1、AN EXPERIMENTAL INVESTIGATION AT LARGESCALE OF SEVERAL CONFIGURATIONS OFAN NACA SUBMERGED AIR INTAKEan J. Mart:Ln and Curt A. Holzhauseri W _ “_ _ _a _t_ca 1 Laboratory,! I_IA_.,AI, _. ,- Moffett Field, Calif.afecting the _ .tiona_ Defellse of the UnitedTz: e, PBOMP_L% States wlthln th meaning of he
2、 Espionage Act :,USC 50:31 and ,S2. its transmission or the._L _ _*_ / _ lation of inany ., unauthorized _son is prohibited by law.Information classified may be impartedonly _o pera in the military and navalservices of t_ United States, appropriate %. _,and employees ofthe Federal _.LGovernrnen_ who
3、 have a legitimate _nterest _.therein, and to United States citizens ofknown _,. , _._ _loyalty and discretion who ofnecessity must be ,- . ,_),_informed thereof. _, -NATIONAL ADVISORY COMMITTEE ,FOR AERONAUTICS WASHINGTON _ :_,_October 19, 1948 “_:Provided by IHSNot for ResaleNo reproduction or net
4、working permitted without license from IHS-,-,-NATIONAL ADVISORY COMMITTEE FOR AERONAUTICSRESEARCH MEMORANDUMAN EXPERIMENTAL INVESTIGATION AT LARGE SCALE OF SEVERALCONFIGURATIONS OF AN NACA SUBMERGED AIR INTAKEBy Norman J. Martin and Curt A. HolzhauserSUMMARYAn investigation of an NACA submerged air
5、 intake was conducted ona full-scale model of a flghter-type airplane. This study was made todetermine the large-scale aerodynamic characteristics of a submergedair intake proposed as the result of small-scale tests and to comparethe pressure-recovery characteristics of the large-and small-scaleinst
6、allations. Additional tests were made to determine the effecton pressure recovery of a systematic variation of ramp divergence.The data obtained at various angles of attack and inlet-velocityratios indicated the same favorable characteristics for the inlet thathave been noted at small scale. The max
7、imum values of entrance pressurerecovery were high (92 percent for the full-scale inlet withoutdeflectors), and the variation of pressure recovery with angle of attackand inlet-velocity ratio was small. Pressure recoveries measured withthe full-scale model were approximately 9 percent higher than th
8、osemeasured with the small-scale model. It is shown that differences ofboundary-layer thickness could account for 3 percent of this amount.The tests in which the amount of ramp divergence was systematicallyvaried indicated that varying the ramp divergence had only a smalleffect on the magnitude of t
9、he maximum pressure recovery measured atthe entrance, but markedly changed the inlet-velocity ratio for maximumrecovery. This change of inlet-velocity ratio resulted in higher maxi-mum pressure recoveries after diffusion for the curved-divergent rampsthan for the parallel-walled ramp.An analysis of
10、the data indicated that the use of deflectors onthis model was not advantageous; the effect of an increased pressurerecovery being outweighed by the external drag increment.INTRODUCTIONThe performance of a Jet-powered or jet-assisted airplane dependsProvided by IHSNot for ResaleNo reproduction or ne
11、tworking permitted without license from IHS-,-,-2 CONFIDENTIAL NACARMNo. A8F216upon the efficiency attained in supplying air to the Jet engine.Several types of inlets are capable of efficiently supplying air toa Jet engine but have one or more of the following disadvantages:1. A ducting system which
12、 severely handicaps the internalarrangement of the airplane2. Large external drag increments3. Insufficient area to handle the large quantities of airrequired for Jet enginesIn an effort to overcome these disadvantages with a minimumsacrifice of efficiency, submerged inlets were developed, and there
13、sults of experimental investigations of these inlets are presentedin references 1 and 2. These references show the results of varyingthe many design parameters of NACA submerged inlets and the use ofthese results in design procedure. These results were obtained atsmall scale using a submerged entran
14、ce installed in one of the wallsof a small wlnd-tunnel test section. A need for investigation ofsuch inlets at large scale was apparent. Presented herein are theresults of an investigation of the design parameters at large scaleof an NACA submerged inlet installed on a model of a fighter-typeairplan
15、e in the Ames 40-by 80-footwind tunnel. The scope of thepresent investigation included the determination of the pressure-recovery characteristics of this submerged installation and thecomparison of these characteristics with results obtained fromsmall-scale tests of a similar air intake. In addition
16、, tests weremade to determine the effect on pressure recovery of a systematicvariation of ramp divergence. Pressure-distribution measurementswere also made from which critical Mach numbers of the variousconfigurations were predicted.SYMBOLSangle of attack referred to fuselage center llne, degreesa v
17、elocity of sound, feet per secondA duct area, square feetd duct depth, inchesCD drag coefficient _ACD change in drag coefficientCONFIDENTIALProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACA RM No. A8F21 CONFIDENTIAL 3H total pressure p + q (i+_) ,
18、 pounds per square footZ_H loss in total pressure, pounds per square footD drag of airplane, poundsM Mach number (V/a)m mass flow through duct (pAV), slugs per secondp static pressure, pounds per square footP pressure coefficient (P-Po_qo #0 mass density of air, slugs per cubic footq dynamic pressur
19、e _2_0V_, pound per square footS wing area, square feetV velocity, feet per secondw duct width, inchesy distance above fuselage surface, inchesz ramp width at beginning of ramp, inches8o,ooo “ “ “)8 boundary-layer thickness (distance from the fuselage where thevelocity differs by 1 percent from the
20、outer velocity at thatstation), inchesramp divergence (1 -w_1)x i00,percentSubscriptso free streami duct entrance (duct station l)2 assumed compressor inlet (duct station 2)or criticalCONFIDENTIALProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-4 CONF
21、IDENTIAL NACA RM No. A8F21Parametersram-recovery ratioHo-poVI inlet-velocity ratioVonD internal duct efficiency _HI-TI_ or i ql(l+nl)DESCRIPTION OF MODEL AND APPARATUSThe submerged entrance was located in one side of a full-scalemodel of a jet-propelled fighter airplane. The center of the sub-merged
22、 entrance was located 16 percent of the wing root chordforward and 21 percent of the wing root chord above the leading edgeof the wing-fuselage juncture. A general view of the model mountedin the tunnel is shown in figure i. A schematic drawing showing thegeneral arrangements, instrumentation, and p
23、rincipal dimensions ispresented in figure 2. Fuselage nose coordinates are presented infigure 3.The geometrical characteristics of the submerged-entranceconfigurations are shown in figures 4, 5, 6, and 7. These character-istics can be defined by means of the following five parameters:1. Width-to-dep
24、th ratio - the ratio of duct entrance width toentrance depth2. Lip shape -the profile of the entrance lip3. Distribution of ramp shape -the variation, with percentramp length, of the nondimensional ordinates definingthe ramp plan form4. Ramp angle -the angle between the floor of the ramp andthe exte
25、nsion of the fuselage contour line5. Ramp divergence -a function of the ratio of the rampwidth at the beginning of the ramp to the width of theduct entrance (i _ z_) i00, percentWlCONFIDENTIALProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACARMNo.
26、A8F21 CONFIDENTIAL 5For all the configurations tested, the entrance area and thewidth-to-depth ratio were held constant at 0.667 square feet and3.8, respectively. The lip shape, as shown in figure 6_ was thesame for all configurations. The distribution of ramp shape wasfixed; that is, at any station
27、, given in percent of the total length,the ratio of the ordinate to the maximum ordinate was constant. T_shape distribution was related to the divergence in that the maximumordinate was taken as the percent divergence, thus the ordinates forany divergence will be a constant percentage of the ordinat
28、es forlO0-percent divergence.For the series of plan forms shown in figure 4, the divergencewas held constant at 91.7 percent, and the ramp length was variedsuch that ramp angles of 5, 7, and 9-1/2 were obtained. These planforms, referred to herein as the standard curved-diverging ramp planforms, hav
29、e the same plan form as the curved-diverging ramp planforms found to be satisfactory at small scale (reference 1). For theseries of plan forms shown in figure 5, the ramp angle was held constantat 7, and the divergence was varied from 0 percent (parallel walls) toa maximum of 98.7 percent.Deflectors
30、 were constructed for the 7 and 9-1/2 ramps withstandard divergence. The deflector coordinates are shown infigure 8. The design of the deflectors was based on shapes foundto be satisfactory from tests on a small-scale model. (Seereference 3.) Views showing the deflectors installed on the modelare sh
31、own in figure 9.The entrance station (duct station l) was located 6-1/2 inchesaft of the submerged-lip leading edge. The duct was of constantarea from a station 3-1/2inches forward to a station 3 inches aftof the entrance station. The pressure recovery was measured at theentrance station by 162 equa
32、lly spaced total-pressure tubes and 25static-pressure tubes. (See fig. lO.)The rake used to measure pressure recovery at an assumed com-pressor inlet of the jet engine (duct station 2) contained 96 equallyspaced total-pressure tubes and 40 static-pressure tubes. The ratioof duct area at this compres
33、sor station to area at the entrance was1.52.Total-pressure rakes were used to measure boundary-layerthickness on the basic fuselage. The basic fuselage contours wereobtained by replacing the ramp and entrance by a filler block. Thebasic fuselage with the boundary-layer rakes installed is shown infig
34、ure ii.CONFIDENTIALProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-6 CONFIDENTIAL NACA RM No. A8F21Static pressure distributions along the ramp and over the lipcontours were obtained by means of flush orifices located along thecenter line of the ramp
35、 and center line of the lip inner and outersurfaces. (See fig. 6 for lip orifice stations.) Additional staticpressure distributions over the lip inner and outer surfaces wereobtained withsimilar flush orifices located 25 percent of the ductwidth (4-3/4 in.) from the center line of the duct.Total-pre
36、ssure tubes, used in obtaining ram recovery, wereconnected to an integrating water-in-glass manometer which providedan arithmetic mean value of loss of total pressure. Individual tubereadings of this integrating manometer and all other manometers wererecorded photographically.The internal-flow syste
37、m included an axial-flow fan which wasnecessary to provide the desired range of inlet-velocity ratios.Flow control was obtained by varying the speed and direction ofrotation of the motors. The quantity of internal air flow wascomputed from the readings of 20 equally spaced total-pressure tubesand 8
38、static-pressure tubes at the air outlet.TESTSIn order to evaluate the effect of entrance conditions on theduct losses, the internal duct efficiency was determined prior toinstallation of the duct in the model. An entrance nozzle wasattached to the duct entrance in place of the ramp and lip to assure
39、satisfactory flow conditions at the entrance. The pressure losseswere measured at an assumed compressor inlet (duct station 2), usingthe rake employed to measure pressure recovery at that station duringthe tunnel tests.The aerodynamic characteristics of the standard curved-divergingramp configuratio
40、ns, with and without deflectors, and of the 7 rampwith no divergence were determined for a large number of flightconditions. Data which included pressure-recovery characteristicsat the entrance and at the assumed compressor inlet, and pressuredistribution over the ramp and lip surfaces were obtained
41、 for aninlet-velocity-ratio range of 0.2 to 1.6 and an angle-of-attackrange of -2 to 9. These data were obtained at free-stream veloci-ties of approximately ii0, 160, and 225 miles per hour to illustratethe effects of Reynolds number. The entrance rake was removed fromthe duct during measurements of
42、 pressure recovery at duct station 2.Drag measurements were made to determine the incremental drag resultingfrom the installation of deflectors.CONFIDENTIALProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACA RM No. ASF21 CONFIDENTIAL 7The effect of
43、varying the divergence of the 7 ramp was deter-mined by making pressure-recovery measurements at the entrancestation throughout an inlet-velocity-ratio range of 0.2 to 1.6 withthe airplane at constant angle of attack (-2) and with a constantair-stream velocity of 160 miles per hour.The effect of a t
44、hickened boundary layer on the pressure-recoveryoharacteristics measured at duct station 2 was investigated by thickening_the boundary layer by means of a quarter-inch cotton rope wrappedaround the fuselage at station 27. The boundary-layer thickness wasdetermined on the basic fuselage at station 1_
45、8.2_. Boundary-layermeasurements were made for both the normal and the thickened boundary-layer conditions.RESULTS AND DISCUSSIONReduction of DataThroughout this report the pressure-recovery values consideredare those obtained from the arithmetic average of the total pressuresindicated by the variou
46、s tubes. As shown in reference l, such valuesare not exact since the true pressure recovery is also a function ofthe mass flow at each point. For the subject tests the pressure-recovery values obtained by using the arithmetic average readingswere lower than the values obtained by weighted integratio
47、n of thetotal pressures, the average deviation for a series of conditionschosen at random being of the order of 2 percent with the maximumdeviation being _ percent. Since the arithmetic average values ofpressure recovery were conservative and their use in making compari-sons and showing trends intro
48、duced only minor errors, it was feltthat the additional work required for the more exact reduction of thedata was not justified.Measurements of entrance ram-recovery ratio at inlet-velocityratios below 0.4 were characterized by wide fluctuations; therefore,values obtained at these low inlet-velocity ratios are not usable.It is not known to what extent these fluctuations mayhave beencaused by the entrance characteristics or by the internal ductcharacteristics. Similar fluctuations were not observed during thesmall-scale tests (r_ferences 1 and 2) indicating that the disturb-ance was caus
