c

●

i,.

‘t

RESEARCH MEMORAND1

FORCE, MOMENT, AND PRESSURE CHARACTERISTICS

OF SEVERAL ANNULAR NOSE INLETS

AT MACH NUMBER 3.85

By James F. Connors and Richard R. Woollett

By .... .,IEB+o

..

)Aq F&a ,.La.. ----...-,-----

~K......... .,..........................

) < l=+,.+.!.. ~D=mQm’r..................---

an matuialCOntad!i&mtfonlffactlcgtin mlfmd m!hsa ofth mibd MatesWlthtnb IEOa!lhqoftinaaPIOIU@laws,l’ltb18,U.&C. 88c0.TG3d ‘/S4,tb trammlaalmorremlamnoi whichfnazv-r b anumutbrimdpersonIs pmkted b law.

TTEENATIONAL ADVISORY COMMFOR AERONAUTICS

WASHINGTONFebruary 3, 1954

Ai---------

— — — -— —. .---- -- -4--

NACA 133E53J09..

●

NATIONAL ADVISORY CCMMI’ITEEFOR AERONAUTICS

TECHLIBRARYKAFB,NM -

IillllllllllllllllllllllilllllilllrllJ432a3

RESEARCH MEMORANDUM

FORCE, MOMENT, AND PRESSURE CHAFWXERISTICS OF SEVERAL ANNuIAR

NGSE INLETS AT MACH NUM81ZR3.85

By James l?.Connors and Richard R. Woollett

SUMMARY

An tivestigation to evaluate the over-all performance character-istics of several annular nose inlets was ●conducted in the Lewis 2- by2-foot supersonic wind tunnel at a Mach number of 3.85. The four experi-mental configurations consisted of a one-cone, a one-cone (low-anglecowl), a two-cone, and an isentropic inlet. Over an angle-of-attackrange frcxn0° to 9°, complete pressure and three-cmnponent force datawere recorded.

For application in a hypothetical ram-~et engine at zero mgle ofattack, the isentroplc tilet indicated the best over-all performance onthe lasls of specific fuel consumptlcm and propulsive thrust as a resultof its ability to attain a high total-pressure recovery without pro-hibitive external drag. The one-cone (low-angle CUW1) inlet had thesmallest external drag and was comparable in performance with thetwo-cone inlet. At the low Reynolds numler of the present tests, theapplication of roughness on the spike tip of both the two-cone and theisentropic inlets eliminated laminar-boundary-layer separation andeffected a reduction in the external drags through reductions h mass-flow spillage.

For the one-cone tilet with varying degrees of supersonic mass-flowspillage, the expertiental values of additive drag agreed quite wellwith theoretical predicticms. At angle of attack, theory tended tounderesthate the pitching-moment coefficient, the normal forcecoefficient, and the angle-of-attack drag rise, particularly at thehigher angles.

INTRODUCTION●

For application at a particular Mach nmnber,design is the attainment of a high total-pressure

8 external drag. Often it is found that there is a

--

the aim of good inletrecovery and a lowconflict between these

2 NACA RM E53J09.-

●

objectives and me can be achieved only at the expense of the other.Consequently, a compromise between the internal and etiezmal flowgeometries must be made. Currently, only limited data are availableon diffuser performance at Mach numbers greater than 3.0. Therefore,in order to obtain further insight into the criteria Involved in thedesign of high Mach number diffusers, the NACA has undertaken anexperimental research progrem that includes the investigation of thevarious conventional annular nose inlets at a Mach number of 3.85.

The inlti.alphase of this research, which is primarily concernedwith the diffuser characteristics of pressure recovery and mass flow,is reported h reference 1. The second phase of the research, coveredin the present report, is ccmcerned chtefly with the drag aspect ofhigh ldachnumber inlets. Accordingly, the expertiental investigate.cmwas directed toward (1) the determtiation of the aerod~amic forcesand moments acting m the various Inlet configurations over a widermge of angles of attack, (2) the evaluation”and ccmpment %reakdownof the external drags at zero angle of attack, and (3) the determina-tion of additive drag coefficient as a function of supercritical flowspillage behtid a conical shock.

The experimental configurations (the same as those used in ref. 1)were axially symmetric emnular nose inlets employing single-shock- andmultlshook-generating centerbodies, or, more speolfically, one-cone,two-ccme, and isentropic inlets. Pressure end three-component balance

a

—

—

—

*—

(normal-force,axial-force, andobtained for eachattack from 0° to

Configln+tion9°,

pltchtig-moment) measurements wereoperating over a ra~e of angles of

—— —

SYMBOIS

The following symbols are used throughout this report: —

A model flow area, sq ft

maxtium frontal area of model

Al maximum capture area defined by mwl-lip dtameter, sq ft—

CA axial-force coefficient, axial force/q&x

% drag coefficient, D/q&

%? thrust coefficient, F/&●

.—

$

xiii.:... .-*...+’

NACA RM E53J09

%?,p

%

D

??

f/a

L

M

m.

m3

N

P

P

a

Y

ez

propulsive thrust coefficient, (F - D)/-

pitching-moment coefficient, ~-/g&ax

normal force coefficient, N/&ax

drag, lb

thrust, lb

fuel-air ratio

over-all length of model (from spike tip to base), f%

Mach number

mass-flow rate through free-stream tube area equal to Al,slugs/see

mass-flow rate through engine, slugs/see

normal force, lb

total pressure, lb/sq ft

static pressure, lb/sq ft

dynamic pressure, Y&I% lb/sq ft

center of pressure locatiun (measured from base), ft

angle of attackj deg

ratio of specific heats for air

kinetic energy efficiency defined as ratio of kinetic energyavailable after diffusion to ktietic ener~ In free stresmj

b 1y-l2 Po Y

1-( Y-1)M2 % ‘1

cowl-position parameter, =@e between ~S =d l~e fiODIspike tip to cowl lip, deg

4 NACA RM E53J09

.

Subscripts:

a additive

c cowl pressure

e external

f friction

o free stream

3 diffuser exit

4 model exit

.

APPARATUS AND PROCEDURE

The experh.uentalInvestigationwas conducted

.

h the NAOA Lewis2- by 2-foot supersonic wind ;unnel, which was operated at a Mach numberof 3.85 and at a simulated premnrre altitude of approximately 108,000feet. The tumnel air was maintained at a stagnation temperature of200°k50 F and at a dew-point temperature of -lOOQ.OO F. Based on themaximum diameter of the cowl (5 in.), the Reynolds number was approx-imately 429,000.

—

As illustrated in figure 1, the experimental model,.which utilizedan adjustable exit plug to very the inlet back pressure, was basicallythe same as the model of reference 1, with the exception of the three-component force-measuring system. Details of the balance link arerevealed in the insert drawing on figure l(a). Mounted on the flexuralmembers of the link are electric resistance-wire strain gages, which,connected in bridge circuits, provide the indications of axial andnormal forces and pitching moment. Some interaction of the forcecomponents was encountered with this balance ‘s’ystem,but the effectswere accounted fcr in the calibration and eliminated from the data.Tare forces acting on the base of the mode~ and within the stingbalance chamber were determined and subtracted out of the axial-forcedata.

Specifications of the various inlet confi~ations are presentedW coordinate form in table Iamd in the sketches of figure 1(%).The one-cone Inlet consisted of a 600-included-angle cone positionedso that, theoretically, the cowl lip would just intercept the tip shockemanating from the centerbody (desi@ et =44.90). A gradual rate of

turning of the flow back toward the axial direction with no internal

—

—

.-

NACA RM E53J09 5

●

contraction was effected by the cowl, which was initially alined in thelocal stream direction. h an attempt to achieve a near-mintium cowl-

*pressure drag, a second single-oblique-shock confi@zration, designatedthe one-ccme (low-angle cowl) inlet, was made with a sharp turn at thecone shoulder and a cylindrical titernal contour on the cowl. Asdescribed in reference 1, the application of local suction immediatelydownstream of the sharp turn was required for attached shocks to existat the cuwl lip. This suctton was accomplished by means of a double rowof l/8-inch-diameterbleed holes on the centerlody, the inside of whichwas vented to ambient tunnel pressure by means of hollow centerbodysupport stznts.

The two-cone inlet was designed with two conical surfaces (40°and 70° included angles) that would, theoretically, locate the resultingshocks at the cowl lip. Additional flow compression was attempted lyappl@ng the maximum permissible fiternal contmction (ref. 2) based onan esttiated average entrance.ldachnumber.

.Of the four Inlet configurations betng considered, theoretloally

the greatest amount of supersonic compression would be obtatied with. the isentropic inlet, whtch utilizes a conttiuously curved centerbody

to produce the desired turntig of the flow. Ih the theoretical charac-teristics solutim, the Mach waves were desi~ed to focus at the cowllip, the titernal contour of which was initially arranged in the localflow direction. The compression was to be carried down to a final Machnumber of approximately 1.5 with no internal contraction.

lh order to circumvent the difficulty of lamlnar-boundary-layerseparation encountered on both the two-cone and the isentropic inlets,an attempt was made to promote an artificially induced transitionfrom a laminar to a turbulent boundary layer. This transition was to heaccomplished by the application of tip roughness in the form of al/2-inch band of (number 60) =borundum grit.

Pressure Instrumentation consisted of eight wall and rake staticorifices plus a 24-tube pitot rake at the end of the subsonic diffuser(see fig. l(f) of ref. 1). Static taps were located on the top, side,and bottom of the base emnulus. A static tap was also used to measurethe pressure inside the sting %alance chsmber, and three rows ofextenal wall static taps were Installed along the cuwl on the top,side, and bottom of eaoh inlet configuration. W order to determinethe boundary-layer profiles along the etiemal shell in the vicinityof the base, a traversing total-pressure probe mounted to the tunnelwall was used to survey the flow field. A static orifice was alsolocated on the etiernal shell in the survey plane..

The total pressure at the diffuser exit was determined through●

an area-weighting of the pitot-rake measurements, while the mass flow

6 NACA FM E53J09

.

passing through the model was calculated with the assumption of one-dh.uenslonalflow from the average static pressure at the rake and thesonic discharge area. An integration ofcthe cowl static-pressure dis-trihution~ yielded values of cowl-pressure drag; friction drag wasobtained from the integrations of boundary-layer-profile data, with aconstant static pressure and total temperature assumed throughout theboundary layer. h the calculations of internal thrust, the evaluationof the exit momentum term was based on the pressure measurements at theexit rake.

Cowl-pressure drags were determined only at zero angle of attack,Otherwise, complete force and pressure data were recorded at anglesof attack of 0°, 3°, 6°, and 9° over a wide range of exit plug posi-tions. A twin-mirror schlieren system provided a means of visualobservation of the inlet flow patterns under all test conditions.

.

—

—

RESULTS AND DISCUSSION ●

Before presenting the results of this investigation it shouldbe emphasized that these experiments dealt with the evaluation of ●-

specific inlet geometries that were believed to be representative ofthe better designs within each category - that is, one-cone, twu-cone,

— —

and isentropic inlets. However, a certain amount of arbitrariness wasinvolved in the designs, for example, in the rate of turning the flowback toward the axial direction, in the rate of subsonic diffusion, orin the manner of coping with shock - boundary-layer interactions.Optimizaticm of the respecting.de.st~s, therefore, may influence the

—

relative over-all performances of these inlets. Further research in . _this direction will be necessary for final evaluation.

Wternal-Flow Performance

Although the diffuser performance (internal-flow)characteristicsof’the several imlet configurationswere extensively detailed in ref-erence 1, they are again included herein (fig. 2) for completeness mdbecause there were minor differences between the values obtained duringthe force tests and those previously presented. Schlieren photographsof the inlet shock structure during supercritical operation are alsoincluded in the figures for angles of attack of 0°, 3°, 6°, and 9°.In figure 2(g), the data are summarized by cross-plotting the optimumpoints of each configuration in order to provide a relative comparisonof the various inlets. At zero angle of attack the isentropic inlethad the highest total-pressure recovery (0.57), corresponding to akinetic-energy efficiency ~ke of 0.94, but fOll off qUiti3 sharply with

increasing angle of attack until at 7.5° the flow separated completely

—

.

.

NACA RM E53J09 7

off the lee side of the spike. This separation oocurred with an attend-ant hysteresis whereti the angle of attack had to be reduced to approx-tiately 6.3° before an attached flcnrwas reestablished. For angles ofattack greater than approximately 7°, the total-pressure recovery of thetwo-cone inlet exceeded that of the isentropic. On both the two-coneand the isentropic spikes the application of tip roughness appeared toeliminate the *’bridginglt(or separation) of the laminar boundary layerdue to the adverse pressure gradient and resulted in improved inletpressure recovery and mass-flow ratio.

At zero angle of attack the supercritical mass-flow ratio for theone-cone filet was varied from 1.00 to 0.73 by controlling the amountof flow spillage behind the conical shock through changes in the valueof the cowl-position parameter el from 45.5° to 42.5°. These changes

were accomplished by inserting spacer rings behind the cowl in orderto vary the spike-tip projection. The resulting diffuser performancecharacteristics are presented in figure 3. With decreasing supercritioalmass-flow ratio, or equivalently increastig spillage, the maxtium total-pressure recovery decreased nearly linearly. Even with the conical shockpassing well ahead of the cowl lip (el = 42.50), there was no indication

of any degree of subcritical flow stability as might have been expectedon the basis of the slipline criterion of reference 3. The aerodynamicinstability or buzz could quite feasibly have been triggered by alocal flow separation occurrtig internally on the centerbody (as illus-trated in ref~ 4), or in the v~cinity of the terminal shockby the criterion of ref. 5), or both.

Force Measurements at Zero Angle of Attack

The force data for the one-cone inlet at zero angle of

(as predicted

attack withseveral values of the cowl-positicm parameter ez are presented in

figure 4, where the variati~s of fite~al t~ust~ Propulsive thrust)and external drag coefficients with outlet-inlet area ratio are plotted.On each set of curves are included the experimentally determined valuesof c~l-presswe drag coefficient ~,c, the friction drag coefficient

~, f, and the theoretical additive drag coefficient ~,a, as given

in reference 6. Based on the experimental profiles, the boundary layeralong the external shell of the model was turbulent with a correspondingskin-friction coefficient of approximately 0.002. The heavy line repre-sents a summation of the preceding components of the total externaldrag. The data points for external drag are experimental values obtainedby subtracting the propulsive (or net) thrust coefficients, as deter-

. mined by actual balance measurements, from the internal thrust coeffi-cients, as calculated from the change in total momentum across the engine.As illustrated by figure 4, a very close agreement was obtained between

* the two methods of arriving at values of the

s-k--aL.%.?,

external drag, that

8,.

NACA RM E53J09.

is, by cmuponent summations and by actual force measurements used in con-junction with internal pressure measurements. The external drag coef- -flcient increased from 0.17 to 0.28 as r3z was changed from 45.5° to

42.5° or, correspondingly,as the capture mass flow decreased from 1.00to 0.73 (fig. 3). This drag ~crease essentially represented theadditive drag contributim resulting from.fluw spillage behind theconical shock. A compmison of the experimental and theoreticaladditive drag coefficients is presented in figure 5 as a function ofsupercriticalmass-fluw ratio. These experimental values were deter- Ccmined by subtracting the measured cowl-pressure and flricticndrag com-

Cu%

ponents from the internal thrust. As shown, the data agreed rather well- ‘-with the predictions of reference 6.

Ih figure 6 the force data are presented for the one-cone \low-anglecowl), the two-cone, and the isentropic inlets at zero angle of attack.Ewept for the two-cone w~th tip roughness and the one-cone (low-anglecowl) inlets, there was no theory readily available for the estimationof additive drags; and, therefore, the horizontal lines of ~,e repre-

.

sent more or less mean values drawn through the balance data. Thetheoretical value of ~,a Ifsted for the one-cme (low-angle cowl) .

inlet is scmwwhat approximate, in that the 1 pedcent of the maximumcapture mass flow mo involved in the suction process (ref. 1) was con-

sidered to have undergone a complete loss of momentum. The greatestscatter in the data occurred with the isentropic inlets near the criticalcondition; and, in these cases, the higher, more supercritical valueswere favored, because it was felt that the calculation of internal thrustmight have been least accurate when the pressure measurements (whichestablish M3) were made with m extremely low dyn~ic Press~e ~39

.—

A tabulation of the foregoing data (table II) provides a directcomparison of the performance of the various inlets at zero angle ofattack. The most significant result was t_hemoderately low value ofexternal drag (%)e = 0.16) achieved with the isentropic inlet with tip

roughness. An examination of the components of this drag showed arelatively low cowl-pressure drag (associatedwith the fact that the designallowed for a small projected area on the cowl) and only a slightly greateramount of additive drag than that for the,one-cone inlet at a comparablemass-flow ratio. This additive drag coefficient for the isentropic inletwith tip roughness amounted to much less than the value for completemomentum loss, assumed as a maximum in reference 1, and =S also s~ewhatless than the minimum calculated from the theoretical characteristicssolution that had the Mach waves coalescing at the cowl lip. The totalexternal drag coefficients for the isentropic with tip roughness, the .

two-cone with tip roughness, endthe same. As a consequence of a

the one-cone inlets were approxhwtelynegligibly small cowl-pressure drag

.

NACA R4E53J09 9

.

.

moNa

&iA.

.

.

.

(~, c = 0.009), the one-c~e (1~-~le cowl) inlet efiibited the lowest

etiernal drag (~, e = 0.09). Again, It is quite evident that for the

one-cone inlet the increase in external drag coefficient with decreasingvalues of 82 was due almost entirely to the increase in additive drag.

Over-All Performance Comparison at Zero Angle of Attack

b order to evaluate the over-all performance of the various inletconfigurations aud to establish a basis of relatfve merit h which thecombtned factors of total-pressure recovery, mass-flow ratio, andexternal drag would be taken into account, the experimental values foreach inlet (table II) were incorporated into the calculations forapplication to a hypothetical ram-jet engine. The assumed operatingconditions for this engine were as follows: flight Mach number of 3.85at 80,000 feet altitude, critical inlet performance at zero angle ofattack, 90-percent combustion efficiency, and complete exit-nozzleexpansion. The 80,000-foot altitude used in this comparison differsfram the actual simulated pressure altitude of the present expertients;however, it was selected because it conforms to a more practical fllghtcondition, as indicated in recent ram-jet analyses. As a result, then,the assumption is made that the Reynolds number effect would be negligiblysmall, at least with respect to the relative performances of the variousinlets. The results of these calculations are presented in figure 7 fora range of fuel-air ratio f/a. On the basis of specific fuel consumption,the one-cone (low-angle cowl) inlet was comparable with the two-cone inletwith tip roughness, and the specific fuel consumption of the Isentropicinlet with tip roughness was 8 percent lower than that obtained with eitherof the precedtng inlets over the entire fuel-air-ratio range. At a fuel-air ratio of 0.03, the isentropic inlet with tip roughness exhibited aspecific fuel consumption approximately 20 percent lower than that of the

.—-.

one-cone inlet. At f~a = 0.03, the propulsive thrust of the one-cone(low-angle cowl) inlet was 15 percent, of the two-cone inlet with tiproughness 28 percent, and of the isentropic Inlet with tip roughness55 percent greater than that of the one-cone inlet. The values of pro-pulsive thrust coefficient CF,P on this figure were based on ~ax

for the engine, which was at the exit. To permit conversion to anyother reference area, the ratio of ~ax/Ao was also included.

Of the ccnf@uraticms studied, then, the isentropic inlet with tiproughness has the best performance at zero angle of attack h terms ofspecific fuel consumption and propulsive thrust. These results are dueto the attainment of a high total-pressure recovery without prohibitiveexternal drag.

10 h~ NACA KM E53J09

.

Angle-of-Attack Force Measurements

Force data obtained for each of the respective inlet configurationsat angle of attack are given h figure 8, where the variations of axial-force, normal force, and pitching-moment cqe.fffcientsare presented aS afunction of outlet-inlet area ratio for angles of attack of 3°, 6°, and90 ● ti general, the pitchtig-moment coefficient ~ was independent of

diffuser back pressure during supercritical operation and increased with --increasing angle of attack. There was considerable variation of thenormal force coefficient ~ with supercritical values of A4/Ap This

may be attributed h part to some distortion of the exit flow conditionsdue to local separation of the internal flow, which was illustrated bythe velocity profiles presented in referenoe 1 for angles of attack of3° and above and which might have caused the mean exit flow direction tobe other than axial. Generally, however, the absolute level of ~

increased with increasing angle of attack. Also included In the fi~esare the values of axial-force (or thrust-minus-drag) coefficients. Themagnitude of axial-force coefficient CA is not Overly si~iff=t hitself, since it simulates the force corresponding to burning and chokingin a constant-area duct with a required heat release beyond the range ofpresent-day hydrocarbon fuels. For critical inlet operation, the axialforce decreased with ticreased angle of attack, as would be expected.

A comparison was made between the angle-of-attack force data andthe theory of reference 7, modified for an open-nose body, bY takinginto account the internal-flow contributions. Pitching-moment coefficients(fig. 9) and normal force coefficients during critical inlet operation(fig. 10) tended to fall above the theoretical values, particularly at thehigher angles of attack.

Attempts to extract the external drag coefficients at angle of attackfrom the data were not very successful. The previously mentioned internal-flow separation at angle of attack prevented a consistent calculation.ofinternal thrust based on the pressure data. Since the method of datareducticm involved a subtraction of the propulsive thrust (or thrust-minus-drag) term from the internal thrust, a laxge amount of scatter wasincurred. The resulting data points, along with correspondingbands ofexperimental scatter, are presented in flgwe 11. Also included is thetheoretical drag rise due to angle of attack (~,e - ~,e,~oo). Byway

of cmparison, a composite curve drawn tWo’iw@ the e~er~enta~ dataindicated the drag rise due to angle of attack to be much more rapidthan that predicted by theory. Based on these crude data, a speci.fic-fuel-consumption comparison of the various inlets appeared to Indicatethat the isentropic inlet would he superior up to approximately 6° angleof attack.

.

—

.-

Ccml0m

.

.-

.

“-

.

NACA RM E53J09 d.~~:ti=; ,. 11,--------.._

The experimental data, locating the center of pressure at angle ofattack, also efiibited considerable scatter but appe=ed to fall fairlywell in the followhg bracket for all inlet cmfigurattons:

z().45> -> 0.60L

where ?? is the distance of the center of pressure measured frcm thebase and L is the over-all length of the modelbase).

SUMMARYOF IWXIEIS

(from spike tip to

An experimental investigation to evaluate the over-all force andpressure characteristics of four annular nose inlets, designated the~e-cone, the one-cone (low-angle cowl), the two-cone, =d the isen-tropic inlets, yielded the following results at a Mach number of 3.85:

1. At zero angle of attack, the isentropic inlet with tip roughnessshowed the most promise of the inlet ccmfigurations studied, as a resultof its ability to attain a high total-pressure recovery (0.57) without aprohibitive external drag (~ = 0.16 based on the maximum cowl area).

The performance of a hypothetical ram-jet engine utilizing this inlet,based on specific fuel consumption and propulsive thrust, exceeded thatobtained with any of the other configurations up to an angle of attackof approximately 6°.

2. At zero ang}e of attack, the low external drag obtained with theone-cone (low-angle cowl) tilet (~,e = 0.09) made it comP~able with

the two-cone inlet with tip roughness on the basis of specific fuelconsumption aud propulsive thrust.

3. At the low Reynolds number of these experiments, the applica-tion of tip roughness on both the two-cone and the isentropic inletsresulted in slightly higher total-pressue recoveries and lowerexternal drags through reduced mass-flow spillage.

4. For the one-cone inlet with various degrees of supersonic mass-flow spillage, the experimental values of additive drag were in gooda~eament with theoretical predictions. In addition, the external dragcoefficients obtained by a summation of components agreed quite wellwith those derived from the balance measurements.

5. At angle of attack, theory tended to underesttiate the pitching-moment coefficient, the normal force coefficient, and the angle-of-attackdrag rise, particularly at the hi@er aWles.

Lewis Flight Propulsion LaboratoryNational Advisory Committee for Aeronautics

Cleveland, Ohio)

12

10

2.

3.

4.

5.

6.

7.

hi—i-” “: NACA EM E53J09

RlzFmENcEs

Connors, James F., and Woollett, Richard R.: Performance Character-.

istics of Several Types of Axially Symmetric Nose Jhlets at MachNumber 3.85. NACA RME52115, 1952.

Kantrowitz, Arthur, and Donaldson, Coleman duP: I?reliminaryIhvestl-gation of Supersonic Diffusers. NACA ~ L-713, 1945. (SupersedesNACA ACR L5D20.)

Ferri, Antonio, and Nucci, Louis M.: The Origin of AerodynamicInstability of Supersonic Inlets at Subcritical Conditions.NACARML50K30, 1951.

Connors, James F., and Woollett, Richard R.: Some Observations ofFlow at the Throat of a Two-Dimensional Diffuser at a Mach Numberof 3.85. NACA RME52104, 1952.

Donaldson, Coleman duP., and Lange, Roy H.: Study of the Pressure.

Rise Across Shock Waves Required to Separate Laminar and TurbulentBoundary Layers. NACA TN 2770, 1952. (SupersedesNACAFML52C21. ) ‘~

Sibulkin, Merwin:Additive Drag.

Allen, H. Julian:Inclined BodiesA9126, 1949.

.

Theoretical and ExperiiientalInvestigation ofNACA RM E51B13, 1951. —

Estimation of the Forces and Moments Acting onof Revolution of High Fineness Ratio. NACA RM

.

NACA RM E53J09

.

.

●

?

TA.3L21 - IP.IETUIKHSI036

—z

S@

‘1

one-Evfke I C5wl

3.2CQ3.4WS.Sal3.6034.Qm4.2034.m4.66U5.aXY5.W6.W6..370

1.6241.8871.926L.9461.6541.9Ea1.9351.9471.9s31.8661.6s01.8~

.Em 2.545 2.434l.cim 2.363 2.463l.m 2s75 2.3&1.500 2.373 cym-1.7= 2.368 Arid2.m 2.360

1-**taper 1

3.7.s3 2.?m Z.m

Two cone

mike(A, 9.852) (B,%%)

x Y z u vo Llss 2.165

s ~z ‘%V 0.m 2.203 2.23.6.2V0 2.Z39 2.266.ma 2.2S9 2.316

d“~ ‘+%.400 2.265 2.332.6W 2.Fsa 2.416.&m 2.3s2 2.459

4.552 1:602 Low 2.3% 2.4-s34.652 1.953 1.250 2.372 2.4974.752 1.963 1.m 2.373 2.5W

2.065 1 %% %%4.932 2.0545.1525.352 2.IJ9 3.ca 2.2735.602 2.133 3.23a 2.3725.652 2.136 3.Y30 23696.102 2.136 3.7W 2.3656.352 2.153 4.000 2.359 t6.652 2.m 5.750 2.3C0 2.5037.352 2.(X57.852 2m468.S52 1.9978.6s2 1.9439.362 1.6239.852 1.820

A

B

lengthof Bpfie h Upto paint Or attachmentto aft bdy, fn.

Mm6-thofcovlmllp~ mtit or sttacb=ntt-aouter nhell,in.

7.1oo 1.820\

Iaentmpic

(A,~i%l)*l

B, 7.75

TXYZ

000.Wo .073 .Oz

LOW .145 .C6c1.YJO .21.2.I.oc2.mo .284 .2CK2.5m .357 .30C3.mo .455 .40CS.5W .52E .6a4.cCXJ .624 .=4.520 .742LCIX5.0&3 .8765.5ca 1.031 1S.om l.ao 5.3X6.3031.433 S.sx7.cm 1.74a 5.75(7.lcm1.630 6.H7.m 1.9227.3C0 2.023 17.4002doo 7.7X7.sm 2.1377.6m 2.156

EEL

u

2.2402.2622.2772.=92.3262.3462.3562.3702.3762.378

2yti-lil-lcal2.3782.3762.3702.W

,trdgll~mr2.333

)v

2.2402.2722.2612.3232.3702.4042.4322.4692.492

$=.6riul

I2.3Q0

13

TABIasII. -8tMrARY oFPERFmMMa P~ IYJRVARIOUS INLET COW180BAlTOliS

AT MACE$OM2ER 3.S5AFU)2ESOAHGLE OF ATI!ACK

Annulsu nose- tipercriticd 2G’l?l-pressure Friction ~tive l’otd

inlet configuration totsl-prewmre -s-flow w w w external dreg

recovery, ratio , coefficient, coefficient, coefficient, coefficient,

P31%3 mJ% %, c %,f %,a %,.

hla.eom:9Z = 45.50 0.317 1.00 0.E9 0.043 0 0.17

44.9°(design) .31J. .99 .127 .043 .003 .17

44.3° .298 .934 .1.36 .043 .025 .20

43.7 .2sl .863 .132 .043 .62 .23

43.1 .257 .@Qo .129 .043 .0s .25

42.5 .233 .728 .1.28 .043 .I.l .28

)ne-cone(low-angle cowl) 0.30 0.925 O.m 0.044 0.037 0.09

m-cone 0.40 0.875 0.I.I.4 0.047 %.070IkO-ccmevith

0.23

tip rOu@neme .44 .963 .114 .047 .008 .11

[Sentropic 0.565 0.91 0.065 0.050 %.075 0.19Lwntroplc vithtip mmghnem .57 .93 .074 .050 *.036 .16

%esignates experimentalvalues for which there was no available theory to check against.

.

WOE ‘ ‘

NACA I?ME53J09

.

15

.

%’8

.

.

—. . . . :. ----- ----—-- ...--—.- —- r .&. --,

-:~_-: ..;.:+-,?.. - -. --- .<.,.-4=.:. _-_:..,.-,$ .+.*. :

.-

—

T

~*Lg-&_F”’-”““g;G:’-”e+.—~*- ------–--,——_:,...,-.’ -~’”=—- --- r.. ——

—-—~ -— ,, -,, ““’’’’”’””‘-==s%-+7’”

(a) Schemtic draw~ of ~-~~h~-ter ~el ‘is~llti ti 2- by 2-fwtsupersonic tunnel.

Figure 1. - Experimental mtiel.

.

16 .NACA FM E53J09

One-cone inlet

rDouble row of l/8”holes10C8ted immediatelybehind break

Holepattern One-cone (low-anglecowl) inlet

+4.352” 1“ 6.625”—q

*1=‘7-4.330’ I

40 Too

c~

Two-cone inlet

Isentropic inlet

(b) Inlet details.

Figure 1. - Concluded. Experimental model.

.

.

.- —

.—

,.=

— .

NACA RM E53J09

6upercriticalflowpattermsat variousanglesof attack

17

Cf,=oo

.-

a=30

-.—.—— .,—

. 1—- (

a. 60 a=9°

Angleof attack,a, deg

o 03

$ 6v 9

---- Pulsh’lgflow

.4

.3

. A /- v

3-” -//”/-

/. 0.2.

/“ “4

/~- “v

/~

.1.4 .5 .6 .7 .8 .9 1.0

Mass-flow ratio, m3/~

(a) One-cone inlet.

Figure 2. - Effect of angle of attack on inlet performance.

~~~i

18 NACA RM E53J09

8upercritical flow patterns at various angles of attack

a= 00 a= 30

,

Angle of attack,a, deg

o 03

: 6v 9

--- Pulsing flow

.4

.3 ---

! .

.2 v

.1

.

“

.

.

U=90

.—.4 .5 .6 .7 .8 .9 1.0 --

Mess-flow ratio,m3/mo

(b)One-cone(low-anglecowl)inlet.Figure2. Continued.Effectof angle of attackon inlet performance.

.

-,

NACA RM E53J09

8upercritical flow patterns at various angles of attack

F4

.5

a,=6°

Angle of attack,a, deg

o 03

: 6v 9

---- Fulsing flowH

1 I I I I I

If

I A- , Im,

.-.4 .5 .6 .7 .8

Mass-flow ratio, m3/mo

(c) Two-cone inlet.

Figure 2. - Continued. Effect of angle of

.9 1.0

attack on inlet performance.

19

20 NACA RM E53J09

oeAm

Supercritical flow patterns at various angles of attack

A.

aii0,.(30

——

r

,.

f’.—-

1

•~

.-— .- —y”.

-—- -—-—. ------— - ..=

a.= 30

a,=6° # a=9°

Angleof attack,a, deg

o 0.5 3

: 6v 9--- Pulsingflow

.4../:

/- /“/ - /

m-/M /0

/ / 0HA 0

/ /‘ 77 <.3- / /

/

/ vdt/

.2 /

—. -—. .—

.1...4 .3’ .6 .7 .8’”.

Mass-floti”ratio,u@@

(d) Two-cone inlet with tip roughness.

.

—

.

.... —.

.

Figure 2. - Continued. Effect of angle ‘~ attack on idet.,~.rformance.-

k~ ‘.:

.

NACA RM E53J09

Supercriticalflowpatternsat variousanglesof attack

21

.

Cl,=oo

g

Pm

.

.

.6

.5

.4

.3

.2

.1

CL=60

CL=30

a=9°

0’ Angle of attack,.- a, deg

/

A3 0 0

I i

. /“I # 1

/ “-0 “ t

b“-1A I i n 3 t

/ / / v0 6

/ “v 9OH

/ --- Pulsingflow0 <>

/ 0 s,/ o

Flow separation// - -v/~ on top of spike.4

r3“

vA

v

1.4 .5 .6 .7 .8 .9 1.(J

Mass-flow ratio, m3/mo

(e) Isentropic inlet

Figure 2. - Continued. Effect of angle of attack on inlet performance.

22 SL’i’ NACA RM

Supercriticalflowpatternsat variousanglesof attack

E53J09

a. 00 a=3°

o-5d’

.5

.4

.3

.2

.1

a=6° a=9°

Angle of attack,) a, deg

o 0 #3 . ///

: 6 /0”=*v 9 ,-0” 0“ ●

)--- Pulsing flow

/0

, Kl”‘c

3 04“

#*//* ‘4‘.>/-

,0/

0 Jd0 /.

0

4 / Flow separation

v “-.

on top of spike)

/?

/b

v

.4 .5 .6 .7 - .8 ,9 1.0

Mass-flowratio,m3/mo

(f)Isentropicinletwithtiproughness.

Figure2. Continued. Effect of angle

hk~” “

of attack on inlet performance.

.

.-

.

.

‘NACA RM E53J09 23

.

n.N

2

.

●

.

.

1.

.

.

.

.

.

0

9

cY

8

Flow separationon top of spike

6

Inlet configuration

One- coneTwo-coneTwo-cone with tip roughness jIsentropicIsentropic with tip roughnessOne-cone (low-angle cowl)

.6 o4~ ~

:Y v.5 \ A

‘A

<>\

~

Flow separation

.2on top of spike

.—0 3 6 9

Angle of attack,a, deg

(g) ~ry cumes showingrelativeeffectsofangle of attackon performanceof variousinletconfigurations.

Figure 2. - Concluded. Effectof angle of’at’tackon inlet performance.

24 NACA RM E53J09

Supercriticalflowpatterns for several values or position parameter

.

.

ez = 42.5°

Position parameter,ez, deg

o 45.544.9(design)

: 44.343.7

x 43.1A 42.5.- ,Pulsingflow

.4

.3 A- .- P

~

c-c ‘.w

/A “ 6Ad “ \ Yr 5 ❑o

.2L A v

A

————

el = 43.70

.l~ I I I I I I I I I I I I 1 .

.4 .5” .6” .7 .8 ‘ .9 1..0 1.1 —Mass-flowratic, m~/~ .

Figure 3. - Effect of position parameter on performance of one-cone inlet .,at zero angle of attack.

—

, . , CZ-4 3028 , ,

..“-~

.6

.2

.9

.4

(l

0 .35 .40 .45 .s4

(b) Ccul-pmitim wmmeter, 44.9°, cowl-premaure dra@, c~ffiobnt, 0.127; tiioticm.iIW ooaffioient, 0.043; thmmotioa.1

additive &W OCUffiOiWIt, O.UIS.

so .35 .40 .45 .M1 .m .35 .40 .46 .baCutlet-lnlat arm ratio, AJAl

(d) Cowl+onitkm PIrametur, 43.7°, wwl- (4) Cowl-position wamk, 4s.1°1 mu. (f) COMI-WBltkm WameW., 42.5°, oml-Prmmm &-aK .awrrieient, O.lzat r.ioti.nW mefflai.nt, 0.043; mm.tid

prmmmm &ag mw-fi.aient, 0.129J friotmndJW 0Mff10151t, 0.043, thmmtiati

msmm bag omffiaimt, 0.12s; fridtim

additive drag mefrkimt, cI.m2. - OWffiOICnt, 0.043; tt.emtianl●flmttve dmg a.m-rioimt, 0.09. UMitive drag coefficient, 0.11.

5/gU3

!,...

,4,

,/, , ,.’, ,, ,1 ,, . ,,, ,

26 NACA RM E53J09

.

.16

.12

.08

.04

0

Experimental—-— Theoretical

.6 .7 .8 .9 1.0Supercritical mass-flow ratio, m3/~

Figure 5. - Comparison of experimental and theoreticalsupersonic additive drags for one-cone inlet.

.

—

.

.

, CZ-4 “ia.ck 3028 ,

1.5

E

d1.0

,5‘j

:

iv

?30 .35 .40 ,45

(a) One-ocne (low-aogle 00.1)

inlet. OOwl-prem.ure drag oOei’-

flaient, 0,009; rriotlnn drag

coefficient, 0.044; theor.tioal

additive drax coefficient. 0.040

.

j2

0

1

L

4

Ida

.“ ‘%z

— z \ \

,2 .3 .4 .5Outlet-inlet area ratio, ~AI

[b) Two-cone inlet. Ccn#l-preame (o) Two-oOne inlet with tip rough-

dm.g coerfioient, 0,114; *iOtiOn neon. Cowl-premum drag aOOf-dmg COorrioient, o ,CMl. fiolent, O .114; friution drag

ooeffioient, O. 047j theoreticaladditivm &m OMffiCi=”t . 0. M8 .

❑ Pmpu18ive thmst,-%,p

v Expml glmg, “

COeffioientIntaPml thmmt, ~

K\ ‘h,. - % - ‘P,p— Exbernal drag,

\

%,. - %,. + OD,f + cD,---- mains flow

; \

s \

I .2 .3 .4 .5Outlet-lnlat arem ratio, A~A1

(d) Inentropio inlet. Cowl.premnura drag (e) I~entropic inlet with tip roughnmm.meffioient, 0.0135J frictim drag mef- Cowl-prmmm drag coefriaient, 0.074;fioient., 0.050. friation drag awffieient, 0.C60.

Rigurt6. - Pome data nhowing ●%parimntal external drag aotffioimtm obtained ~ar W1OUO inlet oonrigwationn

at zero m-qle of attmk,

‘,, ,:!,

,.

28 NACA RM E53J09

4

3

2

.6.

2

Inletconfigurationo

0 One-cone•lOne-cone(low-anglecowl)

3.6 O Two-conewithtiproughnessV Isentropicwithtiproughness

2.8<

c~ - p. — ~2.4 b

~ ~~ *

— _- P ~

2.0.020 .028 .036 .044 .052 .060 .068 .076

Fuel-air ratio, f/a

Figure 7. - Comparison of performance parameters,for engines using variousinlets and operating under following assumed conditions: free-stream Machnumber, 3.85j altitude, 80,000 feetj combustion efficiency, 0.90~ criticalinlet operation zero angle of attackj complete nozzle expansion.

.

.

.

.M-@@@h?- “

. .3026

. ,

(a)One-cone inlet,

Flm 8. - Effect of angle of attack m e.erodynmic force coefficients.

1.2

al0

h .4el

o

~~COef flclent

o Axialforce, CA

Angle of ❑ Howl force, ~attack, u,

Ideg

o Htching IIKlment,~

3 --- PliLaing flow

II Jl-

11 r

.6

w ofJ

attack, m. I J

/delg

6

~ -“I t’ I

I/

Angle of

attack, u,

\I WI

ideg

0 Il. 0

I I d’- 1 v I I

.30 .38 .46 .30 .38 .46 .30 .30 .46

Outlet- inlet area rstio, @Al

(b) One-cone (low-a@e cowl) inlet../!

Figure 8. - Continued. Effect of angle of attack on aemdymmi c force coefllciente.

, , , . 8202 “ ‘

[’:.

● ✎●

3028 , ,,

2.4

0

COefflcient

Nom.%1. force, ~

O Fltching moment, ~

--- Rllstig flow

I

I “\

I‘

— A\

2*

~ — —

t-F-FtdI II I

.2 .3 .4 .5 .2 .3 .4 .5 .2 .3 .4 .5

Outlet-bl.et area ratio, A~A1

(c) Twc-Cone inlet.

Figure 8, - Continued. Effect of angle of attack on aerodynamic force coefficients.

r

II.“.

‘1”

II

Coefficient

2.8Angle of

o Axial force, CA

I attack, u, ❑ Iiormal force, ~

~ ‘

deg

1

0 Fitchlng mment, $

3--- Pulsing flow

2.4

ItII

\;

1-Angle of 1

2.0I attack, a,

I

I 1 &g

i6

b ‘1

I

\

i

1 Angle of

1.6 I attack, a,1 deg

i 9

I

1.2; 2

\ 1’ \- g

I -A

#

– -j

I :w

~ ~

.8I

I & 1* 8.!+ u

;/ - \

1. w‘I

$!

I o-,4 If \

\

❑ <‘+ k

/’0 +

v

0

.2 .3 .4 .5 .2 .2 .3 .4 .5Outlet-in&. area ~~io, Ad<

(d) Two-cone inlet with tiproughnem.

Figure 8. - Continued. Effect of aogl.eof attack on aerodpemic force eoefficienim.

. , 4 3028 ‘ ‘

3.

Ai@e of Coefficientattack, a,

3.5

h

flego hid force, CA

3 ❑ Horiml force, CM

O lHtching moment, ~ —I

,0

—-- PuL3ing flmw

I i

h ;1

2.5

\

1A@Le of

attack} U,

#~

deg

6

2.0! I Flow se~tion

on top of 8pike

1.5

\

.u of

1.0

\

.5

I 4 9 \\ 1 (!

llT1l-hl[+w

o , 1 , , , , ,.1 .2 .3 .4 .5 .2 .3 .4 .5 .2 .3 .4 .5

Piiili Tiiiiii’Kioutlet-inlet aree ratio, A4/Al

(e) Ieentm3plc inlet.

Figure 8. - Continued. Effect of angle of attack on eemdynamic force coefficients. (ACA

.

4.0@@e or

attack, u,

t

3 G3efficient

3.5

J’!\l1

❑ Normal force, $

c.?

--- mlslmg rlnw

; 3“0

c?\

~: 25

Oj’ J

;

u2,0

c%

: \i

Flow reparation

,8 on top of spike

:

j 1.5: Angle of

2

!“

,.1

Lo

3

.5

)\— q

. \

o.1 .2 .3 .4 .5 .2 .3 .4 .5 .2 .3 ,4 .5

Gutlet-ln),et area ratio, ~Al

(f) I.entropic inlet wlthtipro@ne6s.

Figure 8. - Concluded. Effect of angle of attack on aemdymdc force coefficients.

CAI&

i??

.“1 . , 820!2 . .t, ,,.

NACA RM E53J09

.

.

amloF-1

.6

\#

.4

c)/ /

.2 / ‘o No roughness❑ Witht@rO_S.9c — Theoretical

/ v

o

(a)One-coneinlet..6 +

c>

.4

()

.2

o~ J

(b) One-cone (low--e cowl) inlet.

.6

~

.4

E?/ ‘

/ /

.2 / ‘

o

(c)Two-mne inlet..6

8e~tlon on

.4 c< top of spike

@

.2

c1/

f/ ‘

o 3 6 9Angleof attack, a, deg

(d) Isentropic inlet.

Figure 9. - Comparisonof experimentaland theoreticalpitching-mxuentcoefficlents.

35

—

36

1.0

.8

.6

.4

.2

0

- . . NACA RM E53J09

Inlet configuration

o One-cone❑ One-cone (low-angle cowl)O Two-coneA Two-cone with tip roughnessV Isentroplc~ Isentroplc with tip roughness

tion on top-of spike

A

./’/’

/KeoreLical

lHrr[ ./

3 6 9Angleof attack,a, deg

Figure10. - Comparisonof theoreticaland experimentalnormalforcecoefficientsduringcriticalinletoperatton.

.

-1

—.

.

.

NACA RM E53J09 37

.

Rom

i

.

.

I Band of Inlet

m experimental configurationstatter

o .....0....0..One-cone

❑ ~ Two-conewith tiproughness E

~ ~ Isentroplctithtipmu.hess

,5 ,

,4

,3

0 3Angle of attack,

FigureIl. - Comparison of experimentalcoefficients.

6 9a, deg

and theoretical external drag /

NACA.UnglO~-2-S-S4-325