NACA - digital.library.unt.edu

copy “ LRM L54E17

NACA .

RESEARCH MEMORANDUM

AN EXPERIMENTAL INVESTIGATION OF TWO-DIMENSIONAL,

SUPERSONIC CASCADE -TYPE INLETS

AT A MACH NUMBER OF 3.11

By Edward 6ffenhartz

Langley Aeronautical LaboratoryLangley Field, Va+

NATIONAL ADVISORY COMMITTEEFOR AERONAUTICS

WASHINGTONAugust 25, 1954

—

—.. . -—. .—.-_. .-— .-—-—.. . .. ——-. .-.. -.-— .. .. . . ..–. .— . .. . . -—.

.——

lBNACA RM L54E17

TECH LIBRARY KAFB, NM

lluunlMllun!loBlll1441t41? NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS

.RESEARCH MEMORANDUM

AN ~ INVESTIGATION OF TWO-DIMENSIONAL,

SUPERSONIC CASCADE-TYFE INLETS

M A MACH NUMBER OF 3.11

( By Edward Offenhartz

SOMMARY

A prelimimry investigation of two-dimemiontil, supersonic cascade-type inlets at a free-stream Mach number of 3.11 has been conducted atthe Langley Aeronautical Laboratory. Two cascade-type inlets utilizingdifferent methods of internal-flow compression were designed and tested.In one inlet (designated cascade inlet) the passages were contoured togenerate coalesced compressions; whereas those of t~ other inlet (desig-nated stepped-cascade inlet) were contoured to generate noncoslesced com-pressions. Pressure and Mach number distributions as weU. as shadowgraphsare presented for both inlets. With mass-flow ratios of 1, the maximumvalues of total-pressure recovery attained at 0° angle of attack were 0.43

.and 0.45 for the cascade and stepped-cascade inlets respectively, as com-pared to a theoretical value of about 0.64 for both inlets. Individualpassages of the cascade inlet attained maximum values of total-pressurerecovery of O.~. The differences between maximum totsl-pressure recoveryfor the individual passages and the entire cascade inlet are due to thelosses involved h straightening the velocity profiles of the individualpassages and to subsonic mixing of the & from individual passages. Bothinlets were stable in the supercritical region only. The Mach nuniberdis-tributions obtained approximately 40 min.imm width$ downstre~ from theleading edge of the inbosrd passages of~both inlets were fairly uuiformacross the passage width. Separation of’the s@ersoni’c flow ~ch :increases with back pressure t.alcespl+ce within the individual passages ,of both inlets. The separation is asymmetric, and the asymmetry has’nosystematic relationship with back pressure.

INTRODUCTION..

In the development of supersonic inlets, initial emphasiswaa placedon conical-type inlets. These inlets were designed to be housed withinthe nose of a supersonic vehicle or to be carried externally in wing pods.In both cases the inlet and engine are mounted in tandem: Analysis and

:5

~

. . . ... . —.— .— . .—.———- — —

T -.—–

2 NACA FM L54E1.7

tests of conical inlets (refs. 1, 2, and 3) indicated that the largecowl-~p angles necessary in the design of conical inlets result in highvalues of external wave drag at Mach numbers greater than 2.5. However,recent tests of conical inlets (refs. 4 and 5) indicate that it is pos-sible to attain high total-pressure recovery without prohibitive externaldrag.

To overcome effects of external wave drsg associated with conicalinlets operating at their design Mach numbers, two-cllmensional,rectangular-scoop inlets were designed and tested, as reported in ref-erences 6, 7’,and 8. These inlets =e fuselage supported in such amanner so as to add less external wave drag than conical inlets. scoopinlets so mounted (see fig. 1) direct the compressed air (after subsonicturning) to fuselage-housed engines. The inlets reported in these ref-erences attain total-pressure recovery aMost as great as nose inlets.

Two-dimensional cascade-type inlets canbe housed within the wingof a supersonic vehicle in such a manner as to add less external wavedrag than conical inlets. Two versions of such a wing-root inlet areshown in figure 2. The leading edge of the outboard blade tip of eachpassage is similar to the cowl of a conicsl inlet. However, each of theoutboard surfaces of the blades acts as a compression surface for thesucceeding passage. At the design Mach number, the leading edge of thewing and the first compression for each passage are coincident. Theexternal wave drag associated with wing-housed casca& inlets is thenless than that associated with conical inlets.

In both wing-housed cascade inlets shown in figure 2 the turning ofthe air is accomplished sqersonicaldy with the reflected compression CFas shown in the tiboard passsges. Of course, if the air is not to bedelivered in a stresmwise direction, the strength of the reflected com-pression can be reducedby utilizing more cutback on the upper blade sur-face of each passage. If this process is theoretically csrried to thelimit, a cascade scoop inlet having zero thickness at the upper bladetip results, snd the & is directed psrallel to the compression surface.In order to maintain l%nite upper blade thickness, each passage must bebrought forward into the free stresm. The theoretical total-pressurerecovery for any one passage of such a cascade scoop inlet would thencorrespond to the value of approximately 90 percent for the isolatedscoop.

The prhary purpose of this investigation was to study the startingcharacteristics, flow-mixing characteristics, and velocity profiles incascade inlets. To study these characteristics,the two configurationsshown in figure 2 (though not opttiun pressure recovery inlets) wereinvestigated. The data obtained included pressure distributions, total-pressure recoveries, Mach number distributions, and shadowgraphs for aMach number of 3.11 snd 0° angle of attack.

NACA FM L54E17+

3

●

Dr. Antonio Ferri initiated preliminary investigation of cascadeinlets while employed at the Langley Laboratory.

.

MO

Ml

m/q ‘

q

%

P.

Pf

pf/po

SYMBOLS

free-stresm Mach number

subsonic-tiffuserMach number

ratio of measured mass flow to mass flow through a free-stream tube of cross-sectional area equal to projectedinlet frontal area

local.static pressure

static pressure at upstream end of orifice plate

total pressure of free stream

total pressure after diffusion

total-pressure recovery

AERODYNAMIC DESIC3’JOF THE CASCADE INLETS TESTED

Passage contours.- Two cascade-type inlets were designed for opera-tion at a free-stresm Mach number of 3.0. Different internal-flow ccm-pressi.onwas obtained by contouring the inlet passages. The cascadeinlet passages were contoured to generate coalesced compressions; whereasthose of the stepped-cascade inlet were contoured to generate noncoa-lesced compressions.

Consider the cascade inlet shown in figure 2(a). At the design llachnumber the wing leading edge EC and the first compression generated bythe surface BE at B are coincident. The flow is further ccqressedby turning the surface BEF such that all compressions generated coalesceat the point C. Since the surface CD is cut back from the free-streamdirectionby an insufficient smount to follow the streamline, a reflectedcompression CF results. The intensity of this ccmrpressionis determinedby the Ufference between the flow angles at C as generated by the sur-face BE and the cutback angle at point C.

Consider the stepped-cascade inlet shown in figure 2(b). At thedesign Mach number the wing leading edge EC and the first compressiongeneratedby the surface BE( at B are coincident. The remdning com-

. BE..,gmespaced so as to generatepressions generated along the~%~.~c.~ .%*

—.—. -. —.— —.

4 NACA R4 L54E17

noncoalescing compressions. The first compression generated at thepetit B is reflected at the point C where the surface CD is cutback by a smsll amount. The remahing compressions generated by thesurface BE are canceled along the surface CD.

Casca~ Cascading results in a reduction of the adal length ofthe inlet and ~~us the cascade inlet becomes shorter than that of asingle-passage inlet using identical supersonic turning. The cascade-type inlet consists of four identical passages having the required totalfree-stresm-tube capture area. Each passage of the cascade-type inletsdischarges into a cmmaon constant-area plenum chamber which serves as amLxing region. Thus each passage to this point is essentially an inde-pendent inlet.

starting cascade-type inlets.- During the starting process of cas-cade or stepped-cascade inlets, the wss flow corresponding to the free-stream tube based upon the projected passage frontal area mwt be swal-lowed by each passage. The inlet is considered as started when the firstcompression coincides with or is within the inlet swept leading edge, thenormal shock has passed the minimum, and the mass-flow ratio m/~ is 1.

Consider an instantaneous shock position during the starting processof the stepped-cascade inlet as shown in figure 3. The initial shock isassumed to be attached at L and extended a short distance to M. Atthe point M a normal shock occurs. The shock is nomal to the internslflow frcnn M to N and oblique to the free-stream flow as shown by MP.The flow behind the detached shock MP is supersonic such that the obliqueshock OP is attached at the point O but not coincident with the sweptleating edge. The shock PR is assumed to be a strong ob~que shockbehind which the flow is subsonic. For the inboard passage, some flow isspilled around the inlet as indicated by the streamline TS. The flowcan also be spilled to the next passage: This spiUage from one passageto the next is the difference between starting the cascade inlets andsingle scoop inlets as discussed in references 9 and 10. If the mass flowspilled is more than that required for equilibrium, then the shock con-figuration will move in the dowmstresm direction. The shock IM becomescoincident with the swept leading edge IS and the normal shock MN movesdownstream past the mirihum section of the inboard passage. The shock MPRbecomes less oblique and the flow at the point O becomes identical to theconditions originally shown for the inbosrd passage. The process isrepeated for this channel and the others in succession until all passagesstart.

It has been found experimentally that stsrting phenomena limited thecontractim ratio to 2.38 and 2.28 for the cascade and stepped-cascadeinlets, respectively. The entire starting process takes place almostinstantaneously with the inboard passage starting first and the outboardpamage starting last.

--———

NACA RM i54E17

MODELS AND TESTS

.

.—.. - —

5

Models.- The two models tested and their corresponding blade geome-tries are shown in figures 4 and 5. The leading-edge angle of the sweptcover walls in the plane parallel to the free-stream direction was 6° forthe cascade inlet and 12° for the stepped-cascade inlet. The increasedangle for the step~ed-cascade inl.etwas necessary in order to house glassinserts in the cover walls. These inserts enabled visual observation ofthe inlet internal flow.

The surfaces ccmrprisingthe inboard and outbosrd passages of bothinlets were instrumented with static- and total-pressure tties which wereinstalled on the blade center LLnes (see fig. 6). Total-pressure tubeswere also inslxdled in the subsonic _ region. Cover-wW pressureorifices were installed at all total-pressure-tube locations for the cas-cade inlet. The stepped-cascade inlet had no cover-wa12.orifices at thepassage total-pressure-tubelocations because of the glass inserts. Thelocations of u measuring stations are shown in figure 6.

.

Tests aud measurements.- Both inlets were tested at 0° angle ofattack at a free-stream Mach number of 3.11 and Reynolds numbers of

appro-tely 13 and 14.3 x 106, based upon the inlet height for thecascade and stepped-cascade inlets, respectively. Since both inletshad the same frontal dimensions, the difference in Reynolds number isdue to stagnation pressue. The tests were conducted in a supersonicblowdown jet of the Langley Gas Dynamics Branch by using dry air sup-plied frm high-pressure tanks. Figure 7 shows a schematic drawing ofthe test installation.

The tests were perfomned tithe following manner. With the tunneland inlet started, the throttle valve was turned frm fully opened tothe point where flow instability of the inlet occurred. Measurementswere mpde at intermediary conditions up to the point just before theonset of unsteady flow. Gages and mercury manometers were usedto recordpressures. The mass flow through the model was measured with a cali-brated orifice located upstresm of the throttle valve as shown in fig-ure 7. The differential pressure across the orifice was measured on amercury-f~ed U-tube. All pressure measuring instruments were photo-graphed. Total-temperaturemeasurements were made upstresm of the mass-flow orifice plate and in the settling chsniber. &essure measurementswere estimated to be accurate within Al percent and result in pressureand mass-flow ratios accurate to ti percent.

———— - ——

—. ....— ——. 1.

6 NACA RM L54E17

RESULTS AND DISCUSSION

Static- and total-pressure distributions presented were obtained forinlet operation in the supercritical region oIIJ-y.The flow was unstablebelow the critical point and no attempt was made to stabilize the flowfor operation in the subcritical region. The data are presented for threeoperating conditions denoted by the ratio ~/Po. For all conditions

discussed the inlets were started.

At a free-stresmlfachnumber of 3.11, if it is assumed that it ispossible for a normal shock to etist just downstream of the leading edgeof the outboard surface of any one passage and if subsonic losses areneglected, the estimated total-pressure recovery for any one passage is64 percent for both inlets.

Cascade Inlet

Shadowgraphs.- The supersonic flow pattern at thecascade inlet is shown in figure 8. It has been found

entrance of theexperimentally

that the supersonic flow at &e entrance is independent o~ back pressureuntil the onset of unsteady flow. The position of the first shock isseen to be slightly different for each of the passages. This differencebecomes more pronounced in the outboard direction. The disturbance “a”in figure 8(b) is weak and does not alter the flow entering the inlet.

Passage pressure distributions.-Figure 9 shows the variation ofstatic pressure with length for both the inboard and outboard passages.The theoretical distribution for the inboard surfaces 1 and 3 of bothpassages is included for comparison.

‘Mue ‘f %po = 0“36 ‘s ‘her-etically obtainable slightly downstream of the leading edges of sur-faces 2 and 4. The general fluw pattern is qualitatively as follows:For values of

%/poequal to 0.31 and O.%, the flow undergoes com-

pression to the minimum section, supersonic expansion beyond this point,and further compression downstream. ‘or ‘#0

equal to 0.41 there is

compression from the inlet leading edge up to the last measuring station.

The difference in pressure ratio at the first measuring station forsurfaces 1 and 3 is due to the local flow angularity which is estimatedto be +1/2°. ‘or ‘iUes ‘f ‘b/pO equal to 0.31 smd 0.36 the decrease

unmeasured pressure ratio beyond the minhum section is due to the cut-back of surfaces 1 and 3.

The differences between the inboard- and outboard-passageflows arelargely due to boundary-lsyer separation. Figure 10 presents the varia-

~=-<~—.--~

—.

NACA F/ML54E17

—.—— ——.

7

.

tion of cover-wall static-pressure pl P.(/)

and total-pressure recoveries

(/)Pf P. with passage width.

( )Regions of separation pl/Po Z Pf/Po are

apparent. There is asymmetric separation from all surfaces of the inboardand outboard passages, and the asymnetry has no systematic relationshipwith back pressure.

The weighted total-pressure recovery (weightedwith respect to massflow throughout this report) is SO percent for both end passages when

‘b/pOis equal to 0.41. This result is obtained under the assumption

that the mass flow through the separated region is zero and that thestatic pressure measured on the cover wall and the total pressure meas-ured on the blade center line are constant across the passage height.

Plenum-chamber-pressuremeasurement.- Figure ~ presents the total-pressure recoveries obtained b the subsonic plenum chamber at the rakelocation. The outboard portion of the plenum chamber indicates highervalues of total-pressure recovery. This increase is probably due to thedifference of flow travel before reaching the measuring station and isassociated with less mixing losses. There is no distinguishablewakeeffect of the inboard blades at the measuring station. The weightedtotal-pressure recovery is 43 percent when pb/Po is eqti to 0.41.

Comparison of the weighted total-pressure recovery att.ainedby theindividual inboard and outboardpmsages with that attained in theplenum chamber indicates the magnitude of the losses due to strai@t-enlng the individual passage profiles as well as the ndxing of the foursubsonic streams.

The Mach number distribution shown in figure 12 is reasonably uni-form, considering the separation which was present within the individ-ual passages as indicated in figure 10.

Stepped-Cascade wet

Passage pressure distributions.- Figure 13 presents the variation ofpressure with length for both the inboard smd outboard passages of thestewed-cascade inlet. The theoretical distributions are included forcomparison. The general flow pattern for sll operating conditions issimilar to that described for the cascade inlet.

The inboard-surfacepressure distribution for both passages near theleading edge is lower than theoretical. The difference is due to thelocal flow Q@arity of +1/20. ~ figure 13(a) the pressure distribu-tion for tiboard surface 1 as measured is in agreement with theory untilthe minimum section. For dl values of pb Po, the d&crease fi press~e

/

— .—_ -_

8 ~“ NACARML54E17

ratio beyond the minhum section is due to flow expansion around thecutback surface 1. In the absence of stificient data, the data forinboard surface 3 shown in figure 13(b) are.arbitrarily faired to shuwcompression from the point beyond the minimum up to the last measuringstation. ‘

RQure 14 shows the variation of total-pressure recoveries with pas-sage width. The surface values are those obtained frm the pressure dis-tributions shown in figure 13. There is asymmetric separation from sM.surfaces of the inboard and outboard passages, and the asymmetry has nosystematic relationship with back pressure.

Shadowgraphs.- The supersonic flow pattern at the entrance of thestepped-cascade inlet as well as the internal flow in the passages isshown ~t,figure 17 for the values of ~/P. indicated. The disturb-ance b emanates from the lower nozzle block and extends around theinlet. It is partially visible through the glass inserts in the nozzlecover walls and should not be considered in analyzing the passage inter-nal flow.

Figure 15(b) clearly shows the difference in the separation of thesupersonic flow slong the inboard surfaces of passages A and B. Theexpansion at the upstresxaportion of the Moard surfaces of both pas-sages is due to the cutback of these surfaces. The portion of the inter-nal flow visible for the third passage is similar to that of passages “Aand B.

Increasing p~Po moves the region of supersonic flow separation

forward on all the--inboardsurfaces as shown ti figure 15(c). The flowin all passages is similar, with separation of the supersonic flowtaking place asymmetrically.

The flow pattern in passages A and B of figure 15(d) can beconjectured as follows: ~ passage A the flow is subsonic with thepossibility of supersonic flow along the outboard surface. The mainsupersonic flow region has moved upstresm beyond the window. The flowin passage B consists of a supersonic region along the outboard sur-face of the passage and a highly turbulent subsonic region. * P-#oincreases there is no indication of the occurrence of a systematic nomalshock pattern.

Detailed correlation between the flow pictures as determined frathe shaduwgraphs of figure 15 md the pattern which can be inferred fromthe pressure profiles is difficult. The data of figure 14(b) are takenalong the passage center line, whereas the shadowgraphs of figure 15 arepictures through the entire passage and present a cumposite three-cWnensionaL-flow effect.

—. . _- ..-

2B

●

NACA RM L54E17 9

Plenum-chamber-pressuremeasurement.- Figure 16 shows the variationof total-pressure recovery across the width of the plenum chsmber at themeasuring station. The increase of total pressure in the outboard direc-tion is probably due to the difference in flow travel before reaching themeasuring station. The weighted total-pressure recovery is 45 percentwhen ~/p. h eqtiti O.43. The Mach number distributions shown in

figure 17-are relatively uniform over the passage width.

,.

An investigation

CONCLUSIONS

has been made of two-dimensional cascade-typesupersonic i.nietsat a free-stresm Mach number of 3.11 and 00 angle ofattack. Two inlets utilizing different methods of internal-flow com-pression were designed and tested to study primarily starting charac-teristics mixing characteristics, and velocity prof~es fi c=c~einlets. b one inlet (designated cascade inlet) the passages were con-toured to generate coalesced compressions; whereas, those of the otherinlet (designated stepped-cascade inlet) were contoured to generatenoncoslesced compressions. The following results were obtained fromthis investigation:

1. The starting phenomenon of cascade inlets is such that each cas-cade starts by allowing spillage to the sides of the inlet and over thesucceeding cascade blades. Starting takes place almost instantaneouslywith the inboard passage starting first and the outboard passage startinglast. Both inlets were stable in the supercritical region only.

2. With mass-flow ratios of 1, the maximum values of total-pressurerecovery obtained were 0.43 for the cascade inlet and 0.45 for thestepped-cascade inlet, as compared to a theoretical value of about O.&+for both inlets. Individual passages of the cascade inlet attainedmaximum values of total-pressure recoveryof Ox. The differencesbetween msximum total-pressure recovery for the individual passages andthe entire cascade inlet sre due to the losses involved in straighteningthe velocity profiles of the individual passages and to subsonic mixingof the @r from the individual passages. Higher total pressure recoveriescould probably be obtained in cascade-type inlets if the inlets were notrequired to deliver air in the stresmwise direction as in the case of thedesigns tested.

3. The Mach nmiber distributions obtained approximately 40minhumwidths downstream fran the leading edge of the inboard passages of bothinlets were fairly uniform across the passage width.

__—————— -——

10 NACA RM L54E17

4. Separation of the supersonic flow which increases with backpressure takes place within the individual passages of both inlets. Theseparation is asymmetric, and the asynmet~ has no sYst~tic relation-

ship with back pressure.

Langley Aeronautical Laboratory,National Advisory Ctittee for Aeronautics,

LangleyMeld, Vs., ~ 6, 19*.

.- — ——. —— —

—.—. ..— —.-

NACA RM L54E17 >–’ IL

1.

2.

3.

4.

5*

.

6..

7.

8.

9.

10.

Oswatitsch, K1.:pulsion at Highsers). NACA TM

13EFmENcEs

Pressure Recovery for Missiles With Reaction Pro-Supersonic Speeds (The Efficiency of Shock Diffu-1140, 1947.

Ferri, Antonio, and Nucci, Louis M.: Theoretical and llxperimental-sis of Lowm Supersonic Inlets Having Circular Cross Sec-tion and a Central Body at Mach Nmbers of 3.30, Z?.7’j, and 2.45.

NACARM 113~3, 1948.

Ferri, Antonio, and NuCCi, Louis M.:New Type of Supersonic Inlet. NACANACA TN 2286.)

Prelhinary Investigation of aRep. l.l@, 1952. (Supersedes

Connors, Jsmes F., and Woollett, Richard R.: Force, Mcment, and Pres-sure Characteristics of Several Annular Nose Mets at Mach Num-ber 3.85. NACARME53J09, 1954.

Hunczak, HenryR.: Pressure Recovery and Mass-Flow Performance ofFoux Annulsr Nose Inlets Operating in Mach Number Region of 3.1

and Reynolds Number Range of Approximately 0.45 x 106 to 2.20 x 106.NACARME54A07, 1954.

Comenzo, Raymond J., andhlackley, Ernest A.: Preliminary Investiga-tion of a Rectangular Supersonic Scoop Inlet With Swept SidesDesigned for Low Drag at aMach N@er of 2.7. NACARM L52J02, 1952.

Anon.: Design and Analysis of Three Supersonic Side Inlet DiffuserModels. Wright Aero. Rep. No. 1755 (Contract No. AF 33(600)8733Item28, S.A. No. 1), Curtiss-Wright Corp., Wright Aero. Div.(Wood-Ridge,N. J.), Sept. 22, 1953.

Anon.: Quarterly FTogress Report on Ramjet Development. Wright Aero.Rep. No. 171-1(ContractsAF33(038)-9000 and Al?33(600)-8733),Curtiss-Wright Corp., Wright Aero. DLv. (Wood-Ridge,N. J.), Apr. 6, 1953.

Himka, Theodore: Methods of Starting Scoop-Type Ihlets. Wright Aero.Rep. No. 1692, Supersonic Inlet symposium, Curtiss-Wright corp.(Wood-Ridge,N. J.), Jan. 23, 1953.

Ferri, Antonio: Recent Developments in the Design of Sqersonic ScoopDiffusers. Wright Aero. Rep. No. 1692, Supersonic Inlet Symposium,Curtiss-Wright Corp. (Wood-Ridge,N. J.), Jan. 23, 1953.

— .— — ———— —-- ——

‘u

Fuselage

Figwe 1.. Two4imenaion01 scocrp inlet.

I

I

(b) Ste~ed-casctie inlet.

Figme 2.- Two-dimensi.onal. cascade-type inlets.G

I

I

NACA RM L54E17

_—

——

/“& ‘R

L

/

__ ———-

T

T.————-

s

Figure 3.- Starting process of stepped-cascade inlets.

._. - — -.

NACA RM L54E17 15

_—. ——-. –.._ — ——-

L-82229,—.

I/0[I

I

In

x

o.125200246.300.355.440500.555.700.800Soo

1.0001.1251.2501.3701.5001.6351.7501.875

2000

YL

o.021.037.04 I.053.060.075.082.088.1,00.107.110.120.110.107100

:092.080.064.055.048

x

o.625.875

1.0001.2501.3701.5001.6351.7501.875

2.0002. I 242.3752.5502.7523.000

Y“

o.210.290.336.446504560.610.660.700.740.768.800.809.810.795

6°075”R

o

Figure 4.- Cascade-inletblade geometry.

——. —— —.

NACA RM L54E17

1x

o.176235.376.506.6%.700800

YL

o.035.045.088.126.184.200.225

.—— ——.-.. _.i .900 .240

L-82227.11.000 .2501.100 .2521.200 .255

‘T_l

x Y“

001.035 .3401.240 .4151.412 .4851.650 -.5951.910 .6962000 .7202-200 .7582344 .7752.475 .7804.300 780

L-82231.1

.6°

+ !=.

.?5F ‘

\

,2~

Figure 5.- Stepped-cascade-inletblade geometry.

I

I

I

...

La?t

b

8 J)NT

4

,

I

- m-a$wre

(a) Cascade inlet.

I

(b) Stepped-cascade inlet.

Figure 6.- Location of measuring ste.tio~. (All dimensions are in inches. )

‘3!

rCD

I/’”4+

a-------

1

1 1 II

--------

/

y

1 1

Throtlhg votve

~ d ‘

I

(/ \ A ,

L m : : .- “~Slondmd 3T5’ ASME ttwpbte orlflc<

Figure 7.- Schematic hating of test installation.

I

I

NfWARML54E17 19

(a) No flow....—

)utboard1

Y

nboard

l?i.gure 8.-

..—— — —

(b)

Shadowgraphs

%L-83676

—= 0.41.P.

of cascade inlet at ~ . 3.11.

.—— —--— -.-—.

20

Lo\

.—

u“.. UJuL...m NA.CARM L54E17

tion

‘1 I L4s41-%’r I I I I, I --- /

!/ //

20f

1I

{ /,

1 / ./“ d

/ ,/ ‘

I— 4

/1

0

.16L

/

//

\ , /’~.” ------ ;

/ Pbipo

\. /

Surface I

o 0.31 I

I

I

1A

II

I.=-r

II

I

I

I

I

I

II I I I A I I Vn I I I I I I I 1 1

i ❑ .36 I!- I-lEmllllll I II II I I I o .4 I I

.08 w I

.04

0 I 2 3 4 5 6 7 8 9

Passage Iength, in.

(a) Inboard Passage.

Figure 9.- Cascade-inlet passage pressure distribution.

.

NACA RM L54E17 a

.

Passage length, in.

(b) Outboard ~S.SSEl& .

Figure 9.- Concluded.

.. ————— —..—..—

k ~tL

- /.-. :: ~poocoerd

h M ‘ h, — Pf/Po~ .8

r A-0.- 1 I I / ~

%------ p,/po

<‘I \ 1 / ‘

al— —

{ / y I~/P.

g ,4 /1/

070 0,31

m .

L

/o .36

: In boordY ‘ o .4 I

\ \\\\\\\SSSwSSS\\\\\~\\\\\\\\\~\\\\\\\\\\~

(a) Outboard passage.

.16 .20 ,24 .28 .32 .36 .40 .44 .48 .52 .56 .60

Pressure recovery

(b) Ihboard paesege.

Figure 10. - Cascad.e-hlet passage total-pressure-rec-r’y distribution.

1

I

(

I

I

]

NACA RM L54E17 23

-.,

.32 .34 .36 .38 .40 42 .44 .46

Total pressure recover y,pffio

Figure 11.- Cascade-idet total-pressure-recoverydistribution in plenumchsniber.

5

d4.—c--0“s 3

LaJ

:2.-n

I

0

\\\\\\\\\\\\\\\\\\\\\~\\\\\\\\\\\\\\\\~\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\:;; m::d

*P#Po

: 0.3 I.36

0 .41f

c

? y

-

1/

[

“ U

Inboardsurface

h\\\\\\%\\\\\\\~\~\\\\\\\\\\K

.1 .2 .3” .4 .5 .6 -.7

Mach number, Ml

Figure 12.- Cascade-inlet Mach nuniberdistribution in plenum chamber.

.- ———

24 NACA RM L54E17

.

.36 ,

AfI ‘\

.32/’

\II

1’i

.28 4I \I \

,>

I \

<0 24 I \ A ~/ ‘

Ii I ) ~’/

\’\I \ <. Theoreticat- 7 /’ \

0 20 surface 1..-

. .#--.- / +x b-- --- /

0 1.L

\ /@”: .16 I \3 I :1 1 \ ~ Y1--. -m ) \. /In

/,

I Surface: .12 \/

/i

--- _ ------ ;7 )Pbl Po

/_ $’ b d q 0 0.32,08 + *- i .37

: .43k

Theoretics

.04surface 2

J

o I 2 3 4 5 6 7Passage Iength, in.

(a) Inboard passage.

.

Figure 13. - Stepped-cascade-inletpassage pressure distribution.

—

4BNACARML54E17 25

I -\55° >Surfm lrRaketion

A 4/

I 2 3 4 5 6 7 .8 9

Passage Iength, in.

(b) Outboard passage.

Fiwe 13. - Concluded.

.—

!+\\\\\\\\\\\\\\\\\\\\\\\\\\ \\\\\\\\\\\\\\\\\\%\\\\\\\\~\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ y

.- Outboard

~-.8 Su “ face

u.- pb /p.3 %.

Yw -1 0 0.32

g ,4 - / —Q

❑ .370UJ * o

E 1.43

m Inboardv

:o\~\\\\\\\\\\\-\w\\\\ \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

surface\

I (a) (h.ztixerdpe.asage.

>\\\\\\\\\\\\\\\\\\\\\ \\\\\\\\\ \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

c .8wxSS\\\\\\\\\\\\\\\\\\\\\\\\\

.-

_C- W:::d+ 0u.-3

% ~

~ .4 a — — — e ==— +

— — — — —Cl-l / -0m A y09 Inboard

: o“~*

d c)‘-surface

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\m\\\\\\\\\\\\\\\\\\\\\\\\\\\\\m

,12 .16 .20 .24 .28 .32 .36 ,40 .44 .48 .52 .56Total-pressure recover y, PfFo

(b) hbORI’d pS.SSS@.

nJm

I

Figure U.- Stepped-cascade-inlet pe.sssge total-pressure recove”l-y.

.

27

i

t

-fu4te

-1!>

‘__—-————— ““—---—”–

Figure 15”-

~ “—

P’#o= 0.52.

of stepped-cascdeinlet at ~ = 3.D.

—-—---------~””- ‘---

— — _.. ——

28

outboqrd

Inboard

Outboard

Inb rd

-U..ULLI NACARM L54E17

Figure 15.- Concluded.

—.

,

jB

.

NACA RM.L54E17 29

—

/////////l//////J///////A//&//////////////~/~/////=

surfs Ce

<4.-

.c+

:3

L /

:2 f pb/pO

z $0 O::;

I:/ 9 43

\Inboard

v

o7//////////////B///l/////l//////Y/~//AY////~///~///surface

//.

.32 .34 36 .38 .40 .42 .44 .46

Total–pressure recovery, Pf/Po

Figure 16.- Stepped-cascade-inlettotal-pressure-recoverydistributionin plenum chamber.

.1 .2 .3 .4 .5 6

Mach number, Ml

Figure 17.- Stepped-cascade-inletI&ch nuniberdistribution in plenumchaniber.

NACA-U@eY- 8-25-54-360,qj-~$===-:-: :._~lEJ

.—.— ———— -—