1、1 1. . . ?. f ;* FOR AERONAUTICS TECHNICAL NOTE 3393 =xw+XP Reller, Jr., and Frank M. Hamaker, 1952. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-2. NACA TB 3393 ratio 7, the substitution of a 6-caliber ogival boattail (base diameter -. equals 0.6
2、04 maximum diameter) for the cylindrical afterbody resulted in an increase in the base pressure coefficient of approximately 75 percent .- at a Mach number of 1.50 (as determined from tests in the Ames l- by 3-fact supersonic wind tunnel) but only about 22 persent at a Mach number of 4.4 d = 1.25 in
3、., base diameter = 0.604-d). Data obtained on these models were used to evaluate some of the effects of nose and afterbody shape on base pressure. Model number 2 (Z/d and dismeters (see fig. = 5) was used with supports of various lengths 2) to evaluate the effect of support interfer- ence on measure
4、d base pressure. The quality of model surface finish may influence the measured base pressure through its effect on boundary-layer development. The test models had, therefore, a general surface finish of about 10 micro- inches (average deviation from the mean surface). Test Procedure Operating condi
5、tions.- For this investigation the wind tunnel was operated at Mach numbers from 2.73 to 4.98, with a maximum reservoir pressure of 6 atmospheres absolute and reservoir temperatures between 50F and70F. The absolute humidity of the air supplied to the tunnel was maintained between 1.5 X 10m5 and 5.0
6、X 105 pounds of water per pound of air. The Reynolds number of the flow at Mach numbers of 2.73 and 4.98 was approximately 8.2 x 10e per foot and 2.1 x lo8 per foot, respectively. At intermediate Mach numbers a range of Reynolds numbers was available with the maximum range of 3.6 x 1Oe to 8.6 x 1Oe
7、per foot occurring at a Mach number of 4.03. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-6 HACA TN 3393 Methods of promoting a turbulentboundary layer.- In an attempt to extend the range ofReynolds numbers at which a turbulent boundary layer woul
8、d occur, several types of turbulence-promoting devices were investi- gated. Tests were conducted with rings of 0.005- and O.OlO-inch-diameter wire and salt bands of verious widths, located both near the vertex and at the shoulder of a model. -A lampblack coating on the nose of a model was also tried
9、. Several of the turbulence-promoting devices are illustrated in figure 3. It was found that a salt band of approximately 0.020-inch thickness and l/2-inch width, located l/4 inch from the vertex of a model, was the only device that was effective in causing the boundary layer to become turbulent for
10、 the complete range of Mach numbers and Reynolds numbers of this investigation. With this device, the- transition point was fixed at the location of the roughness. The salt- band roughness was therefore used as the turbulence-promoting device in the majority of tests. Some turbulent-boundary-layer d
11、ata were obtained for model 8 with a 0.005-inch-diameter wire ring located close to the vertex. The effectiveness of this device in promoting turbulence was limited to the higher test Reynolds numbers at Mach numbers below k5. INT?ZRPRRTATION AND FUZDUCTION OF TRE DATA Boundary-Layer Identification
12、A representative series of shadowgraph pictures for both leminar- and turburlent-boundary-layer flow is shown in figure 4. Laminar- boundary-layer flow is identified by the characteristic light line tha- is apparentnear the model surface and that extends downstream from the base. Turbulent-boundary-
13、layer flow, on the other hand, is identified by a diffused light region adjacent to the surface and a lack of detail in the expansion region behind the base. The type of boundary-layer flow is also indicated by the location of the trailing shock wave behind the model. For turbulent-flow this shock w
14、ave stands closer to the base than for laminarflow at the same Mach number and Reynolds number. It is necessary to specify the conditions under which the base pressure data of this report correspond to those for.lsminar-, transitional-, and turbulent-boundary-layer flow in the region of the base. Th
15、e data correspond to lsminar-boundary-layer flow when the laminar appearance of the flow (identifiedby the characteristic light line) persists downstream of the base to the location of the trailing shock wave. Similarly, in every case- of turbulent=boundary-layer flow, transition started at least 3-
16、 to h-base diameters upstream of the base. Data that were measured under conditions that fall between these two limits are considered-to be representative of transitional-boundary- layer flow. . Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACA TN
17、 3393 7 Support Interference . Models were supported in the wind tunnel by a cylindrical rod extend- ing from the base. Since this configuration is sQnificantly different from a body with an unobstructed base, the measured values of base pressure may be considerably altered. Tests were conducted ove
18、r the entire Mach number and Reynolds number range to determine the extent of the influence of both support length and support diameter on the base pressure. Typical results are shown in figures 5 and 6. On the basis of these results it seems reasonable to assume that with a ds/d ratio of 0.40 or le
19、ss and an Is/d ratio of 8, .the measured base pressure is essentially free of support interference. Because of the varying loads encountered in the base pressure tests, it was necessary to use ratios as great as 0.625 and 2,/d ratios as low as 6. ThereforZZ was often necessary to apply corrections,
20、based on the results of this inves- tigation, to the measured base pressure coefficients that are presented in the following discussion. The effects of support interference and the correction method are considered in more detail in appendix A. Condensation in the Air Stream As a result of the large
21、flow expansion that takes place in the nozzle of a high supersonic-speed wind tunnel, extremely low static temperatures are realized in the flow passing through the test section. At a settling chsmber temperature of about 600 F, the existing situation in the Ames lo- by lb-inch supersonic wind tunne
22、l, the static temperature in the free stream falls below the liquefaction temperature at Mach numbers somewhat in excess of 4.0. Consequently. as has been shown in reference 8, at these Mach numbers a portion of the air in the wind tunnel will enter the condensed phase and thus the properties of the
23、 stream will be altered. As discussed in appendix B, this phenomenon affects both the boundary-layer flow and the flow field outside of the boundary layer. It is shown in appendix B that, for the purposes of these tests, these influences on the boundary layer can be neglected, but that the alteratio
24、n of the expansion process in the flow downstream of the base may increase f the base pressure coefficFent by as much as I2 percent at the highest test Mach number. (This corresponds to an increase in the base pressure relative to the free-stream static pressure.) Since the method used to evaluate t
25、his effect of condensation is only approximate, the basic data of the present report are.presented both as corrected and uncorrected for condensation effects. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-8 NAcA m 3393 Effect of Transition-Promotin
26、g Device To obtain turbulent boundary layers-on a representative number of : models at several Reynolds numbers, it was necessary, as previously discussed, to locate a transition-promoting device;close to the vertex of each model. However, this device caused a systematic change in the base pressures
27、. This fact is demonstrated.for modLls.4. and 5 in figure 7 where a comparison is made of the b _ of the artificial roughness on the turbulent bound.Land that-the pressure measured on the.model support represents a reasonably good average value. The base pressure data to be discussed subsequently we
28、re therefore obtalned at-this location. Variation of Base Pressure With Reynolds Number l/d = Constant3ody fineness ratio.- Base pressure coefficients for the 5, ogive-cylinder combination arepresented as-a function of Reynolds number in figure 9 for laminar-boundary-layer flow. These coefficients,
29、uncorrected for condensation in the expansion region down- stream of the model, are shown in figure g(a), while those corrected for condensation by the approximate method diBcuBsed in appendix B are shown in figure 9(b).2 AS would be expected, the base pressure coefficient decreases (corresponding t
30、o decreasing base pressurerelative to free- stream pressure) with increasing Reynolds number. It is clear that in general, however, the effect of Reynolds number. on the coefficienr decrease8 as the free-stream Mach number increases. For example, at MO = 3.49 an increase of Reynolds number from 3 x
31、10 to 4.5 x lo8 changes the coefficient about-20 percent, while at M, = 4.48 a similar increase of Reynolds number results in a change of only about 3 percent. 2It will be noted that the trends and relative magnitudes of the corrected data are essentially the same as those for the uncorrected data.
32、This property is characteristic of-all data to be presented; hence the discussion of results may generally be considered to apply to both types of data. .- c Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-. NJUX TN 3393 11 Data are presented in figu
33、re 10 for turbulent-boundary-layer flow. It can be seen that the variation of pressure coefficknt, with Reynolds number is similar for all Mach numbers above 2.73. It is also evident from a comparison with figure 9 that the effect of Reynolds number is somewhat less at lower Mach numbers than for th
34、e laminar boundary layer, which agrees tith the results of other investigators. However, at zG.sry-layer flow for the fineness ratio 5 ogive-cylinder models (?oumber range were obtained from figure 9, while the dashed lines in the high Reynolds number range were obtained by extrapolation from the cu
35、rves of figure 1O.s Also shown in figure 13 are the data of figure 7, at Mach number 3.49, for fully developed turbulent-boundary-layer flow resulting from natural boundary-layer transition. The onset of transition, that is, those conditions for which the transition point in the boundary-layer flow
36、first moves to a position upstream of the trailing shock wave, was found in these tests to occur at Reynolds numbers between approximately 4 x 106 and 5 x lOa. “It will be recalled that the data of figure 10 were obtained with fixed transition resulting from the use of an artificial roughness. Provi
37、ded by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,- - I2 NACA TN 3393 . It cm be seen that a decrease of base pressure coefficient occure in the Reynolds number range of transition at Mach numbers above 2.73. This effect becomes more pronounced as the %chnum
38、ber increases, varying Pram L approximately 15 percent (turbulent flow with fixed transition as compared - to laminar flow) at I a partial verification of the phenomenon, however, is obtained from a consideration of the physical characteristics of the , flow pattern downstream of the base. In partic
39、ular, a difference in the Location of the trailing shock wave, relative to the base, with laminsr- as conrpsred to turbulent-boundary-layer flow (at the-same M. and Re) is shown in the shadowgraph pictures of the present tests. In every ca0e ( see, e.g., fig. 4) the trailing shock wave stands closer
40、 to the base for turbulent flow than for Iaminar, with the difference increastig as the Mach number is increased. In general, then, it would be expected that for-the turbulent case a greater flow expansion occurs around the corner of the base and thus a lower pressure is transmitted into the dead-ai
41、r region. On the other hand, photographs at M. = 1.5 (see fig. 21 of reference 9) for a simflar ogive-cylinder body Indicate that the shock wave stands samewhat closerto the base for laminar than for turbulent flow. General agreement is thus apparent between these limited observa- tions and the tren
42、ds shown in five 13 of the presentmport. Effect of Nose and Afterbody Shapes For the study of nose-shape effects, base press for example, the base pressure coefficients for the blunt-noaed body vary from 7 percent less to about 3 percent less than those for the correspond- ing ogive as the Mach numb
43、er increases from 2.7 to 4.3. It is seen, on the other hand, that when a cylindrical afterbody is added to these bodies to increase their over-all Z/d to 10, no measurable effect of nose shape on base pressure coefficient is observed. That there is a reduction in this effect is not surprising, since
44、 it would be expected that with increasing afterbody length flow conditions in the region of the base (both within and outside the boundary layer) would become less sensitive to nose shape. It is interesting to note, however, that with the two different noses eqloyed, the effect is essentially zero
45、for an after- body length of only 7 diameters. Effects of boattailLng were studied with model 8. A comparison of . the base pressure coefficients for this body and the coefficients for the corresponding unboattailed body is shown in figure 15. Also shown in figure 15 are data obtained with similar m
46、odels in the Ames l- by j-foot supersonic wind tunnel at M. = 1.5 and 2.0. Some of these data sre unpublished; the remainder are interpolated from data of references 4, 9, and 10. It is evident that the boattailed body consistently has the higher base pressures at Mach numbers from 1.5 to 4.5 for bo
47、th laminar- and turbulent-boundary-layer flow in the region of the base. The effect of this amount of boattailing is observed to decrease markedly, however, with increasing corresponding values for laminar-boundary-layer flow are 36 percent and 28 percent, respectively. The results of reference IL,
48、at a Mach number of 3.25, are in substan- tial agreement with those of figure 15. Variation of Base Pressure tith Mach Number Base pressure coefficients for a body of fineness ratio 5 (model 2) with both laminar-boundary-lapr flow and artificially induced turbulent- boundary-layer flow are presented
49、 as a function of free-stream Mach number in figure l-6. The results were obtained from a cros8 plot of the data in figures 9 and 10. The limiting curve of base pressure coefficient (i.e., for a vacuum at the base) is shown for comparison. For the Mach number Results presented here were determined from cross plots of the data in figures 11 and 12 and similar-figures for the boattailed body. Where necessary, the
