NASA-TP-2683-1987 Experimental cavity pressure distributions at supersonic speeds《超音速下实验性空腔压力分布》.pdf

《NASA-TP-2683-1987 Experimental cavity pressure distributions at supersonic speeds《超音速下实验性空腔压力分布》.pdf》由会员分享，可在线阅读，更多相关《NASA-TP-2683-1987 Experimental cavity pressure distributions at supersonic speeds《超音速下实验性空腔压力分布》.pdf（77页珍藏版）》请在麦多课文档分享上搜索。

1、NASA Tec hn ica I Paper 2683 June 1987 runsn Experimental Cavity Pressure Distributions at Supersonic Speeds Robert L. Stallings, Jr., and Floyd J. Wilcox, Jr. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NASA Tec hnica I Paper 2683 1987 National

2、Aeronautics and Space Administration Scientific and Technical Information Office Experiment a1 Cavity Pressure Distributions at Supersonic Speeds Robert L. Stallings, Jr., and Floyd J. Wilcox, Jr. Langley Research Center Hampton, Virginia Provided by IHSNot for ResaleNo reproduction or networking pe

3、rmitted without license from IHS-,-,-Summary An experimental investigation has been con- ducted to define pressure distributions for rectangu- lar cavities over a range of free-stream Mach numbers and cavity dimensions. These pressure distributions together with schlieren photographs were used to de

4、- fine the critical values of cavity length-to-depth ra- tio that separate open type cavity flows from closed type cavity flows. For closed type cavity flow, the shear layer expands over the cavity leading edge and impinges on the cavity floor, whereas for open type cavity flow, the shear layer brid

5、ges the cavity. The tests were conducted by using a flat-plate model that permitted the cavity length to be remotely var- ied from 0.5 to 12 in. Cavity depths and widths were varied from 0.5 to 2.5 in. The flat-plate boundary layer approaching the cavity was turbulent and had a thickness of approxim

6、ately 0.2 in. at the cavity front face for the test range of Mach number from 1.50 to 2.86. Values of (l/h)cT obtained when decreasing cavity length were generally less than those obtained when increasing cavity length. Values of (l/l) ranged from 10 to 13 for the present tests. A large improvement

7、in the correlation of measured cavity centerline pressure distributions for cavities of vari- ous depths was obtained when both the cavity width- to-depth ratio (w/h) and length-to-depth ratio (l/h) were held constant rather than l/h alone. The ef- fects of cavity width on the cavity pressure distri

8、- butions were much greater for cavities having closed or t,ransitional flow fields than cavities having open flow fields. Decreasing cavity width resulted in a re- duction in (l/),-. Three-dimensional effects in the form of large lateral pressure gradients occurred on the rear faces of the cavities

9、 that had closed cavity flow fields. Introduction Numerous irivestigations have been conducted over the past several decades to investigate the flow fields over cavities and to define the resulting local pressure distributions and acoustic levels within the cavities (e.g., refs. 1 through 6). These

10、investigations have been conducted over a speed range from sub- sonic through hypersonic Mach numbers. The results obtained at supersonic speeds are particularly impor- tant for application to cavities on contemporary and future aircraft and missile configurations capable of sustained supersonic fli

11、ght speeds. Some examples of requirements for cavities on these configurat,ions con- sist of weapon bays for high-speed military aircraft and recessed areas on wrap-around-fin missiles that contain the fins before they are deployed. Existing data available in the literature show that cavity flow fie

12、lds can occur that result in large lo- cal turning angles of the shear layer over the cavity; this gives rise to large cavity drag levels (e.g., refs. 7 and 8) as well as large impact pressures on compo- nents within the cavity. Such cavity flow fields can also result in adverse separation character

13、istics for a store being launched from the cavity (e.g., refs. 9 and 10). Large fluctuating pressure levels can also occur in cavities, which sometimes are severe enough to cause component failure of hardware within the cavity (ref. 11). In general, data available in the literature show that at supe

14、rsonic speeds, there are two fundamen- tally different types of cavity flow fields which have been classified as open and closed cavity flows. The type of flow field appears to be primarily a function of cavity length-to-depth ratio (l/h). As illustrated in figure 1, for values of l/h 13, the cavity

15、 flow field is generally of the closed flow type. For this case, the shear layer expands over the cavity leading edge, impinges on the cavity floor and exits ahead of the rear face. Typical cavity floor pressure dis- tributions for this case consist of low pressures oc- curring in the expansion regi

16、on behind the front face followed by an increase in pressure and a pressure plateau occurring in the impingement region. Fur- ther downstream, as the shear layer approaches the cavity rear face, the pressure levels again increase and reach a maximum value just ahead of the rear face. The local flows

17、 over the cavity front and rear faces for the closed cavity flow field are very similar to the flows over rearward-facing and forward-facing steps, respectively. At l/h M 12, the cavity flow field is on the verge of changing from closed cavity flow to open cavity flow (decreasing l/h) and is referre

18、d to as “transitional cavity flow.” For this case, the shear layer turns through an angle to exit from the cavity coincident with impinging on the cavity floor result- ing in the impingement shock and the exit shock col- lapsing into a single wave. The corresponding pres- sure distribution shows tha

19、t the extent of the plateau pressures in the impingement region has diminished and the pressure increases uniformly from the low values in the region aft of the front face to the peak values ahead of the rear face. For l/h 10.5, transitional flow occurred for l/h = 10.5, and open cavity flow occurre

20、d for l/h 1.0 are invariant with yl/h; however, the magnitudes of the pressure measurements are sensitive to the type of cavity flow field and are essentially equal to the pressure level at the most forward instrumented sta- tion on the cavity floor. In general, for the range of l/h shown, the press

21、ure coefficients on the front face increase with decreasing l/h with the greatest changesoccurring for values of l/h at which the flow switches from open to closed cavity flow. On the cav- ity rear face, large pressure gradients exist and large variations in pressure levels occur with varying llh. T

22、hese large gradients, in contrast to the almost con- stant pressures on the front face, result from the fact that the rear face is exposed to the approaching high energy flow similar to a forward-facing step, whereas the front face is exposed to an almost quiescent region similar to a rearward-facin

23、g step. Peak pressures on the rear face for a given value of l/h occurred at the outer edge of the rear face with the exception of the case of transitional cavity flow (llh = 13.0) where a minimum pressure occurred in this region. This trend is observed through the test range of Mach number (figs. 7

24、(a) through (c). With increasing y2/h from the outer edge of the cavities (llh # 13), the pressures decrease to a minimum value at approx- imately mid-depth fdlawed by afi increase in pres- sure with further increases in y2/h toward the cavity floor. The maximum values near the cavity floor are appr

25、oximately equal to the peak values on the cavity floor for those cases where a pressure orifice was lo- cated at xp/l = 1. On the rear plate downstream of the cavity, large pressure gradients occurred for those cavities having the larger values of llh. The large gradients are associated with closed

26、cavity flow fields and occur in a region of flow separation downstream of the outer edge of the rear face that is formed as the flow exits from the cavity and fails to expand around the 90“ corner. For the cavities with open cavity flow fields, the flow essentially bridges the cavity, resulting in m

27、inimal separation at the rear corner and hence only small pressure gradients in this region. The pressure distributions for the 0.5-in-deep cav- ities presented in figures 7(b) and (c) for Mach num- bers 2.16 and 2.86 are somewhat similar to the results shown for M = 1.50. One of the most noticeable

28、 ef- fects of increasing Mach number is to reduce the mag- nitude of the peak pressures. Also at M = 2.86, tran- sition from closed to open cavity flow occurred when decreasing l/h from 11.6 to 11.2 as compared with 13 to 12.6 for the two lower Mach numbers. This trend is consistent with data obtain

29、ed from the schlieren tests presented in figure 6. The data obtained on the rear plate shown in figure 7 indicate that increasing Mach number results in an increase in the extent of the separation region downstream of the rear face. This trend was also observed in the schlieren sys- tem as evidenced

30、 by a downstream movement of the reattachment shock with increasing Mach number. Shown in figures 7(d), (e), and (f) are cavity pres- sure distributions for the l-in-deep cavity at Mach numbers 1.50, 2.16, and 2.86, respectively. For this cavity depth, the maximum cavity length of 12 in. limits the

31、maximum value of l/h to 12 and therefore only transitional and open cavity flow fields would be expected. As discussed previously in the intro- duction and as shown by the data from the 0.5-in- deep cavity in figures 7(a), (b), and (c), the ex- tent of the flow impingement plateau pressures for the

32、transitional flow field diminished; this resulted in monotonically increasing pressures in this region. The pressure distributions presented in figures 7(d) and (e) show that this is also true for the l-in-deep cavity at M, = 1.50 and 2.16. At M, = 2.86, how- ever, the floor pressure distributions f

33、or l/h = 10.5 (fig. 7(f) are very similar to pressure distributions shown previously for closed cavity flow in that the plateau pressures occur over a significant range of x2/1 in the flow impingement region. Since the flow has changed to open cavity flow at l/h = 10.0, it is not clear why the distr

34、ibutions at l/h = 10.5 are not more representative of transitional cavity flow. An- other unanticipated variation in the pressure distri- butions for the 1-in. cavities with l/h = 12.0 and 10.5 occurred on the rear face when increasing Mach num- bers from 1.50 to 2.16 as may be seen by comparing fig

35、ures 7(d) and (e). With increasing Mach number, a large increase in pressure level and pressure gradi- ent occurred as compared with a decrease in pressure level shown for the 0.5-in. cavity (figs. 7(a) and (b). These large pressures also occur on the rear face at Moo = 2.86 (fig. 7(f). The flow fie

36、ld associated with these large pressures also results in a bow shock at the outer edge of the rear face as can be seen in figures 5(a) and (b). This bow shock was not appar- ent for the 0.5-in. cavity flow field (fig. 4(c). The pressure distributions on the forward plate, front 5 Provided by IHSNot

37、for ResaleNo reproduction or networking permitted without license from IHS-,-,-face, and rear plate of the l-in. cavity through the test range of Mach number and l/h are similar to the results shown for the 0.5-in. cavity. Also, the pressure distributions on the cavity floor and rear face of the l-i

38、n. cavity with open cavity flow (yzlh 5 11.2) are similar to the results obtained for the 0.5-in. cavity. Pressure distributions for the 2.0-in-deep cavity were only obtained at Mach numbers 1.50 and 2.16, and these results are shown in figures 7(g) and (h). The maximum value of l/h that could be ob

39、tained at this depth was 6 and therefore all the pressure dis- tributions shown are for open cavity flow. Generally, the trends of the variation of C, with l/h and Mach number that are shown are similar to the open cavity flow results shown previously for the 0.5- and 1.0-in- deep cavities; however,

40、 the peak pressure magnitudes on the cavity floors and rear faces are greater than obtained for the more shallow cavities. This trend is consistent with an observation from reference 1 where it was found that as the ratio S/h increases (6 x Con- stant with Mach number for present tests), pressure gr

41、adients are smoothed out presumably because of the decreased momentum transfer to the cavity. For the deeper cavities of the present tests (h = 2.0 and 2.5 in.) more pressure instrumentation is available on the cavity floor for a given value of l/h simply because 1 is greater and therefore more pres

42、sure orifices are exposed. This more detailed instrumentation on the cavity floor indicates that a different type of flow field occurs for the small- est value of l/h (llh = 1, figs. 7(g) through (j), as compared with l/h = 3.0 and 6.0. This effect of l/h was not apparent for the more shallow caviti

43、es (h = 0.5 and 1.0) because of the reduced number of orifices on the cavity floor. For the deeper cavities with l/h = 1, the data in figures 7(g) through (j) show that a much smaller peak pressure occurs on the cavity floor ahead of the rear face as compared with l/h = 3.0 and 6.0. Additionally for

44、 l/h = 1, lower pressures occurred on the cavity rear face. This change in the pressure distributions may be associ- ated with the flow restructuring from a two-vortex scheme to a single-vortex scheme as observed in ref- erence 13 by flow visualization techniques. For values of l/h ranging from 5.0

45、to 2.5, Shchukin observed two vortices, as shown in the top of sketch A, of approxi- mately the same size. The rear vortex had consider- ably greater circulation intensity and its center was located somewhat above the midsection of the cav- ity. For l/h x 2.0, the flow pattern restructured to form o

46、ne vortex as shown in the bottom of sketch A. The one-vortex pattern was retained with increasing cavity depth to llh = 1. Although pressure distribu- tions are not presented in reference 13, the authors state that for the single-vortex case, the pressure Sketch A patterns become more symmetrical ab

47、out the cavity midlength; this trend is consistent with the present data. Also, heat-transfer distributions presented in reference 13 show a large reduction in heat trans- fer ahead of the rear face for the single-vortex case which could in part be caused by a pressure reduction in this region as me

48、asured in the present tests. Shown in figure 8 are the variations with l/h of the pressure coefficients in the outer-edge regions of the front and rear cavity faces to further illustrate the ef- fect of the cavity flow field on the cavity pressure dis- tributions. The data for the 0.5- and the 1.0-i

49、n. cav- ities (figs. 8(a) and (b) clearly illustrate the increase in pressure on the front face and the decrease in pres- sure on the rear face that occurs as the flow changes from closed to open cavity flow (llh x 10 to 13). The data also show that for all cavity depths, a decrease in pressure occurs on both the front and rear faces at the very low values of llh, which is much more pro- nounced at the larger cavity depths. This decrease in pressure could be associated with the cavity flow re-