NASA NACA-RM-L53K30-1954 Preliminary investigation of the flow in an annular-diffuser-tailpipe combination with an abrupt area expansion and suction injection and vortex-generator .pdf

1、RESEARCH MEMORANDUM PREILMINY INVESTIGATION OF THE FLOW IN AN ANNULSIR-DIFFUSER-TAILPIPE COMBINATION WITH AN ABRUPT AREA EXPANSION AND SUCTION, INJECTION, AND VORTEX-GENEMTOR FLOW CONTROLS By John R. Henry and Stafford W. Wilbur Provided by IHSNot for ResaleNo reproduction or networking permitted wi

2、thout license from IHS-,-,-c NACA RM L53K30 PRFLlNIXARY INVESTIGATION OF TBE AM ABRW AREA EXPANSION AWD SWTION, INJECTION, AND By John R. Henry and StaSford W. Wilbur SUMMARY The performance of an mnula3“diffuser“tailpipe combination with an abrupt area expansion was investigated wlth whereas the ab

3、rupt dump represents the other ex%=, a diffuser with dl the area expansion taken at one station and with a correspondingly inefficient performance. The relative merit of other diffuser designs may be determined through performance campari- sons with these two reference diffusers. The investigation r

4、eported in reference 4 indicated that a performance camparing favorably with the l5O diffuser could be obtained with a shorter dimer of 240 equivalent cone angle and with vortex-generator control. -. “ “ The investigation of the abrupt dump diffuser indicated that the diffusion per unit length in th

5、e constant-area tailpipe was severely penalized by the foiPLation of an extensive vena contracta region down- stream from the abruptly terminated center body. Downstream from the vena contracta region, the diffusion proceeded at a rate camparable to that of the 13O diffuser. These results suggested

6、that an efficient short diffuser design migb.t be obtained by rounding the sharp edge of the abruptly terminated center body in order to elktdnate the vena con- * tracts region anb s“t the diffusion process. Through the use of flow controls, it was anticipated that attached flow could be maintained

7、on a surface of fairly small .raaius. Such a diffuser design has been v investigated, and the results are reported herein. The performance of the modified dump desi- was determined with no controls and with suction, injection, and vortex-generator controls. The inlet conditions corresponded to fully

8、 developed pipe flaw, =an Mach numbers ranging from 0.18 to 0.43, and resulting maximum Reynolds number (based on hydraulic diameter) of approxjmately 1.6 x 10 6 . D B m diffuser outer dieter total pressure- - total-pressure loss . Provided by IHSNot for ResaleNo reproduction or networking permitted

9、 without license from IHS-,-,-MACA RM L53K30 - 3 2 1ongituM distance along tailpipe axis measured from end of center body M Mach number n P P 9, r R U a U V Y 6 s* 0 s*/e tl * exponent in velocity distribution law, = (gr U static pressure static-pressure rise auxiliary air-punping energy coefficient

10、 imgact pressure, E - p radial distance from diffuser center line ratio of auxiliary air-volume flow to main streram-volume flow at inlet station, percent Reynolds number local velocity maxhum velocity occurring in radial velocity distribution perpendicular distance from wall bo-lay thickness bounda

11、ry-layer displacement thickness, s,” (. - 6 boundary-layer momentLmL thickness, :(I- - 5)- boundary-layer shape parameter diffuser effectiveness Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-4 WA RM L53K30 Subscripts: 1 diffuser inlet- statim 2, 3

12、downstream diffuser stations X N R in out S I variable damstream diffuser station no auxiliary flow variable audliary air-flow ratio reference to dLffuser inner wall reference to diffuser outer wall suction injection Bar over symbol indicates a weigh-kd average quantity. APPARATUS m PROCEDURE Test E

13、quipment A diagram of the experimental setup is shown in figure 1. A-tanos- pheric air entered the cylindrical screen section and passed through an inlet bell, which was connected to the test- diffuser by approximately 27 feet of annular ducting having an inner diameter of 1% inches and an outer dia

14、meter varying from 21 to 25 inches. Damstream ducting con- nected the setup to an exhauster. The cen-kr body of the annular approach duct was used as an auxiliary air duct and was connected to a blower or exhauster according to whether injection or suction flow con- trol tests were fn progress. The

15、auxiliary air duct was fitted with a flow-measuring orifice designed and installed according to A.S.M.E. standards (ref. 5). A detailed drawing of the modified dump and ad,-Jacent .ducting is given in figure 2. All internal surfaces and joints were filled and sanded for several feet upstream frm the

16、 diffuser Met station. The plate covering the end of the center body was haxdwood with the outer edge rounded to a 1- - inch radius, as shown, and a groove 1/2 inch deep 1 2 Y a Y Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACA RM L53K30 5 by 1/

17、2 inch wide cut in the downstream face m a 1% - inch-diameter center line. For the auxiliary flow tests, 40 equispaced 3/8-Fnch- diameter holes were drilled 1ongitudLmU.y through the hardwood plate at the base of the groove. Instrumentation Stream total and static pressures were masured by four equa

18、lly spaced, remote-controlled survey rakes at the diffuser inlet station, station 1, and the tdlpipe stations, stations 2 and 3. Flow surveys were made at only one station at a time so that there were no instru- ments in the stream ahead of the measuring statim. Stagnation- temperature measurements

19、were Wen at a point in the approach annulus several diameters upstream from the diffuser Wet, and measurements of the stagnation pressure ad temperature were taken in the amdllary air duct about 1 inner-body diameter from the hardwood plate. Static-pressure orifices extending from upstream of the di

20、ffuser inlet station to station 3 were installed along a single generatrix on the outer wall. At each of the three stations, four equispaced static orifices were located on the outer u“. All pressure measurements were made with multftube manometers con- ta,ining a fluid whose specific gravity was 1.

21、75. The manmeter scales were read to the nearest tenth of a centimeter. Tests Before drj%lli the 3/8-inch-dimeter holes Fn the hazdwood plate (see fig. 2), the performance of the modified better mixing was investigated briefly and the results are presented in fig- ure 5(d). It was found necessary to

22、 move the vortex generators 6 inches upstream to obtain appreciable performance. gains, which effectively lengthened the diffuser undesirably. A gain in static pressure at the l-diameter station of 32 percent over that for no control is indicated for the upstream vortex-generator location. The data

23、of figure 5 are sufficient for sketching rough appro- tions of the flow patterns in the diffuser for the various control sys- tems. Such sketches, figure 6, are useful for illustrating the flow immediately damatream from th whereas at tailpipe lengths of I diameter or more, ug to 2-percent injection

24、 had no effect. Total-pressure-loss coefficient. - Measured total-pressure-loss coef - ficients (not corrected for pumping power) between the inlet and sta- tions 2 and 3 are presented in figure 12 as a function of inlet Mach nmiber. Faired curves at constant values of percent auxiliary sir flaw are

25、 Included. The reduction in loss coefficient with increasing per- centage of suction air flow is indicative of the Ehrmouzlt of low-energy air removed by the suction. The large drop in loss coefficient with increasing injection air flow, however, is largely due to the high energy of the injection ai

26、r raising the mean energy of the stream exclusive of any flow-control effects. A ccmpazison of the injection dab for sta- tions 2 and 3 Fndlcates mixing and friction-loss coefficients in the tailpipe length from stations 2 to 3 of approximately 2 percent. The vortex-generator instkllation which was

27、located 6 inches upstream from the end of the center body produced at an inlet Mach number of 0.26 a loss coefficient about 7 percent higher than that for no control. Coefficients corrected for puqing power.- An accurate assessment of control performance which involves m“b of auxiliary flm cannot be

28、 accomplished unless pumping parers =e considered in the performance panmeters. Pumping-power coefficients calculated Tor the purpose of coefficients according to methoas described in a previous secticm are presented in figure 13 as a function of percent of auxiliary flow. At a given percent auxilia

29、ry air flow, the pump- parers for suction were slightly greater than -those for injection. The pumping powers increase rapidly with increased auxiliary flow and approach values equivslent to the no-control to-t;al-pressure-loss coefficient at -res of R from % to 4 percent. The pumrp pressure rise re

30、quired for a given due of R is readily calculable from the data of figure 13. -Since no effort waa expended in determining optlmum meas for the awdliary air holes, the pumping-power-coefficient values must be regarded as somewbt arbitrary. - correcting the measured static-pressure-rise and total-pre

31、ssure-loss The diffuser effectiveness including the pmging-pmr correction and based on the static-pressure-rfse measurements to station 2 are pre- sented in figure 14 as a function of percent auxiliary drflar. Total- pressure-loss coefficients at station 3 corrected for pmrping power are also includ

32、ed. ,- For purposes of comparison, performance point8 for no Control for - subject-diffuser data indicates that with no control the modified dmg the- 15O diffuser of reference 2 and the sharpedged d- of reference 3 have been indicated in figure 14. Carrparison of these data with the Provided by IHSN

33、ot for ResaleNo reproduction or networking permitted without license from IHS-,-,-12 produced about 80 percent of the loss coefficient of the sharp-edged dtrmp, but exceeded the loss of the diffuser by 170 percent. It is apparent frm figures 5 and 14 that, although injection and suction controls wer

34、e effective, the pumping powers required for the subject confi;uration were excessive and prevented the attainment of the performance of the 15O dif- fuser. These results suggest- that means for reducbg the required purging parers should be investigated mther. PLrmping parers may. be reduced both by

35、 determFning more efficient auxiliary-air-system designs and by reducing the control requirements of the Wfuser by acccnnplishbg more of the dif- fusion in the conventional manner ofincreasing the- cross-sectional area at moderate rates prior to dumping the flow. coNcLusIoNs The performance of an an

36、nular-diffuser-tailpipe combination with an abrupt area expansion was investigated with and without flgw controls in the form of suction, injection, and vortex generators. The diffuser had a 2l-inch-diameter straight outer wall, an area ratio of lg to 1, and fully developed pipe flow at the inlet. m

37、et Mach nmiber was wied between 0.18 and 0.43 with a resulting maximum Reynolds number (based on inlet hydraulic diameter) of appromtely 1.6 x lo6. The. ratio of the auxiliary air flow tu the flow of the main stream was varied from. 0 to approximately 4 percent. The following conclusians are present

38、ed: 1. Rounding the sha-p edge of the- terminus of the center body to a radius of 11 inches was responsible for the elimination of. the vena con- tracta effect resulting in a 200-percent increase in the static-pressure rise across the first diameter of length of tailpipe as campared with that obtain

39、ed with the sharp-edged dump. .“ 2 2. At the same tailpipe station, the rounded dunp produced 69 per- cent of the static-pressure rise of a 15O mnular diffuser previously investigated. The longitudinal static-pressure distributions indicated. that a larger radius at the center-body terminus probably

40、 would produce =her gains in performance. . . . . - .- “ “ 3. Suction and injection control produced 40- and 43-percent increases, respectively, in the measured static-pressure rise to the I-diameter tail- pipe station. A vortex-generator installation produced a 32-percent increase in the static-pre

41、ssure rise to the s8me station; however, it was necessary to locate the generators an appreciable distance upstream from the diffuser which effectively lengthened the diMtzser undesirably. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACA IIM L53K

42、30 13 4. Radial pressure surveys at the tailpipe stations shared that the higher injection rates produced roughly uniform increases in the tow pressure across the sectfon; whereas suction control increased the total pressure and velocity near the tailpipe center line at the expense of the total pres

43、sure in the outer half of the duct. 5. The pumping powers reqared for suction and injection con-h-01 for the configuration tested were excessive and reduced the measured performance gains of approldmately 40 percent to gains corrected for pumping power of 13 and 6 percent for suction and injection,

44、respec- tively. This result suggests that in order to reduce the pmrping parers the basic diffuser design should be *roved and more efficient auxiliary- adr-system designs should be determined through further investigations. Langley -Aeronautical Laboratory, National Advlsory Committee for Aeronauti

45、cs, Langley Field, Va., November 17, 1953. 1. Wood, Charles C.: Prelimhazy hvestigation of the Effects of Rec- tangul-as Vortex Generators on the Performance of a Short 1.9:1 Straight-Wall Annular MfRzser. WA RE9 LglaOg, 1951. 2. Wood, Chmles C., and Eigginbotham, James T.: The Influence of Vortex G

46、enerators on the Performance of a Short 1.9:l Straight- Wall Annular Diffuser With a Whirling Inlet Flow. NACA RM L52Lola, 1-953 3. Wood, Charles C., and Eigginbotham, James T. : Flaw Diffusion in a Constant-Diameter Duct Damstream of an Abruptly Terminated Center Body. WCA RM L53D23, 1953. 4. Wood,

47、 Charles C., and Higginbotham, James T.: Performance Charac- teristics of a 24O Straight-Outer-Wall Annular-Mffuser”Tailpipe CmbFnation Utilizing Rectmgulm Vortex Generators for Flow Control. MACA RM L53B17a, 1953. 5. Anon. : Flow Measurement by Mans of Standardized Nozzles and Orifice Plates. Suppl

48、ement an Instruments and Apparatus, pt. 5, ch. 4, Power Test Codes, A.S.M.E., 1949, pp. 5-62. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-. . . . . . . . . . . . . . . . . . - AUXILURY AIR DUCT Plgure,l.- Diagram of apparatus. Arrowdenote directi

49、on of air flow. . I A . 1 . ! , . I 1: I i . . I s2 P B !? I . . . Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-? Figure 2.- General arrangement of modified dump and tallpipe. All dhennions are In inches. . . . Provided by IHSNot for ResaleNo reproduction or networking permitted without license