《SAE-TPS-741999-01-5601.pdf》由会员分享，可在线阅读，更多相关《SAE-TPS-741999-01-5601.pdf（12页珍藏版）》请在三一文库上搜索。
1、For permission to copy or republish, contact the American Institute of Aeronautics and Astronautics or SAE International. SAE International 400 Commonwealth Drive Warrendale, PA 15096-0001 U.S.A. American Institute of Aeronautics and Astronautics 370 LEnfant Promenade, S.W. Washington, D.C. 20024 19
2、99-01-5601 Force and Moment Measurements with Pressure-Sensitive Paint James H. Bell NASA Ames Research Center 1999 World Aviation Conference October 19-21, 1999 San Francisco, CA Author:Gilligan-SID:1178-GUID:20758466-141.213.232.87 Published by the American Institute of Aeronautics and Astronautic
3、s (AIAA) at 1801 Alexander Bell Drive, Suite 500, Reston, VA 22091 U.S.A., and the Society of Automotive Engineers (SAE) at 400 Commonwealth Drive, Warrendale, PA 15096 U.S.A. Produced in the U.S.A. Non-U.S. purchasers are responsible for payment of any taxes required by their governments. Reproduct
4、ion of copies beyond that permitted by Sections 107 and 108 of the U.S. Copyright Law without the permission of the copyright owner is unlawful. The appearance of the ISSN code at the bottom of this page indicates SAEs and AIAAs consent that copies of the paper may be made for personal or internal u
5、se of specific clients, on condition that the copier pay the per-copy fee through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923. This consent does not extend to other kinds of copying such as copying for general distribution, advertising or promotional purposes, creatin
6、g new collective works, or for resale. Permission requests for these kinds of copying should be addressed to AIAA Aeroplus Access, 4th Floor, 85 John Street, New York, NY 10038 or to the SAE Publications Group, 400 Commonwealth Drive, Warrendale, PA 15096. Users should reference the title of this co
7、nference when reporting copying to the Copyright Clearance Center. ISSN #0148-7191 Copyright 1999 by SAE International and the American Institute of Aeronautics and Astronautics, Inc. All rights reserved. All AIAA papers are abstracted and indexed in International Aerospace Abstracts and Aerospace D
8、atabase. All SAE papers, standards and selected books are abstracted and indexed in the Global Mobility Database. Copies of this paper may be purchased from: AIAAs document delivery service Aeroplus Dispatch 1722 Gilbreth Road Burlingame, California 94010-1305 Phone: (800) 662-2376 or (415) 259-6011
9、 Fax: (415) 259-6047 or from: SAExpress Global Document Service c/o SAE Customer Sales and Satisfaction 400 Commonwealth Drive Warrendale, PA 15096 Phone: (724) 776-4970 Fax: (724) 776-0790 SAE routinely stocks printed papers for a period of three years following date of publication. Quantity reprin
10、t rates are available. No part of this publication may be reproduced in any form, in an electronic retrieval system or otherwise, without the prior written permission of the publishers. Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE or AIAA.
11、 The author is solely responsible for the content of the paper. A process is available by which discussions will be printed with the paper if it is published in SAE Transactions. Author:Gilligan-SID:1178-GUID:20758466-141.213.232.87 1 1999-01-5601 Force and Moment Measurements with Pressure-Sensitiv
12、e Paint James H. Bell NASA Ames Research Center Copyright 1999 by SAE International and the American Institute of Aeronautics and Astronautics. Inc. ABSTRACT The desire to provide integrated surface pressures for aerodynamic loads measurements has been a driving force behind the development of press
13、ure-sensitive paint (PSP). To demonstrate the suitability of PSP for this pur- pose, it is not sufficient to simply show that PSP is accu- rate as compared to pressure taps. PSP errors due to misregistration or temperature sensitivity may be high near model edges, where pressure taps are rarely inst
14、alled. Thus, PSP results will appear good compared to the taps, but will yield inaccurate results when inte- grated. A more stringent technique is to compare inte- grated PSP data over the entire model surface with balance and/or CFD results. This paper describes a sim- ple integration method for PS
15、P data and presents com- parisons of balance and PSP results for three experiments. PSP is shown quite accurate for normal force measurements, but less effective at determining axial force and moments. INTRODUCTION During the 1980s, researchers first in the Soviet Union (Mosharov, et. al. 1988) and
16、later in the US (Kavandi et. al. 1990) began to exploit the capability of oxygen- quenched luminophors to provide pressure measure- ments on aerodynamic surfaces. This technology, known as pressure-sensitive paint (PSP) is now widely used in wind tunnel testing (Kegelman 1999). A major driver for PS
17、P development in the US has been the prospect of using the technique to obtain aerodynamic loads mea- surements. Currently these loads are measured by con- structing a wind tunnel model with a large number of pressure taps. Tap data from wind tunnel tests are inter- polated to estimate the pressure
18、field over the model. The interpolated data are then integrated to determine loads on various aircraft components. The requirement for spatially accurate data leads to models with a large number of pressure taps; upwards of 1000 taps is fairly common. Such a large number of taps leads to a mechanica
19、lly complex, expensive model, which requires a long lead-time to build. As a result, early wind tunnel testing in an aircraft design program is typically done on less-expensive, mechanically simpler “stability and con- trol” models. Only after the aircraft mold lines have been frozen is a “loads” mo
20、del constructed. This procedure has several disadvantages. Loads engineers receive criti- cal design data relatively late in the development pro- gram, when aerodynamic design can only be influenced with difficulty. Much wind tunnel testing is repeated, first for force and moment data, and again for
21、 loads. Finally, the complex construction of the loads model means that it is often physically weaker than the stability and control models, and therefore it cannot duplicate the highest dynamic pressures achieved in testing the simpler model. A practical PSP system would improve this situation sig-
22、 nificantly. A stability and control model could be, in effect, converted to a loads model with a coat of paint. The expense of the loads model, and much duplicate testing, would be eliminated. In addition, loads engineers would receive their data much earlier in the aerodynamic testing cycle. An im
23、portant step toward achieving this goal to demonstrate that PSP data on a surface can be inte- grated to obtain forces and moments. Initial attempts at PSP data integration were reported by Sellers (1995) who made measurements on the horizontal stabilizer of an F-18C model in transonic flow. His res
24、ults showed good agreement between PSP and strain gauge mea- surements of normal force at high transonic Mach num- bers, but poorer agreement at M=0.6. Most other work in this area has been done in connection with specific air- craft development projects, and as such is not reported in the open lite
25、rature. However, in at least one case, engi- neers at the Boeing Co. used PSP to obtain a complete loads database for an advanced fighter configuration (Roger Crites, 1996, private communication). The present paper describes an initial PSP integration capability developed at the NASA Ames Research C
26、en- ter. The integration procedure was applied to results from three wind tunnel tests, including simple and complex models at transonic speeds, and a simple model at low subsonic speeds where PSP signal-to-noise ratio is gen- erally lower. The intent is to use these tests as examples with which to
27、assess the strengths and weaknesses of PSP for loads measurements. DATA REDUCTION PROCEDURE All data reduction was performed using the “Green Boot” PSP software developed by the Boeing Co. and NASA Author:Gilligan-SID:1178-GUID:20758466-141.213.232.87 2 Ames Research Center. Raw PSP image data input
28、 to the Green Boot software were converted to calibrated pressure images as described by McLachlan and Bell (1995). These images were then mapped onto a model surface grid. Bell and McLachlan (1996) gave an over- view of the procedure used for the mapping, but the image resection algorithm described
29、 in that reference was not used in this study. It was replaced by the more accurate method described by Samtaney (1999). For this study, Green Boot was modified to also integrate pres- sure data on the surface grid. Details of the pressure inte- gration procedure are described below. PRESSURE INTEGR
30、ATION The model surface is rep- resented by a structured grid in PLOT3D format (Walatka et. al. 1990). The grid is divided into quadrilateral panels as shown in figure 1. The procedure for force and moment integration closely follows that described by Chan otherwise the summation result would be vul
31、nerable to the usual round- ing errors associated with calculating the small difference of large numbers. Moments are found in a similar way by assuming that the force on a panel acts through the panel center as determined by the mean of the corner posi- tions. Thus the total moment on a surface is:
32、 (2) where is the location of the center of panel I and is the reference location about which moments are calcu- lated. More accurate representations of the force and moment on a panel are possible. For example, an accurate weighting function could be used to estimate the panel pressure and centroid
33、 location. In addition, the quadrilat- eral panels are not guaranteed to be flat. It would be more accurate to divide the surface into triangles. How- ever, it is not clear if these improvements are worthwhile, given that each panel represents a very small patch of the entire surface. In their work
34、on FOMOCO, Chan thus there is some lift associated with the fuselage and the reference area must be increased to account for this fact. While the PSP does not see the lift produced by the fuselage, since this lift is proportionate to the extra area the force coefficients match when appropriately nor
35、mal- ized. At high , the wingtips tend to unload somewhat, and a larger portion of the lift is carried by the inboard sections of the wing. This implies that the lift contribution from the fuselage increases disproportionately to its area. Thus the normalized balance and PSP results no longer match.
36、 This may explain why PSP underpredicts normal force at the higher angles of attack. Other possi- ble explanations could relate to an -dependent inaccu- racy in the PSP measurement. No such error is apparent, however. Attempts to find an -dependence in either the PSP calibration accuracy or the accu
37、racy of the image mapping have not been successful. Figure 4. PSP image showing false-colored Cp values on transonic wing upper surface. -1.4 -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 -1.4-1.2-1-0.8-0.6-0.4-0.200.2 Tap Cp PSP Cp Author:Gilligan-SID:1178-GUID:20758466-141.213.232.87 5 Figure 6.Normal force c
38、oefficient, CN vs. . CN is determined from balance and PSP measurements. Comparison of PSP and balance data for axial force is problematic, since PSP is insensitive to skin friction, which accounts for roughly half the drag on transonic air- craft (Shevell 1989). (And in the present case any bal- an
39、ce/PSP comparison will be made more difficult by the lack of PSP data on the body.) Indeed, as shown in figure 7, the differences between the axial force coefficients reported by the two data sources are quite large. More significantly, PSP reports a slightly negative axial force. This is unrealisti
40、c, since boundary layer growth on the wing should produce a positive pressure drag in any event. While it is not uncommon for section drag on air- foils to be negative, due to leading edge suction, negative total pressure drag on a complete wing is quite rare. The axial force discrepancy can be expl
41、ained by consid- ering a “drag loop” plot of Cp vs. z/t. Just as a Cp vs. x/c plot illustrates section lift by the area between the upper and lower curves, a drag loop plot illustrates section drag by the area between the curves corresponding to the for- ward- and rearward-facing surfaces of the win
42、g. Figure 8 shows such a plot for the midspan portion of the model, with data taken from both PSP and pressure taps. Note that the PSP data differs significantly from the tap results near the stagnation point, where pressure on the forward- facing surface is high. This occurs because neither cam- er
43、a had a good view of the stagnation point. Data on this region of the model were obtained by interpolating valid PSP results from nearby surfaces. However the interpola- tion failed to capture the large pressure spike at the stag- nation point, and thus a significant source of drag was missed by the
44、 PSP. Figure 7.Axial force coefficient, CA vs. . CA is determined from balance and PSP measurements. Figure 8.Drag loop plot of Cp vs. Z. at midspan. Cp is obtained from both PSP and pressure taps. INTEGRATION RESULTS MOMENTS There are sig- nificant discrepancies between moments measured by PSP and
45、those measured by the balance, as shown in figures 9 and 10. In the case of pitching moment (figure 9) this is not unexpected. The body makes a significant contribution to pitching moment, while the PSP data measure moments due to pressure on the wing only. Agreement between balance and PSP results
46、for rolling moment should be better, since the unpainted fuselage has a low moment arm compared to the wing. As shown in figure 10, the PSP captures the overall trend of the roll- ing moment results, but significant discrepancies exist. In particular, the PSP measurements indicate a peak in roll- in
47、g moment at around =2.5, but no corresponding peak exists in the balance data. As discussed in the previous section, one possible source would be an a-dependent error in the PSP calibration. In the present experiment, this possibility could be investigated by examining the agreement between PSP and
48、pressure taps separately for each of the eight chordwise tap rows on the wing. When this was done, the results (not shown) did not indi- cate any trends that might account for the discrepancy in rolling moment between PSP and balance data. Once -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 -10-50510 Angle of A
49、ttack Normal Force Coefficient PSP Cn Balance Cn -0.02 -0.01 0 0.01 0.02 0.03 0.04 -10-50510 Angle of Attack Axial Force Coefficient PSP Ca Balance Ca -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 -0.6-0.4-0.200.20.40.6 Z/t Cp PSP Cp Tap Cp Author:Gilligan-SID:1178-GUID:20758466-141.213.232.87 6 again, the most likely source for the discrepancy lies in the fact that the PSP data does not cover the body. Figure 9.Pitching moment coefficient, CMY vs. . CMY is determined from balance and PSP measurements. Figure 10. Rolling moment coefficient, CMX vs. . CMX is determined