KIWI MARK II - Model EngineerKIWI MARK II Continued from I5 September 1960, popes 32to 324 2 By Edgar T. Westbury Machining the crankshaft and other internal work-ing parts of the

KIWI M A R K IIContinued from I5 September 1960, popes 322 to 324

B y E d g a r T . W e s t b u r y

Machining the crankshaftand other internal work-ing parts of the famouslittle petrol engine nowredesigned for the ’60s

TH E crankshaft’ can be madef rom so l id s t ee l ba r o rfabricated by brazing the

crankpin and main journals intothe webs, according to which isthe more convenient. Both methodshave been described so many timesin ME that it is hardly necessaryto deal with fhem in detail. Formachining from the solid, rect-angular steel bar, 1-3/4 in. x 1/2 in.can be used; these are deadmeasurements which do not allowfor cleaning up the sides or endsof the webs, but the dimensionsare not critical, and no harm willbe done if they finish slightly under-size.

While it would be an advantage touse high tensile steel for the crank-shaft, there is always some difficultyin obtaining it,, and mild steel, asspecified, will give satisfactory resultsfor all except the most exacting duties.Even this is not always readilyavailable in the sizes required; Ihad to cut a piece from a slab of1/2 in. plate. It is advisable to normalisethe steel before machining, by heatingto dull red and cooling slowly.

Some constructors may prefer tomachine the crankshaft from roundbar, with the balance weights integralwith the webs. While this is quitesatisfactory, it involves a good dealmore machining and also waste of

metal. Flat. bar is more economicalin both respects, and a good deal ofthe unwanted metal can be cut awaybefore machining the ‘crankpin andjournals. Marking out of the mainand crank centres, and subsequentcentre-drilling, call for the utmostcare to ensure parallel alignment.The bar should be set up on a surfaceplate with suitable packing,. andpreferably clamped so that it i s notliable to move while marking thecentre lines on both end faces with ascribing block (or surface gauge,according to where you were educated).

1 always recommend machiningthe crankpin first, before cuttingaway metal from the sides of thejournals; the gap can be cut out bydrilling a row of small holes adjacentto the pin and sawing down the sidesof the web; allowance, in each case,being left for cleaning up in themachining operations. It is necessaryto use a tool with a long reach (notless than 1-1/4 in.) to clear the fulldepth of the web, but it should notbe wider than necessary to ensurerigidity, and both sides should havecutting clearance, with a slight tapertowards the back, and the frontcorners slightly rounded. In previousarticles I have described a toolgrooved or bifurcated on the end faceso that it acts virtually as a double-round nosed tool; this relieves cuttingload and reduces the tendency tospring or chatter. Parallel accuracy

and a good tool finish are essential.The surplus metal may now be cut

away from the sides of the journals,and the shaft mounted on its maincentres for the remainder of themachining. To avoid any tendencyto spring, a gap piece should be madea neat push fit between the webs, andclamped in position by any convenientmeans. It is advisable to screwcutthe threads at both ends of thejournals, and all shoulders shouldhave a slight fillet, rather than a sharpcomer, to avoid a possible focus forstarting cracks.

For a brazed crankshaft, the websmay either be made parallel or withintegral balance weights (by makingthem in the form of discs and cuttingaway the part not required). Someconstructors may prefer to make themain journals in one continuouspiece and cut out the gap afterbrazing; I generally make themseparate, with flanges which buttsolidly against the webs, and increasethe surface area of the brazed joint.Oversize journals, which can bemachined between centres after braz-ing, are recommended, but it ispossible to avoid the more awkwardjob of machining the crankpins, ifcare is taken in brazing to avoidsuperfluous metal and the formationof scale. I have not found it ad-vantageous to pin the joints beforebrazing.

One advantage of a fabricated

TAP 6 B.A & PLUG

I n”-a- G&!k_____ 4X

CRANKSHAFT 1 OFF M.S.

MODEL ENGINEER 386

SECTION KK

29 SEPTEMBER 1960

GUDGEON PIN

1 OFF M.S. C.H..A. ALLEN CAPREWS FITTED

shaft is that high tensile steel-morereadily obtainable in small sizes-may be used for the journals, andmore particularly for the crankpin.Do not use high carbon steel (suchas silver steel) which is liable tobecome brittle under shock andvibration. Welding, except possiblyelectric resistance welding, is notrecommended for crankshafts.

Some model petrol engines havemanaged to get by with very sketchylubrication of the so-called splashvariety, in which a certain quantityof oil is injected into the crankcaseat indeterminate intervals and left tofind its way into the bearings as bestit can. But it is obviously much betterto provide some positive means ofsupplying oil direct to the mostheavily loaded bearings. Only thebig end bearing needs to be fed inthis way, as the excess which escapesfrom the ends is flung all over thecylinder walls and other interiorworking parts.

The drilling of the oil passages maybe considered difficult or tedious, butit is not necessarily so if sharp drillsand high speed are used; in the caseof a brazed shaft, the parts can bepre-drilled before assembly, but theyshould be properly cleared afterwards.Short 6BA grubscrews can be usedto plug the holes where required,and the entry hole to the mainjournal should be located as shownso that it lines up with the feed inletat TDC; the opening period of thisport is increased by cross-tiling witha small round file so that it embracesabout a 60 deg. chord of the journal.

Balance weightsMade in bronze as specified, the

balance weights are just about theright size to give reasonably goodbalance in conjunction with the otherworking parts also specified, but ifchanges are made, the difference inthe specific gravity of the metalsshould be allowed for. To ensurethe secure attachment of the weights,the grooves should fit closely overthe sides of the crank webs, and thescrews, of good quality steel, shouldfit both the threads and the counter-bore of the webs, into which the headsare sunk flush. When screwed home,an indent with a centre punch, ad-

29 SEPTEMBER 1960

CONNECTING ROD I OFF BRONZE OR

2 O F F BRONZEBALANCE WEIGHT

jacent to each end of the slot in thescrew head, will produce a burr whichfirmly locks it against unscrewing.The attachment of balance weightsin this way has often been criticised,but though many Kiwi engines havebeen built, I have never heard of onecoming to any harm through thisweakness. After fitting, final machin-ing of the weights may be carried outin situ.

Connecting rodBy making the rod in the form of a

bronze casting, the amount of machin-ing work on it is reduced to the mini-mum. Though bronze is not idealfor this component, experience hasshown that it gives very satisfactoryresults for general purposes, exceptin the hands of users who think itgreat fun to rev. engines flat out withno load-a “test” which few con-necting rods of any kind couldreasonably be expected to stand. Thealternative materials are steel (whichwould need to be bushed at both ends)or duralumin Tin which bushing isoptional); the rod would have to bemachined from the solid.

Assuming that a cast, or other formof semi-finished rod is used. the can ofthe big end bearing is normally madeintegral with the rod for convenience,and must be separated by sawing.Before doing this, it is a good policyto mark out and drill the holes forthe setscrews, to ensure their correctposition and alignment. One sideof the bolting lug on both the cap

387

and the rod should be marked toensure correct assembly. After thecav has been sawn off. the cut surfacesshould be accurately faced by machin-ing or filing, and the two partsfastened together by temporary screws.

The simplest way to ensure thatthe eyes of the rod are bored parallelto each other is to clamp the shankof the rod crosswise to a flat bar.with the ends overhanging each side;and set this up on the faceplate tocentralise each of the eyes in turn.As there may be some roughness onthe edges of the fluted shank, it willprobably be advisable to trim themwith a file, and to check that the rodis parallel to the faceplate whenclamped. The little end will need tobe centre-drilled, followed by anundersize drill, then a boring tooland finallv a reamer: care should betaken not to force the pace to causerisk of distortion or shifting the work.Only boring and reaming is necessaryfor the big end.

As the cap screws will be very heavilystressed in working conditions, theyshould be of good quality steel.Allen screws are ideal except that it isdifficult, in the limited space available,to provide adequate means of lockingthem. I have seen screws of this typewith cross-drilled heads which enablethem to be wired for security, but Ihave been unable to obtain them inthe size required, and owing to theirtoughness, it is very difficult to drillthem, though I have done it success-fully by using drills made from

MODEL ENGINEER

dental burrs, which are exceptionallyhard. I need hardly point out thatif these screws should loosen whilethe engine is running they may causea major disaster; this has happenedto several engines which have cometo my notice.

PistonAluminium alloy is generally pre-

ferred for petrol engine pistons,especially when the maximum speedis required. Apart from being light,they conduct heat from the combustionhead much better than cast ironpistons, though these have their ownadvantages-their wearing propertiesare better, and their expansion is muchlower, so that piston slap can beeliminated and it is possible to fitthem to a sufficiently-line clearanceto dispense with the need for pistonrings.

For a really hard slogging engine,running at speeds not exceeding3,000 to 4,000 r.p.m., a cast ironpiston would be preferable, but Ihave found it very difficult to obtaingood castings in iron for these smallcomponents, and it will probably benecessary to machine them from the

PISTON 1 OFF

L.A. OR C.I.CLEARANCES L.A. C.I. ,

TOP LAND .006" .002"

RING BELT ‘004. .OOlSKIRT ‘003” ‘ooo5*

solid. I have in the past describedmethods of machining pistons, inwhich all operations can be carriedout before parting the piston offfrom the parent bar.

If a cast piston in either metal isavailable, the machining is con-siderably simplified. I do not recom-mend the use of a chucking piece ona piston casting (although this is thecommon practice) for reasons which

The piston may now be set up onan angle plate, and held theretoeither by two clamps, or by a singlestrap and two bolts. for boring thegudgeon pin hole. Before attaching it,a line should be scribed squareacross the angle plate from front toback (this may be deeply incised forpermanence, as it will always beuseful) and the centre lines on thepiston located to coincide with theline on both sides. This will ensure

MODEL ENGINEER 388

After rough machining, the piston i s set up on theangle plate for the gudgeon pin hole to be bored

I have already explained. I usuallyhold the pistons in the four-jaw chuckby the head end, with as much of itslength projecting as is discreetlypracticable, and set it up true by thecored interior surface-not the out-side. The use of a feeler, such as abent scriber held in the toolpost,will help to check the truth of thepart above the gudgeon pin bosses.The outside can then be turned downto about l/32 in. oversize, the end ofthe skirt faced to length, measurementbeing taken from the bosses on theinside, and a witness cut taken in themouth of the skirt, just sufficient toclean it up. A point tool is used tolocate the distance of the gudgeonpin from the end of the skirt, and thecross centre of the bosses can bemarked out with a scribing block setexactly at lathe centre height. Thework i s rotated so that the bosseslie as near horizontal as possible,and the marked line is produced onthe end face and both sides of thepiston, to intersect the lateral linealready marked by the point tool;these intersections are centre-punchedand witness circles 1/4 in. dia. aremarked with dividers: For theseand other delicate marking out jobs,I recommend the use of Spectramarking fluid; it is easily applied,dries almost immediately, and makesscribed lines stand out so clearly asto eliminate all indecision.

that the hole will pass squarelythrough the centre of the diameter,when the work is set up by shiftingthe assembly on the faceplate tocentralise the visible punch mark,as accurately as possible.

The piston mav now be centre-drilled, and then drilled about 3/16 in.dia. through the first boss. In caseof any possible inaccuracy, this holemay be opened out with a smallboring tool. A drill large enough toopen it just a little further-but stillunder 1/4 in. is then put through,with ample lubrication, to followthrough into the other boss, withcaution. Both bosses can then bebored out and reamed, preferablyon the tight side compared with thelittle end-eye of the connecting rod.

Final turning of the outside of thepiston, including the ring grooves,and facing the crown, may now becarried out by a method which relievesit of all chucking stress. A suitablepiece of metal-brass, steel, or lightalloy-is held in the lathe chuck,and a spigot is turned on it to fitthe inside of the piston; it is alsocentrally drilled and tapped about1/4in. dia. Next, an eye bolt, or ascrew with a large cross-drilled head,is made to fit the tapped hole. Adummy gudgeon pin, slightly lessthan 1 in. long, is passed through thepiston and the eye bolt, and byscrewing it up the piston is drawnup firmly and truly on to the spigot.

All the remaining work can nowbe carried out ; the recommendedclearances are shown on the detaildrawing. This method is similar tothat almost universally employed inlarge scale piston manufacture, thougha more elaborate, permanent jig,with a long drawbolt, sometimespneumatically operated, is generallyfitted to the lathe.

* To be continued October 13

29 SEPTEMBER 1960

KIWI M A R K IIBy Edgar T. Westbury Continued from 39 September 1960, papes 386 to 388

MOUNTING THE FLYWHEEL

with a SPLIT COLLET

TH E gudgeon pin, detai ls ofwhich were shown in theprevious article in this series,

is quite a simple machining job.It may be made from a piece of1/4 in. bright mild steel, which willnot need external machining if ithas a good surface and is a tightpush fit; otherwise larger material,machined all over, must be used.

The centre hole should be drilledas truly as possible; the brass oraluminium end pads, the object ofwhich is to prevent scoring of thecylinder walls if the pin moves end-wise, should be tightly fitted, afterthe pin is case-hardened and polished;and the overall length over the padsshould be a few thou less than thecylinder bore diameter.

As flywheels of large diameter aregenerally undesirable for engines in-stalled in boats, owing to the necessityof keeping the shaft line low, the oneshown has been kept small, consistentwith smooth running at the slowestspeed likely to be required. There issome advantage to be gained byincreasing the size of the flywheelwhere it is permissible to do so,especially for a really slow tick-over.

The recommended procedure formachining the flywheel is first to

chuck it pulley-side outwards andrough machine all the external surfacesavailable, including the V-groove,_ towithin about l/32 in. of finished size.It is then reversed, and chucked astruly as it can be by reference to thesurfaces already dealt with, formachining the outer rim, the back face,and recess. The centre hole is centred,drilled and taper bored with greatcare to set the boring tool exactly atcentre height and to produce anaccurate finish. A tapered mandrel,turned in situ, is recommended for thefinal finishing of the pulley grooveand other adjacent surfaces. The sizeand angle of the groove is of someimportance in ensuring a good gripfor the starting cord if the conventionalmode of starting is emnloved: if it isnot, the groove may be suitablymodified.

I may be asked to explain the useof a split collet for mounting theflywheel, instead of the commonermethod of tapering the shaft itself,as in the original Kiwi. There areseveral reasons why I have adoptedthe split collet in most of my laterengines. First, it enables the shaftitself, including the thread on the end,to be kept larger at this vital pointand, therefore, more robust. Secondly,in the event of a serious wrench which

INED ALL OVER

FLYWHEE’L I OFF C.l. or MS.

MODEL ENGINEER 458

may cause scoring or tearing on thetaper, the collet is easily renewed.Thirdly, it enables end play in themain bearing to be eliminated, eventhough the. flywheel itself is notpositively end-located. If end thrustis applied to the shaft (this is notgenerally desirable, but cannot alwaysbe avoided) a hardened thrust washer,or even a ball thrust race, could befitted behind the collet.

Matching the taperThe collet may, if desired, be

machined with the large end outwards,retaining the setting of the top slideas used for boring the flywheel, sothat the exact angle of taper is assured.It is a little more difficult to assessthe diameter in this way as the fly-wheel cannot be offered up for fitting;but by taking a careful caliper measure-ment of the mouth of the taper bore,and making the extreme end of thecollet about 3 thou larger than this,-the correct’ amount of draw will beobtained. The bore of the colletshould be made a press fit on thecrankshaft, and its mouth radiusedor chamfered to clear the fillet andenable it to abut closely against theshoulder. After slitting either byhand, or by a circular saw in thelathe, you should carefully removeany burrs formed both inside andoutside.

The directions which have beengiven for making similar componentsfor previous engines hold equallycorrect for valves and valve guides.Dimensions are the same, whetherthe water-cooled or air-cooled cylinderheads are fitted. Note that with water-cooled heads the valve guide mustprovide a watertight joint on the top

F L Y W H E E L C O L L E T _I OFF M.S,

13 OCTOBER 1960

of the cover plate, and therefore,should be a good tight fit in the head.The shoulder should also abut firmlyagainst the cover plate. When it isinserted, a little varnish or jointingcompound should be applied to thesesurfaces.

Suitable material for the valvescan be obtained from old motor carvalves or aircraft bolts, as discussedin previous articles. Wherever possible,machining of the head and stemshould be carried out at one settingto ensure perfect concentricity. Thefit of the stems in their guides shouldbe as close as is consistent with freeworking; this is particularly importantwith the inlet valve, where a sloppyfit may cause an air leak seriousenough to affect carburation.

The valve collar should also fitnicely on the stem so that, when the

nutted mandrel for filing or machiningthe outer contour. The individualpieces are then mounted each in turnon the mandrel for machining thesides; note that the bosses are offset,right-handed and left-handed respec-tively. A round-nosed tool withplenty of side clearance is used toreduce the web of the rocker, from the5/16 in. dia. of the boss, to a radius of15/32 in. on both sides.

To produce the 3/32 in. recess inthe heel of the rocker, a round-endedflat drill may be made from 1/8 in.silver steel, or an old twist drill maybe ground to shape. The mouth ofthe recess should be well tapered, sothat the push rod does not bind onthe sides at its maximum angularity.After the oil holes have been drilledand countersunk, the rockers arecase-hardened, special care being

V A L V E G U I D E _2 OFF BRONZE

%6 KEYWAYCAMSHAFT 1 OFF

horseshoe collar is snapped into itsrecess, there is no tendency for it to tiltand thereby force the valve out of truth.

Valve rockers are highly stressedparts, which some constructors findrather finicky to make; as a result,they are often made crudely andclumsily. Sometimes gunmetal castrockers are used, but their wearingproperties are not really good enough,unless hardened steel inserts are fittedat both ends. All things considered,I find it just as easy to machine themfrom solid steel.

Solid steel rockersTake pieces of mild steel bar,

3/8 in. x 1/4 in. x 1-3/8 in. (+) long. Thecentre pivot holes are first drilled andreamed in each piece, and the piecesare mounted together on a bolt or

13 OCTOBER 1960

Here is the air-cooled KIWIMark II

in its proper place; allowance beingmade for a shakeproof washer if itis available.

It is important that the pillar shouldnot loosen so as to shift in workingconditions. If this trouble is en-countered, various means of lockingit are possible, including drilling andtapping a hole horizontally in the headfor a grubscrew to penetrate and lockthe screwed end; but this has not beenfound necessary. The original Kiwihad a locknut instead of a squaredshoulder on the shank. Thougheffective, both as a means of heightadjustment and of locking, it was arather clumsy-looking device; butyou may adopt it if you wish.

After locating the position of thesquared end of the pillar when inplace, drill and tap the stud hole.Two studs are then made from 1/8in.

taken to ensure that the bore of thepivot hole, the recess, and the under-side of the toe, get full treatment;these surfaces are finally polished.

The rocker pillar should be madefrom 3/8 in. square mild steel, machinedeither in the chuck or between centres.Be sure that the square and circularparts are concentric. Before finishingthe shoulder of the threaded end, itshould be screwed into place in thehead (with the cover plate fitted, inthe water-cooled type), to makecertain that the square will line up

459

mild steel to screw in from oppositesides, the threads being rather tightso that they are not likely to loosen.Make certain that no burrs arethrown up on the side faces of thesquare when these are screwed in.

The rocker bushes should be madeof drawn gunmetal (not brass) oralternatively of steel, and subse-quently case hardened. If you like,they may, instead of hexagonal headshave large knurled heads so that theycan be adjusted by hand. The centralhole-or, rather, the eccentric one-

MODEL ENGINEER

can be drilled by offsetting the workin the chuck before parting off; asthe distance is not critical, a piece ofstrip metal about 15 thou thick underone jaw of the three-jaw chuck willproduce the desired result.

I may say that as a result of longexperience I have found the eccentricbush method of tappet adjustment tobe not only the simplest but also themost reliable for small engines.Screwed tappets or push rods, withtheir essential lock nuts, are finickyto adjust and generally add to thereciprocating weight of the valveparts. To adjust tappet clearanceon the Kiwi it is only necessary toslacken the nut on the end of the

bushes, from the inside of the crank-case and timing case.

As the extension of the crankshaftnut is not intended to take a closebearing in the timing cover, it isspirally grooved to retard the escapeof oil at this point. Little more thana mere scratch is necessary,, and thepitch is not critical either. Right handis for appropriate engines runninganti-clockwise at the timing end(as is usual for convenience in usingscrewed shaft couplings) and forrunning in the reverse direction, aleft-hand spiral should be applied.This nut could be combined with adriving dog or other form of coupling.

The water flanges are, of course,

these are more readily obtainable.But don’t do as one Kiwi constructordid some years ago, that is, fit gearsof 19 and 40 teeth under the impressionthat they would be near enough--Lhad quite a headache sorting outwhy his engine wouldn’t work !

The pinion should be of steel, butthe spur wheel may be of any othersound metal, as the width of gearface is adequate to cope with workingconditions

It is also possible to vary the meansof securing the gears on their shafts;the pinion, being limited in diameter,is hardly suitable for keying in thenormal way, but as it takes littletoraue load when clamped by tne

pivot stud, to turn the bush, eitherby hand or by a small spanner, asrequired, and then clamp it up again.The valves should, of course, be inthe closed position while this IS done,and a 2 thou feeler gauge or a slipof thin paper may be used to checkthe clearance.

The remaining small details on thedrawing published here are straight-forward, and do not call for muchexplanation. Note that the twocamshaft bushes are of differentlengths, the longer being fitted tothe timing cover and projectingoutside in order to carry the contactbreaker. Both are fitted to about1 thou interference. When the coveris assembled, a reamer may bepassed through both of them toensure perfect alignment. Oil holesare drilled at an angle into both

MODEL ENGINEER

ROCKER PILLAR I OFF M.S.

D I A . E C C E N T RROCKER BUSH2 OFF G.M.

CAMSHAFT BUSH BRONZE

WATER FLANGE 2 OFF BRASS TAPPET 2 OFF_ M.S .C.H.

only necessary for the water-cooledtype of engine, and are intended tobe silver soldered or sweated to theservice pipes. In order to keep downhead room, the outlet will probablyneed a sharp bend or elbow, and theflange may be modified to facilitatethe change in direction, if wanted.

Some latitude is permissible in thespecification of the timing gear. Asthe obvious requirement is that thecamshaft must run at half enginespeed, it is only necessary to ensurethat the gear on this shaft must havetwice as many teeth as the pinionon the crankshaft; but if this entailsany alteration in the pitch diametersof the gears, the distance between theshaft centres must also be varied.

Practical alternatives to the 40 d.p.gears shown are 36 d.p. (18 and 36teeth) or 32 d.p. (16 and 32 teeth), if

460

shaft nut, a 1/16 in. pin or “ snug key ”sunk into the shaft, to engage withthe notch ‘in the boss, will be quitesufficient. Its position may be locatedwhen the complete engine is assembledand timed.

It is desirable, though not absolutelyessential, that the keyway in the spurgear should line up with either atooth or a tooth space, but this willbe explained further when dealingwith the cams.

Incidentally, when turning the cam-shaft. it should be noted that thefront extension of the journal issubject to variation according to thetype of contact breaker and its camfixing; it is best to leave the extensionwell on the long side until the typeof contact breaker has been deter-mined.

* To be continued on October 27

13 OCTOBER 1960

By Edgar T. Westbury Continued from 13 October 1960. pages 458 to 466

TAKE CARE WITH THE CAMS

AMONG the many details whichaffect the success and effi-ciency of an engine, few

are of greater importance than thedesign and production of thevalve-operating cams. They havebeen the subject of much dis-cussion in the past. Many in-experienced constructors have beenconfused by the conflicting ideasexpressed, and have been led tothink that it is very difficult toform and time cams properly.Judging by the letters I receiveeither querying or debating thefiner points of cam design, theremust be many readers who aredeterred from building four-strokeengines by real or imaginarydifficulties in this respect.

Where only moderate speed isrequired, the duty on the valve gearis relatively light, and almost anythinggoes in the matter of cam design.Even in manufactured engines, scienceis often subjugated to convenience inmachining the cams; every enginetuner knows that standard maker’scams are nearly always capable ofsome improvement. In model engines,I have seen some very crude effortsin cam design which have producedmore or less successful results. But

in engines intended for high speed andperformance, cam design is of theutmost importance, and any errors inthe contour and timing angles in-evitably result in loss of mechanicalefficiency, calling for excessively strongvalve springs which still furtherincrease general wear and tear.

There are, of course, as many waysof producing cams as there are ofexterminating a felis domesticus, andI have described most of them atvarious times in ME. Any of thesemethods may be applied to theKiwi, an engine which may be claimed

to be specially adaptable. In itsoriginal form, it was used as a testbench for various cams and valvetiming, though in the publisheddesign the cams specified were selectedmainly with a view to simple produc-tion rather than maximum efficiency;nevertheless, they worked quite wellwithin their limits.

These cams were of the tangenttype; that is, the flanks were straight-line tangents to the base circle, andfaired into the nose with a slight radius,so that they could be formed withreasonable accuracy by filing. The

1/1;: KtYWAY

CUT‘OFF AFTER FORMING & CASE HARDEN & POLISH CAMSCAM BLANK I OFF M.S.

Cam blank, with division plate attached, setup eccentrically for machining the cam flanks

In the present arrangement, I haveconsidered it better to key the camsto the shaft, not only to resist torquestress better, but also to locate themmore positively. The two cams areshown as made in one piece, to locatetheir relative angular positions-for

MODEL ENGINEER 508 27 OCTOBER 1960

followers. or tappets-which shouldalways be considered an integralfactor in the cam &sign-wererounded to a cylindrical curve on thecontact face. This combination worksout fairly well in practice, up to speedswhich were considered high enough formost purposes at the time the enginewas designed. The cams were madeseparately, and after the adjustingof their angular position on the shafttogether with that of the gearwheel,the three components were drilledin situ and pinned together by a steeldowel passing endwise through them.

the adjustment is less easy than itlooks-but they may be made separateif desired. Either the inlet or exhaustcam may be placed adjacent to thegearwheel, as it is possible to operatethe appropriate valve from eithertappet. To reverse the rotation of theengine, it is only necessary to turnthe cams end for end and change thepush rods over. Alternatively, changeover the valves themselves, by revers-ing the positions of the carburettorand exhaust pipe.

The method recommended for form-ing the cams to the contour shown in

should be scribed on the plate forlocating the angular index marks onthe division plate. The eccentricsetting to produce the flank radius of3/4 in. is 3/4 in.- 7/32in. (base circleradius) = 17/32 in. at the stud centre.

If the cams are to be keyed to theshaft as specified, the keyway shouldbe cut in the blank, and lined up asaccurately as possible with the TDCzero mark on the division plate whenthe latter is fixed. The internal key-ways in the blank, and also the gear-wheel, and the external keyway in thecamshaft, can be cut by a suitable

/ FROM TIMING END

-h-x&,!’ \ IN.

\FLA$ RAD.

the drawing is similar to that employedfor the cams of the Dolphin engine-mounting the blank eccentrically onthe faceplate, and indexing it into therequired positions for shaping eachof the flanks by a turning tool. Thesurplus metal around the base circleis then removed in a similar way orby circular milling, and the noseradius is filed by hand to blendsmoothly into the flanks.

As the method involves the necessityof fixing a division plate temporarilyto the cam blank, the detail drawingshows an extension piece for thispurpose; any method of fixing, such as by pressing, soldering, or grub-screwing, is suitable so long as theplate is secure during its period ofservice. The plate, which may bemarked out by a draughtsman’sprotractor! is a replica of the left-hand timmg diagram which is setout in camshaft angles.

To facilitate accurate mounting ofthe blank on the faceplate, a simplejig should be made, consisting of aflat steel plate with a fixed 1/4 in. stud,accurately fitting the blank centre andprovided with a nut and washer forclamping; a plain zero index mark

27 OCTOBER 1960

on the nose. with ample tov rake forclean cutting. The depth of cut forthe first flank can be measured by amicrometer (5/8 in. blank dia. - 3/32 in.lift = 17/32in.) but this is notpossible for the second flank, and soit is advisable, once the setting hasbeen found, to work to the cross-slideindex or a fixed stop.

As it is very easy to make a mistakewhen nibbling away the base circle,I always recommend marking thetips of the cams by a spot of quick-drying paint or Spectra markingfluid so as to make their identification

TIMING DIAGRAMS

TAPPET GUIDE1 OFF

BRONZE

shaping tool set at centre height inthe toolpost, and racked backwardsand forwards by the saddle movement,taking cuts of not more than 5 thoudeep at a time; this process has beendescribed several times in ME.

With the blank mounted on thefaceplate, and indexed to any one ofthe valve opening or closing positions,flank turning can begin. Use a toolnot more than 1/16 in. wide so that itcan run out into the clearance spacebetween the cams; it should berounded

509

quite certain. This advice is born ofbitter-or unfortunate-experience !After the base circles have beenmachined dead concentric with thebore, the nose radius is filed on eachcam, the important thing being toproduce a smooth change of contour,as this part of the cam does not, orshould not, do heavy work. Finally,the extension piece is machined offthe end of the blank and both camsare case-hardened and polished.

Some constructors worry about theexact amount of tappet clearance tobe allowed for, but in the interestsof simplicity I have not specified thisin the design. The main object ofworking to exact figures in full-sizeengines is to ensure that the valvetiming is always as designed, and alsoto ensure quiet operation of thetappets. If nothing is allowed fortappet clearance in the cam design,some loss of the valve opening periodis inevitable; but by keeping this onthe full side, and setting the tappetsto as fine a working clearance aspermissible, very little loss of efficiencyresults.

Constructors who wish to followestablished practice to the letter, may

MODEL ENGINEER

reduce t h e radius of the base circleby the required amount-say 0.005 in.

-and ease the contour to meet thatof the flanks at the opening and closingpoints. I trust that tbis explanationwill take care of the many querieswhich I have received on the subjectof tappet clearance.

In timing the camshaft, it is onlynecessary to locate it so that the key-way is exactly at the top (assumingthat the machining of the cams hasbeen carried out as described) whenthe piston is at top dead centre. Asa check on the accuracy of the timing,the right-hand (crankshaft) diagrammay be marked out on sheet metal orstiff card, and temporarily attached tothe flywheel end of the shaft, or eventransferred, by scribing, to the fly-wheel itself. As already mentioned,a few degrees will be lost in the open-ing periods, owing to clearance andbacklash, and it will be necessary tosplit the difference in errors so caused.

Both the tappets are housed in asingle guide, in order to take up aslittle room as possible in the timingcase, and also to avoid excessiveangularity o f the push rods. Detailsof the tappet were published in thegroup drawing in my last instalment.The guide, which is preferably madein gunmetal or bronze, may bechucked by the flange and turnedparallel to fit the bored hole in thetop of the timing case. It is thenreversed and faced at the top, thepositions of the two bores beinglocated and marked out; after whichit may be set up eccentrically eitherin tbe four-jaw chuck or in a KeatsV-angle plate or similar fixture forcentring, drilling and reaming the holes.

The tongue at the lower end ofthe guide is formed by machining orfiling a step each side so as just tocut away the hole; the object of this,of course, is to keep the T-headedtappets from rotating, as there isnot enough room to fit the usualcircular floating mushroom tappets.Make certain that the tongue is deepenough, when the parts are assembled,to fulfil this purpose when the tappetsare resting on the base circles of thecams. I have encountered caseswhere the tappet heads have droppedbelow the tongue, so that they canpartly rotate and thus lock themechanism when lifted by the cams.

The essential point in machiningthe tappets is that they should be asmooth working fit in the guide-with a sloppy or rough fit they tendto pump oil and make the engineuntidy-and that the heads should bedead flat and smooth. At the top end,a recess is provided similar to thatin the heel of the overhead rocker,

and formed by the same tool. Allworking surfaces are finally case-hardened and polished.MODEL ENGINEER

I-

BREATHER BODY I OFF BRASSASSEMBLY

DISC VALVE1 OFF TUFNOL

T.P.I.CAP 1 OFF BRASS

[,;;,rEYTf$$+ENDS ONLY HARDENED

PUSH ROD 2 OFF S ILVER STEEL

In a single-cylinder four-stroke that the simple lift valve type seemsengme, or otner type in which dis-placement of air takes place, it isdesirable to release the air pressurecaused in this way, as it not onlyresults in quite unnecessary waste ofpower but tends to force oil out of thebearings and past the piston. A plainvent is often used, but it is better tofit a non-return valve, so that slightnegative pressure is maintained in thecrankcase; this helps to keep theoutside of the engine clean and alsoassists the entry of oil when it issup lied by a gravity or suction feed.

ere is not a great amount ofroom in which to fit the breather onthe Kiwi; it is desirable that it shouldbe located in the timing compartmentso that oil mist is conveyed effectivelyto the gears and cams. I have devisedseveral types of breathers, includingsome which were mechanically oper-ated by the camshaft, but nobodyseems inclined to take the trouble tomake them, and it must be admitted

510

to do all that i s necessary.The breather shown incorporates a

disc valve, which is very light inaction and capable of operating atvery high speed. Paper-base bakelite(Tufnol or Paxolin) is suitable for thevalve, and bits of it are to be foundin old radio and electrical junk. Thebest way to make the disc is to turnup a punch, 7/32in. dia., and a diewith wrresponding bore, from anyavailable bits of steel; hardening isnot necessary except in making anumber. With the die in the chuckand the punch in the tailstock,. thediscs can be blanked out qmcklyand accurately.

Other parts of the breather aresimple and call for no special com-ment. The complete assembly isscrewed into the oblique tapped holein the timing case, as shown in thecrankcase detail drawings.

* To be continued on November 10

27 OCTOBER 1960

KIWI M A R K IIB y E d g a r T. \ Westbury

CO N S T R U C T O R S of model petrolengines are nearly alwaysreluctant to spend more

time than they need on fittingsand accessories, and particularlyon the carburettor, which is oftenausterely simple, and sometimescrude, in design. I am all in favourof simplicity where it will producethe desired results. But the con-structor must realise that mostsimple devices have their limita-tions, and also that by reducingconstruction to its simplest limita-tions one does not necessarilymake the adjustment and operationof controls equally simple-quitethe opposite often occurs.

From this aspect there is no suchthing as a simple carburettor. Detailswhich appear insignificant may haveunexpected and far-reaching effects ;the shape and area of fuel and airpassages, the pressure or suction atthe jet orifice, and the throttleposition in relation to other com-ponents, all influence working charac-teristics. In the carburettors whichI have designed and described in ME,I have sought to avoid unnecessarycomplication, and at the same timeto ensure reasonably good controland consistent performance. Butsome constructors have altered themin detail, or built near copies whoseperformance is entirely different.

Various types triedSeveral types of carburettor have

been experimentally employed on theoriginal Kiwi engine. The first onetried was the Atom Baby, castingsfor which were once marketed byTom Senior, was a simplified versionof the Atom series of carburettors,the main characteristics of which weresubmerged jets and annular diffusers,in conjunction with float feed.

In D. H. Chaddock’s brake tests ofthe Kiwi the best results were obtainedwith an Atom Mark III carburettor,but constructors at that time con-sidered its features complicated anddifficult to understand. Sometimes theyattempted to use it with suction feed,not realising that in such conditionsthe jets were no longer submerged;obviouslv. some form of constant-head feed system is a necessity incarburettors of this type, if the com-

MODEL ENGINEER 576

C a r b u r e t t o r s 1

are never simple (

pensation principle is to work asdesigned.

Thr Kiwi carburettor was designedto simplify construction and workingprinciples as much as possible. It isa fairly straightforward plain jet type,with mechanical compensation by abarrel throttle; and though float feedis specified as generally desirable, itmay be adapted to suction feed ifthis is preferred. In the Mark IIversion, n o alteration in workingprinciples or general design has beenmade, but several detail improve-ments have been introduced.

In this carburettor, the position ofthe jet tip, both laterally and vertically,is very important in obtaining goodthrottle control. When any difficultyhas been encountered, I have usuallyfound the cause to be either error ora deliberate change of the jet location.If the tip projects too far into the airpassage, it reaches a zone of higherair velocity when the throttle is wideopen, but-when the throttle is partlyclosed it may be in a much lessfavourable position. The effect ofthe primary air passing up and aroundthe jet also depends on its height, inconjunction with the area of theentry hole and the annular passage.All these details can be varied toaffect the compensation at differentthrottle openings. While the accelera-tion and load adjustment is inferiorto those of the submerged jet type,the carburettor is capable of givingquite good control over a wide rangeof speeds, and has a high flat-outefficiency, as there is little obstructionto streamline air flow.

The carburettor body has beenma& slightly longer than that of -theMark 1, and is provided with acircular flange on the intake end,to enable a choke plate, air cleaner,or trunk to be fitted if this is founddesirable. To machine the casting,it should first be held by the intakeend (a chucking piece on this endwill be provided) for facing, centre-drilling, drilling and taper boring.

Note that the parallel part of the bore,which serves as the throat of a venturiorifice, should not be larger than1/4 in. dia. for the Kiwi engine; if theengine is intended only for moderatespeed. and maximum flexibility ofcontrol, it may with advantage bereduced to 3/16 in. In the prototypeMark II carburettor I have madesome experiments in the size of thethroat, by screwing in nipples ofvarious sizes, and I find that withthe bore reduced to 3/16 in. only thepeak r.p.m. performance is affected.

Streamlined intakeTo face the intake end, and flare

out the entry, a tapered plug mandrelmay be turned and the body mountedon it. The exact shape of the flareis not critical, but it should blendsmoothly into the bore to avoidturbulence. As it is unlikely that asuitable form tool will be available,the operation may be started with acountersink or large centre-drill, andfinished with a half-round or triangularscraper.

The casting may be held on anangle plate or chucked by the rearboss for facing, drilling and boringthe throttle housing. After the backof the boss has been faced, the throttleshould be machined and fitted, onthe tight side at first, to enable thecross bore to be drilled, or at leastfinished, in position. The body maythen be re-chucked, but it should notbe drilled to finished size. or a burrmay be thrown up and make itsremoval difficult, and probably scorethe housing badly.

After boring out with light cuts,the constructor should finally fit thethrottle by lapping with powderedbath brick or other mild abrasive,so that it works quite easily. Usuallyit will be found desirable to fit alarge-headed retaining screw, with afriction washer interposed at the backof the boss, to prevent inadvertentmovement.

Note that the position of the throttle

10 NOVEMBER1960

I O F F_ CORK o r BALSA (VARNISHED) m CA

THE ”KIWI” MARK II CARBURETTOR FOR ENGINES 10 TO 20ccHORIZONTAL, VERTICAL, OR ANGULAR FITTING.

Left: Carburettor body set up on an angle plate for drilling and tapping the jet housing.Right: Float chamber set upon an angle plate for driIling and tapping the connecting boss

lever may be varied to suit controlrequirements, but where rods orcables are not fitted it is usuallyconvenient to set the lever at about30 deg. from the vertical, towardsthe outlet, at full bore. This positionmust, of course, be settled beforecross drilling. The stop screw fittedto the top of the body will limit thefull opening, and also enable theidling speed of the engine to be closelyadjusted.

The jet housing is drilled at anangle of 45 deg.. offset 5/32in. fromthe throttle centre, towards the intakeend of the body. It is advisable toset it up on an angle plate, as thetruth of the tapped hole, and of thejoint face, is highly important.

A modification has been made tothe float chamber casting to simplifyconstruction. In the earlier version.a lug was cast on it, which had to bebored and faced at 45 deg. so that,when connected to the horizontalbody, it was vertically located. Manyconstructors found the operationdifficult; the location was inaccurateand fuel leaked at the joint faces.

The Mark II float chamber has aboss which is drilled and tapped totake a banjo connection, enablingthe angle to be adjusted as required;it is possible to fit the carburettorwith the body horizontal as shown, orin the vertical (updraught) position.The vertical may be found moreconvenient if it is desired to locatethe carburettor at the lowest possibleposition, to facilitate gravity feed orcompactness of installation. This, ofcourse, necessitates fitting a right-angle bend in the induction pipe,which is sometimes iiable to causecondensation or even freezing up. Aseparately mounted float chamber,with a flexible connection to the jet,may be preferred, to avoid risk of

MODEL ENGINEER

upsetting the float action by enginevibration: and a further alternative isto omit the float chamber. and feeddirectly to the jet by suction from atank slightly below jet level. Thisgives less accurate metering than aproperly working float, but is con-sidered good enough for practicalpurposes, if the tank is not morethan about 1 in. deep, to avoid widevariation of fuel level.

Float constructionExtra depth has been provided on

the float chamber to eliminate spillingunder vibration, though I have notfound this necessary in test running.The float may be made of cork or ofbalsa, which is the more buoyant;after the pores have been stoppedwith a suitable filler, it should befinished with a petrol-resisting varnish,such as Phenoglaze, or one of thesynthetic enamels. It is well worthtaking the trouble to make either ametal foil float, or to fabricate afloat of perspex or cellulose-acetate.Those who prefer to avoid suchdelicate operations may fit a ready-made float, as used in the carburettorsof small mopeds or auxiliary engines.It is also possible to adapt the floatneedles of these carburettors; two ormore grooves should be made in themto obtain the best fuel level.

For the home-made float, a clipwill be required. While a wire onemay appear simpler to make, theflat clip shown has been found mostsatisfactory. Tbe holes may be drilled,before the clip is slotted with aminiature hacksaw, while it is tempor-arily soldered or cemented to a suit-able solid backing. The simple bentclip for holding on the float chambercover has also been arrived at as aresult of experience. There are, ofcourse, dozens of ways of effecting

578

the same purpose, but when it becomesnecessary to inspect the float at thepond-side the use of screws or otherfixings which may drop into the works,or be lost, is obviously undesirable.

The brass components are mostlystraightforward to machine. Thespherical part of the banjo may befinished by means of hand tools whileit is held by the screwed end in a nutor chucking piece. I have foundthat a hole of appropriate size drilledin sheet metal, with the unwantedpart cut away, makes a convenientradius gauge. Of course, it is notessential that the part should bespherical, but it is worth while, onthe grounds of appearance., to make itas neat in shape as possible. Aftercross drilling, the sides should befaced by mounting it on a stub mandrel.

For the float needle seating, you willfind it best to make a simple D-bitto form the taper, as it is obviouslyimportant that this should be smoothand accurate to prevent leakage. Thepart is screwed in from the inside ofthe float chamber, and the slot in thehead not only facilitates insertionbut also prevents the float fromblanking off the inflow of fuel whenthe chamber is empty,

In machining the jet tube, the im-portant point is the concentric accur-acy of the deep drilling. I have ex-plained in previous articles how toobtain satisfactory results in this kindof work, using sharp drills, andrunning the lathe at the highestpossible speed. The jet orifice ismost conveniently drilled from the topend, with the screwed part mountedin a chucking piece; the outside ofthe jet may also be machined at thesame setting.

The jet needle may be of compositel Continued on page 585

10 NOVEMBER 1960

Left: Cast gunmetal stern frame complete with rudder in position ready for the plating ofthe hull. Right: Plating begins. Short plates on the shorter spaced frames facilitate shaping

breast stay is, I will explain that itis the piece of plate that can be seenstretching across from the top of thebulwark rail; it is there to strengthenthe bulwarks.

The pin is half-way down from thetop of-the post to the breast plate.There is no hook this time; insteadwe have a small platform which is18 in. high from the deck on Moorcockand 3/8 in. on the model. At full-sizeit is a piece of chequer plate; it couldbe imitated by scratching some lineson the piece of material. The sizeof the finished step is 3/8 in. x 9/32 in.Two brackets support the step, butas no one is going to stand on oursthey can be omitted. Those who arereal sticklers for detail, with TomThumb as their inspector, can amusethemselves.

It will be remembered that MrNorman Kimbrey is building Moor-cock to 1/2 in. scale and has adaptedthe method of making his hull witha bar keel and frame. He has nowfinished his hull.

The stern frame is particularlyinteresting because it is a gunmetalcasting. Mr Kimbrey made his pat-tern to the drawing in this series andthe method has saved him time andmoney. He tells me that his mostanxious moment was in the drillingof the stem tube, for the hole is long,and extreme care must be exercisedto see that the drill does not run off.Luckily his drill came through as nearspot-on as makes no difference.

The bosses for the rudder hingewere not cast on but were silversoldered on afterwards in the manner

which I recommended for the fabri-cating of the frame. The rudder isshaped from a piece of 1/8 in. thickbrass plate and is operated exactlyas on the full-size tug.

In the other photograph, MrKimbrey is just beginning his plating.Each plate has a lapped joint facingaft, which is correct for the bow andthe stem. Amidships the platesshould be butted, but Mr Kimbreypreferred to lap them to save himselfthe bother of lapping a butted jointon the inside of the hull. No one willquarrel with him; the work will not beseen when it is in the water, and whenit is out it looks very smart. Forthose who know the tug extremely wellwill know exactly for what to look.

* To be continued

KIWI MARK II . . .

construction if desired, the head beingscrewed on and riveted, and the pointinserted in a drilled hole in theshank. A plated gramophone needleis suitable; the important thing isthat it should be both secure andconcentric. Note that the thread onthe shank should not extend into thegland, and the tapped hole in thebase should be counterbored to adepth sufficient for the needle toscrew right home.

If preferred, a plain drilled jetorifice may be used instead of theneedle-controlled type, as the jetadjustment should not need to bealtered when once it is properly set.It will be necessary to calibrate thistype of jet by trial and error, as no

10 NOVEMBER 1960

exact dimensions of the hole can bespecified. A number 80 drill-thesmallest in the number size range-should be somewhere near correct,and a fine broach may be used toopen it out if required; conversely,the bore may be reduced by swagingthe tapered end with a burnisher orroller. Instead of making a one-piece jet of the external dimensionsspecified, the jets may be madeseparate and screwed into the topof the main portion, so that a rangeof jets, of different sizes, can be fitted.

In the assembly of the carburettor,the location of the jet tip should becarefully observed, by looking throughthe main air passage with the throttlewide open; it should just, and onlyjust, be visible. The float should thenbe adjusted so that feed is cut offwhen a bead of fuel shows at the jet

585

but does not overflow; note that thejet needle is opened fairly wide, sothat it does not obstruct the flow.It may or may not be necessary tonotch the opening edge of the throttlebarrel, as shown, to obtain the correctmixture for slow running, as this isinterdependent on the primary airintake.

No allowance has been made forthe fitting of fibre joint washers aboveand below the banjo, as they obviouslyaffect the jet location, and theirthickness may be indeterminate. Thereshould be no difficulty in makingsound metal-to-metal joints with theaid of a little joint varnish. Whatevertype of carburettor you employ,details such as these call for meticulousattention if consistent operation andgood control are to be obtained.

* To be continued

10 NOVEMBER 1960

B y E d g a r T . W e s t b u r y

MANY of the troublesencountered with modelpetrol engines, and partic-

ularly the refusal of the engine tostart at all, are caused by somedeficiency in the ignition system.Apart from the actual electricalcomponents, which usually com-prise a battery and coil, the im-portant mechanical item in thesystem is the contact-breaker which,though basically simple, is oftenbadly designed or constructed, andtherefore more troublesome thanit need be.

At the time when the Kiwi was firstintroduced,, there was some differenceof opinion on the respective meritsof trembler and non-trembler ignitioncoils. Many users of model petrolengines preferred the trembler. Itadmittedly had certain points in itsfavour, not the least of which was thatit provided its own circuit tester-an audible buzz when on contact-and a very convincing stream ofsparks at the h.t. lead The verysimplest type. of wine contact wassufficient for timing the spark, andthe coil rarely gave much electricalor mechanical trouble, unless it hadbeen badly made.

But the trembler coil is inherentlyinadequate to cope with the require-ments of modem high-speed engines,and has now become almost entirelyextinct. I doubt whether it would bepossible to obtain one at all, and Imention it only because a few con-structors still confuse the two types,and the distinctive ignition circuitsand equipment associated with them.Most readers are sufficiently wellacquainted with automobile electricsto know that the modem ignitioncoil employs a positive mechanicalmake-and-break device. which pro-duces a spark at the time of the break,and not at the time of the make, aswith the trembler coil.

To simplify the construction of thecontact-breaker for the Kiwi engineas much as possible, and at the sametime to ensure its reliable operationat high speed, I designed it forstandard automobile components,which eliminated much of the finickywork involved in riveting or brazingcontact tips, and so forth, not tomention the obtaining of. suitablecontact materials. So far as I amaware, this form of construction was

24 NOVEMBER 1960

With this instalment the modernversion of the renowned little

petrol engine is ready to givemany years of faithful service

an innovation at the time, though ithas been employed by many designerssince; it is not necessarily the verybest type of contact-breaker for smallhigh-speed engines, and I have nutalways employed it, but at least itoperates with complete certainty.

Some designers have used verysmall “ scale model ” contact-breakers,often of excellent workmanship.. Whilethere is no apparent reason why verysmall breakers should not work quitewell, they are mostly inaccessible,difficult to keep in proper adjustment,and liable to become oiled up: Theinability of spark-ignition two-strokesto attain the high r.p.m. for whichthey are designed can often be traced

such as in the “ 1831 ” contact-breaker, which employs the Lucasbakelite bell-crank type of rocker arm.

Construction of the bracket isquite straightforward; a chuckingpiece is provided on the back of thecasting, enabling it to be held forfacing and. boring the boss, whichshould be made a tight wringing fiton the extended end of the camshaftbush-not relying on the ability ofthe clamping screw to make goodan initiallv loose fit. as this distortsthe bore so that it fits like a ready-made shirt on a gatepost.

The rear face of the boss may bemachined by mounting the casting ona stub mandrel either before or after

I 110KEYWAY

to faults in contact-breaker design,as the spark frequency must be twicethat of a four-stroke engine for agiven speed.

The Kiwi contact-breaker has beenused for several types of engines, bothfour-stroke and two-stroke, and willoperate. at the highest speeds, if thecam is properly made. A speciallylight type of rocker arm is employed,and the electrical connections arepositive; they are taken through thecontact spring and not through pivotswhich may become partially insulatedby a film of oil. It is, of course,possible to modify the design forother types of standard components,

633

the split clamp is finished. It isadvisable to drill and tap the holefor the clamping screw before slittingthe lug, either by hand or by a smallcircular saw. The hole for the rockerpivot screw should be exactly parallelwith the axis of the mounting boss,and it is worth while to set the jobup on the faceplate, using a boltthrough the hole in the boss, andclamping against the back face, forthe facing and drilling operation.

Some care is necessary in locatingboth this and other holes so that theparts, particularly the contact screw,are correctly aligned when assembled.All the surfaces marked F in tbe draw-

MODEL ENGINEER

ingss should be spot-faced or machined square with the axis of the holes. The pivot screw should be an easy fit in the fibre bush of the contact arm, and the screwed end preferably on the tight side; the lockk nut is not absolutely essential, but is an added safeguard against working loose.

Either bakelite, ebonite or vul- canised fibre is suitable for the in- sulating bushes, which are clamped above and below the spring anchorage lug. A standard brass screw may be fitted; it serves to secure the end of the spring, with good electrical contact, and also as the 1.t. terminal for the coil circuit. The cam is formed, after the machining of the blank, by setting it eccentrically and turning away about 900 deg. of the surface as shown; the amount turned away determines the length of time that the coil is on contact. For engines which run at relatively low speeds, the time may be reduced, to economise battery current; but at high speed a certain minimum time is necessary to energise the coil effectively, because of electro-magnetic inertia, or hysteresis. Some contact- breakers are made with just a flat filed on them. Though they appar- ently do all that is necessary, they

cause unnecessary wear and tear of the contact arm, and are liable to cause it to bounce at high speed unless an abnormally strong spring is used.

The grubscrew shown for fixing the camm is effective if it is properly fitted, and a flat or dimple is provided on the camshaft; but an improved method would be to bore it taper and fit it to a correspondinglyc tapered shaft, with a screwed end for a draw nut. Too time the cam, the engine should be set at t.d.c. on the firing stroke, and the cam turned in the direction of rotation, so that it first makes contact, and then just barely breaks it, with the breaker lever in the vertical position; it is then secured permanently on the shaft. There is plenty of latitude for advance and retardd adjustment while the engine is running. The contact clearance is not critical, but for high speed I have found that it is undesirable to exceed 10 thou, and if tbere should be any tendency to bounce, still less may be an advantage.

With the once-through or total loss, method of lubrication, as distinct from continuous oil circulation, some method of metering the oil is highly desirable;. the old method of hit- and-miss feeding would not be toler-

Heree is the lubricator

STD. DELCO-REMY

STD. M.S. SCR

CONTACT BREAKER ASSEMBLY E&

PIVOT SCREW I OFF M.S. C.H.

THE “KIWI” MARK III CONTACT BREAKER ADAPTABLE TO ALL MODEL PETROL ENGINES

24 NOVEMBERR 1960 MODEL ENGINEER 634

ated on modem engines, as it oftenresults in alternate flooding andstarvation of the working parts. Someusers employ the spring-loaded syringetwe of oil injector. which workswell if one does not forget to turn iton-and also off again at the end ofthe run-but it is rather fierce inaction and calls for critical adjustmentof the oil feed valve. With the crank-case pressure kept below atmospheric,oil can be fed quite effectively bysuction, especially with modem oilswhich are much less gummy or viscousthan those used in the early days ofthe Kiwi.

The lubricator shown here is onethat I have used for both suction andgravity feed on several types of engine;it is very simple to make and shouldpreferably be fitted apart from theengine mounting so that it is notaffected by vibration. The body ismade from a transparent plastictest tube; no doubt some of thecontainers in which tablets and othermedical commodities are suppliedcould be adapted for the purpose.Thin metal tubing is practicable asan alternative, but it is obviously anadvantage to be able to see how muchoil is available in the container.Glass needs careful fitting to avoidleakage, and is undesirably fragile.The assembly is held together by alength of 3/16 in. dia. brass tube,screwed at both ends, and cross-drilled for oil at the bottom and airat the top. Two recessed end caps,identical except that one is providedwith a hole for filling, are used toclamp the body endwise; the plastictube is faced accurately on the endsand fits closely in the recesses.

A rotatable disc cover is fitted over

24 NOVEMBER 1960

the top cap, and held down by aspigoted nut so that it is just free tomove somewhat stiffly. The holethrough both cover and cap may bedrilled in situ; it should be as largeas practicable, and may be elongatedif desired to provide free access forfilling. A steel, bronze or Germansilver needle valve, with the knurledhead sweated or otherwise firmlysecured, is fitted to regulate the oilflow. It will be seen that the lowerpart of the head is split, and is squeezedslightly inwards to provide a frictiongrip on the thread; this must be donevery gently and discreetly to avoidcomplete collapse. Altemativelv. acompression spring may be fittedbetween the head and the cover toprovide the necessary friction

Details of the outlet connection.and of the mounting bracket, mayhave to be varied to suit the positionand method of mounting; both partsmay be permanently sweated to thebottom cap when these points havebeen settled The centre hole whichforms the needle valve seating may bevery slightly countersunk by an acute-pointed drill or a graver, and theposition of the needle may be as-certained on assembly and sweatedin position with a well-heated solderingbit

The only difference between suctionand gravity feed is that in one casethe lubricator is fitted below the enginefeed inlet level and in the other aboveit; the former may be somewhat lesspositive, but has the advantage thatthe oil ceases to flow when the enginestops, and in this respect is morecomnletely automatic. In either casethe regulation of oil flow is soon foundby trial, and is not very critical.

635

If the crankshaft is not drilled toconvey oil to the big end bearing, oilmay be fed directly into the crankcase,but it is better to retain the samefeed inlet on the main bearing and toprovide a longitudinal oilway in thebush, so that oil must always passthrough it on the way to the crank-case. In this event, an oil hole mustbe drilled in the underside of the bigend bearing.

I am often severely rapped byreaders for neglecting to give detailsof valve springs, but it is difficult tospecify essential dimensions, becauseultimate strength and durability de-pend on several factors, including notonly the quality and temper of thespring steel, but also the pre-stressingof the material in manufacture. Forthe best results, some experiment iscalled for; the springs which have beenfound satrsfactory are 3/4 in. free length,5/16 in. outside diameter, by 20 s.w.g.,with six complete turns, not includingthe flattened end turns.

Most of the points affecting as-sembly I have dealt with in describingthe details, and no particular problemsshould be met if all parts are accuratelymade and fitted. There is, however,a possibility that the connecting rodmay foul the cylinder skirt on one orboth sides, at its maximum angularity,and this may necessitate filing notchesin the skirt at an appropriate angleso that it just clears. Do not emulatethe engine constructor who providedclearance by filing notches in theconnecting rod !

I have already indicated many waysin which this simple engine may beimproved. To those who have nothad previous experience of engine

l Continued on page 656

MODEL ENGINEER

P

Christmas Shoppintg

contd.

wires are necessary to install it. Price already number several hundred butis 9 gns. Bassett-Lowke will gladly their number is constantly beingsend-full details on request. increased to meet special needs.

Among the popular i t ems a r e a

Shims and gearshand drill at 14s. 6d. and one slightlylarger, at 23s. 6d., table vices a t

Tr-r~i;r~~yo~ in;riFFi$ 13s. 6d. a n d 24s. 6d., light machine

Senier and Co. Ltd, at 115-123 Stvices, 21s. and 35s., a bench grinderwith 4 in. wheel for 33s. 6d., with

John Street, Clerkenwell, London,ECl.

:gy. wheel 39s. 6d. and 6 m. wheel

The policy of this old familybusiness, which has been in existencesince the reign of George III and the

In the workshop

time that the American revolution HOLTS, a firm well known for theirdrew to a close, is to give as much motor car preparations, manu-attention to the small order as the facture several items which can use-large. fully be kept in the home workshop

Their stock covers a vast range of cupboard.materials and includes sheet metals Loy Cold Plastic Metal sets harderin most gauges, brass, copper, steel than lead and proves very effectiveand silver steel in various sections and for a minor ship modelling repair.angles, brass blanks, and ranges of It is sold in tubes of two sixes, 2s.BA screws, with nuts and washers, and 3s.in cheesehead, roundhead and counter- For the small soldering job, Solda-sunk types. paste, in which solder and flux is

Araldite epoxy resins, used in-dustrially for high-grade electricalcomponents and for corrosion-resistantcoatings, have revolutionised the per-manent bonding of metals, glass andother non-porous materials that havein the past proved almost insuperablydifficult to join.

Araldite can also be used forsealing and filling cavities or. inconjunction with- glasscloth, formaking reinforced plastics. It issupplied in 6s. two-tube packs fromironmongers.

Lightweight vice

Power pack

Whynot give your model railway-or your model railway own-

ing friend-the Trainsmaster 622 ?This is not just a new power pack

(good as it is, at that) but is a formof control for two trains of an en-tirely novel conception.

Two trains on one track-sixareas for separate control-two oper-ators or one at immediate choice. Itgives perfect control of locomotivesat all speeds with maximum power.

Every unit is supplied with detailedillustrated instructions, and only seven

It is not generally known thatSeniers carry a tine stock of gearsand that they can supply packets ofshims in assorted sixes.

If a special soldering job crops upa visit to Seniers may prove worthwhile, for they have an excellentvariety of solders, fluxes and spelters.

Their assistants are courteous andready to offer advice.

,.

either on bench or table. It is sturdyand compact and embodies a widerange of features. For example, theclamp for alternative fixing may bebolted or screwed; it has hardenedsteel jaws, pipe grips, hardened steelanvil, and a pipe bender. The Impis ideal for general instrument workor model engineering and, attractivelyfinished in bright red, makes a usefulgift. Price is 37s. 6d

Tools in the Fabrex 400 range

mixed in paste form, is efficient,needing only the heat of a matchflame. Tubes cost 1s. 9d., 8 oz. tins 8s.

Rust, especially this time of year,is the curse of an outdoor workshop.Holts Thixotropic Rust RemoverPaste has a powerful rust removingaction. Being jelly-like it stays whereit is put. Its action not only removesthe rust but primes the metal surfaceready to receive paint. A 4-oz. tincosts 3s., a 12-oz. 6s. 6d.

Ship modellers will find Twinbond,a new unique epoxide resin adhesive,a valuable general purpose cement.It is proof against boiling water,petrol, acids and does not shrinkor expand when setting. Price is 5s.

KIWI . . .

construction, are recommend to buildit in the form described at first, andto proceed thence by easy stages.Many of the possible improvementshave been investigated and describedin the course of experiments whichculminated in the development of theKittiwake 15 c.c. and Kittyhawke 30 c.c.engines. They included the additionof a “ fit and forget ” forced lubri-cation system, with a gravity sumpand a submerged plunger pump, andalso an inclined-valve cylinder headwith larger ports and enclosed offsetrockers. But any advance in engineperformance necessarily imposesheavier duty on all working parts and,therefore, calls for the highest standardof workmanship; and for the beginner,at least, the standard Kiwi Mark IImay be relied upon to give faithfulservice in any sphere of duty.

MODEL ENGINEER 656 24 NOVEMBER 1960