1、RESEARCH MEMORANDUM THE EFFECT OF SURFACE ROUGHNESS ON THE PERFORMANCE OF A 23 CONICAL DIFFUSER AT SUBSONIC UCH NWMBERS By Jerome Persh Langley Aeronautical Laboratory Langley Field, Va. NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS WASHINGTON January 16, 1952 . . “ I ., Provided by IHSNot for ResaleN
2、o reproduction or networking permitted without license from IHS-,-,-1s NACA RM L5LKo9 F NATIONAL ADVTSORY COMkTTEE FOR AERONAUTICS Tm EFFECT OF SURFACE ROUGRNESS OM TRE PFRFORMlwcE OF A 23 comm DIFFUSER m SUBSOKIC MACH mms By Jerome Persh An investigation was conducted to determine the effect of sur
3、face roughness on the performance of a 23 conical diffuser with a 2:l ratio of exit to inlet area and with a constant-area tail pipe about 33 inlet diameters in length. The inlet-boundary-layer thickness was of the order of 5 percent of the inlet diameter. The air flows used in this investigation co
4、ver an inlet Mach number range from 0.10 to 0.64, cor- responding to Reynolds numbers of 10 6 to 6 X 10 based on inlet dime- 6 ter. The surface of the diffuser was coated with cork particles of a downstream edge after each series of pressure measurements were made and the variation of diffuser perfo
5、rmance with percent of diffuser length roughened thereby determined. 1 rr .r controlled .size. Incremental bands of roughness were removed from the The total-pressure losses increased considerably as the extent of the roughness was increased in the downstream direction from the smooth condition to a
6、bout 48. percent of the diffuser length. The total- pressure-loss coefficient,diminished smoothly from a maximum value when about 48 percent of the diffuser length was roughened to a value lesa than that measured at the tail-pipe exit for the smooth-surface diffuser, when the diffiser was almost ful
7、ly roughened. The static-pressure recovery diminished as the extent of the roughness was increased from the smooth condition to the almost fully rough condition. The results indicated a progressive diminution of the static-pressure recovery and a continuous increase in the total-pressure lossea as t
8、he inlet Mach number was increased for all configurations. In contrast to the fluc- . hating flow found in the sane diffuser with smooth walls, the flow W Tor both cases the inlet-boundary- layer thickness is of the order of 5 percent of the inlet diameter. All comparisons between the data presented
9、 herein and those of reference 2 are made for this inlet-boundary-layer thickness. Performance curves are plotted against the inlet pressure ratio . . 1% as in reference 2. A curve of inlet flow characteristics is presented in figure 3, which gives the variation of inlet Mach number with inlet press
10、ure ratio. This correlating parameter was chosen as an approximate index of inlet Mach number and flow rate. Flow in roughened diff“user.- One of the primary purposes of the present. investigationwas to give detailed attention tothe flow in the diffuser. In contrast to the flow in the smooth-surface
11、 diffuser which shifted position from time to time and lacked axial symmetry, the flow in the -roughened diffiser was steady and had approximately symmetrical velocity profiles gt the diffuser exit for all.configurations investi- gated, Velocity profiles .measured at sta$ion 6 with _the rotating pre
12、ssure rake (fig, 4) are shown in figure 6, for configurations I11 and IV. Tuft surveys indicated that the extensive areas of reversed flow found in the smooth-surface diffuser were absent in the roughened diffuser .although very low wall velocities were noted in the downstream “ , ? Y “ Provided by
13、IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-2s r. mD NACA RM L5lKD9 “ SECURITY INFORMATION 9 regions. Pressure surveys at the diffuser exit confirmed this observa- tion, Unlike the smooth-surface diffuser, it was possible to make detailed pressure slirveys a
14、t the end of the conical expansion, because of the steadiness of the flow, and to determine the performance parameters at the diffuser exit. Pressure Survey Results As pointed out in reference 2, d5fficulty in making pressure surveys at the diffuser exit made it lmpractical in that investigation to
15、pre- sent values of Ah/qcl at that point. In the present investigation, however, the steady flow at station 6 made it possible- to make detailed pressure surveys at that point and values of Ah/qc, are presented at both station 6 and station 7. The diffuser effectiveness is the only performance param
16、eter for .which a comison can be made at station 6, between the smooth-surface diffuser results of reference 2 and the roughened diffuser . Weight-flow check.- The weight flow was calculated for each configu- ration at each measuring station to determine whether true mean values of total and static
17、pressure are recorded by the pressure tubes used in this investigation. The results of this check are shown in figure 7, in which the weight flow is plotted as a Function of the inlet pressure ratio, for all configurations at each of the measuring stations. For all configura- tions (figs. 7( a) to 7
18、( d) ) the weight flows calculated from pitot-tube measurements at both the diffuser exit and the tail-pipe exit are slightly-higher than the weight flows calculated from pitot-tube meas- urements at the inlet over the entire speed range. This apparent incon- sistency may partially be traced to the
19、behavior of a total-pressure tube in an air stream containing axial fluctuating velocities. As pointed out in reference 4 the total pre-ssure recorded by 8 pitot tube in such an air stream may be expressed by the following approximate relation: h = p + $(u2 + F2) 2 Because T2 is always positive, the
20、 apparent total pressure is always greater than the mean total pressqre by the amount of h(T2). Since the axial turbulent fluctuating velocities do not contribute to the weight flow, it caa be seen that the weight flow calculated from total- pressure measurements obtained by the pitot-tube method in
21、 a stream with axial fluctuating velocities will always be higher than the- actual weight flow. 2 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-10 Total-pressure-loss coefficient. - The variation of compressibility corrections were not included. *
22、“ c It is of particular interest to compare the diffuser exit (sta- tion 6) velocity profiles measured in the smoth-surface diffuser (refer- ence 2) and those measured in the roughened diffuser. For this compari- son, station 6 velocity profiles have been plotted. in figure 21 for each of the config
23、urations investigated at an approximately constant inlet * pressure ratio of 0.90 along with a station 6 velocity profile measured in the smooth-surface diffuser (reference 2) at the same inlet pressure . “ I Provided by IHSNot for ResaleNo reproduction or networking permitted without license from I
24、HS-,-,-ratio PI/%. .Figure 21 indicates that the flow separation in the smooth- Although detailed pressure surveys were not made for configuration IT,* the rotating rake pressure measurements made at atation 6 did not indi- cate any areas of flow separation. These results when considered with the wa
25、nd tuft surveys made along the length of the diffuser indicate that the roughness suppresses the flow separation that was found in the smooth- surface diffuser. In accordance with results previously discussed in the section entitled “Loss Coefficient,“ .it is agparent that the over-all boundary-laye
26、r thickness 6 is less for configuration I thin the other roughness configurations as well as the smooth-surface configuration. surface diffuser at station 6 is absent for configurations I, 11, and 111. The velocity profiles in the tail pipe (station 7) sham in fig- ures 18(b), lg(b), and X)(b) for c
27、onfigurations I, II, and III, respec- tively, all show evidenceof the distorted velocity profiles measured at station 6 -(figs. 18(a), lg(a), and X)(a) which indicates that fully developed pipe flow is not yet established in the length of tail pipe- provided. This observation is supported by the dat
28、a of Peters (refer- ence 5) which show that the length of tail pipe needed to attain fully developed ptpe. flow is wre than twice that used in the current investigation. CONCLUSIONS From the current investigation regarding the effect of surface roughness OR the performance of a 230 conlcal diffuser
29、with a 2:l ratio of exit to inlet area and with a constant-area tail pipe .3- inlet dim- eters in length, with an inlet-boundary-lEtyer thickness approximately 5 percent of the inlet diameter, the folloufng conc-lusions are drawn,: 1 2 1. The total-pressure-loss coefficients measured at the diffuser
30、 exit and the tail-pipe exit increased considerably as the extent of the roughness was increased in the downstream direction from the smooth * condition to about 48 percent of the diffmer length. The total-pressure- loss coefficients at both measuring stations diminished smoothly from a maximum valu
31、e when about 48 percent of the diffuser length was roughened to a value kasurably less than that measured at the tall-pfpe exit for the smooth-surface diffuser, when the dlffuser was almost fully roughened. 2. The static-pressure recovery diminished as the extent of the roughness was increased from
32、the smooth condition to the amst fully rough condition, at both the diffuser exit and the tail-pipe exit, - 3. For all.configurations investigated, the total-pressure losses Y increased continuously and the static-pressure recovery progressively Provided by IHSNot for ResaleNo reproduction or networ
33、king permitted without license from IHS-,-,-_I SECURITY 1NFORM.TION NACA RM L51KO9 . diminished as the inlet Mach number was increased. For all roughness configurations investigated the rate of increase of the total-pressure- - loss coefficient with increasing Mach number, at both downstream meas- w
34、ing stations, is essentially the same as that found in the smooth- surface diffuser at the tail-pipe exit. . “ - 4. For all roughness configurations investigated the flow in the diffuser was extremely steady over the entire speed range, 5. No evidence of separation was found within the diffuser for
35、any of the roughness configurations- investigated. Reversed flow was not detected at the measurement point- closest to the wall for any of the roughness configurations. Langley Aeronautical Laboratory National Advisory Committee for Aeronautics Langley Field, Va. 1. Jones, R., and Williams, D. H.: T
36、he Effect of Surfsce Roughness on - the Characteristics of the Aerofoils N.A.C.A. 0012 and R.A.F. 34. R. & M. No. 1708, British A.R.C., 1936. b 2. Persh, Jerome: The Effect of the Inlet Mach Number and Inlet-Boundary- Layer Thickness on the Performance of a 23 Conical-Diffuser - Tail- Pipe Combinat
37、ion. NACA RM LgIUO, 1950. . . . L 3. Nikuradse, Johann: Laws of Flow in Rough Pipes. NACA TM 1292, 1950. 4. Tillmann, W.: Investigations of Some Particularities of Turbulent Boundary Layers on Plates. Reps. and Translations No. 45, British M.A.P. Vblkenrode, March 15, 1946. (Issued by Joint Intellig
38、ence - Objectives Agency with File No. B.I.G.S.-19.) 5. Peters, H.: Conversion of Energy in Cross-Sectional Divergences under Different Conditions of Inflow. NACA TM 737, 1934. I Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-. . I e . . . . . . . u
39、1 P . . . . . I Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-. . . . . . . . . . I c, ,I Figure 2.- Photograph of test setup shrmhg inlet pipe, diffuser, and tail pipe. . c .- . . I 1 . . . .I I . . . . . . . . . . . . . . . . . . , ! a s. H Provi
40、ded by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-3s - NACA RM L51Ko9 Figure 3.- Photograph of diffgser with roughness installed. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-18 - EiECURITY INFOFMATION NA
41、CA RM L5W . Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACA RM L51KO9 e6 % .4 .2 0 SeCURITY INFCWATION - “_ 1.0 .96 92 .88 .a .8Q Figure 5.- Variation of inlet Mach number with inlet pressure ratio. Provided by IHSNot for ResaleNo reproduction o
42、r networking permitted without license from IHS-,-,-X) 32 28 26 20 16 rs fn. 12 8 4 0 _ SECURITY INFORMATION NACA RM L51W 0-e (a) Configuration 111. 1 Figure 6 .- Velocity profile at station 6. . ? Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-0 21
43、 0 3.20 l& 200 240 u, ft/aea (b) Configuration IT. Figure 6.- Concluded. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-22 t 80 0 (a) Configurations I and I1 Figure 7.- Variation of weight flow with inlet pressure ratio. Provided by IHSNot for Resal
44、eNo reproduction or networking permitted without license from IHS-,-,-MCA RM w, 0 1.0 .% 92 .88 .84 .80 (b) Configurations III and IV. Figure 7.- Concluded. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-24 .20 .16 e 04 0 I - SECURITY INFORMATION MC
45、A RM L5W (a) Diffuser exit, station 6 “ . . Figure 8.- Variation of tot.al-pr?ssure-loss coefficient with met -. pressure ratio. “ Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-MACA RM L5mg . .20 RD SEXURITY INFORMATION I I configuration 1.0 96 092
46、 .88 084 .80 .76 (b) Tail-pipe et, station 7. Figure 8.- Concluded. . Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-k . El = 0.90 bo Peraent diffuser length mugbned Figure 9.- Variation of total-pressure-loss coefficient with pmcent of diffuser len
47、gth roughened. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-.8 .6 . . I (a) Diffuser exit, station 6. Figure 10.- Variation of diffuser effectiveness with Met pressure ratio. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-. 1.0 .9 *8 .? .6 -0 111 n I“ L 1.0 0% .92 .# (b) Tail-pipe exit, station 7. Figure 10.- Concluded. . Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-Iv Dif Ftlser exit Figure 1l.- Variation of diffuser effectiveness wit
