1、RESEARCH MEMORANDUM PRJ3IJMCNARY INVESTIGATION OF THE EFFECTS OF RECTANGULAR VORTEX GENERfLTORS ON THE PEWORMANCE OF A SHORT L9:l STRAIGHT-WALL ANNULAR DIFFUSER By Charles e. Wood NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS Provided by IHSNot for ResaleNo reproduction or networking permitted without
2、 license from IHS-,-,-1 4 NACA RM LSlGOP PRELIMINARY INVESTIGATION OF THE EFFECTS VORTEX GENERATORS ON THE PERFORp4ANCE CF RECTANGULAR CF A SHORT 1.9 :1 STRIIIGHT-WAIL ANNULAR DIJ?EUSE.R By Charles C. Wood A prelimiqary investigation was conducted in a duct system having fully developed pipe flow to
3、 determine the effectiveness of vortex gen- erators in improving the performance of a 21-inch-diameter straight- outer-wall aular diffuser haviug an over-dl1 equivalent conical expan- sion angle of lSo and a L9:l area ratio. The vortex generators used in this investigation were rectangular noncamber
4、ed airfoils which ware varied in chord, span, angle of attack, number, and location. Without vortex generators, separation occurred at appromtely 4 inches downstream of the cylindercone junction with consequent veloc- ity fluctuat5ons at the diffuser percent- chord station. Some tests were made with
5、 vortex generators on the oker wall as well as on the inner wall. In these tests, the location at sale rows of vortex generators, which were affixed to the psriphery of the outer wall, was varied. These rows were located at 2, a, 10, and 16 inches downstream of the diffuser inlet, 4 complete list of
6、 all vortex-generator arrangements tested is given in table I. Basis of comparison of the effectiveness of vortex generators.- The sepated, rapidly fluctuating flow at the exit of a Kide-angle diffuser prevents measurements necessary to determim the performance of the diffuser. men a tail pipe is at
7、tached to the downstream end of the diffuser, the flow at some point in the tail pipe becomes stable and uniform. The flattening of the velocity profile is accompanied by a static-pressure rise. In this investigation diffuser performance is based on pressure measurements made at a station in the tai
8、l pipe 151 inches (less than one diameter) downstream of the diffuser exit. For those conditions in which stable flow was achieved at the diffuser exit, as for many vortex-generator arrangements, diffuser performance is also referenced to the diffuser,exit station. 2 Provided by IHSNot for ResaleNo
9、reproduction or networking permitted without license from IHS-,-,-6 NACA RM L51GO9 The effectiveness of each vortex-generator configuration on the performance of the annular diffuser has been compared on the basis of the ratio of the actual static-pressure rise APa to the ideal static- pressure rise
10、 Api. The actual static-pressure rise in the diffuser has been calculated as the difference between the average of pressures measured by three equally spaced orifices at the reference station, figure 2, and the average of pressures measured by six equally spaced orifices located on the circumference
11、 of the outer wall at the diffuser exit. The static-pressure rise in the diffuser - tail-pipe combhation was determined in a similar manner, using, however, orifices at the tail- pipe exit rather than at the diffuser exit. The ideal static-pressure rise was calculated using one-dimensional equations
12、. A comparison was also made on the basis of loss coefficient, the ratio of change in weighted total pressure between the diffuser inlet and a downstream station to the mean impact pressure at the diffuser inlet a/%. This comparison is based only on total-pressure measure- ments made at the diffuser
13、 inlet and at the tail-pipe exit. For some test configurations it was impossible to obtain reliable data at the diffuser exit because of unstable flow. RESULTS AM3 DISCUSSION It has been shown in reference 4 that performance of a wide-angle conical diffuser which has large regions of separated flow
14、cam be improved considerably by the use of vortex generators. Since the performance of this annular diffuser, like that of the conical diffuser of reference 4, was strongly affected by flow separation, some measure of imqrovement in diffuser performance was expected from the application of vortex ge
15、ner- ators to this diffuser. Before the performance of a diffuser can be evaluated, the nature of the flow entering the diffuser must be known. Accordingly, pressure surveys were made at three equally spaced stations at the diffuser inlet. The velocity profiles and the tabulated values of boundary-l
16、ayer proper- ties for values of p/g of 0.935 and 0.88 are presented in figure 4. These measurements show that uniform, fully developed pipe flow existed at the diffuser inlet. Diffuser with No Vortex Generators Flow observations.- For this diffuser without vortex generators, visual observation of sm
17、all tufts located on the inner and outer walls of the diffuser when operating at several Mach numbers in the range investigated revealed that the flow separated from the inner wall Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-7 approximately 8 inc
18、hes downstream of the diffuser inlet station but L remained attached to the outer wall throughout the entire diffuser. The line of separation around the body was asymmetrical and unstable. Diffuser performance.- The diffuser effectiveness ApdApi of this bare diffuser is presented in figure 5 as a fu
19、nction of inlet pressure ratio, p/H. The inlet pressure, measured 6 inches upstream of the Met station, gives a somewhat conservative result as some of the pressure drop along the straight pipe is subtracted from the pressure rise con- sidered as occurring in the diffuser. As can be noted from figur
20、e s, the diffuser effectiveness of the diffuser itself is poor; however, when referenced to the tail-pLpe exit station, considerable gain is indicated for the diffuser - tail-pipe combination. Also used to express the Ufuser performance is the diffuser loss coefficient a/ however, velocity fluctuati
21、ons at the diffuser exit station were large. These velocity fluctuations are con- sidered of sufficient magnitude to render this diffuser useless for mst practical applications. - Diffuser with Vortex Generators on Inner Body A number of vortex-generator configurations were investigated in which the
22、 vortex generators were fastened to the inner wall 3 inches downstream of the diffuser inlet station, and effects on diffuser per- fonance of vortex-generator angle of attack, span, chord, and number were determined. Flow observation.- Ebery vortex-generator arrangement on the inner wall resulted in
23、 marked improvement over the bare diffuser, with some arrangements being far superior to others. The tufts indicated the me of separation to be shifted bodily downstream for some vortex-generator arrangements and to be completely eliminated for others. The flow along the outer wall, although more tu
24、rbulent, remained attached. I Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-8 NACA RM 5109 Effect of vortex+enerator angle of attack.- Tests were conducted and values of diffuser effectiveness deternrined for a test configuration consisting of twen
25、ty-four 2-inch-chord, 1-inch-span vgrtex generators in which the angle of attack was varied from to l8i . Results of this phase of the investigation are shown in figure 7(a). Higher performance was obtained with this configuration when vortex generators were set at 15 angle of attack, which value is
26、 in agreement with results of refer- ence 4 determined for a wide-angle conical diffuser. Another arrange- ment consisting of twenty-four pinch-chord, =inch-span vortex gener- ators tested at angles of attack between U.$“ and lso, figure 7(b), indicates no variation in performatlee with angle of att
27、ack in the range investigated. From this limited investigation one might conclude that vortex-generator angle of attack has a rather small effect on diffuser performance;- therefore, unless otherwise noted, all results presented hereafter will be for- configurations in which vortex-generator angle o
28、f attack is 15. 2 1 Effect of vortex-gegerator span.- The variable airfoil spanil appears to be probably the one having the greatest influence on diffuser performance, as determined by this preliminary investigation. Its effects can be readily observed from a cross plot of span as a function of diff
29、user effectiveness, Apa/Api, shown on figure 8. All data for developing this curve were obtained from an installation having twenty- four 2-inch-chord airfoils equally spaced around the inner wall. From this curve it can be noted that the a -inch span and 1 - inch span gave about equal pressure reco
30、veries and about the m.aximm that was obtained by varying the span. These two values of airfoil span are, respectively, 20 and 40 percent of the distance from the inner wall to the point of maximm velocity in the annulus. 2 Effect of vortex-generator chord: Three sets of 24 vortex gener- -4 atom of
31、5 -inch span, having chords of 1 inch, 2 inches, and 3 inches were tested. The effect of vortex-generator chord pn diffuser effec- tiveness is shown in figure 9 as a function of diffuser Lnlet pressure ratio p/g. Results of these tests indicate that variation of generator chord produces no significa
32、nt effect ugon diffuser effectiveness Over the Mach number range investigated. Effect of number of vortex generators.- The variation of diffuser effectiveness with number of vortex generators is presented in figure 10. The construction of this curve is based on limited data. Curves at two values of
33、inlet pressure ratio p/r were faired from the no-generator configuration through values of diffuser effectiveness for 12 and 24 vortex generators of 3-inch chord and for 24 and 48 vortex generators Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-2 NA
34、CA RM LslG-09 9 of l-inch chord. In fairing this curve it was assumed that the effect of these tests. The span of these vortex generators was 1/2 inch and the angle of attack 1So. L of vortex-generator chord was negligible throughout the Mach number range From figure 10 it is seen that the number of
35、 vortex generators has an appreciable effect on diffuser effectiveness. The addition of vortex generators increases diffuser effectiveness, which reaches a maximum when the number is about 24. Longitudinal pressure gradients.- The effects of vortex generators on the longitudinal pressure gradients a
36、long the inner and outer wall of the diffuser are illustrated in figure ll for the diffuser with no generators and for the vortex-generator installation consistlng of twenty-four 3-inch-chord, 1 2-inch-span vortex generators. Upstream of the separation point of the diffuser with no generators, that
37、is, for the first 8 inches downstream ofthe diffuser inlet station, the two curves are practically identical with both configurations indicating a local acceleration region on the inner wall followed by a stronger ratio curve for the diffuser without generators indicates separation, a small amount o
38、f diffusion is still accoqlished in both the remainder of the diffuser and in the tail pipe. It is interesting to note from this figure that the pressure .gradient along the outer wallis less intense than along the inner wall and that the acceleration of air flow noted near the Wer wall does not occ
39、ur near the outer wall. - adverse pressure gradient near the inlet. Even though the pressure- - Total-pressure profiles at diffuser exit station.- Typical total- Pressure Profiles at the diffuser exit station for two vortex-generator - however, this improvement vanishes at higher velocities. 1 A Com
40、parison of Diffuser Performance with and without Vortex Generators By comparison with that of the diffuser having no vortex generators, the performance of the annular diffuser with vortex generators represents a substantial improvement, figure 14. The addition of vortex generators to the inner wall
41、resulted in a gain in diffuser effectiveness of about 15 percent over the larger portion of speed range tested. Use of vortex generators on the outer wall, in combination with those on the inner wall, increased. the diffuser effectiveness about 17 percent with resulting values of diffuser effectiven
42、ess above 90 percent for low speeds. The gains as measured at the tail-pipe station show some improvement, but, - Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-since effective diffusion was obtained in the diffuser with vortex generators, little ad
43、ditional pressure rise could be expected to occur in the tail pipe. . The effect of vortex generators on the diffuser-tail-pipe loss coefficient is also shown in figure lL(b). Because of flow instability, measurements of total-pressure loss could not be made at the diffuser exit station for the diff
44、user without vortex generators. The data obtained at the tail-pipe station show rather large loss in the low-speed range reaching a lrdnimum at a Mach number of about 0.25 and increasing again with further increase in speed. The addition of vortex generators to the inner wall reduced the total-press
45、ure loss coefficient to a minimum of about 3.5 percent at a Mach number of about 0.25. The total- pressure-loss coefficient Over much of the speed range of these tests was less than 5 percent. glthough no data are presented for the con- figuration with vortex generators oa both the inner and outer w
46、alls, it is believed that any further increase due to the outer-wall vortex generators would be small. CONCLUSIONS The following conclusions are drawn as to the effect of various vortex-generator arrangements on the performance of an annular straight- wall diffuser with an outer diameter of 21 inche
47、s and an area ratio of 1.9 to 1 with fully developed pipe flow at the diffuser inlet. Rec- tangular noncambered airfofls were used as vortex generators and were varied in chord, span, angle of attack, number, and location. The results contained herein are preliminary and do not necessarily represent
48、 the maximum pressure recovery and flow stability obtainable with the generators employed, as the optimum station for mounting generators on the inner wall was not determined. It is felt, however, that a high percentage of that obtainable was realized. 1. Ehery vortex-generator configuration tested
49、resulted in improved performance to some degree. 2. The vortex-generator configuration giving the best performance consisted of 2k equally spaced 3-inch-chord, 5-inch-span airfoils at 13g angle of attack located on the inner wall 3 inches downstream of the diffuser inlet station and ,!& equally spaced 1-hch-chord, 5 -inch-span 1 airfoils at lso angle of attack located on the outer wall 8 inches downstream o
