1、NACA RM 5508 llllllillllilllilliCliiiiiiiiiillii Y 3 1176014380449 NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS RESEARCHMEMORANDUM TRANSONIC WIND-TUNNEL INVESTIGATION OF EFFECTS OF WINDSHIELD SHAPE AND CANOPY LOCATION ON THE AERODYNAMIC CHARACTERISTICS OF CANOPY-BODY COMBINATIONS By Elden S. Cornette
2、 and Harold L. Robinson SUMMARY Aerodynamic data have been obtained for a fuselage forebody alone and for canopy-body configurations consisting of four different canopies mounted on a fuselage forebody. Two of the canopies had the same shape and size rearward of the windshield but one had a “flat“ a
3、nd the other a “vee“ windshield. The remaining two canopies were located at differ- ent body stations and were geometrically similar. The data, obtained for a Mach number range from 0.80 to 1.13, an angle-of-attack range from 0 to loo, and an angle-of-sideslip range from -8 to 8, indicated that the
4、drag of the flat-windshield model was consistently lower than that of the vee-windshield model. Of the remaining forces and moments, only the lateral force was significantly affected by windshield shape, the vee-shaped windshield causing increased lateral force. Little effect of canopy location was
5、found for the canopy-body configurations investigated. INTRODUCTION Recently a wind-tunnel test program was undertaken at the Langley Laboratory of the National Advisory Committee for Aeronautics to obtain aerodynamic data in the transonic and supersonic speed ranges for a series of airplane canopy
6、models mounted on a common fuselage forebody. The primary purpose of this program was to investigate the aerodynamic loads experienced by canopy models that simulate present designs used on high-speed aircraft and to evaluate the effects of such design vari- ables as canopy location, size, and winds
7、hield shape. In the present investigation, both pressure-distribution and force data were obtained in the Langley 8-foot transonic tunnel for four dif- ferent canopies mounted on a fuselage forebody. Presented in this paper Provided by IHSNot for ResaleNo reproduction or networking permitted without
8、 license from IHS-,-,-2 NACA RM L55G08 are the six-component force data for the canopy-body configurations as well as the forebody alone. Two of the canopies were mounted well for- ward on the body and had the same shape behind the windshield but one had a “flat“ and the.other a trveetl windshield.
9、The two remaining cano- pies had a smaller maximum cross-sectional area than the first two and both had flat windshields. The two smaller canopies were located at different longitudinal positions on the body. The Mach number range for this investigation was from 0.80 to 1.13 while the angle of attac
10、k was varied from 0 to 10 and the angle of sideslip was varied from -8 to 8. a b cA cD rn n CN cY SYMBOLS one-half major diameter of body cross-section ellipse one-half minor diameter of body cross-section ellipse, a/l. 25 axial-force coefficient, Axial force CIS drag coefficient, Drag/ rolling-mome
11、nt coefficient, Rolling moment qSL pitching-moment coefficient, Pitching momeni qSL . yawing-moment coefficient, Yawlnsnt normal-force coefficient, Normal force ss lateral-force coefficient, Lateral force qs one-half major diameter of canopy cross-section ellipse one-half minor diameter of canopy cr
12、oss-section ellipse distance to top of round canopy, measured from straight center line of body - -. - -.-.-._ - - _._ - _ - _ -_ . . . - _. . . _. . . _- Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACA RM L55G08 3 L M “b P % 9 R r S U V v. X z
13、total body length, 25 inches .streain Mach number base pressure coefficient, RI -P cl stream static pressure static pressure at model base stream dynamic pressure, PV212 Reynolds number per foot of length, W/P radius of cross section of round canopy maximum cross-sectional area of fuselage, 15.71 sq
14、 in. vertical coordinate of canopy cross section horizontal coordinate of canopy cross section stream velocity distance measured from body nose along straight center line (positive rearward) vertical distance from straight center line to drooped center line of body (positive downward) angle of attac
15、k angle of sideslip stream viscosity stream density APPARATUSANDMEASUREMENTS Models The dimensions of the models are presented in figure 1. The fuse- lage is five maximum fuselage depths in length and is of elliptic cross Provided by IHSNot for ResaleNo reproduction or networking permitted without l
16、icense from IHS-,-,-4 NACA RM L55G08 section (b = a/1.25). .(See fig. l(a).) The f uselage center line droops from a station 3.5 maximum fuselage depths rearward of the nose point in an arc so that the maximum droop at the nose is 0.2 maximum fuselage depth. The centers of the fuselage cross-section
17、 ellipses lie along this drooped center.line. The cross sections of canopies 1 and 2 (figs. l(b) and (c) behind the windshield (behind the 6.25-inch station) were ellipses (d = c/2.5) perpendicular to the fuselage horizontal center line with centers located on the fuselage drooped center line. Canop
18、y 1 had a flat windshield while canopy 2 had a vee windshield. Both of these canopies were located at the forward position on the fuselage. Canopies 3 and 4 (figs. l(d) and (e) were geometrically similar and had circular cross sections behind the windshield. Both canopies had flat windshields but ca
19、nopy 4 was located 3.75 inches farther down- stream on the fuselage than canopy 3. Unfortunately, canopy 3 was con- structed slightly inaccurately and the actual measured dimensions are presented in figure l(d). The cross-sectional area distributions of the five models are presented in figure l(f).
20、Tunnel , This investigation was conducted in the Langley 8-foot transonic tunnel, which has a dodecagonal slotted test section and is capable of continuously variable operation through the speed range up to a Mach number of 1.15. The models were mounted on the conventional sting system used in the 8
21、-foot transonic tunnel. Detailed discussions of the design and calibration of this tunnel have been presented in refer- ences 1 and 2. The uniformity of the Mach ,number distribution in the model region is within k0.006. Tunnel-wall constraint and blockage corrections have not been applied to the da
22、ta because such corrections are negligible. The effects of boundary-reflected disturbances are also , considered negligible at the Mach numbers for which data are presented. Tests The models were tested at stream Mach numbers of 0.80, 0.90, 0.95, 0.99, 1.02, 1.08, and 1.13. The maximum random error
23、in measuring the stream Mach number is believed to be about 0.003. At each Mach number the models were tested at nominal angles of attack of O“, 5, and 10 and at nominal angles of sideslip of O“, +4, and +8O. The Reynolds number per foot of length varied during the investigation and the approximate
24、spread is shown in figure 2 plotted against Mach number. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACA RM L55G08 u I Force and Moment Measurements The forces and moments were measured with respect to a body-axis system by means of an internall
25、y mounted electrical strain-gage balance system. The center of the axis system was located at the 14.813-inch body station and the longitudinal axis coincided with the straight center line of the body. (See fig. 3.) The maximum cross-sectional area of the body (15.71 square inches) and the length of
26、 the body (25 inches) were used to reduce these forces and moments to coefficient form. The pressures at the,base of the model were measured and the axial force was adjusted to the condition of free-stream static pressure at the model base. Figure 4 shows the variation of base pressure coeffi- cient
27、 with Mach number at zero angle of attack and sideslip for the five models tested. The base pressure coefficients were estimated to be accurate to within kO.005. A block of wood having the ssme cross- sectional shape as the body at the 25-inch station and 1 foot long was fastened rigidly to the stin
28、g behind the model in order to reduce the flow expansion about the model base and stabilize the base pressure. The gap between the block of wood and the model was approximateiy l/16 inch. The force and moment coefficients presented in this paper are referred to the body-axis system. In addition the
29、forces were resolved along the wind axis to obtain the drag, which also is presented. An estimate of the maximum random error in the data reported herein is presented in the following table: 1 c Axial-force coefficient, CA . io.004 Drag coefficient; CD 20.004 Normal-force coefficient, CN to.02 Later
30、al-force coefficient, CY 50.01 Rolling-moment coefficient, C2 . 20.002 Yawing-moment coefficient, C, +0.002 Pitching-moment coefficient, Cm . f0.002 1 i . Angle-of-Sideslip and Angle-of-Attack Measurements In order to facilitate sideslip-angle measurements, the model was rotated 90 before mounting o
31、n the sting support system. The sideslip Ai 1 I angle was then measured by an electrical strain-gage pendulum device I: mounted internally near the base of the support sting. Sting and model deflections occurring ahead of this point as a result of forces and moments acting on the model were determin
32、ed from static tests. The corrections were applied to the sideslip angle. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-6 NACA BM L55G08 The angle of attack was obtained by inserting O“-, 5O-, and loo- bent couplings in the support sting. The incre
33、mental change in angle of attack due to load was determined from static tests and applied to the zero-load angle of attack. The maximum deflection due to load was approximately 0.3. The angles of sideslip and attack reported herein are accurate within 0.1. RESULTS AND DISCUSSION Basic Data The force
34、 and moment coefficients for the five models are presented in figures 5 to 9. The flagged symbols indicate data at positive side- slip angles and the sense of CY C2, and C, has been reversed for these data points. As mentioned previously, canopy 3 was not accurately constructed and the discrepancy i
35、n the corresponding negative and posi- tive sideslip-angle data is apparent for the axial-force coefficient (fig. 8). Effect of Windshield Shape The effect of windshield shape is illustrated in figure 10, where data for canopy 1 (flat windshield), canopy 2 (vee-shaped windshield), and the body alone
36、 have been plotted against Mach number. It can be seen that the shape of the windshield did not have an important effect on the normal-force coefficient (fig. 10(a). At a sideslip angle of -8O, a reduction in normal force was indicated for all angles of attack when either canopy was added to the bod
37、y. Since CN for the body alone remained constant as the sideslip angle was increased, the loss in normal force at the higher sideslip angle for either canopy-body configuration may have been due to local flow separation on the leeward side of the model induced by the addition of a canopy. At sidesli
38、p angles other than zero, the lateral-force coefficient (fig. 10(b) for canopy 2 was larger than that for canopy 1. The addition of either canopy yielded a greater increase in lateral force with angle of attack. The axial-force coefficient (fig. 10(c) and the drag coefficient (fig. 10(d) were increa
39、sed by the addition of either canopy. The increase was greater at supersonic Mach numbers and amounted to as much as 0.12 indrag coefficient for canopy 2. In figure 10(d), it can be seen that the drag is always less for the flat windshield than for the vee windshield. The difference in drag coeffici
40、ent due to windshield shape amounted to as much as 0.03. Unpublished data for these same models indicate similar measured drag differences at M = 1.4 and 2.0. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACA BM 5508 .-.e . 7 In the light of the t
41、ransonic area rule (ref. 3), it appears that the larger drag of the vee-windshield model may be partially due to the greater rate of area growth between the 6-inch and 8.5-inch stations. (fig. l(f) for that model. Windshield shape did not have an important effect on rolling moment (fig. 10(e), pitch
42、ing moment (fig. 10(f),- or yawing moment (fig. 10(g). At sideslip angles of -4 and -8O, the rolling-moment coefficient was negative for the body alone as well as for either canopy-body combina- tion. This was due to the drooped nose of the body which caused a greater frontal area to be presented be
43、low the roll axis. Since cano- pies 1 and 2 were mounted well forward and near the roll axis of the body; only a very slight positive rolling increment was produced by the canopies at p = -8O, even though the lateral-force increments were rel- atively large. Nearly equal increments in pitching-momen
44、t coefficient (fig. 10(f) were produced by both canopies at all angles of attack and sideslip. Since the increment in normal force at p = O“ and -4O was very near zero, the increase in pitching moment at these sideslip angles was produced by, essentially, a pure couple. At p = -8O, the positive pitc
45、hing-moment increment was maintained despite a negative increment in normal force. The yawing-moment increment due to adding either canopy (fig. 10(g) decreased with increasing angle of attack, even though the lateral force increased. Effect of Canopy Location The variation of zero-lift drag with Ma
46、ch number for the five models tested is shown in figure 11. By comparing the curves for canopy 3 (forward) .and canopy 4 (rearward), it can be seen that the beginning of the transbnic drag rise was delayed to a slightly higher Mach number and the supersonic drag was reduced slightly by placing the c
47、anopy in the forward position. It can also be seen, by comparing the curves for canopies 1 and 2 with those for canopies 3 and 4, that reducing the size of the canopies improves the area distribution (fig. l(f) and yields a considerable reduction in the drag at low supersonic speeds. The basic data
48、for canopies 3 and 4 are presented in figures 6 and 7. Examination of these data indicates that either canopy location produced the same,normal-force and lateral-force characteristics within the accu- racy of the data. The rolling-moment coefficient also showed no signifi- cant change due to canopy
49、location. The positive pitching moment was slightly greater for the forward canopy except near M = 0.99 and a = loo where the rearward canopy exhibited a slightly greater positive pitching tendency. The yawing moment was essentially the same for both canopies except near M = 0.99 and p = -8O, where a slight increase was indicated for the rearward canopy. I , Provi
