imade it can only be made through a conductor offer-

ng an enormous impedance, or through a condenser of xtremely small capacity. This, however, does not do

way with other difficulties.

f the wave length of the impulses is much smaller thanhe length of the wire, then corresponding short waves

will be set up in the conducting coating, and it will be mor less the same as though the coating were directly

onnected to earth. It is therefore necessary to cut up thoating in sections much shorter than the wave length.uch an arrangement does not still afford a perfectcreen, but it is ten thousand times better than none. Ihink it preferable to cut up the conducting coating inmall sections, even if the current waves be much longer

han the coating.

f a wire were provided with a perfect electrostaticcreen, it would be the same as though all objects wereemoved from it at infinite distance. The capacity wouldhen be reduced to the capacity of the wire itself, which

would be very small. It would then be possible to sendver the wire current vibrations of ver hi h fre uencies



t enormous distances, without affecting greatly theharacter of the vibrations. A perfect screen is of courseut of the question, but I believe that with a screen suchs I have just described telephony could be renderedracticable across the Atlantic. According to my ideas, th

utta-percha covered wire should be provided with ahird conducting coating subdivided in sections. On theop of this should be again placed a layer of gutta-perchand other insulation, and on the top of the whole thermor. But such cables will not be constructed, for ereong intelligence—transmitted without wires — will throb

hrough the earth like a pulse through a living organism.he wonder is that, with the present state of knowledgend the experiences gained, no attempt is being made toisturb the electrostatic or magnetic condition of thearth, and transmit, if nothing else, intelligence.

t has been, my chief aim in presenting these results tooint out phenomena or features of novelty, and todvance ideas which I am hopeful will serve as startingoints of new departures. It has been my chief desire thivening to entertain you with some novel experiments.our applause, so frequently and generously accorded,

as told me that I have succeeded.

n conclusion, let me thank you most heartily for yourindness and attention, and assure you that the honor Iave had in

ddressing such a distinguished audience, the pleasure Iave had in resentin these results to a atherin of so



many able men— and among them also some of those inwhose work for many years past I have foundnlightenment and constant pleasure— I shall neverorget.

HAPTER XXVIII. ON LIGHT AND OTHER HIGHREQUENCY PHENOMENA. 1

NTRODUCTORY. SOME THOUGHTS ON THE EYE.

WHEN we look at the world around us, on Nature, we ar

mpressed with its beauty and grandeur. Each thing weerceive, though it may be vanishingly small, is in itself aworld, that is, like the whole of the universe, matter andorce governed by law,—a world, the contemplation of

which fills us with feelings of wonder and irresistibly urgs to ceaseless thought and inquiry. But in all this vast

world, of all objects our senses reveal to us, the mostmarvellous, the most appealing to our imagination,

ppears no doubt a highly developed organism, a thinkineing. If there is anything fitted to make us admire

Nature's handiwork, it is certainly this inconceivabletructure, which performs its innumerable motions of

bedience to external influence. To understand itsworkings, to get a deeper insight into this Nature'smasterpiece, has ever been for thinkers a fascinating aim

nd after many centuries of arduous research men haverrived at a fair understanding of the functions of itsrgans and senses. Again, in all the perfect harmony of it

arts of the arts which constitute the material or



angible of our being, of all its organs and senses, the eyes the most wonderful. It is the most precious, the mostndispensable of our perceptive or directive organs, it ishe great gateway through which all knowledge enters th

mind. Of all our organs, it is the one, which is in the

A lecture delivered before the Franklin Institute,hiladelphia, February* 1893, and before the Nationallectric Light Association, St. Louis, March,

most intimate relation with that which we call intellect. S

ntimate is this relation, that it is often said, the very souhows itself in the eye.

t can he taken as a fact, which the theory of the action ohe eye implies, that for each external impression, that isor each image produced upon the retina, the ends of the

isual nerves, concerned in the conveyance of thempression to the mind, must be under a peculiar stressr in a vibratory state. It now does not seem improbablehat, when by the power of thought an image is evoked, istinct reflex action, no matter how weak, is exertedpon certain ends of the visual nerves, and therefore

pon the retina. Will it ever be within human power tonalyze the condition of the retina when disturbed by hought or reflex action, by the help of some optical orther means of such sensitiveness, that a clear idea of itstate might be gained at any time 2 If this were possiblehen the problem of reading cne's thoughts with precision

ke the characters of an open book, might be much easieo solve than man roblems belon in to the domain of



ositive physical science, in the solution of which many, iot the majority, of scientific men implicitly believe.

Helmholtz, has shown that the fundi of the eye arehemselves, luminous, and he was able to see, in totalarkness, the movement of his arm by the light of his ow

yes. This is one of the most remarkable experimentsecorded in the history of science, and probably only a femen could satisfactorily repeat it, for it is very likely, thahe luminosity of the eyes is associated with uncommonctivity of the brain and great imaginative power. It isuorescence of brain action, as it were.

Another fact having a bearing on this subject which hasrobably been noted by many, since it is stated in populaxpressions, but which I cannot recollect to have foundhronicled as a positive result of observation is, that atmes, when a sudden idea or image presents itself to the

ntellect, there is a distinct and sometimes painfulensation of luminosity produced in the eye, observableven in broad daylight.

he saying then, that the soul shows itself in the eye, iseeply founded, and we feel that it expresses a great

ruth. It has a profound meaning even for one who, like aoet or artist, only following his inborn instinct or love fo

Nature, finds delight in aimless thoughts and in the mereontemplation of natural phenomena, but a still morerofound meaning for one who, in the

pirit of positive scientific investigation, seeks to ascertaihe causes of the effects. It is rinci all the natural



hilospher, the physicist, for whom the eye is the subjectf the most intense admiration.

wo facts about the eye must forcibly impress the mindf the physicist, notwithstanding he may think or say tha

is an imperfect optical instrument, forgetting, that theery conception of that which is perfect or seems so toim, has been gained through this same instrument. Firshe eye is, as far as our positive knowledge goes, the onlyrgan which is directly affected by that subtile medium,

which as science teaches us, must fill all space ; secondly

is the most sensitive of our organs, incomparably moreensitive to external impressions than any other.

he organ of hearing implies the impact of ponderableodies, the organ of smell the transference of detached

material particles, and the organs of taste, and of touch o

orce, the direct contact, or at least some interference of onderable matter, and this is true even in those

nstances of animal organisms, in which some of thesergans are developed to a degree of truly marvelouserfection. This being so, it seems wonderful that thergan of , ^' c sight solely should be capable of being

tirred by that, which all • •>„ our other organs areowerless to detect, yet which plays an essential part inll natural phenomena, which transmits all energy andustains all motion and, that most intricate of all, life, but

which has properties such that even a scientifically rained mind cannot help drawing a distinction between nd all that is called matter. Considerin merel this, and



he fact that the eye, by L its marvelous power, widensur otherwise very narrow range of «•'*' •' perception fareyond the limits of the small world which is our own, tombrace myriads of other worlds, suns and stars in the vinfinite depths of the universe, would make it justifiable

o assert, . that it is an organ of a higher order. Itserformances are beyond ->. comprehension. Nature asar as we know never produced any-t *t« thing more

wonderful. We can get barely a faint idea of its ',rodigious power by analyzing what it does and by omparing. When ether waves impinge upon the human

ody, they produce the sensations of warmth or cold,leasure or pain, or perhaps other #v~a sensations of

which we are not aware, and any degree or intensity /^cWof these sensations, which degrees are infinite inumber, hence an infinite number of distinct sensations.ut our sense of touch, or our sense of force, cannot

eveal to us these differences in degree

r intensity, unless they are very great. Now we caneadily conceive how an organism, such as the human, in

he eternal process of evolution, or more philosophically peaking, adaptation to Nature, being constrained to these of only the sense of touch or force, for instance, mighevelop this sense to such a degree of senstiveness orerfection, that it would be capable of distinguishing the

minutest differences in the temperature of a body even a

ome distance, to a hundredth, or thousandth, or million



art of a degree. Yet, even this apparently impossibleerformance would not begin to compare with that of theye, which is capable of distinguishing and conveying tohe mind in a single instant innumerable peculiarities of he body, be it in form, or color, or other respects. This

ower of the eye rests upon two things, namely, theectilinear propagation of the disturbance by which it isffected, and upon its sensitiveness. To say that the eye ensitive is not saying anything. Compared with it, allther organs are monstrously crude. The organ of smell

which guides a dog on the trail of a deer, the organ of

ouch or force which guides an-insect in its wanderings,he organ of hearing, which is affected by the slightestisturbances of the air, are sensitive organs, to be sure,ut what are they compared with the human eye! Nooubt it responds to the faintest echoes or rveliberations of the medium ; no doubt, it brings us

dings from other worlds, infinitely remote, but in aanguage we cannot as yet always understand. And why ot ? Because we live in a medium filled with air and othases, vapors and a dense mass of solid particles flyingbout. These play an important part in many phenomenthey fritter away the energy of the vibrations before

hey can reach the eye; they too, are the carriers of germf destruction, they get into our lungs and other organs,log up the channels and imperceptibly, yet inevitably,rrest the stream of life. Could we but do away with allonderable matter in the line of sight of the telescope, it

would reveal to us undreamt of marvels. Even thenaided eye, I think, would be capable of distinguishing i



he pure medium, small objects at distances measuredrobably by hundreds or perhaps thousands of miles.

ut there is something else about the eye whichmpresses us still more than these wonderful features

which we observed, viewing it from the standpoint of ahysicist, merely as an optical instrument,—somethingwhich appeals to us more than its marvelous faculty of

eing directly affected by the vibrations of the

medium, without interference of gross matter, and more

han its inconceivable sensitiveness and discerning powet is its significance in the processes of life. No matterwhat one's views oh nature and life may be, he musttand amazed when, for the first time in his thoughts, heealizes the importance of the eye in the physicalrocesses and mental performances of the human

rganism. And how could it be otherwise, when heealizes, that the eye is the means through which theuman race has acquired the entire knowledge itossesses, that it controls all our motions, more still, allur actions.

here is no way of acquiring knowledge except throughhe eye. What is the foundation of all philosophicalystems of ancient and modern times, in fact, of all thehilosophy of man ? / am, I think • I think, therefore Iainut how could I think and how would I know that I exist

f I had not the eye ? For knowledge involves

onsciousness ; consciousness involves ideas, conceptionsonce tions involve ictures or ima es and ima es the



ense of vision, and therefore the organ of sight. But howbout blind men, will be asked ? Yes, a blind man may epict in magnificent poems, forms and scenes from realfe, from a world he physically does not see. A blind man

may touch the keys of an instrument with unerring

recision, may model the fastest boat, may discover andnvent, calculate and construct, may do still greaterwonders — but all the blind men who have done suchhings have descended from those who had seeing eyes.

Nature may reach the same result in many ways. Like awave in the physical world, in the infinite ocean of the

medium which pervades all, so in the world of organismsn life, an impulse started proceeds onward, at times, mae, with the speed of light, at times, again, so slowly thator ages and ages it seems to stay, passing throughrocesses of a complexity inconceivable to men, but in als forms, in all its stages, its energy ever and ever

ntegrally present. A single ray of light from a distant staalling upon the eye of a tyrant in bygone times, may havltered the course of his life, may have changed theestiny of nations, may have transformed the surface of he globe, so intricate, so inconceivably complex are therocesses in Nature. In no way can we get such an

verwhelming idea of the grandeur of Nature, as when wonsider, that in accordance with the law of theonservation of energy, throughout the infinite, the forcere in a perfect balance,- and hence the energy of a singlhought may determine the motion of a Uni-

7"&#3i tVVH/t.



^ ^ \£***^^svt*4&f » -^/-'

erse. It is not necessary that every individual, not evenhat every generation or many generations, should havehe physical instrument of sight, in order to be able to

orm images and to think, that is, form ideas oronceptions ; but sometime or other, during the process volution, the eye certainly must have existed, elsehought, as we understand it, would be impossible ; elseonceptions, like spirit, intellect, mind, call it as you mayould not exist. It is conceivable, that in some other worl

n some other beings, the eye is replaced by a differentrgan, equally or more perfect, but these beings cannot b

men.

Now r what prompts us all to voluntary motions andctions of any kind ? Again the eye. If I am conscious of

he motion, I must have an idea or conception, that is, anmage, therefore the eye. If I am not precisely consciousf the motion, it is, because the images are vague or

ndistinct, being blurred by the superim-position of manut when I perform the motion, does the impulse whichrompts me to the action come from within or from

without ? The greatest physicists have not disdained tondeavor to answer this and similar questions and have mes abandoned themselves to the delights of pure andnrestrained thought. Such questions are generally onsidered not to belong to the realm of positive physicacience, but will before long be annexed to its domain.

Helmholtz has robabl thou ht more on life than an



modern scientist. Lord Kelvin expressed his belief thatfe's process is electrical and that there is a force inhereno the organism and determining its motions. Just as

much as I am convinced of any physical truth I amonvinced that the motive impulse must come from the

utside. For, consider the lowest organism we know —nd there are probably many lower ones— an aggregatiof a few cells only.' If it is capable of voluntary motion itan perform an infinite number of motions, all definite anrecise. But now a mechanism consisting of a finiteumber of parts and few at that, cannot perform an

nfinite number of definite motions, hence the impulseswhich govern its movements must come from thenvironment. So, the atom, the ulterior element of the

Universe's structure, is tossed about in space, eternally, lay to external influences, like a boat in a troubled sea.

Were it to stop its motion it would die. Matter at rest, if

uch a thing could exist, would be matter dead. Death of matter! Never has a sentence of deeper philosophicalmeaning been uttered. This is the way in which Prof.Dewar

orcibly expresses it in the description of his admirable

xperiments, in which liquid oxygen is handled as oneandles water, and air at ordinary pressure is made toondense and even to solidify by the intense cold.xperiments, which serve to illustrate, in his language,

he last feeble manifestations of life, the last quiverings omatter about to die. But human eyes shall not witnessuch death. There is no death of matter, for throughout



he infinite universe, all has t3 move, to vibrate, that is, tve.

have made the preceding statements at the peril of reading upon metaphysical ground, in my desire to

ntroduce the subject of this lecture in a manner notltogether uninteresting, I may hope, to an audience sucs I have the honor to address. But now, then, returningo the subject, this divine organ of sight, this indispensabnstrument for thought and all intellectual enjoyment,

which lays open to us the marvels of this universe,

hrough which we have acquired what knowledge weossess, and which prompts us to, and controls, all ourhysical and mental activity. By what is it affected? By ght! What is light ?

We have witnessed the great strides which have been

made in all departments of science in recent years. Soreat have been the advances that we cannot refrain fromsking ourselves, Is this all true, or is it but a dream ?enturies ago men have lived, have thought, discovered,

nvented, and have believed that they were soaring, whilhey were merely proceeding at a snail's pace. So we too

may be mistaken. But taking the truth of the observedvents as one of the implied facts of science, we mustejoice in the immense progress already made and still

more in the anticipation of what must come, judging fromhe possibilities opened up by modern research. There isowever, an advance which we have been witnessing,

which must be particularly gratifying to every lover of



rogress. It is not a discovery, or an invention, or anchievement in any particular direction. It is an advancen all directions of scientific thought and experiment. I

mean the generalization of the natural forces andhenomena, the looming up of a certain broad idea on th

cientific horizon. It is this idea which has, however, longgo taken possession of the most advanced minds, towhich I desire to call your attention, and which I intend t

lustrate in a general way, in these experiments, as therst step in answering the question "What is light?" and

o realize the modern meaning of this word.

t is beyond the scope of my lecture to dwell upon theubject of light in general, my object being merely to brinresently to your notice a certain class of light effects annumber of phenomena observed in pursuing the study

f these effects. But to be consistent in my remarks it is

ecessary to state that, according to that idea, now ccepted by the majority of scientific men as a positiveesult of theoretical and experimental investigation, thearious forms or manifestations of energy which wereenerally designated as "electric" or more precisely electromagnetic " are energy manifestations of the sam

ature as those of radiant heat and light. Therefore thehenomena of light and heat and others besides these,

may be called electrical phenomena. Thus electricalcience has become the mother science of all and its studas become all important. The day when we shall know xactly what "electricity" is, will chronicle an eventrobably greater, more important than any other



ecorded in the history of the human race. The time willome when the comfort, the very existence, perhaps, of

man will depend upon that wonderful agent. For ourxistence and comfort we require heat, light and

mechanical power. How do we now get all these? We get

hem from fuel, we get them by consuming material.What will man do when the forests disappear, when theoal fields are exhausted ? Only one thing, according tour present knowledge will remain; that is, to transmitower at great distances. Men will go to the waterfalls, tohe tides, which are the stores of an infinitesimal part of

Nature's immeasurable energy. There will they harnesshe energy and transmit the same to their settlements, t

warm their homes by, to give them light, and to keepheir ooedient slaves, the machines, toiling. But how willhey transmit this energy if not by electricity ? Judgehen, if the comfort, nay, the very existence, of man will

ot depend on electricity. I am aware that this view is nohat of a practical engineer, but neither is it that of anlusionist, for it is certain, that power transmission, whict present is merely a stimulus to enterprise, will someay be a dire necessity.

t is more important for the student, who takes up thetudy of light phenomena, to make himself thoroughly cquainted with certain modern views, than to perusentire books on the subject of light itself, as disconnectedrom these views. Were I therefore to make theseemonstrations before students seeking information—an

or the sake of the few of those who may be



resent, give me leave to so assume—it would be my rincipal endeavor to impress these views upon their

minds in this series of experiments.

t might be sufficient for tins purpose to perform a simpl

nd well-known experiment. I might take a familiarppliance, a L3yden jar, charge it from a frictionalmachine, and then discharge it. In explaining to you its

ermanent state when charged, and its transitory ondition when discharging, calling your attention to theorces which enter into play and to the various

henomena they produce, and pointing out the relation ohe forces and phenomena, I might fully succeed inlustrating that modern idea. Xo doubt, to the thinker,his simple experiment would appeal as much as the mos

magnificent display. But this is to be an experimentalemonstration, and one which should possess, besides

nstructive, also entertaining features and as such, aimple experiment, such as the one cited, would not goery far towards the attainment of the lecturer's aim. I

must therefore choose another way of illustrating, morepectacular certainly, but perhaps also more instructive.nstead of the frictional machine and Leyden jar, I shall

vail myself in these experiments, of an induction coil of eculiar properties, which was described in detail by me

n a lecture before the London Institution of Electricalngineers, in Feb., 1892. This induction coil is capable of ielding currents of enormous potential differences,lternating with extreme rapidity. "With this apparatus hall endeavor to show ou three distinct classes of



ffects, or phenomena, and it is my desire that eachxperiment, while serving for the purposes of illustrationhould at the same time teach us some novel truth, orhow us some novel aspect of this fascinating science. Buefore doing this, it seems proper and useful to dwell

pon the apparatus employed, and method of obtaininghe high potentials and high-frequency currents which amade use of in these experiments.

ON THE APPARATUS AND METHOD OFONVERSION.

hese high-frequency currents are obtained in a peculiamanner. The method employed was advanced by me

bout two years ago in an experimental lecture before thAmerican Institute of Electrical Engineers. A number of ways, as practiced in the laboratory, of obtaining these

urrents either from continuous or low frequency lternating currents, is diagramatically indicated in Fig.65, which will be later described in detail. The general

HIGH FREQUENCY AND HIGH POTENTIALURRENTS.



lan is to charge condensers, from a direct or alternate-urrent source, preferably of high-tension, and toischarge them disruptively while observing well-knownonditions necessar to maintain the oscillations of the



urrent. In view of the general interest taken in high-requency currents and effects producible by them, iteems to me advisable to dwell at some length upon this

method of conversion. In order to give you a clear idea ohe action, I will suppose that a continuous-current

enerator is employed, which is often very convenient. Is desirable that the generator should possess such highension as to be able to break through a small air space. his is not the case, then auxiliary means have to beesorted to, some of which will be indicated subsequently

When the condensers are charged to a certain potential,

he air, or insulating space, gives way and a disruptiveischarge occurs. There is then a sudden rush of currentnd generally a large portion of accumulated electricalnergy spends itself. The condensers are thereuponuickly charged and the same process is repeated in morr less rapid succession. To produce such sadden rushes

f current it is necessary to observe certain conditions. Ihe rate at which the condensers are discimrged is theame as that at which they are charged, then, clearly, inhe assumed case the condensers do not come into play. he rate of discharge be smaller than the rate of charginghen, again, the condensers cannot play an important

art. But if, on the contrary, the rate of discharging isreater than that of charging, then a succession of rushesf current is obtained. It is evident that, if the rate at

which the energy is dissipated by the discharge is very much greater than the rate of supply to the condensers,he sudden rushes will be comparatively few, with long-me intervals between. This alwavs occurs when a



ondenser of considerable capacity is charged by means comparatively small machine. If the rates of supply andissipation are not widely different, then the rushes of urrent will be in quicker succession, and this the more,he more nearly equal both the rates are, until limitation

ncident to eacli case and depending upon a number of auses are reached. Thus we are able to obtain from aontinuous-current generator as rapid a succession of ischarges as we like. Of course, the higher the tension ohe generator, the smaller need be the capacity of theondensers, and for this reason, principally, it is of

dvantage to employ a generator of very high tension.esides, such a generator permits the attaining of greateates of vibration.

he rushes of current may be of the same direction undehe conditions before assumed, but most generally there

s an oscillation superimposed upon the fundamentalibration of the current. When the conditions are soetermined that there are no oscillations, the current

mpulses are unidirectional and thus a means is providedf transforming a continuous current of high tension, intodirect current of lower tension, which I think may find

mployment in the arts.

his method of conversion is exceedingly interesting andwas much impressed by its beauty when I firstonceived it. It is ideal in certain respects. It involves themployment of no mechanical devices of any kind, and it

llows of obtaining currents of any desired frequency



rom an ordinary circuit, direct or alternating. Therequency of the fundamental discharges depending onhe relative rates of supply and dissipation can be readilyaried within wide limits, by simple adjustments of theseuantities, and the frequency of the superimposed

ibration by the determination of the capacity, self-nduction and resistance of the circuit. The potential of thurrents, again, may be raised as high as any insulation iapable of withstanding safely by combining capacity andelf-induction or by induction in a secondary, which needave but comparatively few turns.

As the conditions are often such that the intermittence oscillation does not readily establish itself, especially whedirect current source is employed, it is of advantage to

ssociate an interrupter with the arc, as I have, someme ago, indicated the use of an air-blast or magnet, or

ther such device readily at hand. The magnet ismployed with special advantage in the conversion of irect currents, as it is then very effective. If the primarource is an alternate current generator, it is desirable, ahave stated on another occasion, that the frequency hould be low, and that the current forming the arc be

arge, in order to render the magnet more effective.

A form of such discharger with a magnet which has beenound convenient, and adopted after some trials, in theonversion of direct currents particularly, is illustrated inig. 166. N s are the pole pieces of a very strong magnet

which is excited by a coil c. The pole pieces are slotted fo



djustment and can be fastened in any position by screws^ The discharge rods d d t1 thinned down on the ends

n order to allow a closer approach of the magnetic poleieces, pass through the columns of brass b ^ and are

astened in position by screws # 2 $2- Springs r r t and

ollars c c±

re slipped on the rods, the latter serving to set the poinf the rods at [a certain suitable distance by means of crews # 3 s s , and the former to draw the points apart.

When it is desired to start the arc, one of the large rubbe

andles h Ji± is tapped quickly with the [hand, whereby he points of the rods are brought in contact but arenstantly separated by the springs r r^ Such anrrangements-has been found to be often necessary,amely in cases when the E. M. r. was not large enough tause the discharge to break through the gap, and also

when it was desirable to avoid short circuiting of theenerator by the metallic contact of the rods. The rapiditf the interruptions of the current with a magnet dependn the intensity of the magnetic field and on the



IG. ICG.

otential difference at the end of the arc. Thenterruptions are generally in such quick succession as to

roduce a musical sound. Years ago it was observed thatwhen a powerful induction coil is discharged between theoles of a strong magnet, the discharge produces a loudoise, not unlike a small pistol shot, It was vaguely statedhat the spark was intensified by the presence of the

magnetic field. It is now clear that the discharge current,

owing for some time, was interrupted a great number omes by the magnet, thus producing the sound. Thehenomenon is especially marked when the field circuit olarge magnet or dynamo is broken in a powerful

magnetic field.

When the current through the gap is comparatively largeis of advantage to slip on the points of the discharge

ods pieces of very hard carbon and let the arc play etween the carbon pieces. This preserves the rods, andesides has the advantage of keeping the air space hottes the heat is not conducted away as quickly through the

arbons, and the result is that a smaller E. M. F. in the arap is required to maintain a succession of discharges.

Another form of discharger, which may be employed witdvantage in some cases, is illustrated in Fig. 167. In thisorm the discharge rods d d^ pass through perforations i

wooden



IG. 107.

ox B, which is thickly coated with mica on the inside, as

ndicated by the heavy lines. The perforations arerovided with mica tubes m m^ of some thickness, whichre preferably not in contact with the rods d d { . The boas a cover c which is a little larger and descends on theutside of the box. The spark gap is warmed by a small

amp I contained in the box. A plate p above the lamp

llows the draught to pass only through the chimney <? he lamp, the air entering through holes o o in or near thottom of the box and following the path indicated by thrrows. When the discharger is in operation, the door of he box is closed so that the light of the arc is not visibleutside.



t is desirable to exclude the light as perfectly as possibles it interferes with some experiments. This form of ischarger is simple and very effective when properly

manipulated. The air being warmed to a certainemperature, has its insulating power impaired; it

ecomes dielectrically weak, as it were, and theonsequence is that the arc can be established at muchreater distance. The arc should, of course, be sufficientlnsulating to allow the discharge to pass through the gapisruptively. The arc formed under such conditions, whe

ong, may be made extremely sensitive, and the weak

raught through the lamp chimney c is quite sufficient toroduce rapid interruptions. The adjustment is made byegulating the temperature and velocity of the draught.nstead of using the lamp, it answers the purpose torovide for a draught of warm air in other ways. A very imple way which has been practiced is to enclose the arc

n a long vertical tube, with plates on the top and bottomor regulating the temperature and velocity of the airurrent. Some provision had to be made for deadening thound.

he air may be rendered dielectrically weak also by

arefaction. Dischargers of this kind have likewise beensed by me in connection with a magnet. A large tube is

or this purpose provided with heavy electrodes of carbor metal, between which the discharge is made to pass,he tube being placed in a powerful magnetic field. Thexhaustion of the tube is carried to a point at which theischarge breaks through easily, but the pressure should



e more than Y5 millimetres, at which the ordi. nary hread discharge occurs. In another form of discharger,ombining the features before mentioned, the discharge

was made to pass between two adjustable magnetic poleieces, the space between them being kept at an elevate

emperature.

t should be remarked here that when such, ornterrupting devices of any kind, are used and theurrents are passed through the primary of a disruptiveischarge coil, it is not, as a rule, of advantage to produce

number of interruptions of the current per secondreater than the natural frequency of vibration of theynamo supply circuit, which is ordinarily small. It shoullso ba pointed out here, that while the devices mentione

n connection with the disruptive discharge aredvantageous under certain conditions, they may be

ometimes a source of trouble, as they producentermittences and other irregularities in the vibration

which it would be very desirable to overcome.

here is, I regret to say, in this beautiful method of onversion a defect, which fortunately is not vital, and

which I have been gradually overcoming. I will best callttention to this defect and indicate a fruitful line of worky comparing the electrical process with its mechanicalnalogue. The process may be illustrated in this mannermagine a tank with a wide opening at the bottom, whichs kept closed by spring pressure, but so that it snaps off

uddenly when the liquid in the tank has reached a



ertain height. Let the fluid be supplied to the tank by means of a pipe feeding at a certain rate. When the critic

eight of the liquid is reached, the spring gives way andhe bottom of the tank drops out. Instantly the liquid fallhrough the wide opening, and the spring, reasserting

self, closes the bottom again. The tank is now filled, andfter a certain time interval the same process is repeatedt is clear, that if the pipe feeds the fluid quicker than theottom outlet is capable of letting it pass through, theottom will remain off. and the tank will still overflow. Ifhe rates of supply are exactly equal, then the bottom lid

will remain partially open and no vibration of the same"nd of the liquid column will generally occur, though it

might, if started by some means. But if the inlet pipe doeot feed the fluid fast enough for the outlet, then there

will be always vibration. Again, in such case, each time thottom flaps up or down, the spring and the liquid

olumn, if the pliability of the spring and the inertia of thmoving parts are properly chosen, will performndependent vibrations. In this analogue the fluid may bkened to electricity or electrical energy, the tank to theondenser, the spring to the dielectric, and the pipe to thonductor through which electricity is supplied to the

ondenser. To make this analogy quite complete it isecessary to make the assumption, that the bottom, eachme it gives way, is knocked violently against a non-lastic stop, this impact involving some loss of energy ;nd that, besides, some dissipation of energy results dueo frictional losses. In the preceding analogue the liquid iupposed to be under a steady pressure. If the presence



f the fluid be assumed to vary rhythmically, this may baken as corresponding to the case of an alternatingurrent. The process is then not quite as simple toonsider, but the action is the same in principle.

t is desirable, in order to maintain the vibrationconomically, to reduce the impact and frictional losses amuch as possible.

As regards the latter, which in the electrical analogueorrespond to the losses due to the resistance of the

ircuits, it is impossible to obviate them entirely, but thean be reduced to a minimum by a proper selection of thimensions of the circuits and by the the employment ofhin conductors in the form of strands. But the loss of nergy caused by the first breaking through of theielectric—which in the above example corresponds to th

iolent knock of the bottom against the inelastic stop—would be more important to overcome. At the moment ohe breaking through, the air space has a very highesistance, which is probably reduced to a very smallalue when the current has reached some strength, andhe space is brought to a high temperature. It would

materially diminish the loss of energy if the space werelways kept at an extremely high temperature, but thenhere would be no disruptive break. By warming thepace moderately by means of a lamp or otherwise, theconomy as far as the arc is concerned is sensibly ncreased. But the magnet or other interrupting device

oes not diminish the loss in the arc. Likewise, a jet of air



nly facilitates the carrying off of the energy. Air, or a gan general, behaves curiously in this respect. When twoodies charged to a very high potential, dischargeisrupt-ively through an air space, any amount of energy

may be carried off by the air. This energy is evidently

issipated by bodily carriers, in impact and collisionalosses of the molecules. The exchange of the molecules inhe space occurs with inconceivable rapidity. A powerfulischarge taking place between two electrodes, they mayemain entirely cool, and yet the loss in the air may epresent any amount of energy. It is perfectly

racticable, with very great potential differences in theap, to dissipate several horse-power in the arc of theischarge without even noticing a small increase in theemperature of the electrodes. All the frictional lossesccur then practically in the air. If the exchange of the ai

molecules is prevented, as by enclosing the air

ermetically, the gas inside of the vessel is broughtuickly to a high temperature, even with a very smallischarge. It is difficult to estimate how much of thenergy is lost in sound waves, audible or not, in a powerfischarge. When the currents through the gap are large,he electrodes may become rapidly heated, but this is no

reliable measure of the energy wasted in the arc, as theoss through the gap itself may be comparatively small.he air or a gas in general is, 'at ordinary pressure at

east,

learly not the best medium through which a disruptiveischarge should occur. Air or other gas under great



ressure is of course a much more suitable medium forhe discharge gap. I have carried on long-continuedxperiments in this direction, unfortunately lessracticable on account of the difficulties and expense inetting air under great pressure. But even if the medium

n the discharge space is solid or liquid, still the sameosses take place, though they are generally smaller, forusfr as soon as the arc is established, the solid or liquid iolatilized. Indeed, there is no body known which wouldot be disintegrated by the arc, and it is an open questionmong scientific men, whether an arc discharge could

ccur at all in the air itself without the particles of thelectrodes being torn off. When the current through theap is very small and the arc very long, I believe that aelatively considerable amount of heat is taken up in theisintegration of the electrodes, which partially on thisccount may remain quite cold.

he ideal medium for a discharge gap should only crack,nd the ideal electrode should be of some material whichannot be disintegrated. With small currents through theap it is best to employ aluminum, but not when theurrents are large. The disruptive break in the air, or

more or less in any ordinary medium, is not of the naturef a crack, but it is rather comparable to the piercing of

nnumerable bullets through a mass offering greatrictional resistances to the motion of the bullets, thisnvolving considerable loss of energy. A medium which

would merely crack when strained electrostatically—andhis possibly might be the case with a perfect vacuum,



hat is, pure ether—would involve a very small loss in thap, so small as to be entirely negligible, at leastheoretically, because a crack may be produced by annfinitely small displacement. In exhausting an oblongulb provided with two aluminum terminals, with the

reatest care, I have succeeded in producing such aacuum that the secondary discharge of a disruptiveischarge coil would break disrup-tively through the bul

n the form of fine spark streams. The curious point washat the discharge would completely ignore the terminalnd start far behind the two aluminum plates which

erved as electrodes. This extraordinary high vacuumould only be maintained for a very short while. To returo the ideal medium, think, for the sake of illustration, ofiece of glass or similar body clamped in a vice, and the

atter tightened more and

more. At a certain point a minute increase of the pressurwill cause the glass to crack. The loss of energy involvedn splitting the glass may be practically nothing, forhough the force is great, the displacement need be butxtremely smalL Now imagine that the glass wouldossess the property of closing again perfectly the crack

pon a minute diminution of the pressure. This is the wahe dielectric in the discharge space should behave. Butnasmuch as there would be always some loss in the gap,he medium, which should be continuous, shouldxchange through the gap at a rapid rate. In thereceding example, the glass being perfectly closed, it

would mean that the dielectric in the discharge space



ossesses a great insulating power; the glass beingracked, it would signify that the medium in the space is ood conductor. The dielectric should vary enormously iesistance by minute variations of the E. M. F. across theischarge space. This condition is attained, but in an

xtremely imperfect manner, by warming the air space tcertain critical temperature, dependent on the E. M. F.cross the gap, or by otherwise impairing the insulatingower of the air. But as a matter of fact the air does nevereak down disruptively, if this term be rigorously

nterpreted, for before the sudden rush of the current

ccurs, there is always a weak current preceding it, whicises first gradually and then with comparativeuddenness. That is the reason why the rate of change isery much greater when glass, for instance, is brokenhrough, than when the break takes place through an airpace of equivalent dielectric strength. As a medium for

he discharge space, a solid, or even a liquid, would bereferable therefor. It is somewhat difficult to conceive osolid body which would possess the property of closing

nstantly after it has been cracked. But a liquid, especiallnder great pressure, behaves practically like a solid,

while it possesses the property of closing the crack. Henc

was thought that a liquid insulator might be moreuitable as a dielectric than air. Following out this idea, aumber of different forms of dischargers in which aariety of such insulators, sometimes under greatressure, were employed, have been experimented upont is thought sufficient to dwell in a few words upon one ohe forms experimented upon. One of these dischargers



lustrated in Figs. 168« and 168 b.

A hollow metal pulley P (Fig. 16 8«), was fastened uponn arbor #, which by suitable means was rotated at aonsiderable

peed. On the inside of the pulley, but disconnected fromhe same, was supported a thin disc h (which is shownhick for the sake of clearness), of hard rubber in whichhere were embedded two metal segments s s with

metallic extensions e e into which were screwed

onducting terminals t t covered with thick tubes of hardubber 11. The rubber disc h with its metallic segments s9, was finished in a lathe, and its entire surface highly olished so as to offer the smallest possible frictionalesistance to the motion through a fluid. In the hollow of he pulley an insulating liquid such as a thin oil was

oured so as to reach very nearly to the opening left inhe flange/, which was screwed tightly on the front side ohe pulley. The terminals t t, were connected to thepposite coatings of a battery of condensers so that theischarge occurred through the liquid. When the pulley

was rotated, the liquid was forced against the rim of the

ulley and considerable fluid pressure resulted. In thisimple way the discharge gap



IG. 168a.

IG. 168b.

was filled with a medium which behaved practically like olid, which possessed the quality of closing instantly upohe occurrence of the break, and which moreover wasirculating through the gap at a rapid rate. Very powerfuffects were produced by discharges of this kind withquid interrupters, of which a number of different forms

were made. It was found that, as expected, a longer sparor a given length of wire was obtainable in this way thany using air as an interrupting device. Generally thepeed, and therefore also the fluid pressure, was limitedy reason of the fluid friction, in the form of dischargerescribed, but the practically obtainable speed was more

han sufficient to produce a number of breaks suitable fohe circuits ordinarily used. In such instances the metalulley P was provided with a few projections inwardly,nd a definite number of breaks was then produced whicould be computed from the speed of

otation of the pulley. Experiments were also carried onwith liquids of different insulating power with the view o



educing the loss in the arc. When an insulating liquid ismoderately warmed, the loss in the arc is diminished.

A point of some importance was noted in experimentswith various discharges of this kind. It was found, for

nstance, that whereas the conditions maintained in thesorms were favorable for the production of a great sparkength, the current so obtained was not best suited to theroduction of light effects. Experience undoubtedly hashown, that for such purposes a harmonic rise and fall ofhe potential is preferable. Be it that a solid is rendered

ncandescent, or phosphorescent, or be it that energy isransmitted by condenser coating through the glass, it isuite certain that a harmonically rising and fallingotential produces less destructive action, and that theacuum is more permanently maintained. This would beasily explained if it were ascertained that the process

oing on in an exhausted vessel is of an electrolyticature.

n the diagrammatical sketch, Fig. 165, which has beenlready referred to, the cases which are most likely to be

met with in practice are illustrated. One has at his

isposal either direct or alternating currents from aupply station. It is convenient for an experimenter in ansolated laboratory to employ a machine G, such aslustrated, capable of giving both kinds of currents. Inuch case it is also preferable to use a machine with

multiple circuits, as in many experiments it is useful and

onvenient to have at one's dis osal currents of different



hases. In the sketch, D represents the direct and A thelternating circuit. In each of these, three branch circuitsre shown, all of which are provided with double linewitches s s s s s s. Consider first the direct currentonversion; m represents the simplest case. If the E. M.

. of the generator is sufficient to break through a smallir space, at least when the latter is warmed or otherwisendered poorly insulating, there is no difficulty in

maintaining a vibration with fair economy by judiciousdjustment of the capacity, self-induction and resistancef the circuit L containing the devices II m. The magnet

N, s, can be in this case advantageously combined withhe air space. The discharger d d with the magnet may blaced either way, as indicated by the full or by the dottenes. The circuit la with the connections and devices isupposed to possess dimensions such as are suitable for

he maintenance of a vibration. But usually the E. M. r. ohe circuit or branch \a will be something like a 1 00 voltr so, and in this case it is not sufficient to break throughhe gap. Many different means may be used to remedy his by raising the E. M. F. across the gap. The simplest irobably to insert a large self-induction coil in series with

he circuit L. When the arc is established, as by theischarger illustrated in B'ig. 166, the magnet blows therc out the instant it is formed. Now the extra current ofhe break, being of high E. M. F., breaks through the gapnd a path of low resistance for the dynamo current beingain provided, there is a sudden rush of current from thynamo upon the weakening or subsidence of the extra



urrent. This process is repeated in rapid succession, andn this manner I have maintained oscillation with as low 0 volts, or even less, across the gap. But conversion onhis plan is not to be recommended on account of the tooeavy currents through the gap and consequent heating

f the electrodes ; besides, the frequencies obtained inhis way are low, owing to the high self-inductionecessarily associated with the circuit. It is very desirabl

o have the E. M. F. as high as possible, first, in order toncrease the economy of the conversion, and, secondly, tbtain high frequencies. The difference of potential in thi

lectric oscillation is, of course, the equivalent of thetretching force in the mechanical vibration of the springo obtain very rapid vibration in a circuit of some inertiagreat stretching force or difference of potential isecessary. Incidentally, when the E. M. F. is very great,

he condenser which is usually employed in connection

with the circuit need but have a small capacity , and manther advantages are gained. With a view of raising the E

M. F. to a many times greater value than obtainable fromrdinary distribution circuits, a rotating transformer g issed, as indicated at i la, Fig. 165, or else a separate highotential machine is driven by means of a motor operate

rom the generator G. The latter plan is in fact preferabls changes are easier made. The connections from theigh tension winding are quite similar to those in branch

a with the exception that a condenser c, which should bedjustable, is connected to the high tension circuit.

Usually, also, an adjustable self-induction coil in serieswith the circuit has been employed in these experiments



When the tension of the currents is very high, the magnerdinarily used in connection with the discharger is of omparatively small

alue, as it is quite easy to adjust the dimensions of the

ircuit so that oscillation is maintained. The employmentf a steady E. M. F. in the high frequency conversionffords some advantages over the employment of lternating E. M. F., as the adjustments are much simplend the action can be easier controlled. But unfortunatelyne is limited by the obtainable potential difference. The

winding also breaks down easily in consequence of theparks which form between the sections of the armaturer commutator when a vigorous oscillation takes place.esides, these transformers are expensive to build. It haeen found by experience that it is best to follow the planlustrated at ma. In this arrangement a rotating

ransformer g, is employed to convert the low tensionirect currents into low frequency alternating currents,referably also of small tension. The tension of theurrents is then raised in a stationary transformer T. Thecondary s of this transformer is connected to andjustable condenser c which discharges through the gap

r discharger dd, placed in either of the ways indicated,hrough the primary p of a disruptive discharge coil, theigh frequency current being obtained from theecondary s of this coil, as described on previousccasions. This will undoubtedly be found the cheapestnd most convenient way of converting direct currents.



he three branches of the circuit A represent the usualases met in practice when alternating currents areonverted. In Fig. 15 a condenser c., generally of largeapacity, is connected to the circuit L containing theevices Z Z, m m. The devices mm are supposed to be of

igh self-induction so as to bring the frequency of theircuit more or less to that of the dynamo. In this instanche discharger d d should best have a number of makesnd breaks per second equal to twice the frequency of thynamo. If not so, then it should have at least a numberqual to a multiple or even fraction of the dynamo

requency. It should be observed, referring to iJ, that theonversion to a high potential is also effected when theischarger d d, which is shown in the sketch, is omitted.ut the effects which are produced by currents which ris

nstantly to high values, as in a disruptive discharge, arentirely different from those produced by dynamo

urrents which rise and fall harmonically. So, for instancehere might be in a given case a number of makes andreaks at d d equal to just twice the frequency of theynamo,"or in other words, there may be the sameumber of fundamental oscillations as would be pro-

uced without the discharge gap, and there might evenot be any quicker superimposed vibration ; yet theifferences of potential at the various points of the circuihe impedance and other phenomena, dependent uponhe rate of change, will bear no similarity in the two casehus, when working with currents discharging dis-

uptively, the element chiefly to be considered is not the



requency, as a student might be apt to believe, but theate of change per unit of time. With low frequencies in aertain measure the same . effects may be obtained as

with high frequencies, provided the rate of change isufficiently great. So if a low frequency current is raised

potential of, say, 75,000 volts, and the high tensionurrent passed through a series of high resistance lamplaments, the importance of the rarefied gas surroundinhe filament is clearly noted, as will be seen later; or, if aow frequency current of several thousand amperes isassed through a metal bar, striking phenomena of

mpedance are observed, just as with currents of highrequencies. But it is, of course, evident that with low requency currents it is impossible to obtain such rates ohange per unit of time as with high frequencies, hencehe effects produced by the latter are much morerominent. It is deemed advisable to make the precedin

emarks, inasmuch as many more recently describedffects have been unwittingly identified with highrequencies. Frequency alone in reality does not meannything, except when an undisturbed harmonicscillation is considered.

n the branch uib a similar disposition to that in ib islustrated, with the difference that the currentsischarging through the gap d d are used to induceurrents in the secondary s of a transformer T. In suchase the secondary should be provided with an adjustablondenser for the purpose of tuning it to the primary.



b illustrates a plan of alternate current high frequency onversion which is most frequently used and which isound to be most convenient. This plan has been dweltpon in detail on previous occasions and need not beescribed here.

ome of these results were obtained by the use of a highrequency alternator. A description of such machines wile found in my original paper before the Americannstitute of Electrical Engineers, and in periodicals of thaeriod, notably in THE ELECTRICAL ENGINEER of

March 18, 1891.

will now proceed with the experiments.

ON PHENOMENA PRODUCED BY ELECTROSTATICORCE.

he first class of effects I intend to show you are effectsroduced by electrostatic force. It is the force whichoverns the the motion of the atoms, which causes themo collide and develop the life-sustaining energy of heatnd light, and which causes them to aggregate in an in

nite variety of ways, according to Nature's fancifulesigns, and to form all these wondrous structures weerceive around us ; it is, in fact, if our present views berue, the most important force for us to consider in

Nature. As the term electrostatic might imply a steady lectric condition, it should be remarked, that in these

xperiments the force is not constant, but varies at a ratwhich may be considered moderate, about one million



mes a second, or thereabouts. This enables me toroduce many effects which are not producible with annvarying force.

When two conducting bodies are insulated and electrified

we say that an electrostatic force is acting between themhis force manifests itself in attractions, repulsions andtresses in the bodies and space or medium without. Soreat may be the strain exerted in the air, or whatevereparates the two conducting bodies, that it may break own, and we observe sparks or bundles of light or

treamers, as they are called. These streamers formbundantly when the force through the air is rapidly arying. I will illustrate this action of electrostatic force inovel experiment in which I will employ the induction

oil before referred to. The coil is contained in a troughlled with oil, and placed under the table. The two ends*

f the secondary wire pass through the two thick columnf hard rubber which protrude to some height above theable. It is necessary to insulate the ends or terminals of he secondary heavily with hard rubber, because evenry wood is by far too poor an insulator for these currentf enormous potential differences. On one of the termina

f the coil, I have placed a large sphere of sheet brass,which is connected to a larger insulated brass plate, in

rder to enable me to perform the experiments underonditions, which, as you will see, are more suitable forhis experiment. I now set the coil to work and approachhe free terminal with a metallic object held in my hand,his simply to avoid burns. As I approach the metallic



bject to a distance of eight or ten inches, a torrent of urious sparks breaks forth from the end of the secondar

wire, which

asses through the rubber column. The sparks cease

when the metal in my hand touches the wire. My arm isow traversed by a powerful electric current, vibrating about the rate of one million times a second. All around

me the electrostatic force makes itself felt, and the airmolecules and particles of dust flying about are acted

pon and are hammering violently against my body. So

reat is this agitation of the particles, that when the lightre turned out you may see streams of feeble light appean some parts of my body. When such a streamer breaksut on any part of the body, it produces a sensation likehe pricking of a needle. Were the potentials sufficiently igh and the frequency of the vibration rather low, the

kin would probably be ruptured under the tremendoustrain, and the blood would rush out with great force inhe form of fine spray or jet so thin as to be invisible, justs oil will when placed on the positive terminal of

IG. 169.



Holtz machine. The breaking through of the skin thougmay seem impossible at first, would perhaps occur, by

eason of the tissues under the skin being incomparably etter conducting. This, at least, appears plausible,

udging from some observations. I can make thesetreams of light visible to all, by touching with the metallbject one of the terminals as before, and approaching mree hand to the brass sphere, which is connected to theecond terminal of the coil. As the hand is approached, thir between it and the sphere, or in the immediate

eighborhood, is more violently agitated, and you seetreams of light now break forth from my finger tips androm the whole hand (Fig. 169). Were I to approach theand closer, powerful sparks would jump from the brassphere to my hand, which might be injurious. Thetreamers offer no particular inconvenience, except that

n the ends of the finger

ps a burning sensation is felt. They should not beonfounded with those produced by an influence machinecause in many respects they behave differently. I havttached the brass sphere and plate to one of the

erminals in order to prevent the formation of visibletreamers on that terminal, also in order to preventparks from jumping at a considerable distance. Besides,he attachment is favorable for the working of the coil.

he streams of light which you have observed issuing

rom my hand are due to a potential of about 200,000olts alternatin in rather irre ular intervals sometime



ke a million times a second. A vibration of the samemplitude, but four times as fast, to maintain which over,000,000 volts would be required, would be more thanufficient to envelop my body in a complete sheet of ame. But this flame would not burn me up ; quite

ontrarily, the probability is that I would not be injured ihe least. Yet a hundredth part of that energy, otherwiseirected, would be amply sufficient to kill a person.

he amount of energy which may thus be passed into thody of a person depends on the frequency and potentia

f the currents, and by making both of these very great, ast amount of energy may be passed into the body without causing any discomfort, except perhaps, in the

rm, which is traversed by a true conduction current. Theason why no pain in the body is felt, and no injuriousffect noted, is that everywhere, if a current be imagined

o flow through the body, the direction of its flow would bt right angles to the surface; hence the body of thexperimenter offers an enormous section to the current,nd the density is very small, with the exception of therm, perhaps, where the density may be considerable.ut if only a small fraction of that energy would be applie

n such a way that a current would traverse the body inhe same manner as a low frequency current, a shock

would be received which might be fatal. A direct or low requency alternating current is fatal, I think, principallyecause its distribution through the body is not uniform,s it must divide itself in minute streamlets of greatensity, whereby some organs are vitally injured. That



uch a process occurs I have not the least doubt, thougho evidence might apparently exist, or be found uponxamination. The surest to injure and destroy life, is aontinuous current, but the most painful is an alternatingurrent of very low frequency. The expression of these

iews, which are the result of long con-

nued experiment and observation, both with steady anarying currents, is elicited by the interest which is atresent taken in this subject, and by the manifestly rroneous ideas which are daily propounded in journals o

his subject.

may illustrate an effect of the electrostatic force by nother striking experiment, but before, I must call yourttention to one or two facts. I have said before, that

when the medium between two oppositely electrified

odies is strained beyond a certain limit it gives way andtated in popular language, the opposite electric chargesnite and neutralize each other. This breaking down of he medium occurs principally when the force actingetween the bodies is steady, or varies at a moderateate. Were the variation sufficiently rapid, such a

estructive break would not occur, no matter how greathe force, for all the energy would be spent in radiation,onvection and mechanical and chemical action. Thus thepark length, or greatest distance which a spark will jumetween the electrified bodies is the



IG. 170a. FIG. 170b.

maller, the greater the variation or time rate of change.ut this rule may be taken to be true only in a general

way, when comparing rates which are widely different.

will show you by an experiment the difference in theffect produced by a rapidly varying and a steady or

moderately varying force. I have here two large circularrass plates p p (Fig. 170# and Fig. 1706), supported on

movable insulating stands on the table, connected to thends of the secondary of a coil similar to the one usedefore. I place the plates ten or twelve inches apart andet the coil to work. You see the whole space between th

lates, nearly two cubic feet, filled with uniform light, Fig70«. This light is due to the streamers you have seen inhe first experiment, which are now much more intense.ave already pointed out the importance of thesetreamers in commercial apparatus and their still greatemportance in some purely scientific investigations. Ofte

hey are too weak to be visible, but

hey always exist, consuming energy and modifying thection of the apparatus. When intense, as they are atresent, they produce ozone in great quantity, and also,s Professor Crookes has pointed out, nitrous acid. So

uick is the chemical action that if a coil, such as this ones worked for a ver lon time it will make the



tmosphere of a small room unbearable, for the eyes andhroat are attacked. But when moderately produced, thetreamers refresh the atmosphere wonderfully, like ahunderstorm, and exercises unquestionably a beneficialffect.

n this experiment the force acting between the plateshanges in intensity and direction at a very rapid rate. I

will now make the rate of change per unit time muchmaller. This I effect by rendering the discharges throughe primary of the induction coil less frequent, and also b

iminishing the rapidity of the vibration in the secondaryhe former result is conveniently secured by lowering th. M. r. over the air gap in the primary circuit, the lattery approaching the two brass plates to a distance of abouhree or four inches. When the coil is set to work, you seeo streamers or light between the plates, yet the medium

etween them is under a tremendous strain. I still furtheugment the strain by raising the E. M. F. in the primaryircuit, and soon you see the air give way and the hall isluminated by a shower of brilliant and noisy sparks, FigTO&. These sparks could be produced also withnvarying force ; they have been for many years a

amiliar phenomenon, though they were usually obtainedrom an entirely different apparatus. In describing thesewo phenomena so radically different in appearance, Iave advisedly spoken of a " force " acting between thelates. It would be in accordance with accepted views toay, that there was an " alternating E. M. F," actingetween the plates. This term is quite proper and



pplicable in all cases where there is evidence of at least ossibility of an essential inter-dependence of the electritate of the plates, or electric action in their neighborhoout if the plates were removed to an infinite distance, or t a finite distance, there is no probability or necessity

whatever for such dependence. I prefer to use the term lectrostatic force," and to say that such a force is actinground each plate or electrified insulated body in generahere is an inconvenience in using this express/on as the

erm incidentally means a steady electric condition ; but roper nomenclature will eventually settle this difficulty

now return to the experiment to which I have already lluded, and with which I desire to illustrate a strikingffect produced by a rapidly varying electrostatic force. Ittach to the end of the wire, I (Fig. 171), which is inonnection with one of the terminals of the secondary of

he induction coil, an exhausted bulb I. This bulb containthin carbon filament/, which is fastened to a platinum

wire w, sealed in the glass and leading outside of the bulbwhere it connects to the wire I. The bulb may be

xhausted to any degree attainable with ordinary pparatus. Just a moment before, you have witnessed th

reaking down of the air between the charged brasslates. You know that a plate of glass, or any other

nsulating material, would break down in like manner.Had I therefore a metallic coating attached to the outside

f the bulb, or placed near the same, and



IG. 171.

IG. 172a

IG. 172b.

were this coating connected to the other terminal of theoil, you would be prepared to see the glass give way if

he strain were sufficiently increased. Even were theoating not connected to the other terminal, but to annsulated plate, still, if you have followed recentevelopments, you would naturally expect a rupture of he glass.

ut it will certainly surprise you to note that under thection of the var in electrostatic force the lass ives



way when all other bodies are removed from the bulb. Inact, all the surrounding bodies we perceive might beemoved to an infinite distance without affecting theesult in the slightest. Wher^he coil is set to work, thelass is invariably broken through at the seal, or other

arrow channel, and the vacuum is quickly impaired.

uch a damaging break would not occur with a steady orce, even if the same were many times greater. Thereak is due to the agitation of the molecules of the gas

within the bulb, and outside of the same. This agitation,

which is generally most violent in the narrow pointedhannel near the seal, causes a heating and rupture of thlass. This rupture, would, however, not occur, not even

with a varying force, if the medium filling the inside of thulb, and that surrounding it, were perfectly omogeneous. The break occurs much quicker if the top

f the bulb is drawn out into a h'ne fibre. In bulbs usedwith these coils such narrow, pointed channels mustherefore be avoided.

When a conducting body is immersed in air, or similarnsulating medium, consisting of, or containing, small

reely movable particles capable of being electrified, andwhen the electrification of the body is made to undergo a

ery rapid change—which is equivalent to saying that thlectrostatic force acting around the body is varying inntensity,—the small particles are attracted and repellednd their violent impacts against the body may cause a

mechanical motion of the latter. Phenomena of this kind



re noteworthy, inasmuch as they have not beenbserved before with apparatus such as has beenommonly in use. If a very light conducting sphere beuspended on an exceedingly fine wire, and charged to ateady potential, however high, the sphere will remain at

est. Even if the potential would be rapidly varying,rovided that the small particles of matter, molecules ortoms, are evenly distributed, no motion of the spherehould result. But if one side of the conducting sphere isovered with a thick insulating layer, the impacts of thearticles will cause the sphere to move about, generally i

rregular curves, Fig. 172&. In like manner, as I havehown on a previous occasion, a fan of sheet metal, Fig.72&, covered partially with insulating material asndicated, and placed upon the terminal of the coil so as turn freely, on it, is spun around.

All these phenomena you have witnessed and otherswhich will be shown later, are due to the presence of amedium like air, and would not occur in a continuousmedium. The action of the air may be illustrated still

etter by the following experiment. I take a glass tube t,ig. 173, of about an inch in diameter, which has a

latinum wire w sealed in the lower end, and to which isttached a thin lamp filament f. I connect the wire withhe terminal of the coil and set the coil to work. The

latinum wire is now electrified positively and negativelyn rapid succession and the wire and air inside of the tube

s rapidly heated by the impacts of the particles, which



may be so violent as to render the filament incandescentut if I pour oil in the tube, just as soon as the wire isovered with the oil, all action apparently ceases andhere is no marked evidence of heating. The reason of ths that the oil is a practically continuous medium. The

isplacements in such a continuous medium are, withhese frequencies, to all appearance incomparably smallehan in air, hence the work performed in such a mediums insignificant. But oil would behave very differently witrequencies many times as great, for even though theisplacements



IG. 178.

IG. 174.

e small, if the frequency were much greater,

onsiderable work might be performed in the oil.

he electrostatic attractions and repulsions betweenodies of measurable dimensions are, of all the

manifestations of this force, the first so-called electricalhenomena noted. But though they have been known to

s for many centuries, the precise nature of themechanism concerned in these actions is still unknown tos, and has not been even quite satisfactorily explained.

What kind of mechanism must that be ? We cannot helpwondering when we observe two magnets attracting andepelling each other with a force of hundreds of pounds

with apparently nothing between them. We have in ourommercial dynamos magnets capable of sustaining in

mid-air tons of weight. But what are even these

orces acting between magnets when compared with theremendous attractions and repulsions produced by

lectrostatic force, to which there is apparently no limit ao intensity . In lightning discharges bodies are oftenharged to so high a potential that they are thrown away

with inconceivable force and torn asunder or shatterednto fragments. Still even such effects cannot compare

with the attractions and repulsions which exist between

har ed molecules or atoms and which are sufficient to



roject them with speeds of many kilometres a second, shat under their violent impact bodies are renderedighly incandescent and are volatilized. It is of special

mtev- ,t for the thinker who inquires into the nature of hese forces I./ note that whereas the actions between

ndividual molecules ( • atoms occur seemingly under anyonditions, the attractions and repulsions of bodies of measurable dimensions imply a medium possessingnsulating properties. So, if air, either by being rarefied oeated, is rendered more or less conducting, these actionetween two electrified bodies practically cease, while th

ctions between the individual atoms continue to manifehemselves.

An experiment may serve as an illustration and as ameans of bringing out other features of interest. Some

me ago I showed that a lamp filament or wire mounted

n a bulb and connected to one of the terminals of a highension secondary coil is set spinning, the top of thelament generally describing a circle. This vibration wasery energetic when the air in the bulb was at ordinary ressure and became less energetic when the air in theulb was strongly compressed. It ceased altogether when

he air was exhausted so as to become comparatively ood conducting. I found at that time that no vibrationook place when the bulb was very highly exhausted. Buconjectured that the vibration which I ascribed to thelectrostatic action between the walls of the bulb and thelament should take place also in a highly exhausted bulbo test this under conditions which were inore favorable



bulb like the one in Fig. 174, was constructed. Itomprised a globe 5, in the neck of which was sealed alatinum wire w carrying a thin lamp filament/. In the

ower part of the globe a tube t was sealed so as tourround the filament. The exhaustion was carried as far

s it was practicable with .the apparatus employed.

his bulb verified my expectation, for the filament waset spinning when the current was turned on, and becamncandes-

ent. It also showed another interesting feature, bearingpon the preceding remarks, namely, when the filamentad been kept incandescent some time, the narrow tubend the space inside were brought to an elevatedemperature, and as the gas in the tube then becameonducting, the electrostatic attraction between the glass

nd the filament became very weak or ceased, and thelament came to rest. When it came to rest it would glow

ar more intensely. This was probably due to its assuminhe position in the centre of the tube where the moleculaombardment was most intense, and also partly to theact that the individual impacts were more violent and

hat no part of the supplied energy was converted intomechanical movement. Since, in accordance with accepte

iews, in this experiment the incandescence must bettributed to the impacts of the particles, molecules ortoms in the heated space, these particles must thereforn order to explain such action, be assumed to behave as

ndependent carriers of electric charges immersed in an



nsulating medium ; yet there is no attractive forceetween the glass tube and the filament because thepace in the tube is, as a whole, conducting.

t is of some interest to observe in this connection that

whereas the attraction between two electrified bodiesmay cease owing to the impairing of the insulating powerf the medium in which they are immersed, the repulsioetween the bodies may still be observed. This may bexplained in a plausible way. When the bodies are placedt some distance in a poorly conducting medium, such as

lightly warmed or rarefied air, and are suddenly lectrified, opposite electric charges being imparted tohem, these charges equalize more or less by leakagehrough the air. But if the bodies are similarly electrifiedhere is less opportunity afforded for such dissipation,ence the repulsion observed in such case is greater than

he attraction. Repulsive actions in a gaseous medium arowever, as Prof. Crookes has shown, enhanced by

molecular bombardment.

ON CURRENT OR DYNAMIC ELECTRICITY HENOMENA.

o far, I have considered principally effects produced byarying electrostatic force in an insulating medium, suchs air. When such a force is acting upon a conducting bodf measurable dimensions, it causes within the same, orn its surface, displacements of the electricity and gives

ise to electric currents, and these produce another kindf henomena some of which I



hall presently endeavor to illustrate. In presenting thisecond class of electrical effects, I will avail myself rincipally of such as are producible without any returnircuit, hoping to interest you the more by presenting

hese phenomena in a more or less novel aspect.

t has been a long time customary, owing to the limitedxperience with vibratory currents, to consider an electrurrent as something circulating in a closed conductingath. It was astonishing at first to realize that a current

may flow through the conducting path even if the lattere interrupted, and it was still more surprising to learn,hat sometimes it may be even easier to make a currentow under such conditions than through a closed path.ut that old idea is gradually dis appearing, even amongractical men, and will soon be entirely forgotten.

f I connect an insulated metal plate P, Fig. 175, to one ofhe terminals T of the induction coil by means of a wire,hough this

IG. 175.

late be very well insulated, a current passes through th



wire when the coil is set to work. First I wish to give youvidence that there is a current passing through theonnecting wire. An obvious way of demonstrating this iso insert between the terminal of the coil and thensulated plate a very thin platinum or german silver wir

w and bring the latter to incandescence or fusion by theurrent. This requires a rather large plate or else currenmpulses of very high potential and frequency. Another

way is to take a coil c, Fig. 175, containing many turns ofhin insulated wire and to insert the same in the path of he current to the plate. When I connect one of the ends

f the coil to the wire leading to another insulated plate p5 and its other end to the terminal TJ of the inductionoil, and set the latter to work, a current passes throughhe inserted coil c and the existence of the current may b

made manifest in various ways. For instance, I

nsert an iron core * within the coil. The current being onf very high frequency, will, if it be of some strength, sooring the iron core to a noticeably higher temperature, ahe hysteresis and current losses are great witli such higrequencies. One might take a core of some size,aminated or not, it would matter little; but ordinary iron

wire -^th or £th of an inch thick is suitable for theurpose. While the induction coil is working, a currentraverses the inserted coil and only a few moments areufficient to bring the iron wire i to an elevatedemperature sufficient to soften the sealing-wax s, andause a paper washer p fastened by it to the iron wire toall off. But with the a aratus such as I have here, other



much more interesting, demonstrations of this kind cane made. I have a secondary s, Fig 176, of coarse wire,

wound upon a coil similar to the first. In the precedingxperiment the current through the coil c, Fig. 175, wasery small, but there being many turns a strong heating

ffect was, nevertheless,

IG. 176.

roduced in the iron wire. Had I passed that currenthrough a conductor in order to show the heating of theatter, the current might have been too small to producehe effect desired. But with this coil provided with aecondary winding, I can now transform the feebleurrent of high tension which passes through the primarinto a strong secondary current of low tension, and this

urrent will quite certainly do what I expect. In a small

lass tube (t, Fig. 176), I have enclosed a coiled platinumwire, w, this merely in order to protect the wire. On eachnd of the glass tube is sealed a terminal of stout wire to

which one of the ends of the platinum wire w, isonnected. I join the terminals of the secondary coil tohese terminals and insert the primary p, between the

nsulated plate r l5 and the terminal TJ, of the inductionoil as before. The latter being set to work, instantly the



latinum wire w is rendered incandescent and can beused, even if it be verv thick.

nstead of the platinum wire I now take an ordinary 50-olt Ifi c. p. lamp. When I set the induction coil in

peration the lamp filament is brought to highncandescence. It is, however, not necessary to use thensulated plate, for the lamp (7, Fig. 177) is renderedncandescent even if the plate p t be disconnected. Theecondary may also be connected to the primary asndicated by the dotted line in Fig. 177, to do away more

r less with the electrostatic induction or to modify thection otherwise.

may here call attention to a number of interestingbservations with the lamp. First, I disconnect one of theerminals of the lamp from the secondary s. When the

nduction coil plays, a glow is noted which tills the wholeulb. This glow is due to electrostatic induction. It

ncreases'when the bulb is grasped with the hand, and thapacity of the experimenter's body thus added to theecondary circuit. The secondary, in effect, is equivalento a metallic coating, which would be placed near the pri-



IG. 177.

mary. If the secondary, or its equivalent, the coating,were placed symmetrically to the primary, the

lectrostatic induction would be nil under ordinary onditions, that is, when a primary return circuit is useds both halves would neutralize each other. Theecondary is in fact placed symmetrically to the primaryut the action of both halves of the latter, when only onef its ends is connected to the induction coil, is not exactlqual ; hence electrostatic induction takes place, andence the glow in the bulb. I can nearly equalize thection of both halves of the primary by connecting thether, free end of the same to the insulated plate, as in threceding experiment. When the plate is connected, the

low disappears. With a smaller plate it would not entireisappear and then it would contribute to the brightnessf the filament when the secondary is closed, by warminghe air in the bulb.

o demonstrate another interesting feature, I have

djusted the coils used in a certain way. I first connectoth the terminals of the lamp to the secondary, one endf the primary being connected to the terminal TJ of the

nduction coil and the other to the insulated plate p t asefore. When the current is turned on, the lamp glowsrightly, as shown in Fig. 17S&, in which c is a fine wire

oil and s a coarse wire secondary wound upon it. If thensulated late t is disconnected leavin one of the end



of the

IG. 178b.

rimary insulated, the filament becomes dark orenerally it diminishes in brightness (Fig. 1780).onnecting again the plate p t and raising the frequency

he current, I make the filament quite dark or barely redFig. 179J). Once more I will disconnect the plate. One w

f course infer that when the plate is disconnected, theurrent through the primary will be weakened, that



herefore the E. M. F. will fall in the secondary s, and thahe brightness of the lamp will diminish. This might be thase and the result can be secured by an easy adjustmenf the


oils ; also by varying the frequency and potential of theurrents. But it is perhaps of greater interest to note, thahe lamp increases in brightness when the plate isisconnected (Fig. 179#). In this case all the energy the

rimary receives is now sunk into it, like the charge of aattery in an ocean cable, but most of that energy isecovered through the secondary and used to light theamp. The current traversing the primary is strongest athe end b which is connected to the terminal T X of thenduction coil, and



IG 179b.

iminishes in strength towards the remote end a. But theynamic inductive effect exerted upon the secondary s isow greater than before, when the suspended plate wasonnected to the primary. These results might have beenroduced by a number of causes. For instance, the plate! being connected, the reaction from the coil c may beuch as to diminish the potential at the terminal T t of thnduction coil, and therefore weaken the current through

he primary of the coil c. Or the disconnecting

f the plate may diminish the capacity effect with relatioo the primary of the latter coil to such an extent that thurrent through it is diminished, though the potential athe terminal TJ of the induction coil may be the same or

ven higher. Or the result might have been produced byhe change of phase of the primary and secondary urrents and consequent reaction. But the chief etermining factor is the relation of the self-induction anapacity of coil c and plate p t and the frequency of theurrents. The greater brightness of the filament in Fig.

79&, is, however, in part due to the heating of thearefied as in the lam b electrostatic induction which



s before remarked, is greater when the suspended plates disconnected.

till another feature of some interest I may here bring toour attention. When the insulated plate is disconnected

nd the secondary of the coil opened, by approaching amall object to the secondary, but very small sparks cane drawn from it, showing that the electrostatic inductio

s small in this case. But upon the secondary being closedpon itself or through the lamp, the filament glowingrightly, strong sparks are obtained from the secondary

he electrostatic induction is now much greater, becausehe closed secondary determines a greater flow of currenhrough the primary and principally through that half of

which is connected to the induction coil. If now the bulb brasped with the hand, the capacity of the secondary witeference to the primary is augmented by the

xperimenter's body and the luminosity of the filament increased, the incandescence now being due partly to theow of current through the filament and partly to the

molecular bombardment of the rarefied gas in the bulb.

he preceding experiments will have prepared one for

he next following results of interest, obtained in theourse of these investigations. Since I can pass a currenthrough an insulated wire merely by connecting one of itnds to the source of electrical energy, since I can inducey it another current, magnetize an iron core, and, inhort, perform all operations as though a return circuit

were used, clearly I can also drive a motor by the aid of



nly one wire. On a former occasion 1 have described aimple form of motor comprising a single exciting coil, anron core and disc. Fig. 180 illustrates a modified way of perating such an alternate current motor by currentsnduced in a transformer connected to one lead, and

everal other arrangements of circuits

or operating a certain class of alternating motors founden the action of currents of differing phase. In view of thresent state of the art it is thought sufficient to describehese arrangements in a few words only. The diagram,

ig. 180 II., shows a primary coil P, connected with one os ends to the line L leading from a high tensionransformer terminal TJ. In inductive relation to thisrimary P is a secondary s of coarse wire in the circuit of

which is a coil c. The currents induced in the secondary nergize the iron core ?', which is preferably, but not

ecessarily, subdivided, and set the metal disc d inotation. Such a motor M 2 as diagramatically shown inig. 180 II., has been called a " magnetic lag motor," but

his expression may be objected to by those who attributhe rotation of the disc to eddy currents circulating in

minute paths when the core i is finally subdivided. In

rder to operate such a motor effectively on the planndicated, the frequencies should not be too high, not

more than four or five thousand, though the rotation isroduced even with ten thousand per second, or more.

n Fig. 180 I., a motor M t having two energizing circuits

A and B, is diagrammatically indicated. The circuit A is



onnected to the line L and in series with it is a primary pwhich may have its free end connected to an insulated

late p l5 such connection being indicated by the dottednes. The other motor circuit B is connected to theecondary s which is in inductive relation to the primary

. When the transformer terminal T t is alternately lectrified, currents traverse the open line L and alsoircuit A and primary p. The currents through the latternduce secondary currents in the circuit s, which passhrough the energizing coil B of the motor. The currentshrough the secondary s and those through the primary

iffer in phase 90 degrees, or nearly so, and are capable otating an armature placed in inductive relation to theircuits A and B.

n Fig. 180 III., a similar motor M 3 with two energizingircuits A! and B! is illustrated. A primary p, connected

with one of its ends to the line L has a secondary s, whichs preferably wound for a tolerably high E. M. r., and to

which the two energizing circuits of the motor areonnected, one directly to the ends of the secondary andhe other through a condenser c, by the action of whichhe currents traversing the circuit A t and B t are made t

iffer in phase.



n Fig. 180 IV., still another arrangement is shown. Inhis case two primaries p t and P 2 are connected to thene L, one



e

hrough a condenser c of small capacity , and the otherirectly. The primaries are provided witli secondaries s tnd s 2 which are in series with the energizing circuits, A

and B 2 and a motor M 3 , the condenser c again servino produce the requisite difference in the phase of theurrents traversing the motor circuits. As such phase

motors with two or more circuits are now well known inhe art, they have been here illustrated diagrammaticall

No difficulty whatever is found in operating a motor in thmanner indicated, or in similar ways; and although such

xperiments up to this day present only scientificnterest, they may at a period not far distant, be carriedut with practical objects in view.

t is thought useful to devote here a few remarks to theub ect of o eratin devices of all kinds b means of onl



ne leading wire. It is quite obvious, that when high-requency currents are made use of, ground connectionsre—at least when the E. M. F. of the currents is great—etter than a return wire. Such ground connections arebjectionable witli steady or low frequency currents on

ccount of destructive chemical actions of the former andisturbing influences exerted by both on the neighboringircuits; but with high frequencies these actionsractically do not exist. Still, even ground connectionsecome superfluous when the E. M. F. is very high, foroon a condition is reached, when the current may be

assed more economically through open, than throughlosed, conductors. Remote as might seem an industrialpplication of such single wire transmission of energy tone not experienced in such lines of experiment, it will noeem so to anyone who for some time has carried onnvestigations of such nature. Indeed I cannot see why

uch a plan should not be practicable. Nor should it behought that for carrying out such a plan currents of veryigh frequency are expressly required, for just as soon aotentials of say 30,000 volts are used, the single wireransmission may be effected with low frequencies, andxperiments have been made by me from which these

nferences are made.

When the frequencies are very high it has been found inaboratory practice quite easy to regulate the effects inhe manner shown in diagram Fig. 181. Here tworimaries p and p l are shown, each connected with one os ends to the line L and with the other end to the



ondenser plates c and c, respectively. Near these arelaced other condenser plates c x and c,, the former beinonnected to the line L and the latter to an insulatedarger

late P 2 . On the primaries are wound secondaries s andt , of coarse wire, connected to the devices d and Iespectively. By-varying the distances of the condenserlates c and c l5 and c and c t the currents through theecondaries s and s t are varied in intensity. The curiouseature is the great sensitiveness, the slightest change in

he distance of the plates producing considerableariations in the intensity or strength of the currents. Thensitiveness may be rendered extreme by making therequency such, that the primary itself, without any platttached to its free end, satisfies, in conjunction with thelosed secondary, the condition of resonance. In such

ondition an extremely small change in the capacity of three terminal produces great variations. For instance, Iave been able to adjust the conditions so that the merepproach of a person to the coil produces a considerablehange in the brightness of the lamps attached to theecondary. Such observations and experiments possess,

f course, at present, chiefly scientific interest, but they may soon become of practical importance.

ery high frequencies are of course not practicable withmotors on account of the necessity of employing ironores. But one may use sudden discharges of low

requency and thus obtain certain advantages of high-



requency currents without rendering the iron corentirely incapable of following the changes and withoutntailing a very great expenditure of energy in the core. ave found it quite practicable to operate with s.uch low

requency disruptive discharges of condensers,

lternating-current motors. A certain class of such motowhich I advanced a few years ago, which contain closedecondary circuits, will rotate quite vigorously when theischarges are directed through the exciting coils. Oneeason that such a motor operates so well with theseischarges is that the difference of phase between the

rimary and secondary currents is 90 degrees, which isenerally not the case with harmonically rising and fallinurrents of low frequency. It might not be withoutnterest to show an experiment with a simple motor of his kind, inasmuch as it is commonly thought thatisruptive discharges are unsuitable for such purposes.

he motor is illustrated in Fig. 182. It comprises a ratherarge iron core * with slots on the top into which arembedded thick copper washers c c. In proximity to theore is a freely-movable metal disc D. The core isrovided with a primary exciting coil c a the ends a and bf which are connected to

he terminals of the secondary s of an ordinary ransformer, the primary p of the latter being connectedo an alternating distribution circuit or generator o of lowr moderate frequency. The terminals of the secondary re attached to a condenser c which discharges throughn air gap d d which may be placed in series or shunt to



he coil c x . When the conditions are properly chosen theisc D rotates with considerable effort and the iron core oes not get very perceptibly hot. With currents from aigh-frequency alternator, on the contrary, the core getsapidly hot and the disc rotates with a much smaller

ffort. To perform the experiment properly it should berst ascertained that the disc D is not set in rotation whehe discharge is not occurring at d d. It is preferable tose a large iron core and a condenser of large capacity sos to bring the superimposed quicker oscillation to a very

ow pitch or to do away \vith it entirely. By observing

ertain elementary rules I have also found it practicableo operate ordinary series or shunt direct-current motor

with such disruptive discharges, and this can be done witr without a return wire.

MPEDANCE PHENOMENA.

Among the various current phenomena observed,erhaps the most interesting are those of impedanceresented by conductors to currents varying at a rapidate. In my first paper before the American Institute of lectrical Engineers, I have described a few striking

bservations of this kind. Thus I showed that when suchurrents or sudden dischaiges are passed through a thick

metal bar there may be points on the bar only a few nches apart, which have a sufficient potential differenceetween them to maintain at bright incandescence anrdinary filament lamp. I have also described the curiou

ehavior of rarefied gas surrounding a conductor, due to



uch sudden rushes of current. These phenomena haveince been more carefully studied and one or two novelxperiments of this kind are deemed of sufficient intereso be described here.

Referring to Fig. 1830, B and BJ are very stout copperars connected at their lower ends to plates c and c 1?espectively, of a condenser, the opposite plates of theatter being connected to the terminals of the secondary f a high-tension transformer, the primary p of which isupplied with alternating currents from an ordinary low-

requency dynamo & or distribution circuit. The

ondenser discharges through an adjustable gap dd&&sual. By establishing a rapid vibration it was found quiteasy to perform the following curious experiment. Thears B and B t were joined at the top by a low-voltage

amp Z 3 ; a little lower was placed by means of clamps c, a 50-volt lamp 4 ; and still lower another 100-volt lam!; and finally, at a certain distance below the latter lampn exhausted tube T. By carefully determining theositions of these devices it was found practicable to

maintain them



IQB. 183a, 183b and 183c.

ll at their proper illuminating power. Yet they were allonnected in multiple arc to the two stout copper bars anequired widely different pressures. This experimentequires of course some time for adjustment but is quiteasily performed.

n Figs. 1835 and 1836', two other experiments arelustrated which, unlike the previous experiment, do notequire very careful adjustments. In Fig. 1835, twoamps, ^ and 4, the former a

00-volt and the latter a 50-volt are placed in certain

ositions as indicated, the 100-volt lamp being below the0-volt lam . When the arc is la in at d ' d and the



udden discharges are passed through the bars B B,, the0-volt lamp will, as a rule, burn brightly, or at least thisesult is easily secured, while the 100-volt lamp will burnery low or remain quite dark. Fig. 1835. Now the bars B! may be joined at the top by a thick cross bar -^ and it

s quite easy to maintain the 100-volt lamp at full candleower while the 50-volt lamp remains dark, Fig. 183c.hese results, as I have pointed out previously, shouldot be considered to be due exactly to frequency butather to the time rate of change which may be great,ven with low frequencies. A great many other results of

he same kind, equally interesting, especially to those whre only used to manipulate steady currents, may bebtained and they afford precious clues in investigatinghe nature of electric currents.

n the preceding experiments I have already had occasio

o show some light phenomena and it would now beroper to study these in particular ; but to make this

nvestigation more complete I think it necessary to makerst a few remarks on the subject of electrical resonance

which has to be always observed in carrying out thesexperiments.

ON ELECTRICAL RESONANCE.

he effects of resonance are being more and more notedy engineers and are becoming of great importance in thractical operation of apparatus of all kinds with

lternating currents. A few general remarks may herefore be made concernin these effects. It is clear



hat if we succeed in employing the effects of resonanceractically in the operation of electric devices the return

wire will, as a matter of course, become unnecessary, forhe electric vibration may be conveyed with one wire juss well as, and sometimes even better than, with two. Th

uestion first to answer is, then, whether pure resonanceffects are producible. Theory and experiment both showhat such is impossible in Nature, for as the oscillationecomes more and more vigorous, the losses in theibrating bodies and environing media rapidly increasend necessarily check the vibration which otherwise

would go on increasing forever. It is a fortunateircumstance that pure resonance is not producible, for iwere there is no telling what dangers might not lie in

wait for the innocent experimenter. But to a

ertain degree resonance is producible, the magnitude of

he effects being limited by the imperfect conductivity nd imperfect elasticity of the media or, generally statedy f rictional losses. The smaller these tosses, the moretriking are the effects. The same is the case in

mechanical vibration. A stout steel bar may be set inibration by drops of water falling upon it at proper

ntervals; and with glass, which is more perfectly elastic,he resonance effect is still more remarkable, for a goblet

may be burst by singing into it a note of the proper pitchhe electrical resonance is the more perfectly attained,

he smaller the resistance- or the impedance of theonducting path and the more perfect the dielectric. In ae den ar dischar in throu h a short stranded cable of



hin wires these requirements are probably best fulfillednd the resonance effects are therefore very prominent.uch is not the case with dynamo machines, transformernd their circuits, or with commercial apparatus ineneral in which the presence of iron cores complicates

he action or renders it impossible. In regard to Leydenars. with which resonance effects are frequently emonstrated, I would say that the effects observed areften attributed but are seldom due to true resonance, fon error is quite easily made in this respect. This may bendoubtedly demonstrated by the following experiment

ake, for instance, two large insulated metallic plates orpheres which I shall designate A and B; place them at aertain small distance apart and charge them from arictional or influence machine to a potential so high thatust a slight increase of the difference of potential betweehem will cause the small air or insulating space to break

own. This is easily reached by making a few preliminaryrials. If now another plate —fastened on an insulatingandle and connected by a wire to one of the terminals ohigh tension secondary of an induction coil, which is

maintained in action by an alternator (preferably highrequency)— is approached to one of the charged bodies

r B, so as to be nearer to either one of them, theischarge will invariably occur between them ; at least it

will, if the potential of the coil in connection with the plats sufficiently high. But the explanation of this will soon bound in the fact that the approached plate actsnductively upon the bodies A and B and causes a spark tass between them. When this spark occurs, the charges



which were previously imparted to these bodies from thnfluence machine, must needs be lost, since the bodiesre brought in electri-

al connection through the arc formed. Xow this arc is

ormed whether there be resonance or not. But even if he spark would not be produced, still there is anlternating E. M. F. set up between the bodies when thelate is brought near one of thfem ; therefore thepproach of the plate, if it does not always actually, will, any rate, tend to break down the air space by inductive

ction. Instead of the spheres or plates A and B we may ake the coatings of a Ley-den jar with the same result,nd in place of the machine,—which is a high frequency lternator preferably, because it is more suitable for thexperiment and also for the argument, —we may takenother Leyden jar or battery of jars. When such jars are

ischarging through a circuit of low resistance the same iraversed by currents of very high frequency. The plate

may now be connected to one of the coatings of the seconar, and when it is brought near to the first jar justreviously charged to a high potential from an influence

machine, the result is the same as before, and the first ja

will discharge through a small air space upon the secondeing caused to discharge. But both jars and their circuiteed not be tuned any closer than a basso profundo is to

he note produced by a mosquito, as small sparks will beroduced through the air space, or at least the latter wille considerably more strained owing to the setting up ofn alternating K. M. F. by induction, which takes place



when one of the jars begins to discharge. Again anotherrror of a similar nature is quite easily made. If theircuits of the two jars are run parallel and close togethernd the experiment has been performed of dischargingne by the other, and now a coil of wire be added to one o

he circuits whereupon the experiment does not succeedhe conclusion that this is due to the fact that the circuitsre now not tuned, would be far from being safe. For thewo circuits act as condenser coatings and the addition ofhe coil to one of them is equivalent to bridging them, athe point where the coil is placed, by a small condenser,

nd the effect of the latter might be to prevent the sparkrom jumping through the discharge space by diminishinhe alternating E. M. F. acting across the same. All theseemarks, and many more which might be added but forear of wandering too far from the subject, are made withhe pardonable intention of cautioning the unsuspecting

tudent, who might gain an entirely unwarranted opinionf his skill at seeing every experiment succeed; but they re in no way thrust upon the experienced as novelbservations.

n order to make reliable observations of electric

esonance effects it is very desirable, if not necessary, tomploy an alternator giving currents which rise and fallarmonically, as in working with make and break urrents the observations are not always trustworthy,ince many phenomena, which depend- on the rate of hange, may be produced with widely differentrequencies. Even when making such observations with



n alternator one is apt to be mistaken. When a circuit isonnected to an alternator there are an indefinite numbef values for capacity and self-induction which, inonjunction, will satisfy the condition of resonance. Sohere are in mechanics an infinite number of tuning forks

which will respond to a note of a certain pitch, or loadedprings which have a definite period of vibration. But theesonance will be most perfectly attained in that case in

which the motion is effected with the greatest freedom.Now in mechanics, considering the vibration in theommon medium — that is, air— it is of comparatively

ttle importance whether one tuning fork be somewhatarger than another, because the losses in the air are notery considerable. One may, of course, enclose a tuningork in an exhausted vessel and by thus reducing the airesistance to a minimum obtain better resonant action.till the difference would not be very great. But it would

make a great difference if the tuning fork were immersen mercury. In the electrical vibration it is of enormousmportance to arrange the conditions so that the vibratios effected with the greatest freedom. The magnitude of he resonance effect depends, under otherwise equalonditions, on the quantity of electricity set in motion or

n the strength of the current driven through the circuitut the circuit opposes the passage of the currents by eason of its impedance and therefore, to secure the bestction it is necessary to reduce the impedance to a

minimum. It is impossible to overcome it entirely, butmerely in part, for the ohmic resistance cannot be

vercame. But when the frequency of the impulses is ver



reat, the flow of the current is practically determined byelf-induction. Now self-induction can be overcome by ombining it with capacity. If the relation between theses such, that at the frequency used they annul each otherhat is, have such values as to satisfy the condition of

esonance, and the greatest quantity of electricity is mado flow through the external circuit, then the best result btained. It is simpler and safer to join the condenser ineries with the self-induction. It is clear that in such

ombinations there will be, for a given frequency, and

onsidering only the fundamental vibration, values whichwill give the best result, with the condenser in shunt tohe self-induction coil; of course more such values than

with the condenser in series. But practical conditionsetermine the selection. In the latter case in performinghe experiments one may take a small self-induction and

large capacity or a small capacity and a large self-nduction, but the latter is preferable, because it isnconvenient to adjust a large capacity by small steps. Byaking a coil with a very large self-induction the criticalapacity is reduced to a very small value, and the capacitf the coil itself may be sufficient. It is easy, especially by

bserving certain artifices, to wind a coil through whichhe impedance will be reduced to the value of the ohmicesistance only; and for any coil there is, of course, arequency at which the maximum current will be made tass through the coil. The observation of the relationetween self-



IG. 184.

nduction, capacity and frequency is becoming importantn the operation of alternate current apparatus, such asransformers or motors, because by a judiciousetermination of the elements the employment of an

xpensive condenser becomes unnecessary. Thus it isossible to pass through the coils of an alternating curren

motor under the normal working conditions the requiredurrent with a low E. M. F. and do away entirely with thealse current, and the larger the motor, the easier such alan becomes practicable ; but it is necessary for this to

mploy currents of very high potential or high frequency

n Fig. 184 I. is shown a plan which has been followed inhe study of the resonance effects by means of a highrequency alternator. G! is a coil of many turns, which isivided into small separate sections for the purpose of

djustment. The final adjustment was made sometimeswith a few thin iron wires (though this is not always

dvisable) or with a closed secondary. The coil

j is connected with one of its ends to the line L from thelternator G and with the other end to one of the plates c

f a condenser c GI the late (c^ of the latter bein



onnected to a much larger plate PJ. In this manner bothapacity and self-induction were adjusted to suit theynamo frequency.

As regards the rise of potential through resonant action,

f course, theoretically, it may amount to anything since epends on self-induction and resistance and since thesemay have any value. But in practice one is limited in theelection of these values and besides these, there arether limiting causes. One may start with, say, 1,000olts and raise the E. M. F. to 50 times that value, but on

annot start with 100,000 and raise it to ten times thatalue because of the losses in the media which are great,specially if the frequency is high. It should be possible ttart with, for instance, two volts from a high or low requency circuit of a dynamo and raise the E. M. r. to

many hundred times that value. Thus coils of the proper

imensions might be connected each with only one of itsnds to the mains from a machine of low E. M. F., andhough the circuit of the machine would not be closed inhe ordinary acceptance of the term, yet the machine

might be burned out if a proper resonance effect would bbtained. I have not been able to produce, nor have I

bserved with currents from a dynamo machine, suchreat rises of potential. It is possible, if not probable, tha

with currents obtained from apparatus containing iron thisturbing influence of the latter is the cause that theseheoretical possibilities cannot be realized. But if such ishe case I attribute it solely to the hysteresis andoucault current losses in the core. Generally it was



ecessary to transform upward, when the E. M. F. wasery low, and usually an ordinary form of induction coil

was employed, but sometimes the arrangementlustrated in Fig. 184 II., has been found to beonvenient. In this case a coil cis made in a great many

ections, a few of these being used as a primary. In thismanner both primary and secondary are adjustable. Onend of the coil is connected to the line i^ from thelternator, and the other line L is connected to thentermediate point of the coil. Such a coil with adjustablerimary and secondary will be found also convenient in

xperiments with the disruptive discharge. When trueesonance is obtained the top of the wave must of coursee on the free end of the coil as, for instance, at theerminal of the phosphorescence bulb B. This is

asily recognized by observing the potential of a point on

it-wire w near to the coil.

n connection with resonance effects and the problem of ransmission of energy over a single conductor which wareviously considered, I would say a few words on aubject which constantly fills my thoughts and which

oncerns the welfare of all. I mean the transmission of ntelligible signals or perhaps even power to any distance

without the use of wires. I am becoming daily moreonvinced of the practicability of the scheme ; and thougknow full well that the great majority of scientific men

will not believe that such results can be practically and

mmediately realized, yet I think that all consider the



evelopments in recent years by a number of workers toave been such as to encourage thought and experiment

n this direction. My conviction has grown so strong, thato longer look upon this plan of energy or intelligenceransmission as a mere theoretical possibility, but as a

erious problem in electrical engineering, which must bearried out some day. The idea of transmittingntelligence without wires is the natural outcome of the

most recent results of electrical investigations. Somenthusiasts have expressed their belief that telephony tony distance by induction through the air is possible. I

annot stretch my imagination so far, but I do firmly elieve that it is practicable to disturb by means of owerful machines the electrostatic condition of the eartnd thus transmit intelligible signals and perhaps power.n fact, what is there against the carrying out of such acheme ? We now know that electric vibration may be

ransmitted through a single conductor. Why then not tro avail ourselves of the earth for this purpose ? We needot be frightened by the idea of distance. To the weary

wanderer counting the mile-posts the earth may appearery large, but to that happiest of all men, thestronomer, who gazes at the heavens and by their

tandard judges the magnitude of our globe, it appearsery small. And so I think it must seem to the electrician

or when he considers the speed with which an electricisturbance is propagated through the earth all his ideasf distance must completely vanish.

A point of great importance would be first to know what



he capacity of the earth ? and what charge does it contaf electrified ? Though we have no positive evidence of aharged body existing in space without other oppositely lectrified bodies being-near, there is a fair probability hat the earth is such a body, for

y whatever process it was separated from other bodiesnd this is the accepted view of its origin — it must haveetained a charge, as occurs in all processes of mechanicaeparation. If it be a charged body insulated in space itsapacity should be extremely small, less than one-

housandth of a farad. But the upper strata of the air areonducting, and so, perhaps, is the medium in free spaceeyond the atmosphere, and these may contain anpposite charge. Then the capacity might bencomparably greater. In any case it is of the greatestmportance to get an idea of what quantity of electricity

he earth contains. It is difficult to say whether we shallver acquire this necessary knowledge, but there is hopehat we may, and that is, by means of electricalesonance. If ever we can ascertain at what period thearth's charge, when disturbed, oscillates with respect ton oppositely electrified system or known circuit, we sha

now a fact possibly of the greatest importance to thewelfare of the human race. I propose to seek for the

eriod by means of an electrical oscillator, or a source of lternating electric currents. One of the terminals of theource would be connected to earth as, for instance, to thity water mains* the other to an insulated body of largeurface. It is possible that the outer conducting air strata



r free space, contain an opposite charge and that,ogether with the earth, they form a condenser of very arge capacity. In such case the period of vibration may bery low and an alternating dynamo machine might servor the purpose of the experiment. I would then

ransform the current to a potential as high as it would bound possible and connect the ends of the high tensionecondary to the ground and to the insulated body. By arying the frequency of the currents and carefully bserving the potential of the insulated body and

watching for the disturbance at various neighboring

oints of the earth's surface resonance might be detectedhould, as the majority of scientific men in all probabilityelieve, the period be extremely small, then a dynamo

machine would not do and a proper electrical oscillatorwould have to be produced and perhaps it might not be

ossible to obtain such rapid vibrations. But whether thi

e possible or not, and whether the earth contains aharge or not, and whatever may be its period of ibration, it certainly is possible—for of this we have dailvidence—to produce some electrical disturbanceufficiently powerful to be perceptible by suitablenstruments at any point of the earth's surface.

Assume that a source of alternating currentss beonnected, as in Fig. 185, with one of its terminals to earconveniently to the water mains) and with the other to ody of large surface p. When the electric oscillation is sep there will be a movement of electricity in and out of pnd alternating currents will pass through the earth,



onverging to, or diverging from, the point c where theround connection is made. In this manner neighboringoints on the earth's surface within a certain radius will bisturbed. But the disturbance will diminish with theistance, and the distance at which the effect will still be

erceptible will depend on the quantity of electricity set motion. Since the body p is insulated, in order to displaceconsiderable quantity, the potential of the source muste excessive, since there would be limitations as to theurface of p. The conditions might be adjusted so that theenerator or source s will set up the same electrical

movement as though its circuit were closed. Thus it isertainly $2 practicable to impress an electric vibration aeast g of a certain low period upon the earth by means oroper machinery. At what distance such a vibration

might be made perceptible can only be conjectured. Iave on another occasion considered the question how th

arth might behave to electric disturbances. There is nooubt that, since in such an experiment the electricalensity at the surface could be but extremely smallonsidering the size of the earth, the air would not act asery disturbing factor, and there would be not muchnergy lost through the action of the air, which would be

he case if the density were great. Theoretically, then, itould not require a great amount of energy to produce aisturbance perceptible at great distance, or even all ovehe surface of the globe. . tq Now, it is quite certain that ny point within a

ertain radius of the source s a properly adjusted self-



nduction and capacity device can be set in action by esonance. But not only can this be done, but anotherource s,, Fig. 185, similar to s, or any number of suchources, can be set

o work in synchronism with the latter, and the vibrationhus intensified and spread over a large area, or a flow oflectricity produced to or from the source s t if the samee of opposite phase to the source s. I think that beyondoubt it is possible to operate electrical devices in a city hrough the ground or pipe system by resonance from an

lectrical oscillator located at a central point. But theractical solution of this problem would be of ncomparably smaller benefit to man than the realizationf the scheme of transmitting intelligence, or perhapsower, to any distance through the earth or environing

medium. If this is at all possible, distance does not mean

nything. Proper apparatus must first be produced by means of which the problem can be attacked and I have

evoted much thought to this subject. I am firmly onvinced that it can be done and hope that we shall liveo see it done.

ON THE LIGHT PHENOMENA PRODUCED BY HIGH-REQUENCY CURRENTS OF HIGH POTENTIAL AND

GENERAL REMARKS RELATING TO THE SUBJECT.

Returning now to the light effects which it has been thehief object to investigate, it is thought proper to divide

hese effects into four classes : 1. Incandescence of a solid. Phos horescence.



. Incandescence or phosphorescence of a rarefied gas;nd

. Luminosity produced in a gas at ordinary pressure. Th

rst question is : How are these luminous effectsroduced ? In order to answer this question asatisfactorily as I am able to do in the light of acceptediews and with the experience acquired, and to add som

nterest to this demonstration, I shall dwell here upon aeature which I consider of great importance, inasmuch a

promises, besides, to throw a better light upon theature of most of the phenomena produced by high-requency electric currents. I have on other occasionsointed out the great importance of the presence of thearefied gas, or atomic medium in general, around theonductor through which alternate currents of high

requency are passed, as regards the heating of theonductor by the currents. My experiments, describedome time ago, have shown that, the higher the frequencnd potential difference of the currents, the moremportant becomes the rarefied gas in which theonductor is immersed, as a factor of the heating. The

otential difference, however, is, as I then pointed out, amore im-

ortant element than the frequency. When both of thesere sufficiently high, the heating may be almost entirely ue to the presence of the rarefied gas. The experiments

o follow will show the importance of the rarefied gas, or,enerall of as at ordinar or other ressure as re ard



he incandescence or other luminous effects produced byurrents of this kind.

take two ordinary 50-volt 16 c. p. lamps which are invery respect alike, with the exception, that one has bee

pened at the top and the air has filled the bulb, while thther is at the ordinary degree of exhaustion of ommercial lamps. When I attach the lamp which isxhausted to the terminal of the secondary of the coil,

which I have already used, as in experiments illustratedn Fig. 179« for instance, and turn on the current, the

lament, as you have before seen, comes to highncandescence. When I attach the second lamp, which islled with air, instead of the former, the filament stilllows, but much less brightly. This experiment illustratenly in part the truth of the statements before made. Th

mportance of the filament's being immersed in rarefied

as is plainly noticeable but not to such a degree as mighe desirable. The reason is that the secondary of this coil

s wound for low tension, having only 150 turns, and theotential difference at the terminals of the lamp isherefore small. Were I to take another coil with many

more turns in the secondary, the effect would be

ncreased, since it depends partially on the potentialifference, as before remarked. But since the effectkewise depends on the frequency, it may be properly tated that it depends on the time rate of the variation ohe potential difference. The greater this variation, the

more important becomes the gas as an element of eatin . I can roduce a much reater rate of variation in



nother way, which, besides, has the advantage of doingway with the objections, which might be made in thexperiment just shown, even if both the lamps wereonnected in series or multiple arc to the coil, namely,hat in consequence of the reactions existing between th

rimary and secondary coil the conclusions are renderedncertain. This result I secure by charging, from anrdinary transformer which is fed from the alternatingurrent supply station, a battery of condensers, andischarging the latter directly through a circuit of smallelf-induction, as before illustrated in Figs. 183*, 183&,

nd 1836-.

n Figs. 186«, 1865 and 186c, the heavy copper bars BB^re

onnected to the opposite coatings of a battery of

ondensers, or generally in such way, that the highrequency or sudden discharges are made to traversehem. I connect first an ordinary 50-volt incandescentamp to the bars by means of the clamps o c. Theischarges being passed through the lamp, the filament iendered incandescent, though the current through it is

ery small, and would not be nearly sufficient to producevisible effect under the conditions of ordinary use of th

amp. Instead of this I now attach to the bars anotheramp exactly like the first, but with the seal broken off,he bulb being therefore filled with air at ordinary ressure. When the discharges are directed through the

lament, as before, it does not become incandescent. But



he result might still be attributed to one of the many ossible reactions. I therefore connect both the lamps in

multiple arc as illustrated in Fig. 186«. Passing

IG. 186a.

IG. 186b.



IG. 186c.

he discharges through both the lamps, again the filamenn the exhausted lamp I glows very brightly while that inhe non-exhausted lamp Zi remains dark, as previously.

ut it should not be thought that the latter lamp is takingnly a small fraction of the energy supplied to both the

amps; on the contrary, it may consume a considerableortion of the energy and it may become even hotter thahe one which burns brightly. In this experiment theotential difference at the terminals of the lamps varies iign theoretically three to four million times a second. Thnds of the filaments are correspondingly electrified, andhe gas in the bulbs is violently agitated and a largeortion of the supplied energy is thus converted into hean the non-exhausted bulb, there being a few millionmes more gas molecules than in the exhausted one, theombardment, which is most violent at the ends of thelament, in the neck of the bulb, consumes a

arge portion of the energy without producing any visibleffect. The reason is that, there being many molecules,

he bombardment is quantitatively considerable, but thendividual impacts are not very violent, as the speeds of he molecules are comparatively small owing to the smalree path. In the exhausted bulb, on the contrary, thepeeds are very great, and the individual impacts areiolent and therefore better adapted to produce a visible

ffect. Besides, the convection of heat is greater in theormer bulb. In both the bulbs the current traversin th



laments is very small, incomparably smaller than thatwhich they require on an ordinary low-frequency circuit

he potential difference, however, at the ends of thelaments is very great and might be possibly 20,000olts or more, if the filaments were straight and their

nds far apart. In the ordinary lamp a spark generally ccurs between the ends of the filament or between thelatinum wires outside, before such a difference of otential can be reached.

t might be objected that in the experiment before show

he lamps, being in multiple arc, the exhausted lampmight take a much larger current and that the effectbserved might not be exactly attributable to the actionf the gas in the bulbs. Such objections will lose much

weight if I connect the lamps in series, with the sameesult. When this is done and the discharges are directed

hrough the filaments, it is again noted that the filamentn the non-exhausted bulb l^ remains dark, while that inhe exhausted one (7) glows even more intensely thannder its normal conditions of working, Fig. 1865.

According to general ideas the current through thelaments should now be the same, were it not modified b

he presence of the gas around the filaments.

At this juncture I may point out another interestingeature, which illustrates the effect of the rate of change otential of the currents. I will leave the two lampsonnected in series to the bars BB,, as in the previous

xperiment, Fig. 186&, but will presently reduce



onsiderably the frequency of the currents, which wasxcessive in the experiment just before shown. This I

may do by inserting a self-induction coil in the path of thischarges, or by augmenting the capacity of theondensers. When I now pass these low-frequency

ischarges through the lamps, the exhausted lamp I agais as bright as before, but it is noted also that the non-xhausted lamp l± glows, though not quite

s intensely as the other. Reducing the current throughhe lamps, I may bring the filament in the latter lamp to

edness, and, though the filament in the exhausted lamps bright, Fig. I860, the degree of its incandescence ismuch smaller than in Fig. 1865, when the currents were

f a much higher frequency.

n these experiments the gas acts in two opposite ways i

etermining the degree of the incandescence of thelaments, that is, by convection and bombardment. Theigher the frequency and potential of the currents, the

more important becomes the bombardment. Theonvection on the contrary should be the smaller, theigher the frequency. When the currents are steady ther

s practically no bombardment, and convection may herefore with such currents also considerably modify thegree of incandescence and produce results similar tohose just before shown. Thus if two lamps exactly alike,ne exhausted and one not exhausted, are connected in

multiple arc or series to a direct-current machine, the

lament in the non-exhausted lamp will require a



onsiderably greater current to be renderedncandescent. This result is entirely due to convection,nd the effect is the more prominent the thinner thelament. Professor Ayrton and Mr. Kilgour some time agublished quantitative results concerning the thermal

missivity by radiation and convection in which the effecwith thin wires was clearly shown. This effect may betrikingly illustrated by preparing a number of small,hort, glass tubes, each containing through its axis thehinnest obtainable platinum wire. If these tubes beighly exhausted, a number of them may be connected i

multiple arc to a direct-current machine and all of thewires may be kept at incandescence with a smallerurrent than that required to render incandescent a singne of the wires if the tube be not exhausted. Could theubes be so highly exhausted that convection would be nhen the relative amounts of heat given off by convection

nd radiation could be determined without the difficultiettending thermal quantitative measurements. If a sourf electric impulses of high frequency and very highotential is employed, a still greater number of the tubes

may be taken and the wires rendered incandescent by aurrent not capable of warming perceptibly a wire of the

ame size immersed in air at ordinary pressure, andonveying the energy to all df them.

may here describe a result which is still morenteresting, and to which I have been led by thebservation of these phe-



omena. I noted that small differences in the density of he air produced a considerable difference in the degree ncandescence of the wires, and I thought that, since in aube, through which a luminous discharge is passed, theas is generally not of uniform density, a very thin wire

ontained in the tube might be rendered incandescent atertain places of smaller density of the gas, while it wouldemain dark at the places of greater density, where theonvection would be greater and the bombardment lessntense. Accordingly a tube t was prepared, as illustratedn Fig. 187, which contained through the middle a very

ne platinum wire w. The tube was exhausted to amoderate degree and it was found that when it was

ttached to the terminal of a high-frequency coil thelatinum wire w would indeed, become incandescent inatches, as illustrated in Fig. 187. Later a number of thesubes with one or more wires were prepared, each

howing this result. The effect was best noted when thetriated discharge occurred in the tube, but was alsoroduced when the stride were not vi-ible, showing thatven then, the gas in the tube was not of uniform densityhe position of the strirp was generally such, that thearefactions corresponded to the places of incandescence

r greater brightness on the wire w. But in a few instancwas noted, that the bright spots on the wire were

overed by the dense parts of the striated discharge asndicated by / in Fig. 187, though the effect Avas barely erceptible. This was explained in a plausible way by ssuming that the convection was not widely different inhe dense and rarefied places, and that the bombardmen



was greater on the dense places of the striated discharget is, in fact, often observed in bulbs, that under certainonditions a thin wire is brought to higher incandescence

when the air is not too highly rarefied. This is the casewhen the potential of the coil is not high enough for the

acuum, but the result may be attributed to many ifferent causes. In all cases this curious phenomenon of ncandescence disappears when the tube, or rather the

wire, .acquires throughout a uniform temperature.

Disregarding now the modifying effect of convection ther

re then two distinct causes which determine thencandescence of a wire or filament with varying currenthat is, conduction current and bombardment. Withteady currents we have to deal only with the former of hese two causes, and the heating effect is a minimum,ince the resistance is least to steady fiow. When the

urrent is a varying one the resistance is greater, andence

he heating effect is increased. Thus if the rate of changef the current is very great, the resistance may increaseo such an extent that the filament is brought to

ncandescence with inappreciable currents, and we areble to take a short and thick block of carbon or other

material and bring it to bright incandescence with aurrent incomparably smaller than that required to brino the same degree of incandescence an ordinary thinamp filament with a steady or low frequency current.

his result is important, and illustrates how rapidly our



iews on these subjects are changing, and how quickly oueld of knowledge is ex-



IG. 187.

IG. 188.

ending. In the art of incandescent lighting, to view this

esult in one aspect only, it has been commonly onsidered as an essential requirement for practicaluccess, that the lamp filament should be thin and of highesistance. But now we know that the resistance of thelament to the steady flow does not mean anything ; thelament might as well be short and thick ; for if it be

mmersed in rareiied gas it will become incandescent by he passage of a small current. It all depends on therequency and potential of the currents. We may concludrom this, that it

would be of advantage, so far as the lamp is considered, tmploy high frequencies for lighting, as they allow the usf short and thick filaments and smaller currents.

f a wire or filament be immersed in a homogeneousmedium, all the heating is due to true conduction curren

ut if it be enclosed in an exhausted vessel the conditionre entirel different. Here the as be ins to act and the



eating effect of the conduction current, as is shown inmany experiments, may be very small compared withhat of the bombardment. This is especially the case if thircuit is not closed and the potentials are of course very igh. Suppose that a fine filament enclosed in an

xhausted vessel be connected with one of its ends to theerminal of a high tension coil and with its other end to aarge insulated plate. Though the circuit is not closed, thelament, as I have before shown, is brought to

ncandescence. If the frequency and potential beomparatively low, the filament is heated by the current

assing through it. If the frequency and potential, andrincipally the latter, be increased, the insulated plateeed 'be but very small, or may be done away withntirely ; still the filament will become incandescent,ractically all the heating being then due to theombardment. A practical way of combining both the

ffects of conduction currents and bombardment islustrated in Fig. 188, in which an ordinary lamp is showrovided with a very thin filament which has one of thends of the latter connected to a shade serving theurpose of the insulated plate, and the other end to theerminal of a high tension source. It should not be though

hat only rarefied gas is an important factor in the heatinf a conductor by varying currents, but gas at ordinary ressure may become important, if the potentialifference and frequency of the currents is excessive. Onhis subject I have already stated, that when a conductors fused by a stroke of lightning, the current through it

may be exceedingly small, not even sufficient to heat the



onductor perceptibly, were the latter immersed in aomogeneous medium.

rom the preceding it is clear that when a conductor of igh resistance is connected to the terminals of a source

igh frequency currents of high potential, there may ccur considerable dissipation of energy, principally at thnds of the conductor, in consequence of the action of theas surrounding the conductor. Owing to this, the currenhrough a section of the conductor at a point midway etween its ends may be much smaller than

hrough a section near the ends. Furthermore, theurrent passes principally through the outer portions of he conductor, but this effect is to be distinguished fromhe skin effect as ordinarily interpreted, for the latter

would, or should, occur also in a continuous

ncompressible medium. If a great many incandescentamps are connected in series to a source of such currenthe lamps at the ends may burn brightly, whereas thosen the middle may remain entirely dark. This is duerincipally to bombardment, as before stated. But even ihe currents be steady, provided the difference of

otential is very great, the lamps at the end will burnmore brightly than those in the middle. In such case thers no rhythmical bombardment, and the result is •roduced entirely by leakage. This leakage or dissipation

nto space when the tension is high, is considerable whenncandescent lamps are used, and still more considerable

with arcs, for the latter act like flames. Generally, of



ourse, the dissipation is much smaller with steady, thanwith varying, currents.

have contrived an experiment which illustrates in annteresting manner the effect of lateral diffusion. If a ver

ong tube is attached to the terminal of a high frequency oil, the luminosity is greatest near the terminal and fallsff gradually towards the remote end. This is more

marked if the tube is narrow.

A small tube about one-half inch in diameter and twelve

nches long (Fig. 189), has one of its ends drawn out into ne fibre/ nearly three feet long. The tube is placed in arass socket T which can be screwed on the terminal T Xf the induction coil. The discharge passing through theube first illuminates the bottom of the same, which is ofomparatively large section ; but through the long glass

bre the discharge cannot pass. But gradually the rarefieas inside becomes warmed and more conducting and thischarge spreads into the glass fibre. This spreading is slow, that it may take half a minute or more until theischarge has worked through up to the top of the glassbre, then presenting the appearance of a strongly

uminous thin thread. By adjusting the potential at theerminal the light may be made to travel upwards at anypeed. Once, however, the glass fibre is heated, theischarge breaks through its entire length instantly. The

nteresting point to be noted is that, the higher therequency of the currents, or in other words, the greater

elatively the lateral dissipation, at a slower rate may th



ght be made to propagate through the fibre. Thisxperiment

s best performed with a highly exhausted and freshly made tube. When the tube has been used for some timehe experiment often fails. It is possible that the gradualnd slow impairment of the vacuum is the cause. Thislow propagation of the discharge through a very narrowlass tube corresponds exactly to the propagation of hea

hrough a bar warmed at one end. The quicker the heat arried away laterally the longer time it will take for theeat to warm the remote end. When the current of a low

requency coil is passed through the fibre from end tond, then the lateral dissipation is small and the dischargnstantly breaks through almost without exception.

IG. 189.



IG. l»0.

After these experiments and observations which havehown the importance of the discontinuity or atomictructure of the medium and which will serve to explain,

n a measure at least, the nature of the four kinds of lightffects producible with these currents, I may now giveou an illustration of these effects. For the sake of interemay do this in a manner which to many of you might beovel. You have seen before that we may now convey thlectric vibration to a body by means of a single wire or

onductor of any kind. Since the

uman frame is conducting I may convey the vibrationhrough my body.

irst, as in some previous experiments, I connect my

ody with one of the terminals of a high-tensionransformer and take in my hand an exhausted bulb

which contains a small carbon button mounted upon alatinum wire leading to the outside of the bulb, and theutton is rendered incandescent as soon as theransformer is set to work (Fig. 190). I may place a

onducting shade on the bulb which serves to intensify he action, but is not necessary. Nor is it required that thutton should be in conducting connection witli the handhrough a wire leading through the glass,



IG. 192.

or sufficient energy may be transmitted through thelass itself by inductive action to render the button

ncandescent.



Next I take a highly exhausted bulb containing a stronglyhosphorescent body, above which is mounted a smalllate of aluminum on a platinum wire leading to theutside, and the currents flowing through my body excit

ntense phosphorescence in the bulb (Fig. 191). Next

gain I take in my hand a simple exhausted tube, and inhe same manner the gas inside the tube is renderedighly incandescent or phosphorescent (Fig. 192). Finallymay take in my hand a wire, bare or covered with thick

nsulation, it is quite immaterial; the electrical vibration iu intense as to cover the wire with a luminous film (Fig.

93).

A few words must now be devoted to each of thesehenomena. In the first place, I will consider the

ncandescence of a button or of a solid in general, andwell upon some facts which apply equally to all these

henomena. It was pointed out before that when a thinonductor, such as a lamp filament, for instance, isonnected with one of its ends to the terminal of aransformer of high tension the filament is brought toncandescence partly by a conduction current and partlyy bombardment. The shorter and thicker the filament

he more important becomes the latter, and finally,educing the filament to a mere button, all the heating

must practically be attributed to the bombardment. So ihe experiment before shown, the button is renderedncandescent by the rhythmical impact of freely movablemall bodies in the bulb. These bodies may be the

molecules of the residual as articles of dust or lum s



orn from the electrode ; whatever they are, it is certainhat the heating of the button is essentially connected

with the pressure of such freely movable particles, or of tomic matter in general in the bulb. The heating is the

more intense the greater the number of impacts per

econd and the greater the energy of each impact. Yet thutton would be heated also if it were connected to aource of a steady potential. In such a case electricity

would be carried away from the button by the freely movable carriers or particles flying about, and the

uantity of electricity thus carried away might be

ufficient to bring the button to incandescence by itsassage through the latter. But the bombardment couldot be of great importance in such case. For this reason i

would require a comparatively very great supply of nergy to the button to maintain it at incandescence withsteady potential. The higher the frequency of the

lectric impulses the more economically can the button bmaintained at incandescence. One of the chief reasonswhy this is so, is, I believe, that with impulses of very higrequency there is less exchange of the freely movablearriers around the electrode and this means, that in theulb the heated matter is better confined to the

eighborhood of the button. If a double bulb, as illustraten Fig. 194 be made, comprising a large globe B and amall one 5, each containing as usual a filament/"

mounted on a platinum wire w and w t , it is found, that ihe filaments ff be exactly alike, it requires less energy toeep the filament in the globe b at a certain degree of

ncandescence, than that in the globe B. This is due to the



onfinement of the

movable particles around the button. In this case it is alsscertained, that the filament in the small globe 5 is lesseteriorated when maintained a certain length of time at

ncandescence. This is a necessary consequence of the fahat the gas in the small bulb becomes strongly heatednd therefore a very good conductor, and less work ishen performed on the button, since the bombardmentecomes less intense as the conductivity of the gas

ncreases. In this construction, of course, the small bulb

ecomes very hot and when it reaches an elevatedemperature the convection and radiation on the outsidencrease. On another occasion I have shown bulbs in

which this drawback was largely avoided. In thesenstances a very small bulb, containing a refractory utton, was mounted in a large globe and the space be-



IG. 193.

IG. 194.

ween the walls of both was highly exhausted. The outerarge globe remained comparatively cool in such

onstructions. When the large globe was on the pump anhe vacuum between the walls maintained permanent byhe continuous action of the pump, the outer globe wouldemain quite cold, while the button in the small bulb wasept at incandescence. But when the seal was made, andhe button in the small bulb maintained incandescent

ome length of time, the large globe too would becomewarmed. From this I conjecture that if vacuous space (as

rof. Dewar finds) cannot convey heat, it is so merely inirtue of our rapid motion through space or, generally peaking, by the moti n of the medium relatively to us, fopermanent condition could

ot be maintained without the medium bein constantl



enewed. A vacuum cannot, according to all evidence, beermanently maintained around a hot body.

n these constructions, before mentioned, the small bulbnside would, at least in the first stages, prevent all

ombardment a* - against the outer large globe. Itccurred to me then to ascertain how a metal sieve woulehave in this respect, and several bulbs, as illustrated inig. 195, were prepared for this purpose. r . In a globe &

was mounted a thin filament f (or button) upon alatinum wire w passing through a glass stem and leadin

o the outside of the globe. The filament/"was surroundey a metal sieve s. It was found in experiments with suchulbs that a sieve with wide meshes apparently did not ihe slightest affect the bombardment against the globe b

When the vacuum was high, the shadow of the sieve waslearly projected against the globe and the latter would

et hot in a short while. In some bulbs the sieve .s wasonnected to a platinum wire sealed in the glass. Whenhis w r ire was connected to the other terminal of thenduction coil (the E. M. F. being kept low in this case), oro an insulated plate, the bombardment against the outelobe 1) was diminished. By taking a sieve with fine

meshes the bombardment against the globe b was alwayiminished, but even then if the exhaustion was carriedery far, and when the potential of the transformer wasery high, the globe 1} would be bombarded and heateduickly, though no shadow pf the sieve was visible, owingo the smallness of the meshes. But a glass tube or otherontinuous bod mounted so as to surround the filament



id entirely cut off the bombardment and for a while theuter globe b would remain perfectly cold. Of course whehe glass tube was sufficiently heated the bombardmentgainst the outer globe could be noted at once. Thexperiments with these bulbs seemed to show that the

peeds of the projected molecules or particles must beonsiderable (though quite insignificant when comparedwith that of light), otherwise it would be difficult to

nderstand how they could traverse a fine metal sievewithout being affected, unless it were found that suchmall particles or atoms cannot be acted upon directly at

measurable distances. In regard to the speed of therojected atoms, Lord Kelvin has recently estimated it about one kilometre a second or thereabouts in anrdinary Crookes bulb. As the potentials obtainable withisruptive discharge coil are much higher than with or-

HIGH FREQUENCY AND HIGH POTENTIALURRENTS. :',(>:;

inary coils, the speeds must, of course, be much greaterwhen the bulbs are lighted from such a coil. Assuming thpeed to be as high as five kilometres and uniform

hrough the whole trajectory, as it should be in a very ighly exhausted vessel, then if the alternatelectrifications of the electrode would be of a frequency ove million, the greatest distance a particle could getway from the electrode would be one millimetre, and if ould be acted upon directly at that distance, the

xchange of electrode matter or of the atoms would be



ery slow and there would be practically noombardment against the bulb. This at least should be sthe action of an electrode upon the atoms of the residu

as would be such as upon electrified bodies which we caerceive. A hot body enclosed in an exhausted bulb

roduces always atomic bombardment, but a hot body as no definite rhythm, for its molecules performibrations of all kinds.

f a bulb containing a button or filament be exhausted asigh as is possible with the greatest care and by the use o

he best artifices, it is often observed that the dischargeannot, at first, break through, but after some time,robably in consequence of some changes within the bulbhe discharge finally passes through and the button isendered incandescent. In fact, it appears that the highehe degree of exhaustion the easier is the incandescence

roduced. There seem to be no other causes to which thencandescence might be attributed in such case except tohe bombardment or similar action of the residual gas, orf particles of matter in general. But if the bulb bexhausted with the greatest care can these play anmportant part ? Assume the vacuum in the bulb to be

olerably perfect, the great interest then centres in theuestion: Is the medium which pervades all spaceontinuous or atomic ? If atomic, then the heating of aonducting button or filament in an exhausted vessel

might be due largely to ether bombardment, and then theating of a conductor in general through which currentsf high frequency or high potential are passed must be



modified by the behavior of such medium ; then also thekin effect, the apparent increase of the ohmic resistancetc., admit, partially at least, of a different explanation.

t is certainly more in accordance with many phenomena

bserved with high-frequency currents to hold that allpace is pervaded with free atoms, rather than to assumhat it is devoid of these, and dark and cold, for so it muse, if filled with a continuous medium, since in such therean be neither heat nor light.

s then energy transmitted by independent carriers or bhe vibration of a continuous medium ? This importantuestion is by no means as yet positively answered. But

most of the effects which are here considered, especially he light effects, incandescence, or phosphorescence,nvolve the presence of free atoms and would be

mpossible without these.

n regard to the incandescence of a refractory button (orlament) in an exhausted receiver, which has been one o

he subjects of this investigation, the chief experiences,which may serve as a guide in constructing such bulbs,

may be summed up as follows : 1. The button should be mall as possible, spherical, of a smooth or polishedurface, and of refractory material which withstandsvaporation best. 2. The support of the button should beery thin and screened by an aluminum and mica sheet,s I have described on another occasion. 3. The

xhaustion of the bulb should be as high as possible. 4.he fre uenc of the currents should be as hi h as



racticable. 5. The currents should be of a harmonic risend fall, without sudden interruptions. 6. The heat shoule confined to the button by inclosing the same in a smalulb or otherwise. 7. The space between the walls of themall bulb and the outer globe should be highly

xhausted. Most of the considerations which apply to thencandescence of a solid just considered may likewise bepplied to phosphorescence. Indeed, in an exhaustedessel the phosphorescence is, as a rule, primarily excitey the powerful beating of the electrode stream of atomsgainst the phosphorescent body. Even in many cases,

where there is no evidence of such a bombardment, Ihink that phosphorescence is excited by violent impactsf atoms, which are not necessarily thrown off from thelectrode but are acted upon from the same inductively hrough the medium or through chains of other atoms.hat mechanical shocks play an important part in excitin

hosphorescence in a bulb may be seen from the followinxperiment. If a bulb, constructed as that illustrated inig. 1Y4, be taken and exhausted with the greatest careo that the discharge cannot pass, the filament f acts by lectrostatic induction upon the tube t and the latter is sn vibration. If the tube o be rather wide, about an inch o

o, the filament may be so powerfully vibrated thatwhenever it hits the glass tube it excites

hosphorescence. But the phosphorescence ceases whenhe filament comes to rest. The vibration can be arrestednd again started by varying the

re uenc of the currents. Now the filament lias its own



eriod of vibration, and if the frequency of the currents iuch that there is resonance, it is easily set vibrating,hough the potential of the currents be small. I have oftebserved that the filament in the bulb is destroyed by uch mechanical resonance. The filament vibrates as a

ule so rapidly that it cannot be seen and thexperimenter may at first be mystified. When such anxperiment as the one described is carefully performed,he potential of the currents need be extremely small, anor this reason I infer that the phosphorescence is thenue to the mechanical shock of the filament against the

lass, just as it is produced by striking a loaf of sugar withknife. The mechanical shock produced by the projected

toms is easily noted when a bulb containing a button israsped in the hand and the current turned on suddenlybelieve that a bulb could be shattered by observing theonditions of resonance.

n tlie experiment before cited it is, of course, open to sahat the glass tube, upon coming in contact with thelament, retains a charge of a certain sign upon the pointf contact. If now the filament again touches the glass athe same point while it is oppositely charged, the charges

qualize under evolution of light. But nothing of mportance would be gained by such an explanation. It isnquestionable that the initial charges given to the atomr to the glass play some part in excitinghosphorescence. So, for instance, if a phosphorescentulb be first excited by a high frequency coil by onnecting it to one of the terminals of the latter and the



egree of luminosity be noted, and then the bulb be highharged from a Holtz machine by attaching it preferably o the positive terminal of the machine, it is found that

when the bulb is again connected to the terminal of theigh frequency coil, the phosphorescence is far more

ntense. On another occasion I have considered theossibility of some phosphorescent phenomena in bulbseing produced by the incandescence of an infinitesimal

ayer on the surface of the phosphorescent body.ertainly the impact of the atoms is powerful enough toroduce intense incandescence by the collisions, since

hey bring quickly to a high temperature a body of onsiderable bulk. If any such effect exists, then the bestppliance for producing phosphorescence in a bulb, which

we know so far, is a disruptive discharge coil giving annormous potential with but few fundamental dischargesay 25-30 per second, just enough to produce a continu-

us impression upon the eye. It is a fact that such a coilxcites phosphorescence under almost any condition andt all degrees of exhaustion, and I have observed effects

which appear to be due to phosphorescence even atrdinary pressures of the atmosphere, when the

otentials are extremely high. But if phosphorescent lighs produced by the equalization of charges of electrifiedtoms (whatever this may mean ultimately), then theigher the frequency of the impulses or alternatelectrifications, the more economical will be the lightroduction. It is a long known and noteworthy fact that ahe phosphorescent bodies are poor conductors of



lectricity and heat, and that all bodies cease to emithosphorescent light when they are brought to a certainemperature. Conductors on the contrary do not possesshis quality. There are but few exceptions to the rule.arbon is one of them. Becquerel noted that carbon

hosphoresces at at a certain elevated temperaturereceding the dark red. This phenomenon may be easilybserved in bulbs provided with a rather large carbonlectrode (say, a sphere of six millimetres diameter). If he current is turned on after a few seconds, a snow whitlm covers the electrode, just before it gets dark red.

imilar effects are noted with other conducting bodies,ut many scientific men will probably not attribute themo true phosphorescence. Whether true incandescence hnything to do with phosphorescence excited by atomicmpact or mechanical shocks still remains to be decided,ut it is a fact that all conditions, w T hich tend to localize

nd increase the heating effect at the point of impact, arelmost invariably the most favorable for the production ohosphorescence. So, if the electrode be very small, whic

s equivalent to saying in general, that the electric densits great; if the potential be high, arid if the gas be highly arefied, all of which things imply high speed of the

rojected atoms, or matter, and consequently violentmpacts —the phosphorescence is very intense. If a bulbrovided with a large and small electrode be attached tohe terminal of an induction coil, the small electrodexcites phosphorescence while the large one may not doo, because of the smaller electric density and hencemaller speed of the atoms. A bulb provided with a large



lectrode may be grasped with the hand Avhile thelectrode is connected to the terminal of the coil and it

may not phosphoresce ; but if instead of grasping the bulwith the hand, the same be touched with a pointed wire,he phosphorescence at once spreads

hrough the bulb, because of the great density at the poinf contact. .With low frequencies it seems that gases of reat atomic weight excite more intense phosphorescenchan those of smaller weight, as for instance, hydrogen.

With high frequencies the observations are not

ufficiently reliable to draw a conclusion. Oxygen, as iswell-known, produces exceptionally strong effects, whichmay be in part due to chemical action. A bulb with

ydrogen residue seems to be most easily excited.lectrodes which are most easily deteriorated produce

more intense phosphorescence in bulbs, but the condition

s not permanent because of the impairment of theacuum and the deposition of the electrode matter uponhe phosphorescent surfaces. Some liquids, as oils, fornstance, produce magnificent effects of phosphorescenceor fluorescence ?), but they last only a few {seconds. So bulb has a trace of oil on the walls and the current is

urned on, the phosphorescence only persists for a few moments until the oil is carried away. Of all bodies so farried, sulphide of zinc seems to be the most susceptible thosphorescence. Some samples, obtained through theindness of Prof. Henry in Paris, were employed in manyf these bulbs. One of the defects of this sulphide is, that

oses its quality of emitting light when brought to a



emperature which is by no means high. It can thereforee used only for feeble intensities. An observation which

might deserve notice is, that when violently bombardedrom an aluminum electrode it assumes a black color, buingularly enough, it returns to the original condition

when it cools down.

he most important fact arrived at in pursuingnvestigations in this direction is, that in all cases it isecessary, in order to excite phosphorescence with a

minimum amount of energy, to observe certain

onditions. Namely, there is always, no matter what therequency of the currents, degree of exhaustion andharacter of the bodies in the bulb, a certain potentialassuming the bulb excited from one terminal) orotential difference (assuming the bulb to be excited witoth terminals) which produces the most economical

esult. If the potential be increased, considerable energymay be wasted without producing any more light, and if

e diminished, then again the light production is not asconomical. The exact condition under which the bestesult is obtained seems to depend on many things of aifferent nature, and it is to be yet investigated by other

xperimenters, but it will certainly

ave to be observed when such phosphorescent bulbs arperated, if the best results are to be obtained.

oming now to the most interesting of these phenomena

he incandescence or phosphorescence of gases, at low ressures or at the ordinar ressure of the atmos here



we must seek the explanation of these phenomena in theame primary causes, that is, in shocks or impacts of thetoms. Just as molecules or atoms beating upon a solidody excite phosphorescence in the same or render it

ncandescent, so when colliding among themselves they

roduce similar phenomena. But this is a very insufficienxplanation and concerns only the crude mechanism.ight is produced by vibrations which go on at a ratelmost inconceivable. If we compute, from the energy ontained in the form of known radiations in a definitepace the force which is necessary to set up such rapid

ibrations, we find, that though the density of the ethere incomparably smaller than that of any body we knowven hydrogen, the force is something surpassingomprehension. What is this force, which in mechanical

measure may amount to thousands of tons per squarench ? It is electrostatic force in the light of modern view

t is impossible to conceive how a body of measurableimensions could be charged to so high a potential thathe force would be sufficient to produce these vibrationsong before any such charge could be imparted to theody it would be shattered into atoms. The sun emitsght and heat, and so does an ordinary flame or

ncandescent filament, but in neither of these can theorce be accounted for if it be assumed that it is associate

with the body as a whole. Only in one way may weccount for it, namely, by identifying it with the atom. Antom is so small, that if it be charged by coming in contac

with an electrified body and the charge be assumed toollow the same law as in the case of bodies of measurabl



imensions, it must retain a quantity of electricity whichs fully capable of accounting for these forces andremendous rates of vibration. But the atom behavesingularly in this respect—it always takes the same "harge."

t is very likely that resonant vibration plays a mostmportant part in all manifestations of energy in nature.hroughout space all matter is vibrating, and all rates ofibration are represented, from the lowest musical noteo the highest pitch of the chemical rays, hence an atom,

r complex of atoms, no matter what its period, must finvibration with which it is in resonance.

When we consider the enormous rapidity of the lightibrations, we realize the impossibility of producing suchibrations directly with any apparatus of measurable

imensions, and we are driven to the only possible meanf attaining the object of setting up waves of light by lectrical means and economically, that is, to affect the

molecules or atoms of a gas, to cause them to collide andibrate. We then must ask ourselves—How can free

molecules or atoms .be affected ?

t is a fact that they can be affected by electrostatic forces is apparent in many of these experiments. By varyinghe electrostatic force we can agitate the atoms, and caushem to collide accompanied by evolution of heat andght. It is not demonstrated beyond doubt that we can

ffect them otherwise. If a luminous discharge isroduced in a closed exhausted tube do the atoms



rrange themselves in obedience to any other but tolectrostatic force acting in straight lines from atom totom ? Only recently I investigated the mutual actionetween two circuits with extreme rates of vibration.

When a battery of a few jars (c c c c, Fig. 196) is

ischarged through a primary p of low resistance (theonnections being as illustrated in Figs. 183«,83&andl83c), and the frequency of vibration is many

millions there are great differences of potential betweenoints on the primary not more than a few inches apart.hese differences may be 10,000 volts per inch, if not

more, taking the maximum value of the E. M. F. Theecondary s is therefore acted upon by electrostaticnduction, which is in such extreme cases of much greatemportance than the electro-dynamic. To such suddenmpulses the primary as well as the secondary are pooronductors, and therefore great differences of potential

may be produced by electrostatic induction betweendjacent points on the secondary. Then sparks may jumpetween the wires and streamers become visible in theark if the light of the discharge through the spark gap c? be carefully excluded. If now we substitute a closedacuum tube for the metallic secondary s, the difference

f potential produced in the tube by electrostaticnduction from the primary are fully sufficient to exciteortions of it; but as the points of certain differences of otential on the primary are not fixed, but are generally onstantly changing in position, a luminous band isroduced in the tube, apparently not touching the glass,s it should if the oints of maximum and minimum



ifferences of potential were fixed on the primary. I doot exclude the possibility of such a

ube being excited only by electro-dynamic induction, foery able physicists hold this view; but in my opinion,

here is as yet no positive proof given that atoms of a gasn a closed tube may arrange themselves in chains underhe action of an electromotive impulse produced by lectro-dynamic induction in the tube. I have been unablo far to produce striae in a tube, however long, and at

whatever degree of exhaustion, that is, striae at right

ngles to the supposed direction of the discharge or thexis of the tube ; but I have distinctly observed in a largeulb, in which a wide luminous band was produced by assing a discharge of a battery through a wireurrounding the bulb, a circle of feeble luminosity etween two luminous bands, one of which was more

ntense than the other. Furthermore, with my presentxperience I do not think that such a gas discharge in alosed tube can vibrate, that is, vibrate as a whole. I amonvinced that no

V



IG. 196. FIG. 197.

ischarge through a gas can vibrate. The atoms of a gasehave very curiously in respect to sudden electric

mpulses. The gas does not seem to possess any ppreciable inertia to such impulses, for it is a fact, thathe higher the frequency of the impulses, with the greate

reedom does the discharge pass through the gas. If theas possesses no inertia thqui it cannot vibrate, for somenertia is necessary for the free vibration. I conclude fromhis that if a lightning discharge occurs between twolouds, there can be no oscillation, such as would bexpected, considering the capacity of the clouds. But if th

ghtning discharge strike the earth, there is alwaysibration—in the earth, but not in the cloud. In a gasischarge each atom vibrates at its own rate, but there iso vibration of the conducting gaseous mass as a whole.his is an important consideration in the great problem oroducing light economi-

allj, for it teaches us that to reach this result we must usmpulses of very high frequency and necessarily also of igh potential. It is a fact that oxygen produces a more

ntense light in a tube. Is it because oxygen atoms possesome inertia and the vibration does not die out instantly

ut then nitrogen should be as good, and chlorine andapors of many other bodies much better than oxygen,



nless the magnetic properties of the latter enterrominently into play. Or, is the process in the tube of anlectrolytic nature ? Many observations certainly speak or it, the most important being that matter is alwaysarried away from the electrodes and the vacuum in a

ulb cannot be permanently maintained. If such processakes place in reality, then again must we take refuge inigh frequencies, for, with such, electrolytic action shoulde reduced to a minimum, if not ren_ dered entirely

mpossible. It is an undeniable fact that with very highrequencies, provided the impulses be of harmonic natur

ke those obtained from an alternator, there is lesseterioration and the vacua are more permanent. Withisruptive discharge coils there are sudden rises of otential and the vacua are more quickly impaired, forhe electrodes are deteriorated in a very short time. It

was observed in some large tubes, which were provided

with heavy carbon blocks B B l5 connected to platinumwires w w^ (as illustrated in Fig. 197), and which were

mployed in experiments with the disruptive dischargenstead of the ordinary air gap, that the carbon particlesnder the action of the powerful magnetic field in whichhe tube was placed, were deposited in regular fine lines

n the middle of the tube, as illustrated. These lines werettributed to the deflection or distortion of the dischargey the magnetic field, but why the deposit occiirredrincipally where the field was most intense did notppear quite clear. A fact of interest, likewise noted, washat the presence of a strong magnetic field increases theeterioration of the electrodes robabl b reason of the



apid interruptions it produces, whereby there is actuallyhigher E. M. F. maintained between the electrodes.

Much would remain to be said about the luminous effectsroduced in gases at low or ordinary pressures. With the

resent experiences before us we cannot say that thessential nature of these charming phenomena isufficiently known. But investigations in this direction areing pushed with exceptional ardor. Every line of cientific pursuit has its fascinations, but electrical


nvestigation appears to possess a peculiar attraction, forhere is no experiment or observation of any kind in theomain of this wonderful science which Avould notorcibly appeal to us. Yet to me it seems, that of all the

many marvelous things we observe, a vacuum tube,xcited by an electric impulse from a distant source,ursting forth out of the darkness and illuminating theoom with its beautiful light, is as lovely a phenomenon aan greet our eyes. More interesting still it appears wheneducing the fundamental discharges across the gap to a

ery small nuiu-





IG. 198.

er and waving the tube about we produce all kinds of esigns in luminous lines. So by way of amusement I tak

straight long tube, or a square one, or a square attacheo a straight tube, and by whirling them about in theand, I imitate the spokes of a wheel, a Gramme windingdrum winding, an alternate current motor winding, etc

Fig. 198). Viewed from a distance the effect is weak andmuch of its beauty is lost, but being near or holding theube in the hand, one cannot resist its charm.

n presenting these insignificant results I have notttempted to arrange and co-ordinate them, as would beroper in a strictly scientific investigation, in which everyucceeding result should be a logical sequence of thereceding, so that it might be guessed in advance by theareful reader or attentive listener. I have preferred tooncentrate my energies chiefly upon advancing novelacts or ideas which might serve as suggestions to othersnd this may serve as an excuse for the lack of harmony

he explanations of the phenomena have been given inood faith and in the spirit of a student prepared to findhat they admit of a better interpretation. There can beo great harm in a student taking an erroneous view, bu

when great minds err, the world must dearly pay for themistakes.

HAPTEK XXIX.



ESLA ALTERNATING CURRENT GENERATORS FOHIGH FREQUENCY, IN DETAIL.

t lias become a common practice to operate arc lamps blternating or pulsating, as distinguished from continuou

urrents ; but an objection which has been raised to suchystems exists in the fact that the arcs emit a pronounceound, varying with the rate of the alternations orulsations of current. This noise is due to the rapidly lternating heating and cooling, and consequent expansiond contraction, of the gaseous matter forming-the arc,

which corresponds with the periods or impulses of theurrent. Another disadvantageous feature is found in theifficulty of maintaining an alternating current arc inonsequence of the periodical increase in resistanceorresponding to the periodical working of the current.his feature entails a further disadvantage, namely, that

mall arcs are impracticable.

heoretical considerations have led Mr. Tesla to the belihat these disadvantageous features could be obviated bmploying currents of a sufficiently high number of lternations, and his anticipations have been confirmed iractice. These rapidly alternating currents render itossible to maintain small arcs which, besides, possess thdvantages of silence and persistency. The latter qualitys due to the necessarily rapid alternation^ inonsequence of which the arc has no time to cool, and is

lways maintained at a high temperature and low esistance.



At the outset of his experiments Mr. Tesla encounteredreat difficulties in the construction of high frequency

machines. A generator of this kind is described here,which, though constructed quite some time ago, is wellworthy of a detailed description. It may be mentioned, in

assing, that dynamos of this type have been used by Mesla in his lighting researches and experiments withurrents of high potential and high frequency, andeference to them will be found in his lectures elsewhererinted in this volume. 1

. See pages 153-4 5.

n the aecompaning engravings, Figs. 199 and 200 showhe machine, respectively, in side elevation and verticalross-section ; Figs. 201, 202 and 203 showing enlargedetails of construction. As will be seen, A is an annular

magnetic frame, the interior of which is provided with aarge number of pole-pieces D.

Owing to the very large number and small size of theoles and the spaces between them, the field coils arepplied by winding an insulated conductor F zigzaghrough the grooves, as shown in Fig. 203, carrying the

wire around the annulus to form as many layers as isesired. In this way the pole-pieces D will be energized

with alternately opposite polarity around the entire ring

or the armature, Mr. Tesla employs a spider carrying aing



turned down, except at its edges, to form a trough-likeeceptacle for a mass of fine annealed iron wires K, whichre wound in the groove to form the core proper for thermature-coils. Pins L are set in the sides of the ring j an

he coils M are wound over the periphery of thermature-structure and around the pins. The coils M areonnected together in series, and these terminals Narried through the hollow shaft H to contact-rings P P,rom which the currents are taken off by brushes o.

n this way a machine with a very large number of polesmay be constructed. It is easy, for instance, to obtain inhis manner three hundred and seventy-five to fourundred poles in a machine that may be safely driven atpeed of fifteen hundred or sixteen hundred revolutionser minute, which will produce ten

NVENTIONS OF NIKOLA TESLA,



housand or eleven thousand alternations of current perecond. Arc lamps K R are shown in the diagram asonnected up in series with the machine in Fig. 200. If uch a current be applied to running arc lamps, the sounroduced by or in the arc becomes practically inaudible,

or, by increasing the rate of change in the current, andonsequently the number of vibrations per unit of time ohe gaseous material of the arc up to, or beyond, tenhousand or eleven thousand per second, or to what isegarded as the limit of audition, the sound due to such

ibrations will not be audible. The exact number of hanges or undulations necessary to produce this resultwill vary somewhat according to the size of the arc—thats to say, the smaller the arc, the greater the

IGS. 200, 201, 202 and 203.



umber of changes that will be required to render itnaudible within certain limits. It should also be statedhat the arc should not exceed a certain length.

he difficulties encountered in the construction of these

machines are of a mechanical as well as an electricalature. The machines may be designed on two plans: theeld may be formed either of alternating poles, or of polarojections of the same polarity. Up to about 15,000lternations per second in an experimental machine, theormer plan may be followed, but a more efficient

machine is obtained on the second plan.

n the machine above described, which was capable of unning two arcs of normal candle power, the field wasomposed of a

ing of wrought iron 32 inches outside diameter, andbout 1 inch thick. The inside diameter was 30 inches.here were 384 polar projections. The wire was wound iigzag form, but two wires were wound so as toompletely envelop the projections. The distance betweehe pro jections is about T 3 ^ inch, and they are a little

ver j\ inch thick. The field magnet was made relatively mall so as to adapt the machine for a constant current.here are 384 coils connected in two series. It was found

mpracticable to use any wire much thicker than No. 26 Bnd S. gauge on account of the local effects. In such a

machine the clearance should be as small as possible; for

his reason the machine was made onl 1 £ inch wide, so



hat the binding wires might be obviated. The armaturewires must be wound with

IG. 204.

reat care, as they are apt to fly off in consequence of thereat peripheral speed. In various experiments thismachine has been run as high as 3,000 revolutions perminute. Owing to the great speed it was possible to obtai

s high as 10 amperes out of the machine. Thelectromotive force was regulated by means of an

djustable condenser within very wide limits, the limitseing the greater, the greater the speed. This machine

was frequently used to run Mr. Tesla's laboratory lights.

he machine above described was only one of many suchypes constructed. It serves well for an experimental

machine, but if still higher alternations are required andigher efficiency is necessary, then a machine on a plan



hown in Figs. 204 to


07, is preferable. The principal advantage of this type omachine is that there is not much magnetic leakage, andhat a field may be produced, varying greatly in intensityn places not much distant from each other.

n these engravings, Figs. 204 and 205 illustrate amachine in which the armature conductor and field coils

re stationary, while the field magnet core revolves. Fig.06 shows a machine embodying the same plan of onstruction, but having a stationary field magnet andotary armature.

he conductor in which the currents are induced may be

rranged in various ways; but Mr. Tesla prefers theollowing method: He employs an annular plate of copperD, and by



IG. 205.

means of a saw cuts in it radial slots from one edge nearlhrough to the other, beginning alternately from oppositdges. In this way a continuous zigzag conductor isormed. When the polar projections are £ inch wide, the

width of the conductor should not, under any ircumstances, be more than ^ inch wide ; even then the

ddy effect is considerable.

o the inner edge of this plate are secured two rings of onmagnetic metal E, which are insulated from theopper conductor, but held firmly thereto by means of tholts F. Within the rings E is then placed an annular coil

G, which is the energizing coil for the field magnet. Theonductor D and the parts attached thereto are supportey means of the cylindrical shell or

asting A A, the two parts of which are brought togethernd clamped to the outer edge of the conductor D.

he core for the field magnet is built up of two circulararts H H, formed with annular grooves i, which, whenhe two parts are brought together, form a space for theeception of the energizing coil G. The hubs of the coresre trued off, so as to fit closely against one another, whil

he outer portions or flanges which form the polar faces jre reduced somewhat in thickness to make room for the



onductor D, and are serrated on their faces. The numbef serrations in the polar faces is arbitrary ;

IG. 206.

ut there must exist between them and the radialortions of the conductor D certain relation, which will benderstood by reference to Fig. 207 in which N Nepresent the projections or points on one face of the corf the field, and s s the points of the other face. Theonductor D is shown in this figure in section a a'esignating the radial portions of the conductor, and 5 th

nsulating divisions between them. The relative width of

he parts a a' and the space between any two adjacentoints N N or s s is such that when the radial portions a o



he conductor are passing between the opposite points Nwhere the field is strongest, the intermediate radial

ortions a' are passing through the

widest spaces midway between such points and where th

eld is weakest. Since the core on one side is of oppositeolarity to the part facing it, all the projections of oneolar face will be of opposite polarity to those of the otheace. Hence, although the space between any two adjacenoints on the same face may be extremely small, there

will be no leakage of the magnetic lines between any two

oints of the same name, but the lines of force will passcross from one set of points to the other. Theonstruction followed obviates to a great degree theistortion of the magnetic lines by the action of theurrent in the conductor D, in which it will be observedhe current is flowing at any given time from the centre

oward the periphery in one set of radial parts a and inhe opposite direction in the adjacent parts a'.

n order to connect the energizing coil G, Fig. 204, with aource of continuous current, Mr. Tesla utilizes twodjacent radial portions of the conductor D for connectin

he terminals of the coil G with two binding posts M. Forhis purpose the plate D is cut

wwyy/

r m.mmmm ..«.•'•>

IG. 207.



ntirely through, as shown, and the break thus made isridged over by a short conductor c. The plate D is cuthrough to form two terminals d, which are connected toinding posts N. The core H H, when rotated by theriving pulley, generates in the conductors D anlternating current, which is taken off from the bindingosts ST.

When it is desired to rotate the conductor between theaces of a stationary field magnet, the construction shownn Fig. 206, is adopted. The conductor D in this case is or

may be made in substantially the same manner as aboveescribed by slotting an annular conducting-plate andupporting it between two heads o, held together by boltand fixed to the driving-shaft K. The inner edge of thelate or conductor D is preferably flanged to secure a

rmer union between it and the heads o. It is insulatedrom the head. The field-magnet in this case consists of wo annular parts H H, provided with annular grooves ior the reception of the coils. The flanges or facesurrounding

he annular groove are brought together, while the inneranges are serrated, as in the previous case, and form tholar faces. The two parts H H are formed with a base E,pon which the machine rests, s s are non-magneticushings secured or set in the central opening of theores. The conductor D is cut entirely through at one

oint to form terminals, from which insulated conductorsare led through the shaft to collecting-rings v.



n one type of machine of this kind constructed by Mr.esla, the field had 480 polar projections on each side,nd from this machine it was possible to obtain 30,000lternations per second. As the polar projections mustecessarily be very narrow, very thin wires or sheets

must be used to avoid the eddy current effects. Mr. Teslas thus constructed machines with a stationary rmature and rotating field, in which case also the field-oil was supported so that the revolving part consistednly of a wrought iron body devoid of any wire and also

machines with a rotating armature and stationary field.he machines may he either drum or disc, but Mr. Teslaxperience shows the latter to be preferable.

n the course of a very interesting article contributed tohe Electrical World in February, 1891, Mr. Tesla makes

ome suggestive remarks on these high frequency machines and his experiences with them, as well as withther parts of the high frequency apparatus. Part of it isuoted here and is as follows:—

he writer will incidentally mention that any one who

ttempts for the first time to construct such a machinewill have a tale of woe to tell. He will first start out, as amatter of course, by making an armature with theequired number of polar projections. He will then get thatisfaction of having produced an apparatus which is fito accompany a thoroughly Wagnerian opera. It may

esides possess the virtue of converting mechanicalner into heat in a nearl erfect manner. If there is a



eversal in the polarity of the projections, he will get heaut of the machine; if there is no reversal, the heating wie less, but the output will be next to nothing. He willhen abandon the iron in the armature, and he will getrom the Scylla to the Charybdis. He will look for one

ifficulty and will find another, but, after a few trials, hemay get nearly what he wanted.

Among the many experiments winch may be performedwith such a machine, of not the least interest are those

erformed with a high-tension induction coil. The

haracter of the discharge is completely changed. The ars established at much greater distances, and it is so easilffected by the slightest current of air that it often

wriggles around in the most singular manner. It usually mits the rhythmical sound peculiar to the alternateurrent arcs, but the curious point is that the sound may

e heard with a number of alternations far above tenhousand per second, which by many is considered to bebout the limit of audition. In many respects the coilehaves like a static machine. Points impair considerablyhe sparking interval, electricity escaping from themreely, and from a wire attached to one of the terminals

treams of light issue, as though it were connected to aole of a powerful Toepler machine. All these phenomenre, of course, mostly due to the enormous differences ofotential obtained. As a consequence of the self-inductiof the coil and the high frequency, the current is minute

while there is a corresponding rise of pressure. A currenmpulse of some strength started in such a coil should



ersist to flow no less than four ten-thousandths of aecond. As this time is greater than half the period, itccurs that an opposing electromotive force begins to act

while the current is still flowing. As a consequence, theressure rises as in a tube filled with liquid and vibrated

apidly around its axis. The current is so small that, in thpinion and involuntary experience of the writer, theischarge of even a very large coil cannot produceeriously injurious effects, whereas, if the same coil wereperated with a current of lower frequency, though thelectromotive force would be much smaller, the discharg

would be most certainly injurious. This result, however, ue in part to the high frequency. The writer'sxperiences tend to show that the higher the frequency he greater the amount of electrical energy which may bassed through the body without serious discomfort;

whence it seems certain that human tissues act as

ondensers.

One is not quite prepared for the behavior of the coilwhen connected to a Leyden jar. One, of course,

nticipates that since the frequency is high the capacity ohe jar should be small. He therefore takes a very small

ar, about the size of a small wine glass, but he finds thatven with this jar the coil is practically short-circuited. Hhen reduces the capacity until he comes to

bout the capacity of two spheres, say, ten centimetres iiameter and two to four centimetres apart. The

ischarge then assumes the form of a serrated band



xactly like a succession of sparks viewed in a rapidly evolving mirror; the serrations, of course, correspondino the condenser discharges. In this case one may observqueer phenomenon. The discharge starts at the nearesoints, works gradually up, breaks somewhere near the

op of the spheres, begins again at the bottom, and so onhis goes on so fast that several serrated bands are seent once. One may be puzzled for a few minutes, but thexplanation is simple enough. The discharge begins at thearest points, the air is heated and carries the arcpward until it breaks, when it is reestablished at the

earest points, etc. Since the current passes easily hrough a condenser of even small capacity, it will beound (juite natural that connecting only one terminal to ody of the same size, no matter how well insulated,

mpairs considerably the striking distance of the arc.

xperiments with Greissler tubes are of special interest.An exhausted tube, devoid of electrodes of any kind, will

ght up at some distance from the coil. If a tube from aacuum pump is near the coil the whole of the pump isrilliantly lighted. An incandescent lamp approached tohe coil lights up and gets perceptibly hot. If a lamp have

he terminals connected to one of the binding posts of thoil and the hand is approached to the bulb, a very curiound rather unpleasant discharge from the glass to theand takes place, and the filament may become

ncandescent. The diSc"Iiarge resembles to some extenthe stream issuing from the plates of a powerful Toepler

machine, but is of incomparably greater quantity. The



amp in this case acts as a condenser, the rarefied gaseing one coating, the operator's hand the other. By aking the globe of a lamp in the hand, and by bringing th

metallic terminals near to or in contact with a conductoronnected to the coil, the carbon is brought to bright

ncandescence and the glass is rapidly heated. With a00-volt 10 c. p. lanro one may without great discomforttand as much current as will bring the lamp to aonsiderable brilliancy; but it can be held in the hand onlor a few minutes, as the glass is heated in an incredibly hort time. When a tube is lighted by bringing it near to

he coil it may be made to go out by interposing a metallate on the hand between the coil and tube; but if the

metal plate be fastened to a glass rod or otherwisensulated, the tube

may remain lighted if the plate be interposed, or may

ven increase in luminosity. The effect depends on theosition of the plate and tube relatively to the coil, and

may be always easily foretold by assuming thatonduction takes place from one terminal of the coil to thther. According to the position of the plate, it may eitheivert from or direct the current to the tube. In another

ne of work the writer has in frequent experimentsmaintained incandescent lamps of 50 or 100 volts burnin

t any desired candle power with both the terminals of ach lamp connected to a stout copper wire of no morehan a few feet in length. These experiments seemnteresting enough, but they are not more so than theueer experiment of Faraday, which has been revived



nd made much of by recent investigators, and in which ischarge is made to jump between two points of a bentopper wire. An experiment may be cited here which maeem equally interesting. If a Geissler tube, the terminalf which are joined by a copper wire, be approached to th

oil, certainly no one would be prepared to see the tubeght up. Curiously enough, it does light up, and, what ismore, the wire does not seem to make much difference.Now one is apt to think in the first moment that thempedance of the wire might have something to do withhe phenomenon. But this is of course immediately

ejected, as for this an enormous frequency would beequired. This result, however, seems puzzling only at'rst ; for upon reflection it is quite clear that the wire ca

make but little difference. It may be explained in morehan one way, but it agrees perhaps best with observatioo assume that conduction takes place from the terminal

f the coil through the space. On this assumption, if theube with the wire be held in any position, the wire canivert little more than the current which passes throughhe space occupied by the wire and the metallic terminalf the tube ; through the adjacent space the currentasses practically undisturbed. For this reason, if the tub

e held in any position at right angles to the line joininghe binding posts of the coil, the wire makes hardly any ifference, but in a position more or less parallel with thane it impairs to a certain extent the brilliancy of the tubnd its facility to light up. Numerous other phenomena

may be explained on the same assumption. For instancethe ends of the tube be provided with washers of



ufficient size and held in the line joining the terminals ofhe coil, it will not light up, and then nearly the whole of he current, which would otherwise

ass uniformly through the space between the washers,

iverted through the wire. But if the tube be inclinedufficiently to that line, it will light up in spite of thewashers. Also, if a metal plate be fastened upon a glassod and held at right angles to the line joining the bindingosts, and nearer to one of them, a tube held more or lesarallel with the line will light up instantly when one of

he terminals touches the plate, and will go out wheneparated from the plate. The greater the surface of thelate, up to a certain limit, the easier the tube will lightp. When a tube is placed at right angles to the straightne joining the binding posts, and then rotated, its

uminosity steadily increases until it is parallel with that

ne. The writer must state, however, that he does natavor the idea of a leakage or current through the spaceny more than as a suitable explanation, for he isonvinced that all these experiments could not beerformed with a static machine yielding a constantifference of potential, and that condenser action is

argely concerned in these phenomena.

t is well to take certain precautions when operating aRuhm-korff coil with very rapidly alternating currents.

he primary current should not be turned on too long,lse the core may get so hot as to melt the gutta-percha

r paraffin, or otherwise injure the insulation, and this



may occur in a surprisingly short time, considering theurrent's strength. The primary current being turned onhe tine wire terminals may be joined without great risk,he impedance being so great that it is difficult to forcenough current through the fine wire so as to injure it,

nd in fact the coil may be on the whole much safer whenhe terminals of the fine wire are connected than whenhey are insulated; but special care should be taken whehe terminals are connected to the coatings of a Leydenar, for with anywhere near the critical capacity , whichust counteracts the self-induction at the existing

requency, the coil might meet the fate of St. Poly-carpuf an expensive vacuum pump is lighted up by being neao the coil or touched with a wire connected to one of theerminals, the current should be left on no more than aew moments, else the glass will be cracked by the heatinf the rarefied gas in one of the narrow passages—in the

writer's own experience quod erat demonstrandum. 1

It is thought necessary to remark that, although thenduction coil may give quite a good result when operate

with such rapidly alternating currents, yet itsonstruction, quite irrespective of the iron core, makes it

ery unfit for such high frequencies, and to obtain theest results the construction should be greatly modified.

here are a good many other points of interest which mae observed in connection with such a machine.xperiments with the telephone, a conductor in a strong

eld or with a condenser or arc, seem to afford certain



roof that sounds far above the usual accepted limit of earing would be perceived. A telephone will emit notes welve to thirteen thousand vibrations per second; thenhe inability of the core to follow such rapid alternationsegins to tell. If, however, the magnet and core be

eplaced by a condenser and the terminals connected tohe high-tension secondary of a transformer, higher notemay still be heard. If the current be sent around a finelyaminated core and a small piece of thin sheet iron be helently against the •core, a sound may be still heard withhirteen to fourteen thousand alternations per second,

rovided the current is sufficiently strong. A small coil,owever, tightly packed between the poles of a powerful

magnet, will emit a sound with the above number of lternations, and arcs may be audible with a still higherrequency. The limit of audition is variously estimated. Iir William Thomson's writings it is stated somewhere

hat ten thousand per second, or nearly so, is the limit.Other, but less reliable, sources give it as high as twentyour thousand per second. The above experiments haveonvinced the writer that notes of an incomparably higheumber of vibrations per second would be perceivedrovided they could be produced with sufficient power.

here is no reason why it should not be so. Theondensations and rarefactions of the air wouldecessarily set the diaphragm in a correspondingibration and some sensation would be produced,

whatever—within certain limits—the velocity of ransmission to their nerve centres, though it is probablehat for want of exercise the ear would not be able to



istinguish any such high note. With the eye it is differenthe sense of vision is based upon some resonance effect

s many believe, no amount of increase in the intensity ohe ethereal vibration could extend our range of vision onither side of the visible spectrum.

he limit of audition of an arc depends on its size. Thereater the surface by a given heating effect in the arc,he higher the limit of audition. The highest notes aremitted by the high-tension discharges of an induction con which the arc is, so to speak, all surface. If R be the

esistance of an arc, and C the current, and the linearimensions be n times increased, then

he resistance is — , and with the same current density he current would be v?C; hence the heating effect is n*mes greater, while the surface is only n* times as great

or this reason very large arcs would not emit any hythmical sound even with a very low frequency. It

must be observed, however, that the sound emittedepends to some extent also on the composition of thearbon. If the carbon contain highly refractory material,his, when heated, tends to maintain the temperature of

he arc uniform and the sound is lessened; for this reasonwould seem that an alternating arc requires such

arbons.

With currents of such high frequencies it is possible tobtain noiseless arcs, but the regulation of the lamp is

endered extremely difficult on account of the excessivelmall attractions or re ulsions between conductors



onveying these currents.

An interesting feature of the arc produced by theseapidly alternating currents is its persistency. There arewo causes for it, one of which is always present, the othe

ometimes only. One' is due to the character of theurrent and the other to a property of the machine. Therst cause is the more important one, and is due directly

o the rapidity of the alternations. When an arc is formedy a periodically undulating current, there is aorresponding undulation in the temperature of the

aseous column, and, therefore, a correspondingndulation in the resistance of the arc. But the resistancef the arc varies enormously with the temperature of theaseous column, being practically infinite when the gasetween the electrodes is cold. The persistence of the archerefore, depends on the inability of the column to cool.

t is for this reason impossible to maintain an arc with thurrent alternating only a few times a second. On thether hand, with a practically continuous current, the arc

s easily maintained, the column being constantly kept athigh temperature and low resistance. The higher the

requency the smaller the time interval during which the

rc may cool and increase considerably in resistance. Witfrequency of 10,000 per second or more in an arc of

qual size excessively small variations of temperature aruperimposed upon a steady temperature, like ripples11 the surface of a deep sea. The heating effect isractically continuous and the arc behaves like oneroduced b a continuous current with the exce tion



owever, that it may not be quite as easily started, andhat the electrodes are equally

onsumed; though the writer has observed somerregularities in this respect.

he second cause alluded to, which possibly may not beresent, is due to the tendency of a machine of such high

requency to maintain a practically constant current.When the arc is lengthened, the electromotive force risesn proportion and the arc appears to be more persistent.

uch a machine is eminently adapted to maintain aonstant current, but it is very unfit for a constantotential. As a matter of fact, in certain types of such

machines a nearly constant current is an almostnavoidable result. As the number of poles or polar

rojections is greatly increased, the clearance becomes oreat importance. One has really to do with a greatumber of very small machines. Then there is the

mpedance in the armature, enormously augmented by he high frequency. Then, again, the magnetic leakage isacilitated. If there are three or four hundred alternate

oles, the leakage is so great that it is virtually the sames connecting, in a two-pole machine, the poles by a piecf iron. This disadvantage, it is true, may be obviated

more or less by using a field throughout of the sameolarity, but then one encounters difficulties of a differenature. All these things tend to maintain a constant

urrent in the armature circuit.



n this connection it is interesting to notice that even to-ay engineers are astonished at the performance of aonstant current machine, just as, some years ago, they sed to consider it an extraordinary performance if a

machine was capable of maintaining a constant potential

ifference between the terminals. Yet one result is just aasily secured as the other. It must only be rememberedhat in an inductive apparatus of any kind, if constantotential is required, the inductive relation between therimary or exciting and secondary or armature circuit

must be the closest possible ; whereas, in an apparatus fo

onstant current just the opposite is required.urthermore, the opposition to the current's flow in the

nduced circuit must be as small as possible in the formernd as great as possible in the latter case. But oppositiono a current's flow may be caused in more than one way.t may be caused by ohmic resistance or self-induction.

One may make the induced circuit of a dynamo machiner transformer of such high resistance that whenperating devices of considerably smaller resistance

within very wide limits a

early constant current is maintained. But such high

esistance involves a great loss in power, hence it is notracticable. Not so self-induction. Self-induction does notecessarily mean loss of power. The moral is, use self-

nduction instead of resistance. There is, however, aircumstance which favors the adoption of this plan, andhis is, that a very high self-induction may be obtainedheaply by surrounding a comparatively small length of



wire more or less completely with iron, and, furthermorehe effect may be exalted at will by causing a rapidndulation of the current. To sum up, the requirements

or constant current are: Weak magnetic connectionetween the induced and inducing circuits, greatest

ossible self-induction with the least resistance, greatestracticable rate of change of the current. Constantotential, on the other hand, requires : Closest magneticonnection between the circuits, steady induced current,nd, if possible, no reaction. If the latter conditions coulde .fully satisfied in a constant potential machine, its

utput would surpass many times that of a machinerimarily designed to give constant current.

Unfortunately, the type of machine in which theseonditions may be satisfied is of little practical value,wing to the small electromotive force obtainable and thifficulties iii taking off the current.

With their keen inventor's instinct, the now successfulrc-light men have early recognized the desiderata of aonstant current machine. Their arc light machines have

weak fields, large armatures, with a great length of coppwire and few commutator segments to produce great

ariations in the current's strength and to bring self-nduction into play. Such machines may maintain withinonsiderable limits of variation in the resistance of theircuit a practically constant current. Their output is of ourse correspondingly diminished, and, perhaps with thbject in view not to cut down the output too much, aimple device compensating exceptional variations is



mployed. The undulation of the current is almostssential to the commercial success of an arc-light systemt introduces in the circuit a steadying element taking thlace of a large ohmic resistance, without involving a gre

oss in power, and, what is more important, it allows the

se of simple clutch lamps, which with a current of aertain number of impulses per second, best suitable forach particular lamp, will, if properly attended to,egulate even better than the finest clock-work lamps.his discovery has been made by the writer—severalears too late.

-MK) INVENTIONS OF NIKOLA TESLA.

t has been asserted by competent English electricianshat in a constant-current machine or transformer theegulation is effected by varying the phase of the

econdary current. That this view is erroneous may beasily proved by using, instead of lamps, devices eachossessing self-induction and capacity or self-inductionnd resistance— that is, retarding and acceleratingomponents — in such proportions as to not affect

materially the phase of the secondary current. Any

umber of such devices may be inserted or cut out, still iwill be found that the regulation occurs, a constanturrent being maintained, while the electromotive force aried with the number of the devices. The change of hase of the secondary current is simply a result followin

rom the changes in resistance, and, though secondary

eaction is always of more or less importance, yet the rea



ause of the regulation lies in the existence of theonditions above enumerated. It should be stated,owever, that in the case of a machine the above remarkre to be restricted to the cases in which the machine isndependently excited. If the excitation be effected by

ommu-tating the armature current, then the iixedosition of the brushes makes any shifting of the neutralne of the utmost importance, and it may not be thought

mmodest of the writer to mention that, as far as recordso, he seems to have been the first who has successfully egulated machines by providing a bridge connection

etween a point of the external circuit and theommutator by means of a third brush. The armaturend field being properly proportioned and the brusheslaced in th eir determined positions, a constant currentr constant potential resulted from the shifting of theiameter of commutation by the varying loads.

n connection with machines of such high frequencies, thondenser affords an especially interesting study. It isasy to raise the electromotive force of such a machine tour or five times the value by simply connecting theondenser to the circuit, and the writer has continually

sed the condenser for the the purposes of regulation, asuggested by Blakesley in his book on alternate currentsn which he has treated the most frequently occurringondenser problems with exquisite simplicity andlearness. The high frequency allow r s the use of smallapacities and renders investigation easy. But, although

most of the experiments the result may be foretold, som



henomena observed seem at first curious. Onexperiment performed three or four months ago withuch a machine and a condenser may serve as an il-

ustration. A machine was used giving about 20,000

lternations per second. Two bare wires about twenty eet long and two millimetres in diameter, in closeroximity to each other, were connected to the terminalf the machine at the one end, and to a condenser at thether. A small transformer without an iron core, of cours

was used to bring the reading within range of a Cardew

oltmeter by connecting the voltmeter to the secondaryOn the terminals of the condenser the electromotive forcwas about 120 volts, and from there inch by inch it

radually fell until at the terminals of the machine it wasbout 65 volts. It was virtually as though the condenser

were a generator, and the line and armature circuit

imply a resistance connected to it. The writer looked forcase of resonance, but he was unable to augment the

ffect by varying the capacity very carefully andradually or by changing the speed of the machine. A casf pure resonance he was unable to obtain. When aondenser was connected to the terminals of the machin

—the self-induction of the armature being firstetermined in the maximum and minimum position andhe mean value taken —the capacity which gave theighest electromotive force corresponded most nearly tohat which just counteracted the self-induction with thexisting frequency. If the capacity was increased oriminished, the electromotive force fell as expected.



With frequencies as high as the above mentioned, theondenser effects are of enormous importance. Theondenser becomes a highly efficient apparatus capable oransferring considerable energy.

n an appendix to this book will be found a description ofhe Tesla oscillator, which its inventor believes will amonther great advantages give him the necessary highrequency conditions, while relieving him of thenconveniences that attach to generators of the typeescribed at the beginning of this chapter.

HAPTEK XXX.

ALTERNATE CURRENT ELECTROSTATICNDUCTION APPARATUS.*

ABOUT a year and a half ago while engaged in the studyf alternate currents of short period, it occurred to mehat such currents could be obtained by rotating chargedurfaces in close proximity to conductors. Accordingly Ievised various forms



IG. 208.

f experimental apparatus of which two are illustrated inhe accompanying engravings.

n the apparatus shown in Fig. 208, A is a ring of dry hellacked hard wood provided on its inside with two setf tin-foil coatings, a and J, all the a coatings and all the Ioatings being connected together, respectively, butndependent from each other. These two sets of coatings

re connected to two termi-

. Article by Mr. Tesla in The Electrical Engineer, N. Y.,May 6, 1891.

als, T. For the sake of clearness only a few coatings are

hown. Inside of the ring A, and in close proximity to ithere is arranged to rotate a cylinder B, likewise of dry,hellacked hard wood, and provided with two similar setf coatings, a 1 and J 1 , all the coatings a} being connecteo one ring and all the others, S l , to another marked -f-nd —. These two sets, a 1 and J 1 are charged to a high

otential by a Holtz or Wimshurst machine, and may beonnected to a ar of some ca acit . The inside of rin A



oated with mica in order to increase the induction andlso to allow higher potentials to be used.

When the cylinder B with the charged coatings is rotated

IG. 20«J.

ircuit connected to the terminals T is traversed by lternating currents. Another form of apparatus islustrated in Fig. 209. In this apparatus the two sets of n-foil coatings are glued on a plate of ebonite, and aimilar plate which is rotated, and the coatings of whichre charged as in Fig. 208, is provided.

he out ut of such an a aratus is ver small, but some



f the effects peculiar to alternating currents of shorteriods may l>e observed. The effects, however, cannote compared with those obtainable with an induction coil

which is operated by an alternate current machine of higrequency, some of which were described by me a short

while ago.

HAPTER XXXI.

MASSAGE " WITH CURRENTS OF HIGHREQUENCY. 1

TRUST that the present brief communication will not bnterpreted as an effort on my part to put myself onecord as a "patent medicine" man, for a serious workerannot despise anything more than the misuse and abusef electricity which we have frequent occasion to witness

My remarks are elicited by the lively interest whichrominent medical practitioners evince at every realdvance in electrical investigation. The progress in recenears has been so great that every electrician andlectrical engineer is confident that electricity will becomhe means of accomplishing many things that have been

eretofore, with our existing knowledge, deemedmpossible. ]S r o wonder'then that progressive physicianlso should expect to find in it a powerful tool and help inew curative processes. Since I had the honor to bringefore the American Institute of Electrical Engineersome results in utilizing alternating currents of high

ension, I have received many letters from notedh sicians in uirin as to the h sical effects of such



urrents of high frequency. It may be remembered that hen demonstrated that a body perfectly well insulated iir can be heated by simply connecting it with a source oapidly alternating high potential. The heating in this cass due in all probability to the bombardment of the body

y air, or possibly by some other medium, which ismolecular or atomic in construction, and the presence of which has so far escaped our analysis—for according tomy ideas, the true ether radiation with such frequencies

s even a few millions per second must be very small.his body may be a good conductor or it may be a very

oor conductor of electricity with little change in theesult. The human body is, in such a case, a fineonductor, and if a person insulated in a room, or no

matter where, is brought into contact with such a sourcef

apidly alternating high potential, the skin is heated by ombardment. It is a mere question of the dimensionsnd character of the apparatus to produce any degree ofeating desired.

t has occurred to me whether, with such apparatus

roperly prepared, it would not be possible for a skilledhysician to find in it a means for the effective treatmentf various types of disease. The heating will, of course, beuperficial, that is, on the skin, and would result, whethehe person operated on were in bed or walking around aoom, whether dressed in thick clothes or whether

educed to nakedness. In fact, to ut it broadl , it is



onceivable that a person entirely nude at the North Polemight keep himself comfortably warm in this manner.

Without vouching for all the results, which must, of ourse, be determined by experience and observation, I

an at least warrant the fact that heating would occur byhe use of this method of subjecting the human body toombardment by alternating currents of high potentialnd frequency such as I have long worked with. It is onlyeasonable to expect that some of the novel effects will b

wholly different from those obtainable with the old

amiliar therapeutic methods generally used. Whetherhey would all be beneficial or not remains to be proved.

HAPTEE XXXII.

LECTRIC DISCHARGE IN VACUUM TUBES.*

N The Electrical Engineer of June 10 I have noted theescription of some experiments of Prof. J. J. Thomson,n the " Electric Discharge in Vacuum Tubes," and in you

ssue of June 24 Prof. Elihu Thomson describes anxperiment of the same kind. The fundamental idea in

hese experiments is to set up an electromotive force in acuum tube — preferably devoid of any electrodes— bymeans of electro-magnetic induction, and to excite theube in this manner.

As I view the subject I should, think that to any

xperimenter who had carefully studied the problemonfronting us and who attempted to find a solution of it,



his idea must present itself as naturally as, for instance,he idea of replacing the tinfoil coatings of a Leyden jar barefied gas and exciting luminosity in the condenser thubtained by repeatedly charging and discharging it. The

dea being obvious, whatever merit there is in this line of

nvestigation must depend upon the completeness of thetudy of the subject and the correctness of thebservations. The following lines are not penned with anesire on my part to put myself on record as one who haerformed similar experiments, but with a desire to assither experimenters by pointing out certain peculiarities

f the phenomena observed, which, to all appearances,ave not been noted by Prof. J. J. Thomson, who,owever, seems to have gone about systematically in his

nvestigations, and who has been the first to make hisesults known. These peculiarities noted by me wouldeem to be at variance with the views of Prof. J. J.

homson, and present the phenomena in a different ligh

My investigations in this line occupied me principally uring the winter and spring of the past year. During thime many different experiments were performed, and i

my exchanges of ideas

. Article by Mr. Tesla in The Electrical Engineer. N. Y.,uly 1, 1891.

n this subject with Mr. Alfred S. Brown, of the "WesternUnion Telegraph Company, various different disposition

were suggested which were carried out by me in practicei . 210 ma serve as an exam le of one of the man



orms of apparatus used. This consisted of a large glassube sealed at one end and projecting into an ordinary ncandescent lamp bulb. The primary, usually consistingf a few turns of thick, well-insulated copper sheet was

nserted within the tube, the inside space of the bulb

urnishing the secondary. This form of apparatus wasrrived at after some experimenting, and was usedrincipally with the view of enabling me to place aolished reflecting surface on the inside of the tube, andor this purpose the last turn of the primary was covered

with a thin silver sheet. In all forms of apparatus used

IG. 210.

here was no special difficulty in exciting a luminous circlr cylinder in proximity to the primary.

As to the number of turns I cannot uite understand wh



rof. J. J. Thomson should think that a few turns werequite sufficient," but lest I should impute to him anpinion he may not have, I will add that I have gained th

mpression from the reading of the published abstracts ois lecture. Clearly, the number of turns which gives the

est result in any case, is dependent on the dimensions ohe apparatus, and, were it not for various considerationne turn would always give the best result.

have found that it is preferable to use in thesexperiments an alternate current machine giving a

moderate number of alter-

ations per second to excite the induction coil for charginhe Leyden jar which discharges through the primary —hown dia-grammatically in Fig. 211, —as in such case,efore the disruptive discharge takes place, the tube or

ulb is slightly excited and the formation of the luminousircle is decidedly facilitated.

IG. 211.

ut I have also used a Wimshurst machine in somexperiments.

rof. J. J. Thomson's view of the phenomena underonsideration seems to be that they are wholly due to



lectro-magnetic action. I was, at one time, of the samepinion, but upon carefully investigating the subject I waed to the conviction that they are more of an electrostatature. It must be remembered that in thesexperiments we have to deal with primary currents of an

normous frequency or rate of change and of highotential, and that the secondary conductor consists of aarefied



IG. 212.

as, and that under such conditions electrostatic effectsmust play an important part.

n support of my view I will describe a few experimentsmade by me. To excite luminosity in the tube it is not

bsolutely necessary that the conductor should be closedor instance, if

n ordinary exhausted tube (preferably of large

iameter) be surrounded by a spiral of thick copper wireerving as the primary^ a feebly luminous spiral may benduced in the tube, roughly shown in Fig. 212. In one of hese experiments a curious phenomenon was observed amely, two intensely luminous circles, each of them closo a turn of the primary spiral, were formed inside of theube, and I attributed this phenomenon to the existence



f nodes on the primary. The circles were connected by aaint luminous spiral parallel to the primary and in closeroximity to it. To produce this effect I have found itecessary to strain the jar to the utmost. The turns of thpiral tend to close and form circles, but this, of course,

would be expected, and does not necessarily indicate anlectro-magnetic effect; whereas the fact that a glow cane produced along the primary in the form of an openpiral argues for an electrostatic effect.

IG. 213.

n using Dr. Lodge's recoil circuit, the electrostatic actions likewise apparent. The arrangement is illustrated inig. 213. In his experiment two hollow exhausted tubes

H were slipped over the wires of the recoil circuit and

pon discharging the jar in the usual manner luminosity was excited in the tubes.

Another experiment performed is illustrated in Fig. 214.n this case an ordinary lamp-bulb was surrounded by ne or two turns of thick copper wire P and the luminousircle L excited in the bulb by discharging the jar through



he primary. The lamp-bulb was provided with a tinfoiloating on the side opposite to the primary and each timhe tinfoil coating was connected to the ground or to aarge object the luminosity of the circle was considerably ncreased. This was evidently due to electrostatic action.

n other experiments I have noted that when the primarouches the glass the luminous circle is easier producednd is


more sharply defined ; but I have not noted that,enerally speaking, the circles induced were very sharplyefined, as Prof. J. J. Thomson has observed; on theontrary, in my experiments they were broad and oftenhe whole of the bulb or tube was illuminated ; and in on

ase I have observed an intensely purplish



IG. 214.

low, to which Prof. J. J. Thomson refers. But the circleswere always in close proximity to the primary and wereonsiderably easier produced when the latter was very

lose to the glass, much more so than would be expectedssuming the action to be elec-

IG. 215.

romagnetic and considering the distance; and these factpeak for an electrostatic effect.

urthermore I have observed that there is a molecularombardment in the plane of the luminous circle at rightngles to the glass—supposing the circle to be in the planf the primary

—this bombardment being evident from the rapid heatinf the glass near the primary. Were the bombardment nt right angles to the glass the heating could not be so



apid. If there is a circumferential movement of themolecules constituting the luminous circle, I have thoughhat it might be rendered manifest by placing within theube or bulb, radially to the circle, a thin plate of micaoated with some phosphorescent material and another

uch plate tangentially to the circle. If the moleculeswould move circumferentially, the former plate would beendered more intensely phosphorescent. For want of me I have, however, not been able to perform thexperiment.

Another observation made by me was that when thepecific inductive capacity of the medium between therimary and secondary is increased, the inductive effect ugmented. This is roughly illustrated in Fig. 215. In thisase luminosity was excited in an exhausted tube or bulb and a glass tube T slipped between the primary and th

ulb, when the effect pointed out was noted. Were thection wholly electromagnetic no change could possibly ave been observed.

have likewise noted that when a bulb is surrounded by wire closed upon itself and in the plane of the primary, th

ormation of the luminous circle within the bulb is notrevented. But if instead of the wire a broad strip of tinfo

s glued upon the bulb, the formation of the luminousand was prevented, because then the action wasistributed over a greater surface. The effect of the closenfoil was no doubt of an electrostatic nature, for it

resented a much greater resistance than the closed wir



nd produced therefore a much smaller electromagneticffect.

ome of the experiments of Prof. J. J. Thomson alsowould seem to show some electrostatic action. For

nstance, in the experiment with the bulb enclosed in aell jar, I should think that when the latter is exhaustedo far that the gas enclosed reaches the maximumonductivity, the formation of the circle in the bulb and js prevented because of the space surrounding therimary being highly conducting; when the jar is further

xhausted, the conductivity of the space around therimary diminishes and the circles appear necessarily rst in the bell jar, as the rarefied gas is nearer to therimary. But were the inductive effect very powerful,hey would probably appear in the bulb also. If, howeverhe bell jar were exhausted to the highest degree they

would very likely show themselves in the bulb


nly, that is, supposing the vacuous space to be non-onducting. On the assumption that in these phenomena

lectrostatic actions are concerned we find it easily xplicable why the introduction of mercury or the heatinf the bulb prevents the formation of the luminous bandr shortens the after-glow; and also why in some cases alatinum wire may prevent the excitation of the tube.

Nevertheless some of the experiments of Prof. J. J.

homson would seem to indicate an electromagneticffect. I ma add that in one of m ex eriments in which



vacuum was produced by the Torricellian method, I wanable to produce the luminous band, but this may haveeen due to the weak exciting current employed.

My principal argument is the following : I have

xperimentally proved that if the same discharge which arely sufficient to excite a luminous band in the bulbwhen passed through the primary circuit be so directed o exalt the electrostatic inductive effect—namely, by onverting upwards—an exhausted tube, devoid of lectrodes, may be excited at a distance of several feet.

OME EXPERIMENTS ON THE ELECTRICDISCHARGE IN VACUUM TUBES. BY PROF. J. J.

HOMSON, M.A., F.R.S.

he phenomena of vacuum discharges were, Prof.

homson said, greatly simplified when their path waswholly gaseous, the complication of the dark spaceurrounding the negative electrode, and the stratificationo commonly observed in ordinary vaciium tubes, beingbsent. To produce discharges in



IG. 216.

IG. 2V,

ubes devoid of electrodes was, however, not easy toccomplish, for the only available means of producing anlectromotive force in the discharge circuit was by lectro-magnetic induction. Ordinary methods of roducing variable induction were valueless, and recours

was had to the oscillatory discharge of a

Abstract of a paper read before Physical Society of ondon.

eyden jar, which combines the two essentials of a

urrent whose maximum value is enormous, and whoseapidity of alternation is immensely great. The dischargeircuits, which may take the shape of bulbs, or of tubesent in the form of coils, were placed in close proximity tlass tubes filled with mercury, which formed the path ohe oscillatory discharge. The parts thus corresponded tohe windin s of an induction coil the vacuum tubes bein



he sec-ondary, and the tubes filled with mercury therimary. In such an apparatus the Leyden jar need not b

arge, and neither primary nor secondary need havemany turns, for this would increase the self-induction of he former, and lengthen the discharge path in the latter

ncreasing the self-induction of the primary reduces the. M. F. induced in the secondary, whilst lengthening theecondary does not increase the E. M. F. per unit length.he two or three turns, as shown in Fig. 216, in each,

were found to be quite sufficient, and, on discharging theeyden jar between two highly polished knobs in the

rimary



IG. 218.

IG. 219.

ircuit, a plain uniform band of light was seen to pass

ound the secondary. An exhausted bulb, Fig. 217,ontaining traces of oxygen was plaeed within a primary piral of three turns, and, on passing the jar discharge, aircle of light was seen within the bulb inclose proximity o the primary circuit, accompanied by a purplish glow,

which lasted for a second or more. On heating the bulb,

he duration of the glow was greatly diminished, and itould be instantly extinguished by the presence of anlectro-magnet. Another exhausted bulb, Fig. 218,urrounded by a primary spiral, was contained in a bell-ar, and when the pressure of air in the jar was about thaf the atmosphere, the secondary discharge occurred in

he bulb, as is ordinarily the case. On exhausting the jar,owever, the luminous discharge grew fainter, and a poin

was reached at which no secondary discharge was visibleurther exhaustion of the jar caused the secondary ischarge to appear outside of the bulb. The fact of btaining no luminous discharge, either in the bulb or jar

he author



ould only explain on two suppositions, viz.: that underhe conditions then existing the specific inductive capacitf the gas was very great, or that a discharge could pass

without being luminous. '1 he author had also .observedhat the conductivity of a vacuum tube without electrode

ncreased as the pressure diminished, until a certain poinwas reached, and afterwards diminished again, thushowing that the high resistance of a nearly perfectacuum is in no way due to the presence of the electrode

One peculiarity of the discharges was their local nature,he rings of light being much more sharply denned than

was to be expected. They were also found to be mostasily produced when the chain of molecules in theischarge were all of the same kind. For example, aischarge could be easily sent through a tube many feet

ong, but the introduction of a small pellet of mercury inhe tube stopped the discharge, although the conductivit

f the mercury was much greater than that of theacuum. In some cases he had noticed that a very fine

wire placed within a tube, on the side remote from therimary circuit, would prevent a luminous discharge inhat tube.

ig. 219 shows an exhausted secondary coil of one loopontaining bulbs ; the discharge passed along the inneride of the bulbs, the primary coils being placed within thecondary.

In The Electrical Engineer of August 12, I find some

emarks of Prof. J. J. Thomson, which a eared ori inall



n the London Electrician and which have a bearing uponome experiments described by me in your issue of July .

did not, as Prof. J. J. Thomson seems to believe,

misunderstand his position in regard to the cause of thehenomena considered, but I thought that in hisxperiments, as well as in my own, electrostatic effects

were of great importance. It did not appear, from themeagre description of his experiments, that all possible

recautions had been taken to exclude these effects. I di

ot doubt that luminosity could be excited in a closed tubwhen electrostatic action is completely excluded. In fact,t the outset, I myself looked for a purely electrodynamiffect and believed that I had obtained it. But many xperiments performed at that time proved to me thathe electrostatic effects were generally of far greater

mportance, and admitted of a more satisfactory xplanation of most of the phenomena observed.

n using the term electrostatic I had reference rather tohe nature of the action than to a stationary condition,

which is the usual acceptance of the term. To express

myself more clearly, I will suppose that near a closedxhausted tube be placed a small sphere charged to aery high potential. The sphere would act inductively pon the tube, and by distributing electricity over 1.

Article by Mr. Tesla in The Electrical Engineer, N. Y.,August 26, 1891.

he same would undoubtedl roduce luminosit (if the



otential be sufficiently high), until a permanent conditiowould be reached. Assuming the tube to be perfectly welnsulated, there would be only one instantaneous flashuring the act of distribution. This would be due to thelectrostatic action simply.

ut now, suppose the charged sphere to be moved athort intervals with great speed along the exhaustedube. The tube would now be permanently excited, as th

moving sphere would cause a constant redistribution of lectricity and collisions of the molecules of the rarefied

as. We would still have to deal with an electrostaticifect, and in addition an electrodynamic effect would bebserved. But if it were found that, for instance, the effecroduced depended more on the specific inductiveapacity than on the magnetic permeability of the

medium — which would certainly be the case for speeds

ncomparably lower than that of light—then I believe Iwould be justified in saying that the effect produced wasmore of an electrostatic nature. I do not mean to say,

owever, that any similar condition prevails in the case ohe discharge of a Leyden jar through the primary, but Ihink that such an action would be desirable.

t is in the spirit of the above example that I used theerms " more of an electrostatic nature," and havenvestigated the influence of bodies of high specificnductive capacity, and observed, for instance, themportance of the quality of glass of which the tube is

made. I also endeavored to ascertain the influence of a



medium of high permeability by using oxygen. Itppeared from rough estimation that an oxygen tube

when excited under similar conditions —that is, as far asould be determined —gives more light; but this, of ourse, may be due to many causes.

Without doubting in the least that, with the care andrecautions taken by Prof. J. J. Thomson, the luminosityxcited was due solely to electrodynamic action, I woulday that in many experiments I have observed curiousnstances of the ineffectiveness of the screening, and I

ave also found that the electritica. tion through the air iften of very great importance, and may, in some cases,etermine the excitation of the tube.

n his original communication to the Electrician, Prof. J. Jhomson refers to the fact that the luminosity in a tube

ear a wire through which a Leyden jar was dischargedwas noted by Hittorf. I think that the feeble luminous

ffect referred to has

een noted by many experimenters, but in my xperiments the effects were much more powerful than

hose usually noted. The following is the communication eferred to :—

Mr. Tesla seems to ascribe the effects he observed tolectrostatic action, and I have no doubt, from theescription he gives of his method of conducting his

xperiments, that in them electrostatic action plays aery important part. He seems, however, to have



misunderstood my position with respect to the cause of hese discharges, which is not, as he implies, thatuminosity in tubes without electrodes cannot beroduced by electrostatic action, but that it can also beroduced when this action is excluded. As a matter of fac

is very much easier to get the luminosity when theselectrostatic effects are operative than when they are noAs an illustration of this I may mention that the first

xperiment I tried with the discharge of a Leyden jarroduced luminosity in the tube, but it was not until afteix weeks' continuous experimenting that I was able to

et a discharge in the exhausted tube which I wasatisfied was due to what is ordinarily calledlectrodynamic action. It is advisable to have a clear ideaf what we mean by electrostatic action. If, previous tohe discharge of the jar, the primary coil is raised to a higotential, it will induce over the glass of the tube a

istribution of electricity. When the potential of therimary suddenly falls, this electrification will redistribuself, and may pass through the rarefied gas and produc

uminosity in doing so. "Whilst the discharge of the jar isoing on, it is difficult, and, from a theoretical point of iew, undesirable, to separate the effect into parts, one o

which is called electrostatic, the other electromagnetic ;what we can prove is that in this case the discharge is nouch as would be produced by electromotive forceserived from a potential function. In my experiments thrimary coil was connected to earth, and, as a furtherrecaution, the primary was separated from theischarge tube by a screen of blotting paper, moistened



with dilute sulphuric acid, and connected to earth. Wetlotting paper is a sufficiently good conductor to screen ostationary electrostatic effect, though it is not a good

nough one to stop waves of alternating electromotiventensity. When showing the experiments to the Physica

ociety I could not, of course, keep the tubes covered uput, unless my memory deceives me, I stated therecautions which had b«en taken against thelectrostatic effect. To correct misapprehension I may tate that I did not read a formal paper to the Society, mbject being to exhibit a few of the most typical

xperiments. The account of the experiments in thelectrician was from a reporter's note, and was not

written, or even read, by me. I have now almost finishedwriting out, and hope very shortly to publish, an account

f these and a large number of allied experiments,ncluding some analogous to those mentioned by Mr.

esla on the effect of conductors placed near theischarge tube, which I find, in some cases, to produce aiminution, in others an increase, in the brightness of theischarge, as well as some on the effect of the presence oubstances of large specific inductive capacity. Theseeem to me to admit of a satisfactory explanation, for

which, however, I must refer to my paper."

Note by Prof. J. J. Thomson in the London Electrician,uly 24, 1891.

MISCELLANEOUS INVENTIONS AND WRITINGS.

HAPTER XXXIII.



METHOD OF OBTAINING DIRECT FROMALTERNATING CURRENTS.

HIS method consists in obtaining direct from alternatin

urrents, or in directing the waves of an alternatingurrent so as to produce direct or substantially directurrents by developing or producing in the branches of aircuit including a source of alternating currents, eitherermanently or periodically, and by electric, electro-

magnetic, or magnetic agencies, manifestations of energy

r what may be termed active resistances of oppositelectrical character, whereby the currents or currentwaves of opposite sign will be diverted through differentircuits, those of one sign passing over one branch andhose of opposite sign over the other.

We may consider herein only the case of a circuit dividednto two paths, inasmuch as any further subdivisionnvolves merely an extension of the general principle.electing, then, any circuit through which is flowing anlternating current, Mr. Tesla divides such circuit at anyesired point into two branches or paths. In one of these

aths he inserts some device to create an electromotiveorce counter to the waves or impulses of current of oneign and.a similar device in the other branch whichpposes the waves of opposite sign. Assume, for examplehat these devices are batteries, primary or secondary, oontinuous current dynamo machines. The waves or

mpulses of opposite direction composing the mainurrent have a natural tendenc to divide between the



wo branches; but by reason of the opposite electricalharacter or effect of the two branches, one will offer anasy passage to a current of a certain direction, while thether will offer a relatively high resistance to the passagef the same current. The result of this disposition is, that

he waves of current of one sign will, partly or wholly,ass over one of the paths or branches, while those of thpposite sign pass over the other. There may thus bebtained from an alternating current two or more directurrents without the employment of any commutator

uch as it has been heretofore regarded as necessary tose. The current in either branch may be used in theame way and for the same purposes as any other directurrent — that is, it may be made to charge secondary atteries, energize electro-magnets, or for any othernalogous purpose.

ig. 220 represents a plan of directing the alternatingurrents by means of devices purely electrical inharacter. Figs. 221, 222, 223, 224, 225, and 226 areiagrams illustrative of other ways of carrying out the

nvention.

n Fig. 220, A designates a generator of alternatingurrents, and B B the main or line circuit therefrom. Atny given point in this circuit at or near which it is desireo obtain direct currents, the circuit B is divided into twoaths or branches c D. In each of these branches is place

n electrical enerator which for the resent we will



ssume produces direct or continuous cur-

IG. 220.

ents. The direction of the current thus produced ispposite in one branch to that of the current in the otherranch, or, considering the two branches as forming alosed circuit, the generators E F are connected up ineries therein, one generator in each part or half of theircuit. The electromotive force of the current sources End F may be equal to or higher or lower than the

lectromotive forces in the branches c D, or between theoints x and Y of the circuit B B. If equal, it is evident thaurrent waves of one sign will be opposed in one branchnd assisted in the other to such an extent that all the

waves of one sign will pass over one branch and those of pposite sign over the other. If, on the other hand, the

lectromotive force of the sources E F be lower than thatetween x and Y, the currents in both branches will belternating, but the waves of one sign will preponderate.

One of the generators or sources of current E or F may bispensed with ; but it is preferable to employ both, if

hey offer an appreciable resistance, as the two brandieswill be thereby better balanced. The translating or other

evices to be acted upon by the current are designated bhe letters G, and they are inserted in the branches c D iny desired manner ; but in order to better preserve anven balance between the branches due regard should, o

ourse, be had to the number and character of theevices.



igs. 221, 222, 223, and 224 illustrate what may termedelectro-magnetic" devices for accomplishing a similaresult—that is to say, instead of producing directly by aenerator an electromotive force in each branch of the

ircuit, Mr. Tesla establishes a field or fields of force andeads the branches through the same in such manner than active opposition of opposite effect or direction will beeveloped therein by the passage, or tendency to pass, ohe alternations of current. In Fig. 221, for example, A is

IG. 221.

he generator of alternating currents, B B the line circuitnd c D the branches over which the alternating current

re directed. In each branch is included the secondary ofransformer or induction coil, which, since they orrespond in their functions to the batteries of therevious figure, are designated by the letters E F. Therimaries H H' of the induction coils or transformers areonnected either in parallel or series with a source of

irect or continuous currents i, and the number of



onvolutions is so calculated for the strength of theurrent from i that the cores J j' will be saturated. Theonnections are such that the conditions in the tworansformers are of opposite character—that is to say, thrrangement is such that a current wave or impulse

orresponding in direction with that of the direct currentn one primary, as H, is of opposite direction to that in thther primary H'. It thus results that while one secondarffers a resistance or op-


osition to the passage through it of a wave of one sign,he other secondary similarly opposes a wave of oppositeign. In consequence, the waves of one sign will, to areater or less extent, pass by way of one branch, whilehose of opposite sign in like manner pass over the other

ranch.

n lieu of saturating the primaries by a source of ontinuous current, we may include the primaries in theranches c D, respectively, and periodically short-circuity any suitable mechanical devices—such as an ordinary

evolving commutator—their secondaries. It will benderstood, of course, that the rotation and action of theommutator must be in synchronism or in proper accord

with the periods of the alternations in order to secure thesired results. Such a disposition is represented



IG. 222.

iagrammatically in Fig. 222. Corresponding to therevious figures, A is the generator of alternatingurrents, B B the line, and c D the two branches for theirect currents. In branch c are included two primary oils E E', and in branch D are two similar primaries F F'he corresponding secondaries for these coils and which

re on the same subdivided cores j or j', are in circuits therminals of which connect to opposite segments K K', an

i/, respectively, of a commutator. Brushes b I bear upohe commutator and alternately short-circuit the plates nd K', and L and L', through a connection c. It is obvioushat either the magnets and commutator, or the brushes

may revolve. The operation will be understood from aonsideration of the effects of closing or short-circuitinghe secondaries. For example, if at the instant when aiven wave of current passes, one

et of secondaries be short-circuited, nearly all the

urrent flows through the corresponding primaries; buthe secondaries of the other branch bein o en-circuited



he self-induction in the primaries is highest, and hencettle or no current will pass through that branch. If, as thurrent alternates, the secondaries of the two branchesre alternately short-circuited, the result will be that theurrents of one sign pass over one branch and those of th

pposite sign over the other. The disadvantages of thisrrangement, which would seem to result from themployment of sliding contacts, are in reality very slightnasmuch as the electromotive force of the secondaries

may be made exceedingly low, so that sparking at therushes is avoided.

ig. 223 is a diagram, partly in section, of another plan ofarrying out the invention. The circuit B in this case isivided, as before, and each branch includes the coils of oth the fields

IG. 223.

nd revolving armatures of two induction devices. Thermatures o P are preferably mounted on the same shaf

nd are adjusted relatively to one another in such mannehat when the self-induction in one branch, as c. is



maximum, in the other branch D it is minimum. Thermatures are rotated in synchronism with thelternations from the source A. The winding or position ohe armature coils is such that a current in a givenirection passed through both armatures would establish

n one, poles similar to those in the adjacent poles of theeld, and in the other, poles unlike the adjacent fieldoles, as indicated by n n s s in the diagram. If the likeoles are presented, as shown in circuit D, the condition ihat of a closed secondary upon a primary, or the positiof least inductive resistance; hence a given alternation of

urrent will pass mainly through D. A half revolution of he armatures produces an opposite effect and theucceeding


urrent impulse passes through c. Using this figure as anlustration, it is evident that the fields N M may beermanent magnets or independently excited and thermatures o P driven, as in the present case, so as toroduce alternate currents, which will set up alternately

mpulses of opposite direction in the two branches D c,

which in such case would include the armature circuitsnd translating devices only.

n Fig. 224 a plan alternative with that shown in Fig. 222s illustrated. In the previous case illustrated, each brancand D contained one or more primary coils, the

econdaries of which were periodically short circuited innchronism with the alternations of current from the



main source A, and' for this purpose a commutator wasmployed. The latter may, however, be dispensed withnd an armature with a closed coil substituted.

Referring to Fig. 224 in one of the branches, as c, are two

oils

IG. 224.

M', wound on laminated cores, and in the other branchesD are similar coils N'. A subdivided or laminated armatur

7 , carrying a closed coil R', is rotatably supportedetween the coils M' N', as shown. In the position shownhat is, with the coil K' parallel with the convolutions of he primaries N' M' —practically the whole current will

ass through branch D, because the self-induction in coilM' M' is maximum. If, therefore, the armature and coil botated at a proper speed relatively to the periods orlternations of the source A, the same results arebtained as in the case of Fig. 222.

ig. 225 is an instance of what may be called, in



istinction to the others, a " magnetic " means of securinhe result, v and w are two strong permanent magnetsrovided with armatures v' w', respectively. Thermatures are made of thin laminae of soft iron or steel,nd the amount of magnetic metal which they

ontain is so calculated that they will be fully or nearly aturated by the magnets. Around the armatures are coi F, contained, respectively, in the circuits c and D. Theonnections and electrical conditions in this case areimilar to those in Fig. 221, except that the current sourc

f i, Fig. 221, is dispensed with and the saturation of theore of coils E F obtained froth the permanent magnets.

he previous illustrations have all shown the tworanches or paths containing the translating or inductionevices as in derivation one to the other; but this is not

lways necessary. For example, in Fig. 226, A is anlternating-current generator; B B, the line wires orircuit. At any given point in the circuit let us form twoaths, as D D', and at another point two paths, as c c' '.ither pair or group of paths is similar to the previous di



IG. 225.

ositions with the electrical source or induction device inne branch only, while the two groups taken togetherorm the obvious equivalent of the cases in which an

nduction device or generator is included in bothranches. In one of the paths, as D, are included theevices to be operated by the current. In the otherranch, as D', is an induction device that opposes theurrent impulses of one direction and directs themhrough the branch D. So, also, in branch c are translatin

evices o, and in branch c' an induction device or itsquivalent that diverts through c impulses of oppositeirection to those diverted by the device in branch D'. Thiagram shows a special form of induction device for thisurpose, .r / are the cores, formed with pole-pieces, upo

which are wound the coils M N. Between these pole-ieces are mounted at right angles to one another the

magnetic armatures o P, preferably mounted on the samhaft and


esigned to be rotated in synchronism with thelternations of current. When one of the armatures is inne with the poles or in the position occupied by armatur, the magnetic circuit of the induction device isractically closed; hence there will be the greatest

pposition to the passage of a current through coils N N.he alternation will therefore ass b wa of branch D. A



he same time, the magnetic circuit of the other inductionevice being broken by the position of the armature o,here will be less opposition to the current in coils M,

which will shunt the current from branch c. A reversal ofhe current being attended by a shifting of the armature

he opposite effect is produced.

Other modifications of these methods are possible, buteed not be pointed out. In all these plans, it will bebserved, there

IG

s developed in one or all of these branches of a circuitrom a source of alternating currents, an active (asistinguished from a dead) resistance or opposition to th

urrents of one sign, for the purpose of diverting theurrents of that sign through the other or another path,



ut permitting the currents of opposite sign to passwithout substantial opposition.

Whether the division of the currents or waves of currentf opposite sign be effected with absolute precision or no

s immaterial, since it will be sufficient if the waves arenly partially diverted or directed, for in such case thereponderating influence in each branch of the circuit of he waves of one sign secures the same practical results

many if not all respects as though the current were direcnd continuous.

An alternating and a direct current have been combinedo that the waves of one direction or sign were partially o

wholly overcome by the direct current; but by this plannly one set of alternations are utilized, whereas by theystem just described the entire current is rendered

vailable. By obvious applications of this discovery Mr.esla is enabled to produce a self-exciting alternatingynamo, or to operate direct current meters onlternating-current circuits or to run various devices—uch as arc lamps —by direct currents in the same circui

with incandescent lamps or other devices operated by

lternating currents.

t will be observed that if an intermittent counter orpposing force be developed in the branches of the circuind of higher electromotive force than that of theenerator, an alternating current will result in each

ranch, with the waves of one sign preponderating, whileconstantl or uniforml actin o osition in the



ranches of higher electromotive force than the generatowould produce a pulsating current, which conditionswould be, under some circumstances, the equivalent of hose described.

HAPTER XXXIY.

ONDENSERS WITH PLATES IN OIL.

N experimenting with currents of high frequency andigh potential, Mr. Tesla has found that insulating

materials such as glass, mica, and in general those bodieswhich possess the highest specific inductive capacity , arenferior as insulators in such devices when currents of thind described are employed compared with thoseossessing high insulating power, together with a smallerpecific inductive capacity ; and he has also found that it

ery desirable to exclude all gaseous matter from thepparatus, or any ac-



IG. 227.

IG. 228.

ess of the same to the electrified surfaces, in order torevent heating by molecular bombardment and the lossr injury consequent thereon. He has therefore devised a

method to accomplish these results and produce highly fficient and reliable condensers, by using oil as the

ielectric 1 . The plan admits of a particular con-

Mr. Tesla's experiments, as the careful reader of hishree lectures will perceive, have revealed a very mportant fact which is taken advantage of in thisnvention. Namely, he has shown that in a condenser a

onsiderable amount of energy may be wasted, and theondenser may break down merely because gaseousmatter is present between the surfaces. A number of

xperiments are described in the lectures, which bringut this fact forcibly and serve as a guide in the operationf high tension apparatus. But besides bearing upon this

oint, these experiments also throw a light uponnvestigations of a purely scientific nature and explainow the lack of harmony among the observations of arious investigators. Mr. Tesla shows that in a fluid sucs oil the losses are very small as compared with thosencurred in a gas.



truction of condenser, in whicli the distance between thelates is adjustable, and of which he takes advantage.

n the accompanying illustrations, Fig. 227 is a section ofondenser constructed in accordance with this principlend having stationary plates; and Fig. 228 is a similariew of a condenser with adjustable plates.

Any suitable box or receptacle A may be used to containhe plates or armatures. These latter are designated by

nd c and are connected, respectively, to terminals i> an, which pass out through the sides of the case. The platerdinarily are separated by strips of porous insulating

material F, which are used merely for the purpose of maintaining them in position. The space within the can is

lled with oil G. Such a condenser will prove highly

fficient and will not become heated or permanently njured.

n many cases it is desirable to vary or adjust the capacitf a condenser, and this is provided for by securing thelates to adjustable supports —as, for example, to rods n

—passing through stuffing boxes K in the sides of case A nd furnished with nuts L, the ends of the rods beinghreaded for engagement with the nuts.

t is well known that oils possess insulating properties,nd it it has been a common practice to interpose a body

f oil between two conductors for purposes of insulation ut Mr. Tesla believes he has discovered eculiar



roperties in oils which render them very valuable in thiarticular form of device.

HAPTER XXXY.

LECTROLYTIC REGISTERING METER.

AN ingenious form of electrolytic meter attributable toMr. Tesla is one in which a conductor is immersed in aolution, so arranged that metal may be deposited fromhe solution or taken away in such a manner that the

lectrical resistance of the conductor is varied in a definitroportion to the strength of the current the energy of which is to be computed, whereby this variation inesistance serves as a measure of the energy and also

may actuate registering mechanism, whenever theesistance rises above or falls below certain limits.

n carrying out this idea Mr. Tesla employs an electrolytell, through which extend two conductors parallel and inlose proximity to each other. These conductors heonnects in series through a resistance, but in such

manner that there is an equal difference of potential

etween them throughout their entire extent. The freends or terminals of the conductors'are connected eithern series in the circuit supplying the current to the lampsr other devices, or in parallel to a resistance in the circund in series with the current consuming devices. Underuch circumstances a current passing through the

onductors establishes a difference of potential betweenhem which is proportional to the strength of the current



n consequence of which there is a leakage of current fromne conductor to the other across the solution. Thetrength of this leakage current is proportional to theifference of potential, and, therefore, in proportion to thtrength of the current passing through the conductors.

Moreover, as there is a constant difference of potentialetween the two conductors throughout the entire extenhat is exposed to the solution, the current density hrough such solution is the same at all correspondingoints, and hence the deposit is uniform along the wholef one of the conductors, while the metal is taken away

niformly from the other. The resistance of one conductos by this means diminished, while that of the other is

LECTROLYTIC REGISTERING METER.

21

ncreased, both in proportion to the strength of theurrent passing through the conductors. From sucliariation in the resistance of either or both of theonductors forming the positive and negative electrodesf the cell, the current energy expended may be readily

omputed. Figs. 229 and 230 illustrate two forms of suchmeter.

n Fig. 229 G designates a direct-current generator. L Lre the conductors of the circuit extending therefrom. A tube of glass, the ends of which are scaled, as by means

f insulating plugs or caps B B. c c' are two conductorsxtending through the tube A, their ends passing out



hrough the plugs B to

IG. 229.

erminals thereon. These conductors may be corrugatedr formed in other proper ways to offer the desiredlectrical resistance. K is a resistance connected in series

witli the two conductors c c', which by their free terminare connected up in circuit with one of the conductors L.

he method of using this device and computing by meanhereof the energy of the current will be readily nderstood. First, the resistances of the two conductors , respectively, are accurately measured and noted. Theknown current is passed through the instrument for a-

iven time, and by a second measurement the increase

nd diminution of the resistances of the two conductorsre respectively taken. From these data the constant is




btained — that is to say, for example, the increase of esistance of one conductor or the diminution of theesistance of the other per lamp hour. These two

measurements evidently serve as a check, since the gainf one conductor should equal the loss of the other. A urther check is afforded by measuring both wires ineries with the resistance, in which case the resistance ofhe whole should remain constant.

n Fig. 230 the conductors c c' are connected in parallel,he current device at x passing in one branch iirst througresistance R' and then through conductor c, while on thther branch it passes iirst through conductor c', and thehrough resistance



IG. 280.

R". The resistances R' R" are equal, as also are the

esistances of the conductors c c'. It is, moreover,referable that the respective resistances of theonductors c c' should be a known and convenient fractiof the coils or resistances R' R". It will be observed that ihe arrangement shown in Fig. 230 there is a constantotential difference between the two conductors c c'

hroughout their entire length.

t will be seen that in both cases illustrated, theroportionality of the increase or decrease of resistance he current strength will always be preserved, for whatne conductor gains the other loses, and the resistances

he conductors c c' being small as

ompared with the resistances in series with them. It wile understood that after.each measurement oregistration of a given variation of resistance in one oroth conductors, the direction of the current should -be

hanged or the instrument reversed, so that the depositwill be taken from the conductor which has gained and

dded to that which has lost. This principle is capable of many modifications. For instance, since there is a section

f the circuit — to wit, the conductor c or c'— that variesn resistance in proportion to the current strength,.such

ariation maybe utilized, as is done in many analogousases to effect the o eration of various automatic device



uch as registers. It is better, however, for the sake of implicity to compute the energy by measurements of esistance.

he chief advantages of this arrangement are, first, that

s possible to read off directly the amount of the energy xpended by means of a properly constructed ohm-metend without resorting to weighing the deposit; secondly is not necessary to employ shunts, for the whole of theurrent to be measured may be passed through thenstrument; third, the accuracy of the instrument and

orrectness of the indications are but slightly affected byhanges in temperature. It is also said that such metersave the merit of superior economy and compactness, as

well as of cheapness in construction. Electrolytic meterseem to need every auxiliary advantage to make themermanently popular and successful, no matter how muc

ngenuity may be shown in their design.

HAPTEK XXXVI.

HERMO-MAGNETIC MOTORS AND PYRO-MAGNETIC GENERATORS.

No electrical inventor of the present day dealing with theroblems of light and power considers that he has doneimself or his opportunities justice until he has attackedhe subject of thermo-magiietism. As far back as theeginning of the seventeenth century it was shown by D

William Gilbert, the father of modern electricity, that aoadstone or iron bar when heated to redness loses its



magnetism ; and since that time the influence of heat onhe magnetic metals has been investigated frequently,hough not with any material or practical result.

or a man of Mr. Tesla's inventive ability, the problems i

his field have naturally had no small fascination, andhough he has but glanced at them, it is to be hoped hemay find time to pursue the study deeper and further.

or such as he, the investigation must undoubtedly bearruit. Meanwhile he has worked out one or two operativeevices worthy of note. 1 He obtains mechanical power b

reciprocating action resulting from the joint operationsf heat, magnetism, and a spring or weight or other force— that is to say he subjects a body magnetized by nduction or otherwise to the action of heat until the

magnetism is sufficiently neutralized to allow a weight orpring to give motion to the body and lessen the action o

he heat, so that the magnetism may be sufficiently estored to move the

It will, of course, be inferred from the nature of theseevices that the vibration obtained in this manner is verlow owing to the inability of the iron to follow rapid

hanges in temperature. In an interview with Mr. Teslan this subject, the compiler learned of an experiment

which will interest students. A simple horseshoe magnets taken and a piece of sheet iron bent in the form of an Ls brought in contact with one of the poles and placed inuch a position that it is kept in the attraction of the

pposite pole delicately suspended. A spirit lamp is place



nder the sheet iron piece and when the iron is heated tocertain temperature it is easily set in vibrationscillating as rapidly as 400 to 500 times a minute. Thexperiment is very easily performed and is interestingrincipally on account of the very rapid rate of vibration

lIK II MO-MAGNETISM AND PYRO MAGNKTIHM.

-25

ody in tlie opposite direction, and again subject the sam

o the demagnetizing power of the heat.Use is made of either an electro-magnet or a permanentmagnet, and the heat is directed against a body that ismagnetized by induction, rather than directly against a

ermanent magnet, thereby avoiding the loss of

magnetism that might result in the permanent magnet bhe action of heat. Mr. Tesla also provides for lesseninghe volume of the heat or for intercepting the same durinhat portion of the reciprocation in which the coolingction takes place.

n the diagrams are shown some of the numerousrrangements that may be made use of in carrying outhis idea. In all of these figures the magnet-poles are

marked N s, the armature A, the Bunsen burner or otheource of heat H, the axis of mo-



IG. 232.

IG. 231.

IG. 233.

on M, and the spring or the equivalent thereof—namelyweight— is marked w.

n Fig. 281 the permanent magnet N is connected with a

rame, F, supporting the axis M, from which the arm Pangs, and at the lower end of which the armature A isupported. The stops 2 and :-J limit the extent of motionnd the spring w tends to draw the armature A away rom the magnet N. It will now be understood that the

magnetism of >• is sufficient to overcome the spring w

nd draw the armature A toward the magnet N. The heacting upon the armature A neutralizes its induced

magnetism sufficiently for the spring w to draw thermature A away from the magnet M and also from theeat at ir. The armature now cools, and the attraction of he magnet N overcomes the spring w and draws the

rmature A back again above the burm-i-




, so that the same is again heated and the operations arepeated. The reciprocating movements thus obtainedre employed as a source of mechanical power in any

esired manner. Usually a connecting-rod to a crank upofly-wheel shaft would be made use of, as indicated inig. 240.

ig. 232 represents the same parts as before described;ut an

ie. 234.



IG. 235.

lectro-magnet is illustrated in place of a permanentmagnet. The operations, however, are the same.

n Fig. 233 are shown the same parts as in Figs. 231 and32, but they are differently arranged. The armature A,

nstead of swinging, is stationary and held by arm p', and

he core N s of the electro-magnet is made to swing withhe helix Q, the core being suspended by the arm p fromhe pivot M. A shield, R, is connected with the magnet-ore and swings with it, so that after the heat hasemagnetized the armature A to such an extent that thepring w draws the core N s away from the armature A,

he shield K comes between the flame H and armature Ahereby intercepting the action of the heat and allowinghe armature to cool, so that the magnetism, againreponderating, causes the movement of the core N soward the armature A and the removal of the shield R rom above the flame, so that the heat again acts to lesse

r neutralize the magnetism. A rotary or other movemenma be obtained from this reci rocation.



ig. 234 corresponds in every respect with Fig. 233,xcept that a permanent horseshoe-magnet, N s isepresented as taking the place of the electro-magnet inig. 233.

n Fig. 235 is shown a helix, Q, with an armature adapteo swing toward or from the helix. In this case there maye a soft-

HERMO-MAGNETI8M AND PJEO-MAGNETI8M.

-21

ron core in the helix, or the armature may assume theorm of a solenoid core, there being no permanent core

within the helix.

ig. 23tf is an end view, and Fig. 237 a plan view,lustrating the method as applied to a swinging armature

A, and a stationary permanent magnet, N s. In thisnstance Mr. Tesla applies the heat to an auxiliary rmature or keeper, T, which is adjacent to and

referably in direct contact with the magnet. Thisrmature T, in the form of a plate of sheet-iron, extendscross from one pole to the other and is of sufficientection to practically form a keeper for the magnet, sohat when the armature T is cool nearly all the lines of orce pass over the same and very little free magnetism

xhibited. Then the armature A, which swings freely onhe pivots M in front of the poles N s, is very little



ttracted and the spring w pulls the same way from theoles into the position indicated in the diagram. The heat

s directed upon the iron plate T at some distance fromhe magnet, so as to allow the magnet to keepomparatively cool. This heat is applied beneath the plat

y means of the burners H, and there is a connection frohe armature A or its pivot to the gas-cock 6, or otherevice for regulating the heat. The heat acting upon the

middle portion of the plate T, the magnetic conductivity he heated portion is diminished or destroyed, and a greumber of the lines of force are deflected over the

rmature A, which is now

IG. 287.

IG. 238.

IG.



owerfully attracted and drawn into line, or nearly so,with the poles N s. In so doing the cock 6 is nearly closed

nd the plate T cools, the lines of force are again deflectever the same, the attraction exerted upon the armature

A is diminished, and the spring w pulls the same away

rom the magnet into the position shown by full lines, andhe operations are repeated.

he ar-

NVENTIONS OF NIKOLA TE8LA

angement shown in Fig. 236 has the advantages that thmagnet and armature are kept cool and the strength of he permanent magnet is better preserved, as the

magnetic circuit is constantly closed.

n the plan view, Fig. 238, is shown a permanent magnetnd keeper plate, T, similar to those in Figs. 236 and 237with the burners H for the gas beneath the same; but th

rmature is pivoted at one end to one pole of the magnetnd the other end swings toward and from the other polef the magnet. The spring w acts against a lever arm tha

rojects from the armature, and the supply of heat has te partly cut oif by a connection to the swingingrmature, so as to lessen the heat acting upon the keepelate when the armature A has been attracted.



.;N

IG. 240.

IG. 241.

ig. 239 is similar to Fig. 238, except that the keeper T iot made use of and the armature itself swings into andut of the range of the intense action of the heat from theurner H. Fig. 240 is a diagram similar to Fig. 231, excephat in place of using a spring and stops, the armature is

hown as connected by a link, to the crank of a fly-wheelo that the fl -wheel will be revolved as ra idl as the



rmature can be heated and cooled to the necessary xtent. A spring may be used in addition, as in Fig. 231. Iig. 241 the armatures A A are connected by a link, so

hat one will be heating while the other is cooling, and thettraction exerted to move the cooled armature is availe

f to draw away the heated armature instead of using apring.

Mr. Tesla has also devoted his attention to theevelopment of a pyromagnetic generator of electricity 1ased upon the following laws: First, that electricity or

lectrical energy is developed in any conducting body by ubjecting such body to a varying magnetic influence ; anecond, that the magnetic properties of iron or other

magnetic substance may be partially or entirely estroyed-or caused to disappear by raising it to a certaemperature, but restored and caused to reappear by

gain lowering its temperature to a certain degree. Thesaws may be applied in the production of electricalurrents in many ways, the principle of which is in allases the same, viz., to subject a conductor to a varying

magnetic influence, producing such variations by thepplication of heat, or, more strictly speaking, by the

pplication or action of a varying temperature upon theource of the magnetism. This principle of operation maye illustrated by a simple experiment: Place end to end,nd preferably in actual contact, a permanently

magnetized steel bar and a strip or bar of soft iron.Around the end of the iron bar or plate wind a coil of nsulated wire. Then apply to the iron between the coil



nd the steel bar a flame or other source of heat whichwill be capable of raising that portion of the iron to an

range red, or a temperature of about 600° centigrade.When this condition is reached, the iron somewhatuddenly loses its magnetic properties, if it be very thin,

nd the same effect is produced as though the iron hadeen moved away from the magnet or the heated sectionad been removed. This change of position, however, isccompanied by a shifting of the magnetic lines, or, inther words, by a variation in the magnetic influence to

which the coil is exposed, and a current in the coil is the

esult. Then remove the flame or in any other way reduche temperature of the iron. The lowering of itsemperature is accompanied by a return of its magneticroperties, and another change of magnetic conditionsccurs, accompanied by a current in an opposite directionn the coil. The same operation may be

The chief point to be noted is that Mr. Tesla attackedhis problem in a way which was, from the standpoint of heory, and that of an engineer, far better than that from

which some earlier trials in this direction started. Thenlargement of these ideas will be found in Mr. Tesla's

work on the pyromagnetic generator, treated in thishapter. The chief effort of the inventor was to economizhe heat, which was accomplished by inclosing the iron insource of heat well insulated, and by cooling the iron by

means of steam, utilizing the steam over again. Theonstruction also permits of more rapid magnetic changeer unit of time, meaning larger output.




epeated indefinitely, the effect upon the coil being similao that which would follow from moving the magnetizedar to and from the end of the iron bar or plate.

he device illustrated below is a means of obtaining thisesult, the features of novelty in the invention being, firshe employment of an artificial cooling device, and,econd, inclosing the source of heat and that portion of th

magnetic circuit exposed to the heat and artificially

ooling the heated part.

hese improvements are applicable generally to theenerators constructed on the plan above described—thas to say, we may use an artificial cooling device inonjunction with a variable or varied or uniform source o

eat.

ig. 242 is a central vertical longitudinal section of theom-



IG. 242.

IG. 243.

lete apparatus and Fig. 243 is a cross-section of themagnetic armature-core of the generator.

et A represent a magnetized core or permanent magnehe poles of which are bridged by an armature-coreomposed of a casing or shell B inclosing a number of ollow iron tubes c. Around this core are wound theonductors E E', to form the coils in which the currentsre developed. In the circuits of these coils are current-onsuming devices, as F F'.

is a furnace or closed fire-box, through which theentral portion of the core B extends. Above the fire is aoiler K, containing water. The flue L from the fire-box

may extend up through the boiler.

G is a water-supply pipe, and H is the steam-exhaustipe, which communicates with all the tubes c in thermature B, so that steam escaping from the boiler willass through the tubes.

n tlie steam-exhaust pipe H is a valve v, to which isonnected the lever i, by the movement of which the

alve is opened or closed. In such a case as this the heat he fire ma he utilized for other ur oses after as much



f it as may be needed has been applied to heating theore u. There are special advantages in the employmentf a cooling device, in that the metal of the core B is not suickly oxidized. Moreover, the difference between theemperature of the applied heat and of the steam, air, or

whatever gas or fluid be applied as the cooling medium,may be increased or decreased at will, whereby theapidity of the magnetic changes or fluctuations may beegulated.

HAPTEK XXXVII. ANTI-SPAKKING DYNAMO

RUSH AND COMMUTATOR.

N direct current dynamos of great electromotive force—uch, for instance, as those used for arc lighting—whenne commutator bar or plate comes out of contact withhe collecting-brush a spark is apt to appear on the

ommutator. This spark may be due to the break of theomplete circuit, or to a shunt of low resistance formed bhe brush between two or more commutator-bars. In thrst case the spark is more apparent, as there is at the

moment when the circuit is broken a discharge of themagnets through the field helices, producing a great spar

r Hash which causes an unsteady current, rapid wear ofhe commutator bars and brushes, and waste of power.he sparking may be reduced by various devices, such aroviding a path for the current at the moment when theommutator segment or bar leaves the brush, by short-ircuiting the field-helices, by increasing the number of

he commutator-bars, or by other similar means ; but al



hese devices are expensive or not fully available, andeldom attain the object desired.

o prevent this sparking in a simple manner, Mr. Teslaome years ago employed with the commutator-bars and

ntervening insulating material, mica, asbestos paper orther insulating and incombustible material, arranged toear on the surface of the commutator, near to andehind the brush.

n the drawings, Fig. 244 is a section of a commutator

with an asbestos insulating device; and Fig. 245 is aimilar view, representing two plates of mica upon theack of the brush.

n Fig. 244, c represents the commutator and interveninnsulating material; B B, the brushes, d d are sheets of

sbestos paper or other suitable non-conducting materia/ are springs, the pressure of which may be adjusted bymeans of the screws

V 9-

n Fig. 245 a simple arrangement is shown with twolates of mica or other material. It will be seen thatwhenever one com-

mitator segment passes out of contact with the brush, thormation of the arc will be prevented by the intervening

nsulating material coming in contact with the insulatingmaterial on the brush.



Asbestos paper or cloth impregnated with zinc-oxide,magnesia, zirconia, or other suitable material, may be

sed, as the

IG. 244.

IG. 245.

aper and cloth are soft, and serve at the same time towipe and polish the commutator ; but mica or any other

uitable material can be employed, provided the materiae an insulator or a bad conductor of electricity.

A few years later Mr. Tesla turned his attention again tohe same subject, as, perhaps, was very natural in view ohe fact that the commutator had always been prominen

n his thoughts, and that so much of his work was evenimed at dispensing with it entirely as an objectionablend unnecessary part of dynamos and motors. In theseater efforts to remedy commutator troubles, Mr. Teslaonstructs a commutator and the collectors therefor inwo parts mutually adapted to one another, and, so far a

he essential features are concerned, alike in mechanicaltructure. Selectin as an illustration a commutator of tw



egments adapted for use with an armature the coils oroil of which have but two free ends, connectedespectively to the segments, the bearing-surface is theace of a disc, and is formed of two metallic quadrantegments and two insulating segments of the same

imensions, and the face of the disc is smoothed off, sohat the metal and insulating segments are flush. The pawhich takes the place of the usual brushes, or the "ollector," is a disc of the same character as theommutator and has a surface similarly formed with twonsulating and two metallic segments. These two parts ar

mounted with their faces in contact and in such mannerhat the rotation of the armature causes the commutatoro turn upon the collector, whereby the currents inducedn the


oils are taken off by the collector segments and thenceonveyed off by suitable conductors leading from theollector segments. This is the general plan of theonstruction adopted. Aside from certain adjuncts, theature and functions of which are set forth later, this

means of commutation will be seen to possess many mportant advantages. In the first place the short-ircuiting and the breaking of the armature coil connecteo the commutator-segments occur at the same instant,nd from the nature of the construction this will be done

with the greatest precision ; secondly, the duration of

oth the break and of the short circuit will be reduced to



minimum. The first results in a reduction which amountsractically to a suppression of the spark, since the breaknd the short circuit produce opposite effects in thermature-coil. The second has the effect of diminishinghe destructive effect of a spark, since this would be in a

measure proportional to the duration of the spark; whileessening the duration of the short circuit obviously ncreases the efficiency of the machine.



IG. 246.

IG. 247.

he mechanical advantages will be better understood byeferring to the accompanying diagrams, in which Fig. 24s a central longitudinal section of the end of a shaft withhe improved commutator carried thereon. Fig. 247 is aiew of the inner or bearing face of the collector. Fig. 248

s an end view from the armature side of a modified formf commutator. Figs.

49 and 250 are views of details of Fig. 248. Fig. 251 is aongitudinal central section of another modification, andig. 252 is a sectional view of the same. A is the end of thrmature-shaft of a dynamo-electric machine or motor. s a sleeve of insulating material around the shaft, securen place by a screw a'.

IG. 248 FIG. 249. FIG. 250.



he commutator proper is in the form of a disc which ismade up of four segments n D' G G', similar to thosehown in Fig. 248. Two of these segments, as D D', are of

metal and are in electrical connection with the ends of thoils on the armature. The other two segments are of

nsulating material. The segments are held in place by aand, B, of insulating material. The disc is held in place briction or by screws, y' g', Fig. 248, which secure the disrmly to the sleeve A'.

he collector is made in the same form as the

ommutator. It is composed of the two metallic segment E' and the two insulating segments r F', bound togethey a band, c. The metallic segments E E' are of the samer practically the same width or extent as the insulatingegments or spaces of the commutator. The collector isecured to a sleeve, B', by screws g g, and the sleeve is

rranged to turn freely on the shaft A. The end of theleeve B' is closed by a plate, /, upon which presses aivot-pointed screw, /i, adjustable in a spring, H, whichcts to maintain the collector in close contact with theommutator and to compensate for the play of the shaft.he collector is so fixed that it cannot turn with the shaft

or example, the diagram shows a slotted plate, K, whichs designed to be attached to a stationary support, and anrm extending from the collector and carrying a clampincrew, L, by which the collector may be adjusted and seto the desired position.

Mr. Tesla refers the form shown in Fi s. 2 6 and 2 to



t


he insulating segments of both commutator and collectooosely and to provide some means—as, for example, lighprings, e e, secured to the bands A' B', respectively, andearing against the segments— to exert a light pressurepon them and keep them in close contact and toompensate for wear. The metal segments of theommutator may be moved forward by loosening the

crew a'.

he line wires are fed from the metal segments of theollector, being secured thereto in any convenient

manner, the plan of connections being shown as applied tmodified form of the commutator in Fig. 251. The

ommutator and the collector in thus presenting two flatnd smooth bearing surfaces prevent most effectually bymechanical action the occurrence of sparks.

he insulating segments are made of some hard materiaapable of being polished and formed with sharp edges.

uch materials as glass, marble, or soapstone may bedvantageously used. The metal segments are preferablf copper or brass ; but they may have a facing or edge ourable material —such as platinum or the like — wherehe sparks are liable to occur.

n Fig. 248 a somewhat modified form of the invention ishown, a form designed to facilitate the construction and



eplac-

IG. 251.

IG. 252.

ng of the parts. In this modification the commutator andollector are made in substantially the same manner asreviously described, except that the bands B o aremitted. The four segments of each part, however, are

ecured to their respective sleeves by screws g' g', andne edge of each segment is cut away, so that small plate



b may be slipped into the spaces thus formed. Of

hese plates a a are of metal, and are in contact with themetal segments D D', respectively. The other two, b &,

re of glass or marble, and they are all better square, as

hown in Figs. 249 and 250, so that they may be turnedo present new edges should any edge become worn by se. Light springs d bear upon these plates and presshose in the commutator toward those in the collector,nd insulating strips c c are secured to the periphery of he discs to prevent the blocks from being thrown out by

entrifugal action. These plates are, of course, useful athose edges of the segments only where sparks are liableo occur, and, as they are easily replaced, they are of reat advantage. It is considered best to coat them withlatinum or silver.

n Figs. 251 and 252 is shown a construction where,nstead of solid segments, a fluid is employed. In this cashe commutator and collector are made of two insulatingiscs, s T, and in lieu of the metal segments a space is cutut of each part, as at K K', corresponding in shape andize to a metal segment. The two parts are iitted smooth

nd the collector T held by the screw h and spring ngainst the commutator s. As in the other cases, theommutator revolves while the collector remainstationary. The ends of the coils are connected to bindingosts $ -v, which are in electrical connection with metallates t 2 within the recesses in the two parts s T. These

hambers or recesses are filled with mercury, and in the



ollector part are tubes w w, with screws w w, carryingprings x and pistons x', which compensate for thexpansion and contraction of the mercury under varyingemperatures, but which are sufficiently strong not toield to the pressure of the fluid due to centrifugal action

nd which serve as binding-posts.

n all the above cases the commutators are adapted foraingle coil, and the device is particularly suited to suchurposes. The number of segments may be increased,owever, or more than one commutator used with a

ingle armature. Although the bearing-surfaces are shows planes at right angles to the shaft or axis, it is evidenthat in this particular the construction 'may be very reatly modified.

HAPTER XXXVIII. AUXILIARY BKUSH

REGULATION OF DIRECT CURRENT DYNAMOS.

AN interesting method devised by Mr. Tesla for theegulation of direct current dynamos, is that which liasome to be known as the "third brush" method. In

machines of this type, devised by him as far back as 188

e makes use of two main brushes to which the ends of he field magnet coils are connected, an auxiliary brush,nd a branch or shunt connection from an intermediateoint of the iield wire to the auxiliary brush. 1

he relative positions of the respective brushes are

aried, either automatically or by hand, so that the shunecomes inoperative when the auxiliary brash has a



ertain position upon the commutator; but when theuxiliary brush is moved in its relation to the mainrushes, or the latter are moved in their relation to theuxiliary brush, the electric condition is disturbed and

more or less of the current through the field-helices is

iverted through the shunt or a current is passed overhe shunt to the field-helices. By varying the relativeosition upon the commutator of the respective brushesutomatically in proportion to the varying electricalonditions of the working-circuit, the current developedan be regulated in proportion to the demands in the

working-circuit.

ig. 253 is a diagram illustrating the invention, showingne core of the field-magnets with one helix wound in theame direction throughout. Figs. 254 and 255 areiagrams showing one core of the field-magnets with a

ortion of the helices wound in opposite directions. Figs.56 and 257 are diagrams illustrating

. The compiler has learned partially from statementsmade on several occasions in journals and partially by

ersonal inquiry of Mr. Tesla, that a great deal of work in

his interesting line is unpublished. In these inventions awill be seen, the brushes are automatically shifted, but inhe broad method barely suggested here the regulation iffected without any change in the position of the brushehis auxiliary brush invention, it will be remembered,

was very much discussed a few years ago, and it may be

f interest that this work of Mr. Tesla, then unknown in



his field, is now brought to light

he electric devices that may be employed forutomatically adjusting the brushes, and Fig. 258 is aiagram illustrating the positions of the brushes when th

machine is being energized at the start.

and 5 are the positive and negative brushes of the mainr working-circuit, and c the auxiliary brush. The

working-circuit i) extends from the brushes a and b, assual, and contains electric lamps or other devices, D',

ither in series or in multiple arc.

M M' represent the field-helices, the ends of which areonnected to the main brushes a and 5. The branch orhunt wire c' extends from the auxiliary brush c to theircuit of the field-helices, and is connected to the same a

n intermediate point, v.

H represents the commutator, with the plates of ordinaron-

IG. 2 .



truction. When the auxiliary brush c occupies such aosition upon the commutator that the electro-motiveorce between the brushes a and c is to the electro-motivorce between the brushes c and b as the resistance of th

ircuit a M c' c A is to the resistance of the circuit b M' c' , the potentials of the points x and Y will be equal, ando current will flow over the auxiliary brush; but when

he brush c occupies a different position the potentials ofhe points x and Y will be different, and a current will flover the auxiliary brush to and from the commutator,

ccording to the relative position of the brushes. If, fornstance, the commutator-space between the brushes and c, when the latter is at the neutral point, isiminished, a current will flow from the point Y over thehunt c to the brush J, thus strengthening the current inhe part M', and partly neutralizing the current in part M

but if the space between the brushes a and c isncreased, the cur-


ent will flow over the auxiliary brush in an opposite

irection, and the current in M will be strengthened, andn M' partly neutralized.

y combining with the brushes a, I, and c any usualutomatic regulating mechanism, the current developedan be regulated in proportion to the demands in the

working circuit. The parts M



IG. 254.

nd M' of the Held wire may be wound in the same

irection. In this case they are arranged as shown in Fig.53 ; or the part M may be wound in the oppositeirection, as shown in Figs.

54 and 255.

t will be apparent that the respective cores of the tield-nag-nets are subjected to neutralizing or intensifyingffects of the current in the shunt through c', and the

magnetism of the cores will be partially neutralized, or thoints of greatest magnetism shifted, so that it will be

more or less remote from or approaching to the armatur

nd hence the aggregate energizing actions of the fieldmagnets on the armature will be correspondingly varied

n the form indicated in Fig. 253 the regulation is effectey shifting the point of greatest magnetism, and in Figs.54 and

the same effect is roduced b the action of the



urrent in the shunt passing through the neutralizingelix.

he relative positions of the respective brushes may bearied by moving the auxiliary brush, or the brush c may

emain stationary and the core P be connected to themain-brush holder A, so as to adjust the brushes a b inheir relation to the brush c. If, however, an adjustment pplied to all the brushes, as seen in Fig. 257, the solenoihould be connected to both a and c, so as to move themoward or away from each other.

here are several known devices for giving motion inropor-

A UXfLIARY BRUSH It KG ULAT1ON.

41

on to an electric current. In Figs. 25»i and 257 themoving cores are shown as convenient devices for

btaining the required extent of motion with very slighthanges in the current passing through the helices. It is

nderstood that the adjustment of the main brushesauses variations in the strength of the currentndependently of the relative position of those brushes tohe auxiliary brush. In all cases the adjustment should beuch that no current flows over the auxiliary brush whenhe dynamo is running with its normal load.

n Figs. 256 and 257 A A indicate the main-brush holder



arrying the main brushes, and c the auxiliary-brusholder, carrying the auxiliary brush. These brush-holderre movable in arcs concentric with the centre of theommutator-shaft. An iron piston, p, of the solenoid s, Fi5(5, is attached to the auxiliary-brush holder c;. The

djustment is effected by means of a spring and screw orghtener.

n Fig. 257 instead of a solenoid, an iron tube inclosing aoil is shown. The piston of the coil is attached to bothrush-holders A A and c. When the brushes are moved

irectly by electrical devices, as shown in Figs. 25(5 and57, these are so constructed that the force exerted fordjusting is practically uniform through the whole lengthf motion.

t is true that auxiliary brushes have been used in

onnection with the helices of the field-wire; but in thesenstances the

IG. 255.

elices receive the entire current through the auxiliary



rush or brushes, and these brushes could not be takenff without breaking the circuit through the field. Theserushes cause, move-over, heavy sparking at theommutator. In the present case the auxiliary brushauses very little or no sparking, and can be taken off

without breaking the circuit through the field-


elices. The arrangement lias, besides, the ad vantage ofacilitating the self-excitation of the machine in all cases

where the resistance of the field-wire is very greatomparatively to the resis-' tance of the main circuit athe start—for instance, on arc-light

IG. 256.

machines. In this case the auxiliary brush e is placed neao, or better still in contact with, the brush £, as shown inig. 258. In this manner the part M' is completely cut ou

nd as the part M has a considerably smaller resistancehan the whole length of the field-wire the machine



xcites itself, whereupon the auxiliary brush is shiftedutomatically to its normal position.

n a further method devised by Mr. Tesla, one or moreuxiliary brushes are employed, by means of which a

ortion or the whole of the field coils is shunted. Accordino the relative position upon the commutator of theespective brushes more or less current is caused to passhrough the helices of the field, and the current developey the machine can be varied at will by varying theelative positions of the brushes.

n Fig. 259, a and 1) are the positive and negative brushef the main circuit, and c an auxiliary brush. The mainircuit D

IG. 258.

xtends from the brushes a and b, as usual, and containshe helices M of the field wire and the electric lamps orther working devices. The auxiliary brush c is connecteo the point x of the main circuit by means of the wire c'

H is a commutator

f ordinary construction. It will have been seen from wha



was said already that when the electro-motive forceetween the brushes a and c is to the electromotive forceetween the brushes c and b as the resistance of theircuit a M c' c A is to the resistance of the circuit b c B c

D, the potentials of the points a? and y will be equal, and

o current will pass over the auxiliary brush c; but if tharush occupies a different position relatnely to the mainrushes the electric condition is disturbed, and current

will flow either from y to x or from a? to y, according tohe relative position of the brushes. In the first case theurrent through the field-helices will be partly neutralize

nd the magnetism of the field magnets will beiminished. In the second case the current will be

ncreased and the magnets gain strength. By combiningwith the brushes a 1) c any automatic regulatingmechanism, the current developed can be regulated

utomatically in proportion to the demands of the

working circuit.

n Figs. 264 and 265 some of the automatic means areepresented that may be used for moving the brushes.he core P, Fig. 264, of the solenoid-helix s is connected

with the brush c to move the same, and in Fig. 265 the

ore P is shown as within the helix s, and connected withrushes a and c, so as to move the same toward or fromach other, according to the strength of the current in thelix, the helix being within an iron tube, s', that become

magnetized and increases the action of the solenoid.

n practice it is sufficient to move only the auxiliary



rush, as shown in Fig. 264, as the regulation is very ensitive to the slightest changes; but the relative positiof the auxiliary brush to the main brushes may be variedy moving the main brushes, or both main and auxiliary rushes may be moved, as illustrated in Fig. 265. In the

atter two cases, it will be understood, the motion of themain brushes relatively to the neutral line of the machinauses variations in the strength of the currentndependently of their relative position to the auxiliary rush. In all cases the adjustment may be such that whehe machine is running with the ordinary load, no curren

ows over the auxiliary brush.

he field helices may be connected, as shown in Fig. 25!>r a part of the field helices may be in the outgoing andhe other part in the return circuit, and two auxiliary rushes may be employed as shown in Figs. 261 and 262

nstead of shunting the whole of the field helices, a portionly of such helices maybe shunted, as shown in Figs. 26nd 262.


he arrangement shown in Fig. '2ti*2 is advantageous, adiminishes the sparking upon the commutator, the ma

ircuit being closed through the auxiliary brushes at themoment of the break of the circuit at the main brushes.



IG. 259.

IG. 261.

IG. 262.

IG. 263.

he field helices may be wound in the same direction, or



art may be wound in opposite directions.

he connection between the helices and the auxiliary rush or brushes may be made by a wire of smallesistance, or a resistance may be interposed (R, Fig.

63,) between the point ./• and the

uxiliary brush or brushes to divide the sensitivenesswhen the 'brushes are adjusted.

he accompanying sketches also illustrate improvement

made by Mr. Tesla in the mechanical devices used toffect the shifting of the brushes, in the use of an auxiliarrush. Fig. 266 is an elevation of the regulator with therame partly in section ; and Fig. 267 is a section at thene a? a?, Fig. 266. c is the commutator; B and B', therush-holders, B carrying the main brushes a a', and B'

he auxiliary or shunt brushes b b. The axis of the brusholder B is supported by two pivot-screws, JP />. Thether brush-holder, B', has a sleeve, d, and is movableround the axis of the brush-holder B. In this way bothrush-holders can turn very freely, the friction of thearts being reduced to a minimum. Over the brush-

olders is mounted the solenoid s, which rests upon aorked column, c. This column



IG. 264. Fro. 265.

lso affords a support for the pivots pp, and is fastened

pon a solid bracket or projection, p, which extends fromhe base of the machine, and is cast in one piece with theame. The brush-holders B B' are connected by means ohe links e e and the cross-piece F to the iron core i, whiclides freely in the tube T of the solenoid. The iron core ias a screw, s, by means of which it can be raised and

djusted in its position relatively to the solenoid, so thathe pull exerted upon it by the solenoid is practically niform through the whole length of motion which isequired to effect the regulation. In order to effect thedjustment with greater precision, the core i is provided

with a small iron screw, s'. The core being first broughtery nearly in the required position relatively to theolenoid by means of the screw s, the small screw s' ishen adjusted until the magnetic attraction upon the cores the same when the core is in any posi. tion. A onvenient stop, £, serves to limit the upward movemen

f the iron core.


o check somewhat the movement of the core i, a dash-ot, K, is used. The piston L of the dash-pot is provided

with a vahê, v, which opens by a downward pressurend allows an eas -downward movement of the iron cor



but closes and checks the movement of the core when is pulled up by the action of the solenoid.

o balance the opposing forces, the weight of the movingarts, and the pull exerted by the solenoid upon the iron

ore, the weights w w may be used. The adjustment isuch that when the solenoid is traversed by the normalurrent it is just strong enough to balance the downwardull of the parts.

he electrical circuit-connections are substantially the

ame as

IG. 266

ndicated in the previous diagrams, the solenoid being in



eries with the circuit when the translating devices are ineries, and in shunt when the devices are in multiple arche operation of the device is as follows: When upon aecrease of the resistance of the circuit or for some othereason, the current is increased, the solenoid s gains in

trength and pulls up the iron core i, thus shifting themain brushes in the direction of rotation and the auxiliarrushes in the opposite way. This diminishes the strengtf the current until the opposing forces are balanced andhe solenoid is traversed by the normal current ; but if rom any cause the current in the circuit is diminished,

hen the weight of the moving parts overcomes the pull ohe solenoid, the iron

ore i descends, thus shifting the brashes the oppositeway and increasing the current to the normal strength.

he dash-pot connected to the iron core i may he of

rdinary construction ; but it is better, especially inmachines for arc lights, to provide the piston of the dash

ot with a valve, as indicated in the diagrams. This valveermits a comparatively easy downward movement of he iron core, but checks its movement when it is drawnp by the solenoid. Such an arrangement has the

dvantage that a great number of lights may be put onwithout diminishing the light-power of the lamps in theircuit, as the brushes assume at once the proper positio

When lights are cut out, the dash-pot acts to retard themovement; but if the current is considerably increasedhe solenoid gets abnormally strong and the brushes arehifted instantly. The regulator being properly adjusted,



ghts or other devices may be put on or out with scarcelyny perceptible difference. It is obvious that instead of he dash-pot any other retarding device may be used.

HAPTER XXXIX.

MPROVEMENT IN THE CONSTRUCTION OFDYNAMOS AND MOTORS.

HIS invention of Mr. Tesla is an improvement in theonstruction of dynamo or magneto electric machines or

motors, consisting in a novel form of frame and fieldmagnet which renders the machine more solid andompact as a structure, which requires fewer parts, and

which involves less trouble and expense in itsmanufacture. It is applicable to generators and motors

enerally, not only to those which have independent

ircuits adapted for use in the Tesla alternating currentystem, but to other continuous or alternating currentmachines of the ordinary type generally used.

ig. 268 shows the machine in side elevation. Fig. 269 is ertical sectional view of the field magnets and frame an

n end view of the armature ; and Fig. 270 is a plan viewf one of the parts of the frame and the armature, aortion of the latter being cut away.

he field magnets and frame are cast in two parts. Thesearts are identical in size and shape, and each consists of

he solid plates or ends A B, from which project inwardlyhe cores c D and the side bars or bridge pieces, E F. The



recise shape of these parts is largely a matter of choice—hat is to say, each casting, as shown, forms anpproximately rectangular frame ; but it might obviouslye more or less oval, round, or square, without departurrom the invention. It is also desirable to reduce the widt

f the side bars, E F, at the center and to so proportionhe parts that when the frame is put together the spacesetween the pole pieces will be practically equal to thercs which the sur-. faces of the poles occupy.

he bearings G for the armature shaft are cast in the sid

ars E F. The field coils are either wound on the poleieces or on a form and then slipped on over the ends of he pole pieces. The lower part or casting is secured to thase after being finished off. The armature K on its shaft

s then mounted in

MPROVEMENTS IN DYNAMOS AND MOTORS.

49

he bearings of the lower casting and the other part of thrame placed in position, dowel pins L or any other mean

eing used to secure the two parts in proper position.





his machine possesses several advantages. For examplwe magnetize the cores alternately, as indicated by the

haracters y s, it will be seen that the magnetic circuitetween the poles of each part of a casting is completedhrough the solid iron side bars. The bearings for the sha

re located at the neutral points of the field, so that thermature core is not affected by the magnetic condition ohe field.

he improvement is not restricted to the use of four poleieces, as it is evident that each pole piece could be

ivided or more than four formed by the shape of theasting.

HAPTER XI,

KSLA DIRECT CURRENT ARC LIGHTING SYSTKM.

AT one time, soon after his arrival in America, Mr. Teslawas greatly interested in the subject of arc lighting, whichen occupied public attention and readily enlisted theupport of capital. He therefore worked out a system

which was confided to a company formed for its

xploitation, and then proceeded to devote his energies the perfection of the details of his more celebrated "otary field" motor system. The Tesla arc lightingpparatus appeared at a time when a great many otheramps and machines were in the market, but itommanded notice by its ingenuity. Its chief purpose wa

o lessen the manufacturin cost and sim lif the



rocesses of operation.

We will take up the dynamo first. Fig. 271 is a longitudinection, and Fig. 272 a cross section of the machine. Fig.73 is a top view, and Fig. 274 a side view of the magneti

rame. Fig. 275 is an end view of the commutator bars,nd Fig. 276 is a section of the shaft and commutatorars. Fig. 277 is a diagram illustrating the coils of thermature and the connections to the commutator plates.

he cores c c c c of the field-magnets are tapering in both

irections, as shown, for the purposes of concentrating thmagnetism upon the middle of the pole-pieces.

he connecting-frame F F of the field-magnets is in theorm indicated in the side view, Fig. 274, the lower parteing provided with the spreading curved cast legs e e, s

hat the machine will rest firmly upon two base-bars, r r

o the lower pole, s, of the field-magnet M is fastened, bmeans of babbitt or other fusible diamagnetic material,he base B, which is provided with bearings b for thermature-shaft H. The base B has a projection, p, which

upports the brush-holders and the regulating devices,which are of a special character devised by Mr. Tesla.

NVENTIONS OF NIKOLA TEHLA.

mum the loss of power due to Foucault currents and to

he change of polarity, and also to shorten as much asossible the length of the inactive wire wound upon the



rmature core.

t is well known that when the armature is revolvedetween the poles of the field-magnets, currents areenerated in the iron body of the armature which develo

eat, and consequently cause

IG. 271.

waste of power. Owing to the mutual action of the lines

f force, the magnetic properties of iron, and the speed ohe different portions of the armature core, theseurrents are generated principally on and near theurface of the armature core, diminishing in strengthradually toward the centre of the core. Their quantity inder some conditions proportional to the length of the

ron body in the direction in which these currents areenerated. By subdividing the iron core electrically in thiirection, the generation of these currents can be reduceo a great extent. For instance, if the length of thermature-core is twelve inches, and by a suitableonstruction it is subdivided electrically, so that there arn the generating direction six inches of iron and six inch



f intervening air-spaces or insulating material, the wasturrents will be reduced to fifty per cent.

As shown in the diagrams, the armature is constructed ohin iron discs n D D, of various diameters, fastened upon

he armature-shaft in a suitable manner and arrangedccording to their sizes, so that a series of iron bodies, i i s formed, each of which diminishes in thickness from theentre toward the periphery. At both ends of thermature the inwardly curved discs d d, of cast iron, areastened to the armature shaft.

he armature core being constructed as shown, it will beasily seen that on those portions of the armature thatre the most remote from the axis, and where theurrents are principally developed, the length of iron inhe generating direction is only a

DIRECT CURRKNT ARC LIGHTING SYSTEM.

53

mall fraction of the total length of the armature core, an

esides this the iron body is subdivided in the generatingirection, and therefore the Foueault currents are greatleduced. Another cause of heating is the shifting of theoles of the armature core. In consequence of theubdivision of the iron in the armature and the increasedurface for radiation, the risk of heating is lessened.

he iron discs D D D are insulated or coated with some



nsulating-paint, a very careful insulation beingnnecessary, as an electrical contact between severaliscs can only occur at places where the generatedurrents are comparatively weak. An armature coreonstructed in the manner described may be revolved

etween the poles of the field magnets without showinghe slightest increase of temperature.

he end discs, d d, which are of sufficient thickness and,or the sake of cheapness, of cast-iron, are curvednwardly, as indicated in the drawings. The extent of the

urve is dependent on the amount of wire to be woundpon the armatures. In this machine the wire is woundpon the armature in two superimposed parts, and theurve of the end discs, dd, is so calculated that the firstart—that is, practically half of the wire—just fills



ro. 273.

p the hollow space to the line xx; or, if the wire is woundn any other manner, the curve is such that when the

whole of the wire is wound, the outside mass of wires, Mnd the inside mass of wires, w', are equal at each side ofhe plane x x. In this case the passive or electrically-nactive wires are of the smallest length practicable. Therrangement has further the advantage


hat the total lengths of the crossing wires at the two sidf the plane x x are practically equal.

o equalize further the armature coils at both sides of thlates that are in contact with the brushes, the windingnd connecting up is effected in the following manner :he whole wire is wound upon the armature-core in twouperimposed parts,



which are thoroughly insulated from each other. Each of hese two parts is composed of three separated groups ooils. The first group of coils of the first part of wire being

wound and connected to the commutator-bars in the

sual manner, this group is insulated and the secondroup wound ; but the coils < -f this second group, insteaf being connected to the next following commutator barre connected to the directly opposite bars of theommutator. The second group is then insulated and thehird group wound, the coils of this group being connecte

o those bars to which they would be connected in thesual way. The wires are then thoroughly insulated andhe second part of wire is wound and connected in theame manner.

uppose, for instance, that there are twenty-four coils—

hat is, twelve in each part—and consequently twenty-our commutator plates. There will be in each part threeroups, each containing four coils, and the coils will beonnected as follows:

Groups. Commutator J><ir*\

First 1—5

irst part of wire I Second 17 —21

Third 9—13

First 13—17



econd part of wire •< Second 5— 9

Third 21— 1

n constructing the armature core and winding and

onnecting the coils in the manner indicated, the passiver electrically in-

DIRECT CURRENT ARC LIGHTING SYSTEM.

55

ctive wire is reduced to a minimum, and the coils at eacide of the plates that are in contact with the brushes areractically equal. In this way the electrical efficiency of he machine is increased.

he commutator plates t are shown as outside theearing b of

IG. 275.

IG. 276.



he armature shaft. The shaft H is tubular and split at thnd portion, and the wires are carried through the samen the usual manner and connected to the respectiveommutator plates. The commutator plates are upon aylinder, w, and insulated, and this cylinder is properly

laced and then secured by expanding* the split end of he shaft by a tapering screw plug, v.

IG. 277.

he arc lamps invented by Mr. Tesla for use on the

ircuits from the above described dynamo are those inwhich the separation and feed of the carbon electrodes oheir equivalents is accomplished by means of electro-

magnets or solenoids in connection with suitable clutchmechanism, and were designed for the purpose

f remedying certain faults common to arc lamps.



He proposed to prevent the frequent vibrations of themovable carbon "point" and flickering of the light arisingherefrom; to prevent the falling into contact of thearbons; to dispense with the dash pot, clock work, orearing and similar devices; to render the lamp extreme

ensitive, and to feed the carbon almost imperceptibly,nd thereby obtain a very steady and uniform light.

n that class of lamps where the regulation of the arc isffected by forces acting in opposition on a free, movableod or lever directly connected with the electrode, all or

ome of the forces being dependent on the strength of thurrent, any change in the electrical condition of theircuit causes a vibration and a corresponding flicker inhe light. This difficulty is most apparent when there arenly a few lamps in circuit. To lessen this difficulty lampsave been constructed in which the lever or armature,

fter the establishing of the arc, is kept in a fixed positionnd cannot vibrate during the feed operation, the feed

mechanism acting independently; but in these lamps,when a clamp is employed, it frequently occurs that thearbons come into contact and the light is momentarily xtinguished, and frequently parts of the circuit are

njured. In both these classes of lamps it has beenustomary to use dash pot, clock work, or equivalentetarding devices; but these are often unreliable andbjectionable, and increase the cost of construction.

Mr. Tesla combines two electro-magnets —one of low

esistance in the main or lamp circuit, and the other of



omparatively high resistance in a shunt around the arc movable armature lever, and a special feed mechanism

he parts being arranged so that in the normal workingosition of the armature lever the same is kept almostigidly in one position, and is not affected even by

onsiderable changes in the electric circuit ; but if thearbons fall into contact the armature will be actuated byhe magnets so as to move the lever and start the arc, anold the carbons until the arc lengthens and the armatur

ever returns to the normal position. After this the carbood holder is released by the action of the feed

mechanism, so as to feed the carbon and restore the arco its normal length.

ig. 278 is an elevation of the mechanism made use of inhis arc lamp. Fig. 279 is a plan view. Fig. 280 is anlevation of the balancing lever and spring; Fig. 281 is a

e-

DIRECT CURRENT ARC LIGHTING 8YSTKM.

f>7

aclied plan view of the pole pieces and armatures uponhe friction clamp, and Fig. 282 is a section of thelamping tube.

M is a helix of coarse wire in a circuit from the lowerarbon holder to the negative binding screw— . N is a

elix of fine wire in a shunt between the positive bindingcrew -\- and the negative binding screw — . The upper



arbon holder s is a parallel rod sliding through the plate s 2 of the frame of the lamp, and hence the electricurrent passes from the positive binding



IG. 279,

IG. 281.

IG. 280.

ost _j_ through the plate s 2 , carbon holder s, and uppe

arbon to the lower carbon, and thence by the holder andmetallic connection to the helix M.

he carbon holders are of the usual character, and tonsure electric connections the springs I are made use of o grasp the upper carbon holding rod s, but to allow theod to slide freely through the same. These springs / mae adjusted in their pressure by the screw m, and thepring / may be sustained upon

ny suitable support. They are shown as connected withhe upper end of the core of the magnet N.



Around the carbon-holding rod s, between the plates s' sthere is a tube, R, which forms a clamp. This tube is

ounter-bored, as seen in the section Fig. 282, so that itears upon the rod s at its upper end and near the middlnd at the lower end of this tubular clamp K there are

rmature segments r of soft iron. A frame or arm, n,xtending, preferably, from the core N 2 , supports theever A by a fulcrum-pin, o. This lever A has a hole,hrough which the upper end of the tubular clamp Easses freely, and from the lever A is a link, q, to the lev, which lever is pivoted at y to a ring upon one of the

olumns s 8 . This lever t has an opening or bow urrounding the tubular clamp K, and there are pins orivotal connections w between the lever t and this clamp

R, and a spring, r, serves to support or suspend theweight of the parts and balance them, or nearly so. Thispring is adjustable.

At one end of the lever A is a soft-iron armature block, «ver the core M' of the helix M, and there is a limitingcrew, c, passing through this armature block «, and athe other end of the lever A is a soft iron armature block, with the end tapering or wedge shaped, and the same

omes close to and in line with the lateral projection e onhe core N 2 . The lower ends of the cores M' N 2 are

made with laterally projecting pole-pieces M 3 N 3 ,espectively, and these pole-pieces are concave at theiruter ends, and are at opposite sides of the armatureegments ;• at the lower end of the tubular clamp R.



he operation of these devices is as follows : In theondition of inaction, the upper carbon rests upon theower one, and when the electric current is turned on itasses freely, by the frame and spring /, through the rodnd carbons to the coarse wire and helix M, and to the

egative binding post v and the core M' thereby isnergized. The pole piece M 3 attracts the armature r,nd by the lateral pressure causes the clamp R to grasphe rod s', and the lever A is simultaneously moved fromhe position shown by dotted lines, Fig. 278, to the normosition shown in full lines, and in so doing the link q and

ever t are raised, lifting the clamp R and s, separating tharbons and forming the arc. The magnetism of the poleiece e tends to hold the lever A level, or nearly so, theore N 2 being energized by the current in the shunt

which contains the helix N. In this position the lever A isot

moved by any ordinary variation in the current, becausehe armature b is strongly attracted by the magnetism o?, and these parts are close to each other, and the

magnetism of e acts at right angles to the magnetism of he core M'. If, now, the arc becomes too long, the curren

hrough the helix M is lessened, and the magnetism of thore N 3 is increased by the greater current passinghrough the shunt, and this core N 3 , attracting theegmental armature /•, lessens the hold of the clamp R pon the rod s, allowing the latter to slide and lessen the

ength of the arc, which instantly restores the magneticquilibrium and causes the clamp R to hold the rod s. If i



appens that the carbons fall into contact, then themagnetism of N 2 is lessened so much that the attraction

f the magnet M will be sufficient to move the armature nd lever A so that the armature b passes above theormal position, so as to separate the carbons instantly;

ut when the carbons burn away, a greater amount of urrent will pass through the shunt until the attraction ohe core N 2 will overcome the attraction of the core M'nd bring the armature lever A again into the normalorizontal position, and this occurs before the feed canake place. The segmental armature pieces '/• are shown

s nearly semicircular. They are square or of any otheresired shape, the ends of the pole pieces M 3 , N 3 being

made to correspond in shape.

n a modification of this lamp, Mr. Tesla provided meansor automatically withdrawing a lamp from the circuit, or

utting it out when, from a failure of the feed, the arceached an abnormal length; and also means forutomatically reinserting such lamp in the circuit whenhe rod drops and the carbons come into contact.

ig. 283 is an elevation of the lamp with the case in

ection. Fig. 284 is a sectional plan at the line x .r. Fig. 28s an elevation, partly in section, of the lamp at rightngles to Fig. 283. Fig. 286 is a sectional plan at the line yof Fig. 283. Fig. 287 is a section of the clamp in about fu

ize. Fig. 288 is a detached section illustrating theonnection of the spring to the lever that carries the

ivots of the clamp, and Fig. 289 is a diagram showing th



ircuit-connections of the lamp.

n Fig. 283, M represents the main and N the shuntmagnet, both securely fastened to the base A, which with

s side columns, s s, are cast in one piece of brass or othe

iamagnetic material. To the magnets are soldered ortherwise fastened the brass washers or discs a a a a.imilar washers, b &, of fibre or other insii-

KVKNTIOXS OF NIKOLA TKSLA.

ating material, serve to insulate the wires from the braswashers. The magnets M and N are made very flat, sohat their width exceeds three times their thickness, orven more. In this way a comparatively small number ofonvolutions is sufficient to produce the required

magnetism, while a greater surface is offered for cooling

ff the wires.



IG. 284.

IG. 287. FIG. 288.

he upper pole pieces, /// », of the magnets are curved,

s indicated in the drawings, Fig. 283. The lower poleieces/// /i', are brought near together, tapering towardhe armature g, as shown in Figs. 284 and 286. The objecf this taper is to concentrate the greatest amount of theeveloped magnetism upon the armature, and also tollow the pull to be exerted always upon the middle of th

rmature y. This armature yisa piece of iron

n the shape of a hollow cylinder, having on each side aegment cut away, the width of which is equal to the

width of the pole pieces m' n'.

he armature is soldered or otherwise fastened to thelam - which is formed of a brass tube rovided with



ripping-jaws e Fig. 287. These jaws are arcs of a circle ohe diameter of the rod R, and are made of hardened

German silver. The guides /"/', through which thearbon-holding rod E slides, are made of the same

material. This has the advantage of reducing greatly the

wear and corrosion of the parts coming in frictionalontact with the rod, which frequently causes trouble.he jaws e e are fastened to the inside of the tube r, so

hat one is a little lower than the other. The object of thiss to provide a greater opening for the passage of the rod

when the same is released by the clamp. The clamp r is

upported on bearings w w, Figs. 283, 285 and 287, whicre just in the middle between the jaws e e. The bearings

w w are carried by a lever, t, one end of which rests uponn adjustable support, §-, of the side columns, s, the othend being connected by means of the link e' to thermature-lever L. The armature-lever L is a flat piece o

ron in |sj shape, having its ends curved so as toorrespond to the form of the upper pole-pieces of the

magnets M and N. It is hung upon the pivots v v, Fig. 28which are in the jaw x of the top plate B. This plate B, withe jaw, is cast in one piece and screwed to the sideolumns, s s, that extend up from the base A. To partly

alance the overweight of the moving parts, a spring, «',igs. 284 and 288, is fastened to the top plate, B, andooked to the lever t. The hook o is toward one side of th

ever or bent a little sidewise, as seen in Fig. 288. By thismeans a slight tendency is given to swing the armatureoward the pole-piece m' of the main magnet.



he binding-posts K K' are screwed to the base A. A manual switch, for short-circuiting the lamp when thearbons are renewed, is also fastened to the base. Thiswitch is of ordinary character, and is not shown in therawings.

he rod E is electrically connected to the lamp-frame bymeans of a flexible conductor or otherwise. The lamp-caeceives a removable cover, s 2 , to inclose the parts.

he electrical connections are as indicated

iagrammatically in Fig. 289. The wire in the mainmagnet consists of two parts, a?' and p'. These two partsmay be in two separated coils or in


ne single helix, as shown in the drawings. The part ,//eing normally in circuit, is, with the fine wire upon thehunt-magnet, wound and traversed by the current in thame direction, so as to tend to produce similar poles, N r s s, on the corresponding pole-pieces of the magnets Mnd N. The part p' is only in circuit when the lamp is cut

ut, and then the current being in the opposite directionroduces in the main magnet, magnetism of the oppositeolarity.

he operation is as follows : At the start the carbons areo be in contact, and the current passes from the positive

inding-post K to the lamp-frame, carbon-holder, uppernd lower carbon, insulated return- wire in one of the sid



ods, and from there through the part x' of the wire onhe main magnet to the nega-

ro. 289.

ve binding-post. Upon the passage of the current themain magnet is energized and attracts the clamping-

rmature g, swinging the clamp and gripping the rod by means of the gripping jaws e e. At the same time the

rmature lever L is pulled down and the carbons areeparated. In pulling down the armature lever L the mai

magnet is assisted by the shunt-magnet N, the lattereing magnetized by magnetic induction from the magne

M.

t will be seen that the armatures L and g are practicallyhe keepers for the magnets M and N, and owing to thisact both magnets with either one of the armatures L an



may be considered as one horseshoe magnet, which wemight term a " compound magnet." The whole of the softron parts M, m' t g, n' y N and i, form a compound

magnet.

he carbons being separated, the fine wire receives aortion of the current. Now, the magnetic induction fromhe magnet M is such as to produce opposite poles on theorresponding ends of the magnet N ; but the currentraversing the helices tends to produce similar poles onhe corresponding ends of both magnets, and therefore a

oon as the fine wire is traversed by sufficient current thmagnetism of the whole compound magnet is diminished

With regard to the armature g and the operation of theamp, the pole 77i' may be considered as the " clamping "nd the pole /// as the " releasing " pole.

As the carbons burn away, the fine wire receives moreurrent and the magnetism diminishes in proportion. Thauses the armature lever L to swing and the armature go descend gradually under the weight of the movingarts until the end/>, Fig. 283, strikes a stop on the top

late, B. The adjustment is such that when this takeslace the rod K is yet gripped securely by the jaws ee.he further downward movement of the armature levereing prevented, the arc becomes longer as the carbonsre consumed, and the compound magnet is weakened

more and more until the clamping armature g releases

he hold of the gripping-jaws e e upon the rod R, and theod is allowed to dro a little thus shortenin the arc. Th



ne wire now receiving less current, the magnetismncreases, and the rod is clamped again and slightly aised, if necessary. This clamping and releasing of theod continues until the carbons are consumed. In practiche feed is so sensitive that for the greatest part of the

me the movement of the rod cannot be detected withouome actual measurement. During the normal operationf the lamp the armature lever L remains practically tationary, in the position show T n in Fig. 283.

hould it happen that, owing to an imperfection in it, the

od and the carbons drop too far, so as to make the arcoo short, or even bring the carbons in contact, a very mall amount of current passes through the fine wire, anhe compound magnet becomes sufficiently strong to acts at the start in pulling the armature lever L down andeparating the c'arbons to a greater distance.

t occurs often in practical work that the rod sticks in theuides. In this case the arc reaches a great length, until itnally breaks. Then the light goes out, and frequently thne wire is

njured. To prevent such an accident Mr. Tesla provideshis lamp with an automatic cut-out which operates asollows : When, upon a failure of the feed, the arc reachescertain predetermined length, such an amount of

urrent is diverted through the fine wire that the polaritf the compound magnet is reversed. The clamping

rmature g is now moved against the shunt magnet Nntil it strikes the releasin ole n'. As soon as the contac



s established, the current passes from the positiveinding post over the clamp />, armature g, insulatedhunt magnet, and the helix p' upon the main magnet Mo the negative binding post. In this case the currentasses in the opposite direction and changes the polarity

f the magnet M, at the same time maintaining by magnetic induction in the core of the shunt magnet theequired magnetism without reversal of polarity, and thermature g remains against the shunt magnet pole n'. Thamp is thus cut out as long as the carbons are separatedhe cut out may be used in this form without any furthe

mprovement ; but Mr. Tesla arranges it so that if the rorops and the carbons come in contact the arc is startedgain. For this purpose he proportions the resistance of art j!/ and the number of the convolutions of the wirepon the main magnet so that when the carbons come inontact a sufficient amount of current is diverted through

he carbons and the part x' to destroy or neutralize themagnetism of the compound magnet, Then the armature

, having a slight tendency to approach to the clampingole m' t comes out of contact with the releasing pole n'.

As soon as this happens, the current through the part j?'s interrupted, and the whole current passes through theart x. The magnet M is now strongly magnetized, thermature g is attracted, and the rod clamped. At the samme the armature lever L is pulled down out of its normosition and the arc started. In this way the lamp cutsself out automatically when the arc gets too long, and

einserts itself automatically in the circuit if the carbonsrop together.



HAPTER XLI.

MPROVEMENT IN "UNIPOLAR" GENERATORS.

ANOTHER interesting class of apparatus to which Mr.

esla has directed his attention, is that of " unipolar "enerators, in which a disc or a cylindrical conductor is

mounted between magnetic poles adapted to produce anpproximately uniform field. In the disc armature

machines the currents induced in the rotating conductorow from the centre to the periphery, or conversely,

ccording to the direction of rotation or the lines of forces determined by the signs of the magnetic poles, andhese currents are taken off usually by connections orrushes applied to the disc at points on its periphery andear its centre. In the case of the cylindrical armature

machine, the currents developed in the cylinder are take

ff .by brushes applied to the sides of the cylinder at itsnds.

n order to develop economically an electromotive forcevailable for practicable purposes, it is necessary either totate the conductor at a very high rate of speed or to us

disc of large diameter or a cylinder of great length; butn either case it becomes difficult to secure and maintain ood electrical connection between the collecting brushesnd the conductor, owing to the high peripheral speed.

t has been proposed to couple two or more discs togethen series with the ob ect of obtainin a hi her electro-



motive force ; but with the connections heretofore usednd using other conditions of speed and dimension of discecessary to securing good practicable results, thisifficulty is still felt to be a serious obstacle to the use of his kind of generator. These objections Mr. Tesla has

ought to avoid by constructing a machine with two fieldach having a rotary conductor mounted between itsoles. The same principle is involved in the case of both

orms of machine above described, but the descriptionow given is confined to the disc type, which Mr. Tesla is

nclined to favor for that machine. The discs are formed

with flanges, after tho


manner of pulleys, and are connected together by flexiblonducting bands or belts.

he machine is built in such manner that the direction ofmagnetism or order of the poles in one tield of force is

pposite to that in the other, so that rotation of the discsn the same direction develops a current in one fromentre to circumference and in the other from

ircumference to centre. Contacts applied therefore to thhafts upon which the discs are mounted form theerminals of a circuit the electro-motive force in which ishe sum of the electro-motive forces of the two dises.

t will be obvious that if the direction of magnetism in

oth



ro. 290.

IG. 291.

elds be the same, the same result as above will bebtained by driving the discs in opposite directions androssing the connecting belts. In this way the difficulty of

ecuring and maintaining good contact with theeripheries of the discs is avoided and a cheap andurable machine made which is useful for many purpose

—such as for an exciter for alternating currentenerators, for a motor, and for any other purpose for

which dynamo machines are used.

ig. 290 is a side view, partly in section, of this machine.



ig. 291 is a vertical section of the same at right angles tohe shafts.

n order to form a frame with two fields of force, aupport, A, is cast with two pole pieces u B' integral with

. To this are joined by bolts E a casting D, with twoimilar and corresponding pole pieces c c'. The pole pieceB' are wound and connected to produce a field of force

iven polarity, and the pole pieces c c' are wound so as toroduce a field of opposite polarity. The driving shafts F ass through the poles and are journaled in insulating

earings in the casting A u, as shown.

H K are the discs or generating conductors. They areomposed of copper, brass, or iron and are keyed orecured to their respective shafts. They are provided witroad peripheral flanges j. It is of course obvious that the

iscs may be insulated from their shafts, if so desired. A exible metallic belt L is passed over the flanges of the

wo discs, and, if desired, maybe used to drive one of theiscs. It is better, however, to use this belt merely as aonductor, and for this purpose sheet steel, copper, orther suitable metal is used. Each shaft is provided with

riving pulley M, by which power is imparted from ariving shaft.

N N are the terminals. For the sake of clearness they arehown as provided with springs p, that bear upon thends of the shafts. This machine, if self-exciting, would

ave copper bands around its poles ; or conductors of anind —such as wires shown in thexlrawin s —ma be



sed.

t is thought appropriate by the compiler to append hereome notes on unipolar dynamos, written by Mr. Tesla,n a recent occasion.

t is characteristic of fundamental discoveries, of greatchievements of intellect, that they retain anndiminished power upon the imagination of the thinkerhe memorable experiment of Faraday with a discotating between the two poles of a magnet, which has

orne such magnificent fruit, has long passed into every-ay experience; yet there are certain features about thismbryo of the present dynamos and motors which eveno-day appear to us striking, and are worthy of the mostareful study.

onsider, for instance, the case of a disc of iron or othermetal

Article by Mr. Tesla, contributed to The Electricalngineer, N. Y., Sept. 2, 1891.

evolving between the two opposite poles of a magnet,nd the polar surfaces completely covering both sides of he disc, and assume the current to be taken off oronveyed to the same by contacts uniformly from alloints of the periphery of the disc. Take first the case of

motor. In all ordinary motors the operation is dependent

pon some shifting or change of the resultant of thema netic attraction exerted u on the armature this



rocess being effected either by some mechanicalontrivance on the motor or by the action of currents of he proper character. We may explain the operation of uch a motor just as we can that of a water-wheel. But inhe above example of the disc surrounded completely by

he polar surfaces, there is no shifting of the magneticction, no change whatever, as far as we know, and yetotation ensues. Here, then, ordinary considerations doot apply ; we cannot even give a superficial explanations in ordinary motors, and the operation will be clear to unly when we shall have recognized the very nature of th

orces concerned, and fathomed the mystery of thenvisible connecting mechanism.

onsidered as a dynamo machine, the disc is an equally nteresting object of study. In addition to its peculiarity oiving currents of one direction without the employment

f commutat-ing devices, such a machine differs fromrdinary dynamos in that there is no reaction betweenrmature and field. The armature current tends to set umagnetization at right angles to that of the field currenut since the current is taken off uniformly from all poinf the periphery, and since, to be exact, the external

ircuit may also be arranged perfectly symmetrical to theld magnet, no reaction can occur. This, however, is trunly as long as the magnets are weakly energized, for

when the magnets are more or less saturated, bothmagnetizations at right angles seemingly interfere with

ach other.



or the above reason alone it would appear that theutput of such a machine should, for the same weight, be

much greater than that of any other machine in which thrmature current tends to demagnetize the field. Thextraordinary output of the Forbes unipolar dynamo and

he experience of the writer confirm this view.

Again, the facility with which such a machine may bemade to excite itself is striking, but this may be due—

esides to the absence of armature reaction—to theerfect smoothness of the current and non-existence of

elf-induction.

UNIPOLAR GENERATORS.

69

f the poles do not cover the disc completely on bothides, then, of course, unless the disc be properly ubdivided, the machine will be very inefficient. Again, inhis case there are points worthy of notice. If the disc beotated and the field current interrupted, the currenthrough the armature will continue to flow and the field

magnets will lose their strength comparatively slowly.he reason for this will at once appear when we considerhe direction of the currents set up in the disc.

Referring to the diagram Fig. 292, d represents the discwith the sliding contacts B B' on the shaft and periphery.

N and s represent the two poles of a magnet. If the pole Ne above, as indicated in the diagram, the disc being



upposed to be in the

io. 292.

lane of the paper, and rotating in the direction of therrow D, the current set up in the disc will flow from theentre to the periphery, as indicated by the arrow A.ince the magnetic action is more or less confined to thepace between the poles N s, the other portions of the dis

may be considered inactive. The current set up willherefore not wholly pass through the external circuit F,ut will close through the disc itself, and generally, if theisposition be in any way similar to the one illustrated, b

ar the greater portion of the current generated will notppear externally, as the circuit F is practically short-

ircuited by the inactive portions of the disc. The directiof the resulting currents in the latter may be assumed to



e as indicated by the dotted

nes and arrows HI and n / and tlie direction of thenergizing field current being indicated by the arrows a bd, an inspection of the figure shows that one of the two

ranches of the eddy current-, that is, A B' m B, will tendo demagnetize the field, while the other branch, that is, ' n B, will have the opposite effect. Therefore, the branc

A B' m B, that is, the one which is approaching the field,will repel the lines of the same, while branch A B' n B, thas, the one leaving the field, will gather the lines of force

pon itself.

n consequence of this there will be a constant tendency o reduce the current flow in the path A B' m B, while onhe other hand no such opposition will exist in path A B' n, and the effect of the latter branch or path will be more

r less preponderating over that of the former. The jointffect of both the assumed branch currents might beepresented by that of one single current of the sameirection as that energizing the field. In other words, theddy currents circulating in the disc will energize the fiel

magnet. This is a result quite contrary to what we might

e led to suppose at first, for we would naturally expecthat the resulting effect of the armature currents woulde such as to oppose the field current, as generally occur

when a primary and secondary conductor are placed innductive relations to each other. But it must beemembered that this results from the peculiar

isposition in this case, namely, two paths being afforded



o the current, and the latter selecting that path whichffers the least opposition to its flow. From this we seehat the eddy currents flowing in the disc partly energizehe field, and for this reason when the field current isnterrupted the currents in the disc will continue to flow,

nd the field magnet will lose its strength withomparative slowness and may even retain a certaintrength as long as the rotation of the disc is continued.

he result will, of course, largely depend on the resistancnd geometrical dimensions of the path of the resulting

ddy current and on the speed of rotation; theselements, namely, determine the retardation of thisurrent and its position relative to the field. For a certainpeed there would be a maximum energizing action ; thet higher speeds, it would gradually fall off to zero andnally reverse, that is, the resultant eddy current effect

would be to weaken the field. The reaction would be bestemonstrated experimentally by arranging the fields N s

N' s', freely movable on an axis concentric with the shaftf the

isc. If the latter were rotated as before in the direction

he arrow D, the field would be dragged in the sameirection with a torque, which, up to a certain point, wouo on increasing with the speed of rotation, then fall off,nd, passing through zero, finally become negative ; thats, the field would begin to rotate in opposite direction tohe disc. In experiments with alternate current motors i

which the field was shifted by currents of differing phase



his interesting result was observed. For very low speedf rotation of the field the motor would show a torque of 00 Ibs. or more, measured on a pulley 12 inches iniameter. When the speed of rotation of the poles was

ncreased, the torque would diminish, would finally go

own to zero, become negative, and then the armaturewould begin to rotate in opposite direction to the field.

o return to the principal subject ; assume the conditiono be such that the eddy currents generated by theotation of the disc strengthen the field, and suppose the

atter gradually removed while the disc is kept rotating an increased rate. The current, once started, may then bufficient to maintain itself and even increase in strengthnd then we have the case of Sir William Thomson'scurrent accumulator." But from the above consideration would seem that for the success of the experiment the

mployment of a disc not subdivided 1 would be essentiaor if there should be a radial subdivision, the eddy urrents could not form and the self -exciting action

would cease. If such a radially subdivided disc were used would be necessary to connect the spokes by aonducting rim or in any proper manner so as to form a

ymmetrical system of closed circuits.

he action of the eddy currents may be utilized to excitemachine of any construction. For instance, in Figs. 293

nd 294 an arrangement is shown by which a machinewith a disc armature might be excited. Here a number of

magnets, N s, N s, are placed radially on each side of a



metal disc D carrying on its rim a set of insulated coils, c he magnets form two separate fields, an internal and anxternal one, the solid disc rotating in the

Mr. Tesla here refers to an interesting article which

ppeared in July, 1865, in the Phil. Magazine, by Sir W.homson, in which Sir William, speaking of his " uniformlectric current accumulator," assumes that for self-xcitation it is desirable to subdivide the disc into annfinite number of infinitely thin spokes, in order torevent diffusion of the current. Mr. Tesla shows that

iffusion is absolutely necessary for the excitation andhat when the disc is subdivided no excitation can occur.

eld nearest the axis, and the coils in the field furtherrom it. Assume the magnets slightly energized at thetart ; they could be strengthened by the action of the

ddy currents in the solid disc so as to afford a strongereld for the peripheral coils. Although there is no doubt

hat under proper conditions a machine might be excitedn this or a similar manner, there being sufficientxperimental evidence to warrant such an assertion, sucmode of excitation would be wasteful.

ut a unipolar dynamo or motor, such as shown in Fig.92, may be excited in an efficient manner by simply roperly subdividing the disc or cylinder in which theurrents are set up, and it is practicable to do away withhe field coils which are usually employed. Such a plan is

lustrated in Fig. 295. The disc or



IG. 293. FIG. 294.

ylinder D is supposed to be arranged to rotate betweenhe two poles N and s of a' magnet, which completely over it on both sides, the contours of the disc and poles

eing represented by the circles d and d 1 respectively,he upper pole being omitted for the sake of clearness.he cores of the magnet are supposed to be hollow, thehaft c of the disc passing through them. If the unmarkedole be below, and the disc be rotated screw fashion, theurrent will be, as before, from the centre to theeriphery, and may be taken off by suitable slidingontacts, B B', on the shaft and periphery respectively. Inhis arrangement the current flowing through the disc andxternal circuit will have no appreciable effect on the field

magnet.

ut let us now suppose the disc to be subdivided spirally,

s



UNIPOLAR GENERATORS.

78

ndicated by the full or dotted lines, Fig. 295. Theifference of potential between a point on the shaft and a

oint on the periphery will remain unchanged, in sign aswell as in amount. The only difference will be that theesistance of the disc will be augmented and that there

will be a greater fall of potential from a point on the shafto a point on the periphery when the same current israversing the external circuit. But since the current is

orced to follow the lines of subdivision, we see that it willend either to energize or de-energize the field, and this

will depend, other things being equal, upon the directionf the lines of subdivision. If the subdivision be as

ndicated by the full lines in Fig. 295, it is evident that if he current is of the same direction as before, that is, from

entre to periphery, its effect will be to strengthen theeld magnet; whereas, if the subdivision be as in-

IG. 295.



IG. 296.

icated by the dotted lines, the current generated willend to weaken the magnet. In the former case the

machine will be capable of exciting itself when the disc isotated in the direction of arrow D ; in the latter case the

irection of rotation must be reversed. Two such discsmay be combined, however, as indicated, the two discsotating in opposite fields, and in the same or oppositeirection.

imilar disposition may, of course, be made in a type of

machine in which, instead of a disc, a cylinder is rotated.n such unipolar machines, in the manner indicated, thesual field coils and poles may be omitted and the

machine may be made to consist only of a cylinder or of wo discs enveloped by a metal casting.

nstead of subdividing the disc or cylinder spirally, asndicated in Fig. 295, it is more convenient to interposene or more turns

etween the disc and the contact ring on the periphery, aslustrated in Fig. 296.

Forbes dynamo may, for instance, be excited in such amanner. In the experience of the writer it has been foundhat instead of taking the current from two such discs by iding contacts, as usual, a flexible conducting belt may e employed to advantage. The discs are in such caserovided with large flanges, affording a very great contact

urface. The belt should be made to bear on the flangeswith spring pressure to take up the expansion. Several



machines with belt contact were constructed by thewriter two years ago, and worked satisfactorily ; but forwant of time the work in that direction has beenemporarily suspended. A number of features pointed outbove have also been used by the writer in connection

with some types of alternating current motors.

ART IV.

PPENDIX.-EARLY PHASE MOTORS AND THE TESLA MECHANICAL AND ELECTRICAL OSCILLATOR.

HAPTER XLII.

MR. TESLA'S PERSONAL EXHIBIT AT THE WORLD'SAIR.

WHILE the exhibits of firms engaged in the manufacturef electrical apparatus of every description at the Chicago

World's Fair, afforded the visitor ample opportunity foraining an excellent knowledge of the state of the art,here were also numbers of exhibits which brought out introng relief the work of the individual inventor, whiches at the foundation of much, if not all, industrial or

mechanical achievement. Prominent among such personal

xhibits was that of Mr. Tesla, whose apparatus occupiedart of the space of the Westinghouse Company, inlectricity Building.

his apparatus represented the results of work andhought covering a period of ten years. It embraced a

arge number of different alternating motors and Mr.esla's earlier high frequency apparatus. The motor



xhibit consisted of a variety of fields and armatures forwo, three and multiphase circuits, and gave a fair idea of he gradual evolution of the fundamental idea of theotating magnetic field. The high frequency exhibitncluded Mr. Tesla's earlier machines and disruptiveischarge coils and high frequency transformers, which hesed in his investigations and some of which are referredo in his papers printed in this volume.

ig. 297 shows a view of part of the exhibits containinghe motor apparatus. Among these is shown at A a largeng intended to exhibit the phenomena of the rotating

magnetic field. The field produced was very powerful andxhibited striking effects, revolving copper balls and eggsnd bodies of various shapes at considerable distancesnd at great speeds. This ring was wound for two-phaseircuits, and the winding was so distributed that aractically uniform field was obtained. This ring was

repared for Mr. Tesla's exhibit by Mr. C. F. Scott,lectrician of the Westinghouse Electric and

Manufacturing Company.

NVENTION* OF NIKOLA 7/:s/..|.



smaller ring, shown at B, was arranged like the one

xhibited at A but designed especially to exhibit theotation of an armature in a rotatin field. In connection



with these two rings there was an interesting exhibithown by Mr. Tesla which consisted of a magnet with aoil, the magnet being arranged to rotate in bearings.

With this magnet he first demonstrated the identity etween a rotating field and a rotating magnet ; the latter,

when rotating, exhibited the same phenomena as thengs when they were energized by currents of differinghase. Another prominent exhibit was a model illustratedt c which is a two-phase motor, as well as an induction

motor and transformer. It consists of a large outer ring of aminated iron wound with two superimposed, separated

windings which can be connected in a variety of ways.

his is one of the first models used by Mr. Tesla as annduction motor and rotating transformer. The armature

was either a steel or wrought iron disc with a closed coil.When the motor was operated from a two phase

enerator the windings were connected in two groups, assual. When used as an induction motor, the current

nduced in one of the windings of the ring was passedhrough the other winding on the ring and so the motorperated with only two wires. When iised as aransformer the outer winding served, for instance, as aecondary and the inner as a primary. The model shownt D is one of the earliest rotating field motors, consisting

f a thin iron ring wound with two sets of coils and anrmature consisting of a series of steel discs partly cutway and arranged on a small arbor.

t E is shown one of the first rotating field or inductionmotors used for the regulation of an arc lamp and for

ther purposes. It comprises a ring of discs with two setsf coils having different self-inductions, one set being of



German silver and the other of copper wire. Thermature is wound with two closed-circuited coils at rightngles to each other. To the armature shaft are fastened

evers and other devices to effect the regulation. At F ishown a model of a magnetic lag motor ; this embodies aasting with pole projections protruding from two coilsetween which is arranged to rotate a smooth iron body.

When an alternating current is sent through the two coilshe pole projections of the field and armature within it aremilarly magnetized, and upon the cessation or reversalf the current the armature and field repel each other andotation is produced in this way.

nother interesting exhibit, shown at G, is an early modelf a two field motor energized by currents of differenthase. There are two independent fields of laminated iron

oined by brass' bolts ; in each field is mounted anrmature, both armatures being on the same shaft. The

rmatures were originally so arranged as to be placed inny position relatively to each other, and the fields also

were arranged to be connected in a number of ways. Themotor has served for the exhibition of a number of eatures; among other things, it has been used as aynamo for the production of currents of any frequency

etween wide limits. In this case the field, instead of beingnergized by direct current, was energized by currentsiffering in phase, which



IG. 298.

roduced a rotation of the field ; the armature was thenotated in the same or in opposite direction to the

movement of the field; and so any number of alternationsf the currents induced in the armature, from a small to aigh number, determined by the frequency of thenergizing field coils and the speed of the armature, wasbtained.

he models H, i, j, represent a variety of rotating field,ynchronous motors which are of special value in longistance transmission work. The principle embodied inhese motors was enunciated by Mr. Tesla in his lectureefore the American Institute of Electrical Engineers, in

May, 1888 1 . It involves the production

See Part I, Chap. Ill, page 9.

f the rotating field in one of the elements of the motor by ur rents differing in phase and energizing the otherlement by direct currents. The armatures are of the twond three phase type. K is a model of a motor shown in annlar ed view in Fi . 2 8. This machine, to ether with



hat shown in Fig. 299, was exhibited at the same lecture,n May, 1888. They were the first rotating field motors

which were independently tested, having for that purposeeen placed in the hands of Prof. Anthony in the winter of 88T-88. From these tests it was shown that thefficiency and output of these motors was quiteatisfactory in every respect.

t was intended to exhibit the model shown in Fig. 299,ut it was unavailable for that purpose owing to the facthat it was

IG. 299.

ome time ago handed over to the care of Prof. Ayrton inngland. This model was originally provided with twelve

ndependent coils; this number, as Mr. Tesla pointed outn his first lecture, being divisible by two and three, waselected in order to make various connections for two andhree-phase operations, and during Mr. Tesla'sxperiments was used in many ways with from two to six



hases. The model, Fig. 298, consists of a magnetic framef laminated iron with four polar projections between

which an armature is supported on brass bolts passinghrough the frame. A great variety of armatures was usedn connection with these two and other fields. Some of thermatures are shown in front on the table, Fig. 297, andeveral are also shown enlarged in Figs. 300 to 310. Annteresting exhibit is that shown at L, Fig. 297.. This is anrmature of hardened steel which was used in a demon-


tration before the Society of Arts in Boston, by Prof.nthony. Another curious exhibit is shown enlarged inig. 301. This consists of thick discs of wrought ironlaced lengthwise, with a-mass of copper cast aroundhem. The discs were arranged longitudinally to afford anasier starting by reason of the induced current formed in

he iron discs, which differed in phase from those in theopper. This armature would start with a single circuitnd run in synchronism, and represents one of thearliest types of such an armature. Fig. 305 is anothertriking exhibit.

IG. 303.



IG. 304.

IG. 305.

IG. 306

IG. 307

IG. 308.

IG. 309.

IG. 310.

his is one of the earliest types of an armature with holeseneath the periphery, in which copper conductors are

mbedded. The armature has eight closed circuits and was

sed in many different ways. Fig. 304 is a type of ynchronous armature consisting of a block of soft steel

wound with a coil closed upon itself. This armature wassed in connection with the field shown in Fig. 298 andave excellent results.

ig. 302 represents a synchronous armature with a largeoil around a bod of iron. There is another ver small coil



t right angles to the first. This small coil was used for theurpose of

ncreasing the starting torque and was found very ffective in this connection. Figs. 306 and 308 show aavorite construction of armature ; the iron body is made

p of two sets of discs cut away and placed at right angleso each other, the interstices being wound with coils. Thene shown in Fig. 308 is provided with an additionalroove on each of the projections formed by the discs, forhe purpose of increasing the starting torque by a wire

wound in these projections. Fig. 307 is a form of armature

milarly constructed, but with four independent coilswound upon the four projections. This armature was usedo reduce the speed of the motor with reference to that of he generator. Fig. 300 is still another armature with areat number of independent circuits closed uponhemselves, so that all the dead points on the armature

re done away with, and the armature has a large startingorque. Fig. 303 is another type of armature for a four-ole motor but with coils wound upon a smooth surface. A umber of these armatures have hollow shafts, as they ave been used in many ways. Figs. 309 and 310epresent armatures to which either alternating or direct

urrent was conveyed by means of sliding rings. Fig. 309onsists of a soft iron body with a single coil wound around, the ends of the coil being connected to two sliding rings

o which, usually, direct current was conveyed. Thermature shown in Fig. 310 has three insulated rings on ahaft and was used in connection with two or three phaseircuits.



ll these models shown represent early work, and thenlarged engravings are made from photographs takenarly in 1888. There is a great number of other models

which were exhibited, but which are not brought outharply in the engraving, Fig. 297. For example at M is a

model of a motor comprising an armature with a hollow haft wound with two or three coils for two or three-hase circuits; the armature was arranged to betationary and the generating circuits were connectedirectly to the generator. Around the armature isrranged to rotate on its shaft a casting forming six closedircuits. On the outside this casting was turned smooth

nd the belt was placed on it for driving with any desiredppliance. This also is a very early model.

n the left side of the table there are seen a large variety f models, N, o, P, etc., with fields of various shapes. Eachf these models involves some distinct idea and they all

epresent gradual

evelopment chiefly interesting as showing Mr. Tesla'sfforts to adapt his system to the existing highrequencies.

n the right side of the table, at s, T, are shown, oneparate supports, larger and more perfected armaturesf commercial motors, and in the space around the table aariety of motors and generators supplying currents tohem was exhibited.

he high frequency exhibit embraced Mr. Tesla's first

riginal apparatus used in his investigations. There wasxhibited a glass tube with one layer of silk-covered wire



wound at the top and a copper ribbon on the inside. Thiswas the first disruptive discharge coil constructed by him.

t u is shown the disruptive

IG. 811.

ischarge coil exhibited by him in his lecture before themerican Institute of Electrical Engineers, in May, 1891.At v and w are shown some of the first high frequency

ransformers. A number of various fields and armatures

f small models of high frequency apparatus as shown at xnd Y, and others not visible in the picture, werexhibited. In the annexed space the dynamo then used by

Mr. Tesla at Columbia College was exhibited ; alsonother form of high frequency dynamo used.

n this space also was arranged a battery of Leyden jarsnd his lar e disru tive dischar e coil which was used for



xhibiting

See Part II, Chap. XXVI., page 145.

he light phenomena in the adjoining dark room. The coilwas operated at only a small fraction of its capacity , as the

ecessary condensers and transformers could not be hadnd as Mr. Tesla's stay was limited to one week ;otwithstanding, the phenomena were of a strikingharacter. In the room were arranged two large plateslaced at a distance of about eighteen feet from eachther. Between them were placed two long tables with all

orts of phosphorescent bulbs and tubes; many of thesewere prepared with great care and marked legibly withhe names which would shine with phosphorescent glow.mong them were some with the names of Helmholtz,araday, Maxwell, Henry, Franklin) etc. Mr. Tesla hadlso not forgotten the greatest living poet of his own

ountry, Zmaj Jovan ; two or three were prepared withnscriptions, like " Welcome, Electricians," and produced aeautiful effect. Each represented some phase of this

work and stood for some individual experiment of mportance. Outside the room was the small battery seenn Fig. 311, for the exhibition of some of the impedancend other phenomena of interest. Thus, for instance, ahick copper bar bent in arched form was provided withlamps for the attachment of lamps, and a number of amps were kept at incandescence on the bar ; there waslso a little motor shown on the table operated by theisruptive discharge.

s will be remembered by those who visited thex osition the Westin house Com an made a fine



xhibit of the various commercial motors of the Teslaystem, while the twelve generators in Machinery Hall

were of the two-phase type constructed for distributingght and power. Mr. Tesla, also exhibited some models of is oscillators.

HAPTER XLIII.

HE TESLA MECHANICAL AND ELECTRICALSCILLATORS.

N the evening of Friday, August 25, 1893, Mr. Tesla

elivered a lecture on his mechanical and electricalscillators, before the members of the Electrical Congress,

n the hall adjoining the Agricultural Building, at theWorld's Fair, Chicago. Be r sides the apparatus in theoom, he employed an air compressor, which was driveny an electric motor.

Mr. Tesla was introduced by Dr. Elisha Gray, and begany stating that the problem he had set out to solve was toonstruct, first, a mechanism which would producescillations of a perfectly constant period independent of he pressure of steam or air applied, within the widestmits, and also independent of frictional losses and load.econdly, to produce electric currents of a perfectly onstant period independently of the working conditions,nd to produce these currents with mechanism whichhould be reliable and positive in its action withoutesorting to spark gaps and breaks. This he successfully ccomplished in his apparatus, and with this apparatus,

ow, scientific men will be provided with the necessariesor carr in on investi ations with alternatin currents





pproximately 20 pounds at the rate of about 80 perecond and with a stroke of about f- inch, but by hortening the stroke the weight could be vibrated many undred times, and has been, in other experiments.

o start the vibrations, a powerful blow is struck, but the

djustment can be so made that only a minute effort isequired to start, and, even without any special provision

will start by merely turning on the pressure suddenly.he vibration being, of course, isochronous, any change of ressure merely produces a shortening or lengthening of he stroke. Mr. Tesla showed a number of very clear

rawings, illustrating the construction of the apparatusrom which its working was plainly discernible. Specialrovisions are made so as to equalize the pressure withinhe dash pot and the outer atmosphere. For this purposehe inside chambers of the dash pot are arranged toommunicate with the outer atmosphere so that no

matter how the temperature of the enclosed air mightary, it still retains the same mean density as the outertmosphere, and by this means a spring of constantgidity is obtained. Now, of course, the pressure of thetmosphere may vary, and this would vary the rigidity of he spring, and consequently the period of vibration, and

his feature constitutes one of the great beauties of thepparatus; for, as Mr. Tesla pointed out, this mechanicalystem acts exactly like a string tightly stretchedetween two points, and with fixed nodes, so that slighthanges of the tension do not in the least alter the periodf oscillation.

he applications of such an apparatus are, of course,



umerous and obvious. The first is, of course, to producelectric currents, and by a number of models andpparatus on the lecture platform, Mr. Tesla showed how his could be carried out in

ractice by combining an electric generator with his

scillator. He pointed out what conditions must bebserved in order that the period of vibration of •thelectrical system might not disturb-the mechanicalscillation in such a way as to alter the periodicity, but

merely to shorten the stroke. He combines a condenserwith a self-induction, and gives to the electrical system

he same period as that at which the machine itself scillates, so that both together then fall in step andlectrical and mechanical resonance is obtained, and

maintained absolutely unvaried.

sText he showed a model of a motor with delicate

wheelwork, which was driven by these currents at aonstant speed, no matter what the air pressure appliedwas, so that this motor could be employed as a clock. He

lso showed a clock so constructed that it could bettached to one of the oscillators, and would keepbsolutely correct time. Another curious and interestingeature which Mr. Tesla pointed out was that, instead of ontrolling the motion of the reciprocating piston by

means of a spring, so as to obtain isochronous vibration,e was actually able to control the mechanical motion by he natural vibration of the electro-magnetic system, ande said that the case was a very simple one, and was quitenalogous to that of a pendulum. Thus, supposing we hadpendulum of great weight, preferably, which would be



maintained in vibration by force, periodically applied ;ow that force, no matter how it might vary, although it

would oscillate the pendulum, would have no control overs period.

Mr. Tesla also described a very interesting phenomenon

which he illustrated by an experiment. By means of thisew apparatus, he is able to produce an alternatingurrent in which the E. M. F. of the impulses in oneirection preponderates over that of those in the other, sohat there is produced the effect of a direct current. Inact he expressed the hope that these currents would be

apable of application in many instances, serving as directurrents. The principle involved in this preponderating E.

M. F. he explains in this way: Suppose a conductor ismoved into the magnetic field and then suddenly withdrawn. If the current is not retarded, then the work

erformed will be a mere fractional one; but if the current

retarded, then the magnetic field acts as a spring.magine that the motion of the conductor is arrested by he current generated, and that at the instant when ittops to move into the field, there is still the

maximum current flowing in the conductor; then thisurrent will, according to Lenz's law, drive the conductorut of the field again, and if the conductor has noesistance, then it would leave the field with the velocity itntered it. Now it is clear that if, instead of simply epending on the current to drive the conductor out of he field, the mechanically applied force is so timed that itelps the conductor to get out of the field, then it might

eave the field with higher velocity than it entered it, and



hus one impulse is made to preponderate in E. M. r. overhe other.

With a current of this nature, Mr. Tesla energizedmagnets strongly, and performed many interesting

xperiments bearing out the fact that one of the current

mpulses preponderates. Among them was one in whiche attached to his oscillator a ring magnet with a small airap between the poles. This magnet was oscillated up andown 80 times a second. A copper disc, when inserted

within the air gap of the ring magnet, was brought intoapid rotation. Mr. Tesla remarked that this experiment

lso seemed to demonstrate that the lines of flow of urrent through a metallic mass are disturbed by theresence of a magnet in a manner quite independently of he so-called Hall effect. He showed also a very nteresting method of making a connection with thescillating magnet. This was accomplished by attaching to

he magnet small insulated steel rods, and connecting tohese rods the ends of the energizing coil. As the magnet

was vibrated, stationary nodes were produced in the steelods, and at these points the terminals of a direct currentource were attached. Mr. Tesla also pointed out that onef the uses of currents, such as those produced in his

pparatus, would be to select any given one of a numberf devices connected to the same circuit by picking out theibration by resonance. There is indeed little doubt that

with Mr. Tesla's devices, harmonic and synchronouselegraphy will receive a fresh impetus, and vastossibilities are again opened up.

Mr. Tesla was very much elated over his latest



chievements, and said that he hoped that in the hands of ractical, as well as scientific men, the devices describedy him would yield important results. He laid specialtress on the facility now afforded for investigating theffect of mechanical vibration in all directions, and alsohowed that he had observed a number of facts inonnection with iron cores.

he engraving, Fig. 312, shows, in perspective, one of theorms of apparatus used by Mr. Tesla in his earliernvestigations in this field of work, and its interioronstruction is made plain, by the sectional view shown in

ig. 313. It will be noted that the piston P is fitted into theollow of a cylinder c which is provided with channel portso, and i, extending all around the inside surface. In thisarticular apparatus there are two channels o o



or the outlet of the working fluid and one, i, for the inlet.he piston P is provided with two slots s s' at a carefully etermined distance, one from the other. The tubes T T

which are sere wed into the holes drilled into the piston,stablish communication between the slots s s' andhambers on each side of the piston, each of thesehambers connecting with the slot which is remote from. The piston P is screwed tightly'on a shaft A

MECHANICAL AND ELECTRICAL OSCILLATORS.

91

which passes through fitting boxes at the end of theylinder c. The boxes project to a carefully determinedistance into the hollow of the cylinder c, thusetermining the length of the stroke.

urrounding the whole is a jacket j. This jacket acts chiefly o diminish the sound produced by the oscillator and as aacket when the oscillator is driven by steam, in whichase a somewhat different arrangement of the magnets ismployed. The apparatus here illustrated was intendedor demonstration purposes, air being used as mostonvenient for this purpose.



magnetic frame .M M is fastened so as to closely urround the oscillator and is provided with energizingoils which establish

IG. 313.

wo strong magnetic fields on opposite sides. Themagnetic frame is made up of thin sheet iron. In thentensely concentrated field thus produced, there are

rranged two pairs of coils H H supported in metallicrames which are screwed on the shaft A of the piston andave additional bearings in the boxes B B on each side.he whole is mounted on a metallic base resting on two

wooden blocks.

he operation of the device is as follows: The workinguid being admitted through an inlet pipe to the slot i and



he piston being supposed to be in the position indicated,is sufficient, though not necessary, to give a gentle tap

n one of the shaft

nds protruding from the boxes B. Assume that themotion imparted be such as to move the piston to the left

when looking at the diagram) then the air rushes throughhe slot s' and tube T into the chamber to the left. Theressure now drives the pie-ton towards the right and,wing to its inertia, it overshoots the position of quilibrium and allows the air to rush through the slot snd tube T into the chamber to the right, while the

ommunication to the left hand chamber is cut off, the airf the latter chamber escaping through the outlet o on the

eft. On the return stroke a similar operation takes placen the right hand side. This oscillation is maintainedontinuously and the apparatus performs vibrations fromscarcely perceptible quiver amounting to no more than l

f an inch, up to vibrations of a little over | of an inch,ccording to the air pressure and load. It is indeednteresting to see how an incandescent lamp is kepturning with the apparatus showing a scarcely erceptible quiver.

o perfect the mechanical part of the apparatus so thatscillations are maintained economically was one thing,nd Mr. Tesla hinted in his lecture at the great difficultiese had first encountered to accomplish this. But toroduce oscillations which would be of constant period

was another task of no mean proportions. As already ointed out, Mr. Tesla obtains the constancy of period inhree distinct ways. Thus, he provides properly calculated



hambers, as in the case illustrated, in the oscillator itself ;r he associates with the oscillator an air spring of onstant resilience. But the most interesting of all,erhaps, is the maintenance of the constancy of oscillationy the reaction of the electromagnetic part of theombination. Mr. Tesla winds his coils, by preference, forigh tension and associates with them a condenser,

making the natural period of the combination fairly pproximating to the average period at which the piston

would oscillate without any particular provision beingmade for the constancy of period under varying pressure

nd load. As the piston with the coils is perfectly free to

move, it is extremely susceptible to the influence of theatural vibration set up in the circuits of the coils H H.he mechanical efficiency of the apparatus is very highwing to the fact that friction is reduced to a minimumnd the weights which are moved are small; the output of he oscillator is therefore a very large one.

heoretically considered, when the various advantageswhich Mr. Tesla holds out are examined, it is surprising,onsidering the simplicity of the arrangement, thatothing was done in this

irection before. No doubt many inventors, at one time orther, have entertained the idea of generating currents by ttaching a coil or a magnetic core to the piston of a steamngine, or generating currents by the vibrations of auning fork, or similar devices, but the disadvantages of uch arrangements from an engineering standpoint muste obvious. Mr. Tesla, however, in the introductory emarks of his lecture, pointed out how by a series of



onclusions he was driven to take up this new line of work y the necessity of producing currents of constant periodnd as a result of his endeavors to maintain electricalscillation in the most simple and economical manner.

95

High Potential.—Continued.

zone, Production of 171

hosphorescence 367

hysiological Effects... 162, 394

Resonance 340

pinning Filament . 168 Streaming Discharges of

High Tension Coil.. ..155, 163 Telegraphy without Wires46 Impedance, Novel Phenomena.

94, 338

mprovements in Unipolar Generators 465

mproved Direct Current Dynamos and Motors 448

nduction Motors 92

nstitution Electrical Engineers

ecture 189



amps and Motor operated on

Single Wire 330

amps with Single Straight

iber 183

amps containing only a Gas.. 188 Lamps with Refractory utton 177, 239, 360

amps for Simple Phosphorescence... 187, 282, 364

ecture, Tesla before :

merican Institute Electrical Engineers 145

Royal Institution 124

nstitution Electrical Engineers 189

ranklin Institute and National Electric Light Association94

lectrical Congress,Chicago 486 Lighting Lamps Throughhe

ody 359

ight Phenomena with High

requencies. 349

uminous Effects with Gases at



ow Pressure 368

Magnetic Lag " Motor 67

Massage" with Currents of

High Frequency 394

Mechanical and Electrical Oscillators 486

Method of obtaining Direct from Alternating currents 409

Method of obtaining Difference of Phase by Magnetichielding 71

Motors :

With Circuits of Different

Resistance 79

With Closed Conductors... 9 Combination of Synchronizingnd Torque 94

With Condenser in Armature Circuit 100

With Condenser in one of

he Field Circuits 106

With Coinciding Maxima of Magnetic Effect in Armature

nd Field 83



With "Current Lag " Artificially Secured 58

arly Phase 477

With Equal Magnetic Energies in Field and Armature 81

r Generator, obtaining Desired Speed of 36

mproved Direct Current... 448

nduction 92

Magnetic Lag " 67

No Wire" 235

With Phase Difference in Magnetization of Inner anduter Parts of Core.. 88 Regulator for Rotary Current 45

ingle Circuit, Self-starting

ynchronizing .... 50

ingle Phase 76

With Single Wire to Generator 234, 330

ynchronizing 9

hermo-Magnetic 424

Utilizing Continuous Current Generators .. 31

National Electric Light Association Lecture 294



No Wire " Motor 285

bservations on the Eye. 294

il, Condensers with Plates in . 418 Oil Insulation of

nduction Coils

73, 221

scillators, Mechanical and Electrical... .. 486

zone, Production of 171

henomena Produced by Electrostatic Force 319

hosphorescence and Sulphide

f Zinc 367

hysiological Effects of High

requency 162, 394

olyphase Systems 26

olyphase Transformer 109

yromagnetic Generators 429

Regulator for Rotary Current

Motors 45



Resonance, Electric, Phenomena of 340

Resultant Attraction " 7

Rotating Field Transformers... 9

Rotating Magnetic Field .9

Royal Institution Lecture 124

cope of Lectures 119

ingle Phase Motor 76

ingle Circuit, Self-Starting

ynchronizing Motors 50

pinning Filament Effects 168

treaming Discharges of High Tension Coil 155, 163

ynchronizing Motors 9

elegraphy without Wires. ... 346 Transformer withhield between Primary and Secondary 113

hermo-Magnetic Motors 424

homson, J. J., on Vacuum

ubes 397,402, 406

Thomson, Sir W., Current Ac-



umulator 471

ransformers :

lternating 7'

Magnetic Shield 113

olyphase, 109

Rotating Field 9

ubes :

oated with Yttria, etc 187

oated with Sulphide of

inc, etc 290, 367

Unipolar Generators 465

Unipolar Generator, Forbes,468, 474

ttria, Coated Tubes 187

inc, Tubes Coated with Sulphide of 367



teh



000

••I •

61 176 1 b J

(9 ^^^X:



OFCAlIF'



?

^



The Inventions, Researches and Writings - Martin, Thomas Commerford 1856-1924

Documents