Document
NASA/CP-2004-213028/PT1
COMSAC: Computational Methods for
Stability and Control
Compiled by C. Michael Fremaux and Robert M. Hall Langley Research Center, Hampton, Virginia
April 2004
The NASA STI Program Office . . . in Profile Since its founding, NASA has been dedicated to the • CONFERENCE PUBLICATION. Collected advancement of aeronautics and space science. The papers from scientific and technical NASA Scientific and Technical Information (STI) conferences, symposia, seminars, or other Program Office plays a key part in helping NASA meetings sponsored or co-sponsored by NASA.
maintain this important role.
• SPECIAL PUBLICATION. Scientific, The NASA STI Program Office is operated by technical, or historical information from NASA Langley Research Center, the lead center for NASA’s programs, projects, and missions, often scientific and technical information. The NASA STI concerned with subjects having substantial Program Office provides access to the NASA STI public interest.
Database, the largest collection of aeronautical and space science STI in the world. The Program Office is • TECHNICAL TRANSLATION. English- also NASA’s institutional mechanism for language translations of foreign scientific and disseminating the results of its research and technical material pertinent to NASA’s mission.
development activities. These results are published by NASA in the NASA STI Report Series, which Specialized services that complement the STI includes the following report types: Program Office’s diverse offerings include creating custom thesauri, building customized databases, • TECHNICAL PUBLICATION. Reports of organizing and publishing research results ... even completed research or a major significant phase providing videos.
of research that present the results of NASA programs and include extensive data or For more information about the NASA STI Program theoretical analysis. Includes compilations of Office, see the following: significant scientific and technical data and information deemed to be of continuing • Access the NASA STI Program Home Page at reference value. NASA counterpart of peer- http://www.sti.nasa.gov reviewed formal professional papers, but having less stringent limitations on manuscript length • E-mail your question via the Internet to and extent of graphic presentations.
help@sti.nasa.gov • TECHNICAL MEMORANDUM. Scientific • Fax your question to the NASA STI Help Desk and technical findings that are preliminary or of at (301) 621-0134 specialized interest, e.g., quick release reports, working papers, and bibliographies that contain • Phone the NASA STI Help Desk at minimal annotation. Does not contain extensive (301) 621-0390 analysis.
• Write to: • CONTRACTOR REPORT. Scientific and NASA STI Help Desk technical findings by NASA-sponsored NASA Center for AeroSpace Information contractors and grantees.
7121 Standard Drive Hanover, MD 21076-1320
NASA/CP-2004-213028/PT1
COMSAC: Computational Methods for
Stability and Control
Compiled by C. Michael Fremaux and Robert M. Hall Langley Research Center, Hampton, Virginia Proceedings of a symposium sponsored by the National Aeronautics and Space Administration, Washington, DC, and held at the Holiday Inn and Conference Center, Hampton, Virginia September 23-25, 2003 National Aeronautics and Space Administration Langley Research Center Hampton, Virginia 23681-2199
April 2004
Acknowledgments The organizing committee acknowledges the highly professional administrative contributions of Ms. Susan N. Price of the NASA Langley Management Support Office. In addition, the committee thanks Mr. Douglas N. Ball (Boeing Commercial Airplanes), Dr. Pradeep Raj (Lockheed Martin), and Mr. John W. Clark, Jr. (NAVAIR) for their participation in the COMSAC planning sessions, and for sharing their personal observations and opinions in the symposium.
The use of trademarks or names of manufacturers in the report is for accurate reporting and does not constitute an official endorsement, either expressed or implied, of such products or manufacturers by the National Aeronautics and Space Administration.
Available from: NASA Center for AeroSpace Information (CASI) National Technical Information Service (NTIS) 7121 Standard Drive 5285 Port Royal Road Hanover, MD 21076-1320 Springfield, VA 22161-2171 (301) 621-0390 (703) 605-6000 Executive Summary A group of nearly 100 technical professionals from government, industry, and academia met in Hampton, Virginia on September 23-25, 2003, for a NASA-sponsored symposium on Computational Methods for Stability and Control (COMSAC) to discuss the status, opportunities, and challenges of applying Computational Fluid Dynamics (CFD) methodology to current and future issues in the field of aircraft stability and control (S&C). The unprecedented advances now being made in CFD technology have demonstrated the powerful capabilities of codes in applications to civil and military vehicles. Used in conjunction with wind-tunnel and flight investigations, many codes are now routinely used by designers in diverse applications such as aerodynamic performance predictions and propulsion integration. Typically, these codes are most reliable for attached, steady, and predominantly turbulent flows. As a result of increasing reliability and confidence in CFD, wind-tunnel testing for some new configurations has been substantially reduced in key areas, such as wing trade studies for mission performance guarantees.
Interest is now growing in the application of CFD methods to other critical design challenges.
One of the most important disciplinary elements for civil and military aircraft is S&C.
Experience has shown that predictions and analyses of aerodynamic S&C characteristics for full- scale aircraft can be in serious error because of Reynolds number effects, configuration sensitivities, dynamic motion effects, and other issues. Existing experimental facilities may not even be capable of replicating the motions required for aerodynamic measurements. As a result of these shortcomings, a major portion of aircraft development wind-tunnel time (about 60-70%) is typically devoted to S&C testing, especially for various off-design conditions ranging from takeoff and landing to cruise and maneuver. Even with an enormous amount of experimental work, pre-flight aerodynamic prediction errors result in unacceptable increases in program costs, “fly and try” approaches to fixing deficiencies, and extensive developmental delays.
Unfortunately, applications of current and emerging CFD codes to engineering analysis in the field of aircraft S&C have been extremely limited. Although isolated examples of success have been demonstrated for certain configurations, the more global issues in S&C – which may involve massive flow separation, unsteady and nonlinear phenomena, dynamic effects, and other extremely complex factors – have not yet been significantly addressed by the CFD community.
The current lack of COMSAC-related activities has been further aggravated by the fact that, in contrast to the areas of CFD and performance, very little cross-cultural interaction and communication appears to occur between participants in the areas of CFD and S&C. Within the aerospace community, it is generally agreed that the field of CFD has rapidly matured to the point that the next high payoff applications could occur in S&C. In particular, CFD offers the potential for significantly increasing the basic understanding, prediction, and control of flow phenomena associated with requirements for satisfactory aircraft handling characteristics.
The objectives of the 3-day symposium were to: 1. Discuss the unique aerodynamic phenomena and issues of S&C 2. Define the current characteristics, capabilities, and limitations of CFD codes 3. Define additional or new code requirements for S&C applications iii 4. Identify potential approaches to develop validated codes 5. Discuss the potential contents and funding opportunities for a COMSAC program The scope of technical discussions covered civil and military aircraft, including commercial transports, business jets, fighter and attack aircraft, military transports, and bombers. Discussions were limited to fixed-wing aircraft. All sessions were unclassified, and all non-proprietary presentations were collated in the form of PowerPoint presentations with note pages for post- meeting distribution to attendees.
Presentations by speakers described numerous examples of severe impacts of erroneous aerodynamic predictions on the stability and control characteristics of civil and military aircraft.
Typically, resolving and mitigating unexpected aerodynamic behavior involved laborious “cut and fly” approaches required during critical flight test programs. These shortcomings resulted in significant program delays, costs, mission limitations, non-optimum configurations (weight, capabilities, etc.), and severe scrutiny by stakeholders and customers.
In-depth discussions of specific experiences with actual applications of various levels of computational methods to S&C indicated a wide range of success and an overriding sense of skepticism by the attendees. After individual presentations were made to provide organizational and individual perspectives on CFD for S&C, the attendees were briefed on NASA’s vision of a COMSAC program. Comments were solicited to identify and prioritize technology areas for such a program. Finally, the Director of the NASA-Langley Aerospace Vehicle Systems Technology Office shared his view of a potential strategy to augment funding and program priority in this area.
The general findings of the workshop were: 1. Inaccurate prediction of aerodynamic stability and control parameters continues to have major cost and programmatic impacts in virtually every vehicle class. These impacts include unacceptable increases in program costs, “fly and try” approaches to fixing deficiencies, extensive developmental delays and profit losses due to delayed deliveries.
2. Prediction of the character of separated flows across the speed range (with the attendant issues of transition prediction, turbulence modeling, unsteady flows, etc.)
and the impact of separated flow on aircraft S&C should receive priority in a COMSAC program.
3. A pervasive attitude of skepticism regarding the success of CFD applications to aircraft S&C issues (especially for preliminary and conceptual design) exists within the CFD community, as well as the S&C community.
4. The application of advanced and emerging CFD methods as design tools will be dependent on the accumulation and demonstrated success of experiences for both generic and specific aircraft configurations.
iv 5. Issues regarding the CFD process (cost, time required, adaptive gridding requirements, error quantification, etc.) should be high priority targets for COMSAC efforts.
6. One of the most valuable contributions of the symposium was the mechanism to share perspectives and experiences between the diverse CFD specialists and S&C specialists. Prior to this meeting, communication between these two groups was extremely poor, resulting in a major barrier to the acceleration and acceptance of CFD methods for S&C applications.
Joseph R. Chambers ViGYAN, Inc.
Hampton, Virginia v Contents Executive Summary …………………………………………………………………………….. iii Attendees …………………………………………………………………………………………. x Part 1 Introductory Remarks …………………………………………………………………………... 1 Darrel R. Tenney Aerospace Vehicle Systems Technology Office, NASA Langley Research Center, Hampton, Virginia Introduction to Computational Methods for Stability and Control (COMSAC) …………… 7 Robert M. Hall and C. Michael Fremaux NASA Langley Research Center, Hampton, Virginia Joseph R. Chambers ViGYAN, Inc., Hampton, Virginia Stability & Control Challenges for COMSAC: a NASA Langley Perspective ……………. 28 C. Michael Fremaux NASA Langley Research Center, Hampton, Virginia Emerging CFD Capabilities and Outlook – A NASA Langley Perspective ………………... 48 Robert T. Biedron, S. Paul Pao, and James L. Thomas NASA Langley Research Center, Hampton, Virginia The Role for Computational Fluid Dynamics for Stability and Control – Is it Time? ……. 69 Douglas N. Ball Boeing Commercial Airplanes, Renton, Washington Northrop Grumman Perspective on COMSAC ……………………………………………… 98 Dale Lorincz Northrop Grumman Air Combat Systems, El Segundo, California Boeing Integrated Defense Systems Perspective on COMSAC …………………………… 118 David M. Evans Boeing Integrated Defense Systems – Tactical Aircraft and Weapons, St. Louis, Missouri Computational Methods in Stability and Control – WPAFB Perspective ………………... 124 William Blake AFRL Air Vehicles Directorate, Wright Patterson Air Force Base, Ohio William Thomas Aeronautical Systems Center, Wright Patterson Air Force Base, Ohio vi Perspective: Raytheon Aircraft Company ………………………………………………….. 144 Neal J. Pfeiffer and Dana C. Herring Raytheon Aircraft Company, Wichita, Kansas A Greybeard’s View of the State of Aerodynamic Prediction ……………………………... 181 Tom Lawrence Aeromechanics Division, Naval Air Systems Command, Patuxent River, Maryland Computational Methods for Stability and Control: A Perspective ……………………….. 214 Pradeep Raj Lockheed Martin Aeronautics Company, Marietta, Georgia Boeing TacAir Stability and Control Issues for Computational Fluid Dynamics ………... 227 William B. Hollingsworth Boeing TacAir, St. Louis, Missouri NAVAIR S&C Issues for CFD ……………………………………………………………….. 239 Steve Donaldson Flight Dynamics Branch, Naval Air Systems Command, Patuxent River, Maryland An S&C Perspective on CFD ………………………………………………………………… 248 Russ D. Killingsworth JSF Stability and Control, Lockheed Martin Aeronautics, Fort Worth, Texas Issues, Challenges & Payoffs: A Boeing User’s Perspective on CFD for S&C …………... 273 D. R. Bogue, T. R. Lines, and R. D. Doll Boeing Commercial Airplanes, Seattle, Washington Stability and Control in Computational Simulations for Conceptual and Preliminary Design: the Past, Today, and Future? ………………………………………... 309 William H. Mason Department of Aerospace and Ocean Engineering, Virginia Tech, Blacksburg, Virginia Computational Methods for Dynamic S&C Derivatives …………………………………… 341 Lawrence L. Green and Patrick C. Murphy NASA Langley Research Center, Hampton, Virginia Angela M. Spence Department of Aerospace Engineering, Mississippi State University, Starkville, Mississippi Part 2 Boeing TacAir CFD Capabilities/Issues …………………………………………………….. 365 David Stookesberry and Frank C. Berrier Boeing TacAir, St. Louis, Missouri vii TetrUSS Capabilities for S&C Applications ………………………………………………... 378 Neal T. Frink and Paresh C. Parikh NASA Langley Research Center, Hampton, Virginia Computational Simulations for Stability and Control – BCA State-of-the-Art …………. 396 N. J. Yu, T. J. Kao, D. R. Bogue, and T. R. Lines Boeing Commercial Airplanes, Seattle, Washington Time-Accurate Computational Simulation …………………………………………………. 417 S. Paul Pao and Pieter G. Buning NASA Langley Research Center, Hampton, Virginia Application of CFD to Abrupt Wing Stall Using RANS and DES ………………………… 433 James R. Forsythe Cobalt Solutions, LLC, Springfield, Ohio Application of Computational Stability and Control Techniques Including Unsteady Aerodynamics and Aeroelastic Effects …………………………………………… 489 David M. Schuster and John W. Edwards NASA Langley Research Center, Hampton, Virginia Hinge Moment Predictions Using CFD ……………………………………………………… 511 M. J. Grismer, D. Kinsey, and D. Grismer Air Vehicles Directorate, Air Force Research Laboratory, Wright Patterson Air Force Base, Ohio CFD Simulation of Aircraft in Coning Motion ……………………………………………... 533 Syta Saephan and C. P. van Dam Department of Mechanical and Aeronautical Engineering, University of California, Davis, California Use of CFD-Generated Aerodynamic Data for an F-15E/SLV Flying/Handling Qualities Analysis ……………………………………………………………………………... 565 Jeffery A. Batte and W. Shawn Westmoreland Jacobs Sverdrup, Eglin Air Force Base, Florida Rapid Euler CFD for High-Performance Aircraft Design …………………………………. 593 Eric F. Charlton Lockheed Martin Aeronautics Company, Fort Worth, Texas Quantitative Prediction of Computational Quality ………………………………………… 638 Michael J. Hemsch, James M. Luckring, and Joseph H. Morrison NASA Langley Research Center, Hampton, Virginia Best Practices System to Enhance CFD Use in Stability and Control Applications ……… 656 Michael R. Mendenhall Nielsen Engineering & Research, Inc., Mountain View, California viii Dynamic Water Tunnel Testing for Code Benchmarking …………………………………. 671 Brooke C. Smith and John Hodgkinson AeroArts LLC, Palos Verdes Peninsula, California COMSAC: Vision and Potential Program Planning ………………………………………. 692 Robert M. Hall and C. Michael Fremaux NASA Langley Research Center, Hampton, Virginia Joseph R. Chambers ViGYAN, Inc., Hampton, Virginia COMSAC Feedback ………………………………………………………………………….. 718 Pradeep Raj Lockheed Martin Aeronautics Company, Marietta, Georgia COMSAC Plan: NAVAIR Comments ……………………………………………………… 722 John W. Clark Naval Air Systems Command, Patuxent River, Maryland COMSAC Feedback – Boeing Commercial Airplanes ……………………………………... 730 Douglas N. Ball Boeing Commercial Airplanes, Renton, Washington ix NASA Langley Symposium on Computational Methods for Stability and Control Hampton Holiday Inn and Conference Center Hampton, Virginia September 23 – 25, 2003 Attendee List 1. Anderson, Dr. William K. 6. Berrier, Mr. Frank C.
UT SimCenter at Chattanooga Boeing TacAir Two Union Square, Suite 300 PO Box 516, MC S1066450 Chattanooga, TN 37402 St. Louis, MO 63166 kyle-anderson@utc.edu frank.berrier@boeing.com 423-648-0343 314-232-5774 2. Arabshahi, Dr. Abdollah 7. Biedron, Dr. Robert T.
UT SimCenter at Chattanooga M.S. 128 Two Union Square, Suite 300 NASA Langley Research Center Chattanooga, TN 37402 Hampton, VA 23681 Abi-Arabshahi@utc.edu robert.t.biedron@nasa.gov 423-648-0342 757-864-2156 3. Ashbaugh, Mr. William H. 8. Blake, Mr. William ASC/AAAV Air Force Research Laboratory 2145 Monahan Way, Bldg 28 AFRL/VACA Wright-Patterson AFB, OH 45433 WPAFB, OH 45433 william.ashbaugh@wpafb.af.mil william.blake2@wpafb.af.mil 937-255-7210,x-3864 937-255-6764 4. Ball, Mr. Douglas N. 9. Bogue, Mr. David Boeing Commercial Airplanes Boeing Commercial Airplanes 535 Garden Avenue N., Mail Stop 15530 Bothell Way NE #111 67-LH Renton, WA 98055 Seattle, WA 98155 douglas.n.ball@boeing.com david.r.bogue@boeing.com 425-234-1016 425-234-1078 5. Batte, Mr. Jeffery A. 10. Bower, Dr. Daniel R.
Jacobs Sverdrup National Transportation Safety 308 West D Ave, Suite 1 Board P.O. Box 1935 490 L'Enfant Plaza East Eglin AFB, FL 32542 Washington, DC 20594 batte@eglin.af.mil bowerd@ntsb.gov 850-882-0398 202-314-6562 x 11. Chambers, Mr. Joseph R. 17. Dreyer, Dr. James J.
ViGYAN, Inc. Applied Research Laboratory 205 Old Dominion Rd. Penn State University Yorktown, VA 23692 P.O. Box 30 jrchambers@cox.net State College, PA 16804 757-898-6080 jjd@wt.arl.psu.edu 814-863-3018 12. Chung, Dr. James J.
NAVAIR 18. Edwards, Dr. John W.
Bldg. 2187, Suite 1320B, Unit5 MS 340 48110 Shaw Road NASA Langley Research Center Patuxent River, MD 20670 Hampton, VA 23681 chungjj@navair.navy.mil john.w.edwards@nasa.gov 301-342-8547 757-864-2273 13. Clark, Jr., Mr. John W. 19. Evans, Mr. David M.
NAVAIR Boeing 3040 Blackberry Ln. Box 516, MS 2703760 Prince Frederick, MD 20678 St. Louis, MO 63166 clarkjw1@navair.navy.mil david.m.evans@boeing.com 301-342-8550 314-233-6290 14. Craft, Mr. John W. 20. Forsythe, Dr. James R.
Boeing, Huntington Beach Cobalt Solutions, LLC 5301 Bolsa Ave., MC H013-B318 4636 New Carlisle Pike Huntington Beach, CA 92647 Springfield, OH 45504 john.w.craft@boeing.com forsythe@cobaltcfd.com 714-235-8261 937-620-5938 15. Crider, Mr. Dennis A. 21. Fremaux, Mr. Charles M.
National Transportation Safety MS 153 Board NASA Langley Research Center 490 L'Enfant Plaza East, SW Hampton, VA 23681 RE-60 charles.m.fremaux@nasa.gov Washington, D.C. 20594 757-864-1193 criderd@ntsb.gov 202-314-6564 22. Frink, Dr. Neal T.
MS 499 16. Donaldson, Mr. Steve A. NASA Langley Research Center NAVAIR Hampton, VA 23681 B2187, Suite 1390A neal.t.frink@nasa.gov 48110 Shaw Rd., Unit 5 757-864-2864 Patuxent River, MD 20670 donaldsonsa@navair.navy.mil 301-342-0282 xi 23. Garb, Mr. Slava Z. 29. Grismer, Dr. Matthew J.
Gulfstream Aerospace AFRL/VAAC, B146 R225 PO Box 2206, M/S D-04 2210 Eighth Street Savannah, GA 31402 Wright-Patterson AFB, OH 45433 slava.garb@gulfaero.com matthew.grismer@wpafb.af.mil 912-965-3527 937-255-3876 24. Garcia, Mr. Garrett M. 30. Grove, Mr. Darren V.
U,S, Air Force NAVAIR, 75 Vandenburg Drive, Bldg 1630 Bldg. 2187, Suite 1320-D4 ESC/MAV 48110 Shaw Rd.
Hanscom AFB, MA 01731 Patuxent River, MD 20670 garrett.garcia@hanscom.af.mil grovedv@navair.navy.mil 719-964-7377 301-342-8562 25. Ghaffari, Mr. Farhad 31. Guruswamy, Dr. Guru MS 286 T27B-1 NASA Langley Research Center NASA Ames Research Center Hampton, VA 23681 Moffett Field, CA 94035 Farhad.Ghaffari-1@nasa.gov Guru.p.guruswamy@nasa.gov 757-864-2856 650-604-6329 26. Goble, Dr. Brian D. 32. Hall, Dr. Robert M.
Lockheed Martin Aeronautics Co. MS 499 101 Academy Blvd., Mail Zone 8656 NASA Langley Research Center P.O. Box 748 Hampton, VA 23681 Fort Worth, TX 76101 robert.m.hall@nasa.gov brian.d.goble@lmco.com 757-864-2883 817-935-4707 33. Heim, Mr. Eugene H.
27. Gopalarathnam, Dr. Ashok MS 153 North Carolina State University NASA Langley Research Center Mech. and Aerospace Engineering Hampton, VA 23681 Box 7910 eugene.h.heim@nasa.gov Raleigh, NC 27695 757-864-9638 ashok_g@ncsu.edu 919-515-5669 34. Hemsch, Dr. Michael J.
MS 286 28. Green, Mr. Lawrence L. NASA Langley Research Center MS 159 Hampton, VA 23681 NASA Langley Research Center michael.j.hemsch@nasa.gov Hampton, VA 23681 757-864-2882 Lawrence.L.Green@nasa.gov 757-864-2228 xii 35. Herring, Mr. Dana C. 41. Kerho, Dr. Michael F.
Raytheon Aircraft Company Rolling Hills Research Corporation MS 971-B6, PO Box 85 3425 Lomita Blvd.
9709 E Central Torrance, CA 90505 Wichita, KS 67201 MKerho@RollingHillsResearch.com dana_herring@rac.ray.com 310-257-9578 316-676-6718 42. Killingsworth, Mr. Russ D.
36. Hickey, Mr. Herbert J. (Skip) Lockheed Martin Aeronautics WJN Consulting PO Box 748, MZ 6468 5137 Croftshire Drive Fort Worth, TX 76101 Kettering, OH 45440 russ.d.killingsworth@lmco.com skiphickey@aol.com 817-7632915 937-433-5391 43. Kokolios, Mr. Alex 37. Hollingsworth, Mr.William B. NAVAIR Boeing TacAir B2187, Suite 1390A PO Box 516, MC S1066450 48110 Shaw Rd., Unit 5 St. Louis, MO 63166 Patuxent River, MD 20670 william.b.hollingsworth- kokoliosa@navair.navy.mil iii@boeing.com 301-342-8574 314-234-2221 44. Kolly, Dr. Joseph 38. Hoyle, Mr. David L. National Transportation Safety Lockheed Martin Board M/C 0685 490 L'Enfant Plaza East, SW 86 S. Cobb Drive Washington, D.C. 20594 Marietta, GA 30063 kollyj@ntsb.gov david.l.hoyle@lmco.com 202-3146622 770-494-8279 45. Komerath, Dr. Narayanan M.
39. Jiang, Dr. Minyee J. Georgia Institute of Technology U.S. Navy, NSWCCD School of Aerospace Engineering 9500 MacArthur Blvd. Atlanta, GA 30332 West Bethesda, MD 20817 narayanan.komerath@ae.gatech.edu jiangm@nswccd.navy.mil 404-894-3017 301-227-6090 46. Kramer, Mr. Brian R.
40. Karman, Dr. Steve L. Rolling Hills Research Corporation UT SimCenter at Chattanooga 3425 Lomita Blvd.
Two Union Square, Suite 300 Torrance, CA 90505 BKramer@RollingHillsResearch.com Chattanooga, TN 37402 310-257-9578 Steve-Karman@utc.edu 423-648-0595 xiii 47. Kumar, Dr. Ajay 53. Luckring, Dr. James M.
MS 285 MS 286 NASA Langley Research Center NASA Langley Research Center Hampton, VA 23681 Hampton, VA 23681 Ajay.Kumar-1@nasa.gov James.M.Luckring@nasa.gov 757-864-3520 757-864-2869 48. Laiosa, Mr. Joseph P. 54. Malone, Dr. John B.
Naval Air Warfare Center MS 285 Aircraft Division NASA Langley Research Center Bldg 2187, Suite 1320-E2 Hampton, VA 23681 48110 Shaw Road, Unit #5 John.B.Malone@nasa.gov Patuxent River, MD 20670 757-864-8988 LaiosaJP@navair.navy.mil 301-342-5723 55. Mangalam, Dr. Siva M.
Tao Systems 49. Lawrence, Mr. Tom 471 McLaws Circle NAVAIR Williamsburg, VA 23185 Aeromechanics Division siva@taosystem.com Patuxent River, MD 20670 757-220-5040 lawrencejt@navair.navy.mil 301-342-8551 56. Mason, Dr. William H.
Virginia Tech 50. Leavitt, Mr. Laurence D. 215 Randolph Hall, MC 0203 MS 499 Blacksburg, VA 24061 NASA Langley Research Center whmason@vt.edu Hampton, VA 23681 540-231-6740 Laurence.D.Leavitt@nasa.gov 757-864-3017 57. Masters, Mr. James Jacobs Sverdrup, AEDC Group 51. Lee-Rausch, Ms. Elizabeth M. Arnold AFB, TN 37389 MS 128 james.masters@arnold.af.mil NASA Langley Research Center 931-454-7739 Hampton, VA 23681 Elizabeth.M.Lee-Rausch@nasa.gov 58. McNamara, Mr. William G.
757-864-8422 NAVAIR/Naval Air Warfare Center Bldg 2035 52. Lorincz, Mr. Dale 48183 Switzer Road Northrop Grumman Corp. Patuxent River, MD 20670 One Hornet Way mcnamaraWG@navair.navy.mil 9L30/W6 301-757-0848 El Segundo, CA 90245 dale.lorincz@ngc.com 310-332-9966 xiv 59. Mehrotra, Dr. Sudhir C. 65. O'Callaghan, Mr. John ViGYAN, Inc. National Transportation Safety 30 Research Drive Board Hampton, VA 23666 490 L'Enfant Plaza E., S.W.
mehrotra@vigyan.com Washington, D.C. 20594 757-865-1400 ocallaj@ntsb.gov 202-314-6560 60. Mendenhall, Mr. Michael R.
Nielsen Engineering & Research 66. Om, Dr. Deepak 605 Ellis St., Suite 200 Boeing Commercial Airplanes Mountain View, CA 94043 Mail Code 67-LF, P. O. Box 3707 mrm@nearinc.com Seattle, WA 98124 650-968-9457 deepak.om@boeing.com 425-234-1116 61. Morrison, Mr. Joseph H.
MS 128 67. Ozoroski, Ms. Lori P.
NASA Langley Research Center MS 348 Hampton, VA 23681 NASA Langley Research Center joseph.h.morrison@nasa.gov Hampton, VA 23681 757-864-2294 Lori.P.Ozoroski@nasa.gov 757-8645992 62. Murphy, Dr. Patrick C.
MS 132 68. Pao, Dr. S. Paul NASA Langley Research Center MS 499 Hampton, VA 23681 NASA Langley Research Center Patrick.C.Murphy@nasa.gov Hampton, VA 23681 757-864-4071 s.p.pao@nasa.gov 757-864-3044 63. Murri, Mr. Daniel G.
MS 153 69. Parikh, Dr. Paresh C.
NASA Langley Research Center MS 499 Hampton, VA 23681 NASA Langley Research Center Daniel.G.Murri@nasa.gov Hampton, VA 23681 757-864-1160 paresh.c.parikh@nasa.gov 757-864-2244 64. Nelson, Dr. Robert C.
University of Notre Dame 70. Park, Mr. Michael A.
Department of Aerospace and MS 128 Mechanical Engineering NASA Langley Research Center Notre Dame, IN 46556 Hampton, VA 23681 robert.c.nelson.1@nd.edu Michael.A.Park@nasa.gov 574-631-4733 757-864-6604 xv 71. Pfeiffer, Dr. Neal J. 77. Roberts, Mr. Michael Raytheon Aircraft Company University of California, Davis MS 999-B12 , PO Box 85 Dept. of Mechanical & Aeronautical 9709 E. Central Engineering Wichita, KS 67201 1 Shields Ave., neal_pfeiffer@rac.ray.com Davis, CA 95616 316-676-6775 mike_m_roberts@yahoo.com 530-752-2261 72. Pirzadeh, Dr. Shahyar Z.
MS 499 78. Saephan, Mr. Syta NASA Langley Research Center University of California, Davis Hampton, VA 23681 Dept. of Mechanical & Aeronautical Shahyar.Z.Pirzadeh@nasa.gov Engineering 757-864-2245 1 Shields Ave., Davis, CA 95616 73. Player, Ms. Jennifer ssaephan@ucdavis.edu Bihrle Applied Research, Inc. 530-752-2261 18 Research Drive Hampton, VA 23666 79. Schuster, Dr. David M.
jplayer@bihrle.com MS 340 757-766-2416 NASA Langley Research Center Hampton, VA 23681 74. Raj, Dr. Pradeep david.m.schuster@nasa.gov Lockheed Martin Aeronautics Co. 757-864-2259 86 S. Cobb Dr.
D/1AEM, Z/0660 80. Shah, Mr. Gautam H.
Marietta, GA 30063 MS 153 pradeep.raj@lmco.com NASA Langley Research Center 770-494-3801 Hampton, VA 23681 gautam.h.shah@nasa.gov 75. Ralston, Mr. John 757-864-1163 Bihrle Applied Research, Inc.
18 Research Drive 81. Skujins, Dr. Ojars Hampton, VA 23666 Aeronautical Systems Center jralston@bihrle.com ASC/ENFT 757-766-2416 2530 Loop Road West Wright-Patterson AFB, OH 45433 76. Ratvasky, Mr. Thomas P. ojars.skujins@wpafb.af.mil 21000 Brookpark Rd, Mail Stop 11-2 937-255-4166 NASA Glenn Research Center Cleveland, OH 44135 82. Slomski, Dr. Joe thomas.p.ratvasky@nasa.gov NSWC Carderock Division 216-433-3905 9500 MacArthur Boulevard West Bethesda, MD 20817 slomskijf@nswccd.navy.mil 301-227-1935 xvi 83. Smith, Mr. Brooke C. 89. Tenney, Dr. Darrel R.
AeroArts LLC MS 208 P.O. Box 2909 NASA Langley Research Center Palos Verde Peninsula, CA 90274 Hampton, VA 23681 Brooke.Smith@AeroArts.com Darrel.R.Tenney@nasa.gov 310-547-0927 757-864-6033 84. Stookesberry, Mr. David C. 90. Thacker, Mr. Michael Boeing Cessna Aircraft Company P.O. Box 516, MC S106-6420 5701 E. Pawnee, D363P St. Louis, MO 63166 Wichita, KS 67218 david.c.stookesberry@boeing.com mthacker@cessna.textron.com 314-233-6346 316-831-2807 85. Stuckert, Dr. Gregory K. 91. Thomas, Dr. James L.
Fluent Inc. MS 128 10 Cavendish Court NASA Langley Research Center Lebanon, NH 03766 Hampton, VA 23681 gks@fluent.com James.L.Thomas@nasa.gov 603-643-2600, x243 757-8645578 86. Stuever, Mr. Robert A. 92. Van Dam, Dr. Case The Boeing Company University of California, Davis Wichita Development & Dept. of Mechanical & Aeronautical Modification Center Engineering P. O. Box 7730, MC K05-14 1 Shields Ave., Wichita, KS 67277 Davis, CA 95616 Robert.A.Stuever@boeing.com cpvandam@ucdavis.edu 316-523-4826 530-752-7741 87. Sung, Dr. Chao Ho 93. Vassberg, Dr. John C.
Dept of the Navy The Boeing Company 9500 MacArthur Blvd 5301 Bolsa Avenue West Bethesda, MD 20817 MC H013-B318 sungch@nswccd.navy.mil Huntington Beach, CA 92647 301-227-1865 john.c.vassberg@boeing.com 714-896-1607 88. Tello, Mr. Richard J.
The MITRE Corporation 94. Wahls, Dr. Richard A.
75 Vandenberg Drive MS 499 Attn: Joint Stars NASA Langley Research Center Hanscom AFB, MA 01731 Hampton, VA 23681 rtello@hansom.af.mil richard.a.wahls@nasa.gov 781-377-9919 757-864-5108 xvii 95. Westmoreland, Mr. William S.
Jacobs Sverdrup 98. Yip, Mr. Long P.
205 West D Ave. MS 254 Eglin AFB, FL 32542 NASA Langley Research Center westmore@eglin.af.mil Hampton, VA 23681 850-8820918 Long.P.Yip@nasa.gov 757-864-1700 96. Wong, Dr. Tin-Chee U. S. Army, AMSAM-RD-AE-A 99. Yu, Dr. Neng J.
Bldg 4488 Boeing Commercial Airplanes Redstone Arsenal, AL 35898 Enabling Technology & Research TinChee.Wong@us.army.mil M/S 67-LF, P.O.Box 3707 256-705-9605 Seattle, WA 98124 neng.j.yu@pss.boeing.com 97. Yang, Dr. Cheng I. 425-234-1192 Code 5400, Bldg 16 Naval Surface Warfare Center Bethesda, MD 20084 yangci@nswccd.navy.mil 301-227-4658 xviii
COMSAC SYMPOSIUM
Introductory Remarks
September 23, 2003
Dr. Darrel R. Tenney Director for Aerospace Vehicle Systems Technology Office NASA Langley Research Center Hampton, VA NASA/DoD Aerodynamic Flight Prediction Workshop Nov. 19-21, 2002 in Williamsburg, VA Ojars Skujins Brian Kramer Joe Laiosa Russ Killingsworth Steve Cook Charlie Wilson Tom Rudowsky Mike Fremaux Dan Murri Dave Evans Brian Lundy Paul Waters Frank Lynch Joe Chambers Rich Wahls Ed Kraft Dale Lorincz Lawrence Ash Mike Hemsch Neal Pfeiffer Mark Potsdam Dan Banks Steve Donaldson Mark Booher John Rundquist Pradeep Raj Jim Thomas Marge Larry Leavitt John Roberts Doug Ball Bill Baker Sue Crotts John Liu Rick Hooker Ajay Kumar Draper-Donley Dave Schuster Tom Lawrence Frank Berrier John Malone Bill Sellers Terry Britt Bob Hall Deepak Om Donna Speller Long Yip Hugo Gonzales Greg Addington James Chung Tim Naumomicz Bob Stuever Tom Ratvasky Jennifer Hansen Megan McCluer Jack Ralston Tony Antani Mark Potapczuk Roger Clark Al Paris Mike Bonner
General Findings
Aerodynamic Flight Prediction Workshop Nov. 19-21, 2002 in Williamsburg, VA 1. Pr ediction of the onset of separated flows across 1. Pr ediction of the onset of separated flows across the speed range (with the attendant issues of the speed range (with the attendant issues of transition prediction, turbulence modeling, unsteady transition prediction, turbulence modeling, unsteady flows, etc.) and the character and impact of separated flows, etc.) and the character and impact of separated flow on aircraft capabilities is the si ngle most critical flow on aircraft capabilities is the single most critical fundamental issue to be addressed and should fundamental issue to be addressed and should receive a very high pr iority in aerodynamic R&D receive a very high pr iority in aerodynamic R&D progr ams .
progr ams .
2. The issu e of Reynolds number impact s on 2. The issu e of Reynolds number impact s on aerodynamic prediction s continues t o p ose aerodynamic prediction s continues t o p ose significant barrier s to advances in the state of the significant barrier s to advances in the state of the art. The issues leading t o this situation (cost, art. The issues leading t o this situation (cost, accuracies, operational difficulties, etc.) should be accuracies, operational dif ficulties, etc.) should be addr essed with high pri ority.
addressed with high pri ority.
3. T he loss of corpor a te knowledge and 3. T he loss of corpor ate knowledge and docum entation of lessons lear ned in aerodynamic docum entation of lessons lear ned in aerodynamic predictions is a maj or area of concern. As a result predi ctions is a major area of concern. As a result of corporate mergers, large turnovers in staffs within of corporate mergers, large turnovers in staffs within government an d industry, and fewer aircra ft program s, gov ernment and industry, and fewer aircraft program s, the nation is rapidly los ing its cornerstone experience the nation is rapidly losing its cornerstone experience base for the fut ure.
base for the future.
Concluding Remarks
• Future vehicle designs will see a paradigm shift from
– Steady to the unsteady world (e.g. flow control, adaptive morphing), – Passive to active, – Rigid designs to exploitation of flexibility and adaptability – Few discrete to numerous distributed (e.g. sensors, control surfaces) – To obtain a vehicle that is always at optimum performance.
• Therefore, future designs will be inherently multidisciplinary,
and the greatest technical challenges and opportunities
occur at the intersection of disciplines
• COMSAC appears to be a step towards enabling the future
vision
This Symposium is intended to bring together the often distinct cultures of the Stability and Control (S&C) community and the Computational Fluid Dynamics (CFD) community. The COMSAC program is itself a new effort by NASA Langley to accelerate the application of high- end CFD methodologies to the demanding job of predicting stability and control characteristics of aircraft. This talk is intended to set the stage for needing a program like COMSAC. It is not intended to give details of the program itself.
While there are many reasons to have this Symposium, a direct motivation for this event was the Flight Prediction Workshop.
This chart, by Doug Ball of Boeing Commercial, highlights the large amount of wind tunnel resources that are dedicated to determining stability and control characteristics, certification requirements, and low-speed lines. CFD has not generally penetrated these needs areas.
Impacts occur across of vehicle classes--767,F/A-18E, C130J, T-45,X-43 Stack, 777, Lear 23, AV-8B, and 737NG.
• 767--Stall for 767-400 model with raked tips more rapid than expected--vortilon pattern had to be developed • F/A-18E--wing drop at transonic speeds. Impact: program almost canceled.
• C-130J--wing drop due to propeller induced effects. Impact: delayed deliveries, increased development costs • T-45--low speed approach wing drop. Impact: redesigned wing • X-43 Stack--inaccuracies of S&C aero data base. Impact: lost research vehicle • 777--missed horizontal tail effectiveness. Impact: larger than needed horizontal • Lear 23--Laminar separation bubble breakdown leading to wing drop on approach.
Impact: safety of flight, development costs • AV8B--wing drop and wing rock. Impact on operational envelopes (considered minimal) • 737--737NG (400 to 800) sensitivity to wing rigging with unacceptable number of aircraft not passing acceptance flights. Impact: production expenses and development costs Existing tools and methods for predicting characteristics when flow is primarily attached are adequate. However, when separation becomes significant, analytical tools are inadequate and CFD methods have not been calibrated, in general.
While wind tunnel availability is decreasing, needs for aero data bases are increasing.
Computational tools will be needed to complement wind tunnel data to an increasing extent in the future.
As will be reported in this Symposium, current and emerging CFD methods offer the exciting promise of new approaches to address the S&C needs. This will be even more important as emerging flow-control concepts are brought on line.
The pyramid shows the general evolution of algorithms and computer power as a function of decade. The level V is labeled RANS+ because of the addition of methodologies such as either Large Eddy Simulation (LES) or Detached Eddy Simulation (DES). The bottom line is that there are new developments in algorithms which, when combined with increasing availability of computer resources, will enable the community to address problems that previously were untenable. The challenge now facing the CFD community is to take the latest levels of technology and begin making the sort of impacts in the stability and control arena that it has already made in the performance arena.
This list shows just a few of the many applications that have been addressed by the authors reporting during this Symposium. This is merely to communicate that a lot of work has already been done by a lot of organizations.
I would like to show one example with which I am familiar that comes from the Abrupt Wing Stall (AWS) program. This work was by Jim Forsythe and utilized a Detached Eddy Simulation (DES) implementation. The insight into the flow physics of this example changed the thinking of the S&C folks.
While there are examples of successes in applying CFD to S&C problems, it is still unclear within and outside of the CFD community that the current state-of-the-art is up to the task of predicting the very complicated, sometimes time dependent, flows associated with massively separated flows. What is clear, however, that it was appear that if separation is a large player in the flow field, it will be necessary to bring to the problem RANS or RANS+ levels of technology. This means that large resources will be required to address these problems. So ways will have to be found apply these codes with as much automation and robustness as possible. Of course, CFD credibility must be established in the S&C community by demonstrating that the codes can predict the answer before knowing it. Finally, while cultural differences are a challenge in bringing together the two disciplines, some of the reduced accuracy requirements associated with S&C may reduce some of the resource requirements.
This chart contrasts the differences between the two communities.
NASA has been involved with trips to different organizations to make sure we understood the level of technology and the needs of the communities.
This presentation is designed as a limited-scope “tutorial” and is aimed primarily at the CFDer who has not been exposed to stability and control problems. Examples of some classic S&C problems are used for illustration.
S&C is a fundamental technology for enabling flight, but significant problems with the prediction of S&C characteristics persists, especially where separated flow is involved.
Even after 100 years of flight, experimental methods still have significant limitations.
Experimental and computational tools can and must be complementary.
NASA Flight Prediction Workshop (Williamsburg, Virginia, November 2002) brought together experts from government, industry, and academia to discuss problems associated with state-of- the-art flight prediction. Among the concerns highlighted were deficiencies in S&C prediction lack of calibrated CFD tools for aerodynamic prediction in general.
Some problem areas highlighted at the Flight Prediction Workshop, plus a few added by the author.
For illustration purposes, problem areas for four “vehicle classes” are examined.
Some issues typically associated with large transports. Items in red are highlighted in the example on the following page.
As illustrated by NASA Aviation Safety Program data, roll damping for a large jet transport predicted by the forced-oscillation technique in a wind tunnel is significantly different from that obtained by analytical or handbook methods (e.g. DATCOM), as illustrated by the “Simulation Model” curve. Wind tunnel data indicate that this configuration will have slightly unstable roll damping at stall and will be highly unstable in roll above about 40 degrees angle of attack.
Training pilot for stalls and dealing with “out of control” upset conditions may be greatly improved by having better roll damping predictions for simulation.
High performance airplanes can have many of the same issues as transports, but there are differences due to the configuration (e.g., sharp leading edge wings, highly swept leading-edge extensions (LEX) or strakes, and close-coupled control surfaces. The fact that these vehicles routinely maneuver at post-stall angles of attack means that flying with separated and vortical flow is the rule, not the exception. Transonic phenomena such as shock-induced wing drop or low-speed wing rock are also not uncommon.
The F-4 was originally designed as a “missile shooter”, not a high- α fighter. During the Vietnam conflict, they were engaged as close-in dogfighters and began suffering significant losses due to spin accidents resulting from loss of directional control at elevated angles of attack.
Over 100 Navy and Air Force 100 F-4s were lost before the cause of the problem was identified and resolved by modifications to the leading edge of the wing (slats) to delay stall and improve stall warning. Adverse sidewash at the tail as a contributing factor to loss of directional stability was identified through wind tunnel tests.
In this case, adding area to the vertical tail to improve directional stability helps for pre-stall angles of attack (i.e., prior to formation of the large wake from the stalled wing), as anticipated, but actually makes the directional instability worse at high angles of attack due to the adverse sidewash at the tail.
Video of F-4 experiencing directional departure during flight-test wind up turn and entering flat spin illustrates how rapidly the airplane goes from controlled to uncontrolled flight.
Again, many S&C issues in common with large transports and high-performance fighters, but business jets tend to have T-tails and commonly do not have leading edge devices, potentially leading to issues with deep stall and laminar separation bubbles, respectively.
Wind tunnel data (NOT for configuration in photo at left) show that some T-tail airplanes do not have enough nose-down control authority at high angles of attack to recover from a deep stall.
Animation shows laminar separation “bubble” at leading edge at elevated angle of attack (e.g. in landing configuration) progressing to sudden full wing stall on one side after the bubble “lets go” and the entire surface separates abruptly. Large rolling moments are then induced by the asymmetric stall pattern, which is potentially catastrophic if the airplane is at low altitude.
Unconventional configurations such as flying wings are illustrated by the Blended Wing Body (BWB). Flying wings have many distinct S&C characteristics, depending on the geometry, but may include reduced longitudinal and directional stability due to the lack of a tail, highly non- linear control surface interactions if there are multiple control surfaces, and the potential for entering a tumble mode (i.e., autorotation in pitch).
The Northrop YB-49 (and earlier XB-35) were advanced all-wing bombers produced in the late 1940s. Longitudinal stability in general (and tumbling in particular) were identified as potential problems for flying wings early on, and experimental studies were conducted to identify potential problem areas. The plot shows wind tunnel pitching moment data for another flying wing which shows that the vehicle is statically unstable in pitch (I.e., the slope of the curve is positive near zero angle of attack), which could lead to a pitch departure if the dynamic pitch damping is such that rotation is sustained over a complete 360 degree cycle.
COMSAC goals include increasing the acceptance of CFD as a viable tool for S&C predictions, as well as to focus CFD development and improvement towards the needs of the S&C community. We view this as a symbiotic relationship, with increasing improvement of CFD promoting increasing acceptance by the S&C community, and increasing acceptance spurring further improvements.
In this presentation we want to provide an overview for the non CFD expert of current CFD strengths and weaknesses, as well as to highlight a few emerging capabilities that we feel will lead toward increased usefulness in S&C applications.
“CFD” can imply different things to different people. To put everyone on the same footing for this presentation we are going to restrict the definition of CFD to imply the numerical solution of the Navier-Stokes equations, with the Euler equations as a subset.
There are of course many levels of approximation to the Navier-Stokes equations, principally distinguished by how turbulence is treated. In principle one can use a fine enough mesh and a small enough time step to resolve and track all the important scales of the turbulence. Such solutions for full aircraft geometries are many decades away. So for practical applications we must resort to some level of turbulence modeling.
At the highest level of turbulence modeling lie the Reynolds Averaged Navier-Stokes (RANS) solvers. These solvers require O(10 ) points for a complete configuration in order to have an adequate resolution of the flow field. In a RANS code, the effect of turbulence is entirely modeled; the grid is not dense enough to realistically track individual turbulent eddies anywhere in the field. Many RANS codes can simulate both steady and unsteady flows; those that simulate unsteady flows are often referred to as URANS. One might label RANS solvers as “state-of-the- art” for engineering applications.
RANS codes tend to do a poor job with massively separated flows. Detached Eddy Simulation (DES) is one of a number of hybrid methods that have been proposed to better deal with these flows. A well-resolved DES calculation may require O(10 ) points. Large, detached eddies in the separated flow region are computed and tracked, while the remaining turbulent length scales (near the body) are modeled with the RANS approach. A DES simulation is fundamentally unsteady. The need for very large numbers of grid points and time-accurate simulation puts DES out of the engineering realm at this time, into the “Grand Challenge” category.
All CFD codes employ some sort of grid or mesh, and may be categorized into two general types by the kind of grid used.
Structured grids are comprised of body-fitted hexahedral cells. In order to fit around complex geometries, structured grids must usually be made up of multiple blocks or zones of points.
These zones may either abut against one another, or may overlap. Overlapped grids are often referred to as overset grids. Flow solvers utilizing structured grids make use of the inherent connectivity (or structure) between the cells, resulting in relatively fast flow solvers.
The other major category of flow solvers utilize unstructured grids. Such grids are also body fitted, but are usually comprised of tetrahedral cells, although prisms, pyramids and hexahedra are often used. Prisms are particularly well suited for use within the boundary layer.
Unstructured grids can readily handle complex geometry, easing the grid generation problem.
The lack of any inherent connectivity between cells means that nearby neighbors of each cell must be explicitly spelled out for the solver, resulting in a slower speed than a comparable structured-grid solver. However, the ease of grid generation, as well as the relative ease of adaptation, discussed in subsequent slides, makes unstructured solvers very attractive – it’s a tradeoff of more CPU time for less human effort.
To set the stage for later discussion it is worthwhile to give a very broad overview of the CFD solution process in today’s environment. First, it is important to emphasize that the quality of a CFD result hinges on the quality of the grid.
When generating the grid (typically a substantial undertaking), the CFD expert or grid expert (ideally one person expert in both areas) decides where to place points. Typically points are clustered to resolve geometrical features and flow features using established “best practices”.
After the grid is generated, the flow solver is run and the process stops with a solution on this grid.
Adaptation carries the process one or more steps further. Given a solution on the baseline grid, obtained as described above, points are added to better resolve “important” features, typically indicated by regions of strong gradients. Then the solution/adaptation process is repeated until the user is satisfied (or gives up…). A fundamental question, sometimes not easily answered even by the expert, is what features and regions are important to resolve in order to get the desired results for the problem of interest?
Adaptation may be done in with formal process as described above. More often however, there is a less formal approach to “adaptation” that occurs when the original grid fails to produce the desired result. In these cases the grid is changed in a manner deemed to result in a better CFD prediction from subsequent simulations. Unstructured grids offer the best path for adaptation as points can more easily be inserted into the grid as needed.
This slide and the next are intended to give an overview of some of the current capabilities of CFD.
Here we show a solution-adapted grid for a Modular Transonic Vortex Interaction (MTVI) configuration. This modular wind-tunnel model was used to provide detailed experimental data for CFD validation with a wide variety of vortical flows. The image in the lower right shows the grid in the vortical flow regions after refinement. The refinement was based on the locations of strong off-body vorticity as computed by the flow solver. The chine and wing leading edge vortices are indicated. The image in the upper left shows the corresponding flow visualization from the experiment. In this case the adaptation process was critical to the prediction of vortex bursting, which in turn was critical to predicting pitch up.
This slide shows a sampling of the wide range of problems that have been tackled using CFD.
Today’s CFD codes are capable of simulating a wide variety of steady and unsteady flows over complex configurations. An example of a steady flow simulation is that shown for the S3 Viking in the lower left.
Complex flow interactions may arise even for simple geometries, but are the norm for complex configurations. An example is the full-stack Space Shuttle simulation in the upper left. Here we see the multiple shock waves formed during ascent.
Unsteady simulations may include bodies in relative motion. In the upper right is shown a computation of the V22 Tilt Rotor with the spinning blades in the cruise mode. The complex wake structure is made visible by particle traces emanating from the rotors and wing tips.
Although not shown, analysis capabilities have been extended into the design environment in limited applications. More precisely, the use of CFD in design has been largely limited to design improvement, rather than for conceptual design. However, efforts are underway to bring CFD earlier into the design process.
Now let’s look at two applications of CFD to stability and control of the F/A 18 aircraft, in the areas of Forebody Controls and Abrupt Wing Stall.
In the High Alpha Technology Program, a number of novel control effectors for maneuvering at high angles of attack were studied. Among them was an actuated nose strake, designed to be deployed on an as-needed basis. Early on in the wind tunnel program, using a generic strake design, it was observed that the direction of the yawing moment produced by the strake changed with deployment angle. CFD was used to confirm the experimental observation for the final strake design intended for use on the aircraft. The image on the top right shows (left to right) the vortical structure with the strake retracted, extended 10 degrees, and 90 degrees (full extension).
The image on the lower right shows the computed yawing moments (symbols) for those three strake positions, as well as the wind tunnel measurements for a range of deflections (line). It is seen that CFD simulation has correctly predicted the yawing moment reversal, and has done a reasonable job at predicting the magnitude of the yawing moment.
The image in the top left shows in-flight visualization of the vortex generated by the nose strake (fully deployed); the image in the lower left shows the corresponding visualization from the computation (green trace). Also shown are the traces of the LEX vortices (red and blue traces); the asymmetrical LEX vortex breakdown induced by the nose strake deployment is evident.
This slide shows a more recent application, by Jim Forsythe using the Cobalt code, for the prediction of Abrupt Wing Stall on the F/A-18E. The “E” variant exhibited an abrupt wing drop for a certain range of Mach number and angle of attack. Initial attempts using the RANS approach with either the Spalart-Allmaras (SA) model or the Shear Stress Transport (SST) model missed the shock location and thus gave incorrect predictions for the lift coefficient. In the case of the SA model the lift coefficients were in general too large, and though the break in the lift curve slope was correctly predicted near nine degrees angle of attack, the reduction in slope was under predicted. Conversely, using the SST model, the lift coefficients in the attached flow region were well predicted, but the break was predicted to occur too soon.
Subsequently, the DES approach was employed, using the SA model near the body. Time averaged output showed a more accurate prediction of the shock location compared to the standard RANS model with SA. Improvement was also seen in the lift coefficient near the break, but the break occurred too early. Finally, the grid was solution-adapted to the initial DES result, and the DES solutions were re-run. Time averaged lift coefficients from the DES simulations showed excellent agreement with the data, predicting the break near nine degrees correctly, as well as the correct reduction in slope, and the subsequent increase in lift curve slope beyond 12 degrees.
Now that we have shown a sampling of CFD capabilities, including some directly addressing S&C, lets consider some important CFD issues that can directly impact S&C calculations.
Massively separated flows will be quite common for S&C applications. Such flows require a grid of high resolution to adequately capture the slow-moving wake regions, which tends to slow the convergence of the CFD solution even for nominally steady flows. But in reality such flows are usually unsteady and we may need to do time dependant simulations in order to properly capture the relevant effects.
Then the question arises, is URANS sufficient or do we need DES? The preceding slide gave evidence that the DES is required in at least some situations. This leads to greater computational cost.
In many cases we will be dealing with transonic flows, and so we may need to resolve complex flows involving shock-shock interactions, shock-boundary layer interactions, and shock-vortex interactions.
Another potentially large issue involves transitional flows, especially ones involving laminar separation, transition to turbulent flow, and subsequent reattachment. Physics-based prediction of transition is not generally available in Navier-Stokes solvers. Even for high Reynolds Number flows based on vehicle length scales, there may be localized transitional phenomenon on control surfaces or near wing leading edges that can have a huge impact on S&C.
We will often have to model the entire configuration. This of course requires double the number of grid points needed compared to situations where symmetry can be assumed.
Such things as differential control surface deflections, sideslip or roll, as well as lateral flow asymmetries arising in a nominally symmetric configuration will dictate a full grid.
Derivatives are often of interest. Thus we need to be able to calculate these reliably for both static and dynamic situations. It should be noted that often derivatives change sign over very small ranges of flow conditions, so simple finite differencing over (say) angle of attack ranges of a degree or so may not yield sufficient accuracy. Other methods for evaluating derivatives, such as complex arithmetic or differentiated source code can provide more reliable derivatives, but are not available in all solvers.
There are many cases where vehicle dynamics are important. These can range from situations in which aeroelastic effects are important to 6 DOF motion.
Finally, at some point many years from now we would like to handle all these situations in a design environment, so that S&C considerations can be designed in from the start.
This slide lists some additional technology barriers that inhibit engineering applications of CFD to S&C.
As things stand today, the whole process of obtaining a CFD solution, but particularly the grid generation aspect, requires a high degree of expertise. This applies to the use of the grid generation software and to the experience required to judiciously place grid points for an accurate CFD result.
The time to obtain a solution is also currently too long for day-to-day engineering calculations.
Flow codes are not nearly as efficient at solving the equations as they could be, and the increasing need to simulate unsteady flows just compounds the problem.
The issues of transition and turbulence modeling, especially for separated flows, is one that causes considerable debate even among CFD experts Solvers are not robust enough. Not every attempt to get a CFD solution is successful, especially at extreme conditions. Even on a good day, if the solver runs 90% of the cases, the remaining 10% seem to require 90% of the user’s time.
When a solution is obtained, there may be questions that arise as to it’s accuracy, as well as the uncertainty associated with the result. Mike Hemsch will cover these issues in more detail in a separate presentation.
Finally, the whole process is fairly complex, even for the CFD expert, so much work needs to be done to simplify the process for “routine” engineering applications.
Here we list a few capabilities that are emerging and should have a beneficial impact on S&C applications.
The first is error-based adaptation, which should go a long way to automating the CFD process.
We will cover this topic in more detail in the next few slides.
There has been some very promising work to couple high-fidelity, Navier-Stokes solvers with lower order, potential or Euler solvers in a design environment. In the long term, this will allow inclusion of CFD earlier in the design process than is currently practical.
A significant effort is underway to increase the basic flow solver efficiencies toward their full potential. This work is particularly targeted towards the multigrid methods. Impressive results have been obtained for simple flows, but application to complete configurations with complex flows is some years away.
Progress has been made in the computation of unsteady flows, using dual time step methods, as well as the efficient evaluation of rate derivatives for constant rotation rate cases.
The DES method described earlier, has been applied to a wide range of massively separated flows, and may eventually become a routinely used tool for such cases. Likewise, the inclusion of aeroelastic effects, particularly static deformations, is becoming more widely used.
Finally, cheaper faster computers will help bring CFD into the S&C world, regardless of advances on the algorithmic front.
This slide illustrates some trends in computing resources at NASA Langley Research Center – no doubt similar trends can be observed elsewhere.
In the early 1990’s we were doing all of out CFD simulations on Cray vector supercomputers. At that time we had approximately 20,000 hours available for the center, and they cost around $100 per hour. By 1995 the SGI Origin class machine was beginning to be widely used, perhaps doubling the available hours and halving the computing costs. Parenthetically it should be noted that a great deal of human effort was required to make the change from the single processor vector machines to the parallel processors of the SGI. Nonetheless, the Origin class machines took over as the primary computing platform by the late 1990’s. Both the Cray and the SGI machines were developed with high-end computing as their primary market niche.
Meanwhile, Linux “Beowulf” clusters of commodity machines appeared, and have been steadily gaining ground. Price drops in CPUs, memory chips and storage have made the Linux clusters very compelling. Having made the effort in converting to parallel processing for the SGI Origin, the effort necessary to adopt the Linux clusters was comparatively minimal. Today there are roughly the equivalent of 10 million Cray C90 hours available, at a cost of about $0.10 per hour.
Now we want to turn attention to a newly emerging technology that shows a tremendous potential as a way to move forward to what might be termed “automated CFD”.
As discussed earlier, current adaptive methods can be somewhat ad-hoc. Recall that the CFD expert must identify the key feature or features to adapt to. Usually, human intervention is required to control the process and decide when to stop. Furthermore, we are still left with the question of what features are important for the problem at hand.
Ultimately, what we want is to provide engineers with a tool that is useful for timely engineering trade studies. So, we must ask, is there a less ad-hoc means of adaptation, and one that can be automated?
Not surprisingly, we believe the answer is “yes” – error-based adaptation.
Error-based adaptation method described here is the result of some pioneering work in two dimensions by Darmofal and Venditti a MIT. It is currently being extended to three dimensions by Mike Park at NASA Langley. As applied to CFD, the method is quite new; though a similar methodology seems to have been used for some time in computational structures.
This methodology is based upon a solution of the adjoint (dual) equations for the Navier-Stokes or Euler (primal) equations, which is used to determine a computable error estimate. In this approach, the engineer defines an error tolerance on an integral quantity of interest. For example, he may want to know the drag to within one count. (It should be noted that “within one count” refers to the best solution that can be obtained given the choice of numerical scheme, turbulence model, geometrical fidelity, etc. – not necessarily to within 1 count of the “true” answer.) Given a solution on a baseline grid, the method adapts the grid to minimize the primal and dual equation errors, and thus dictates where the mesh is refined. Furthermore, it provides a means to automatically terminate the process when the error is less than the specified tolerance.
This slides shows an application of error-based adaptation to supersonic inviscid flow past a staggered pair of airfoils – a Mach 3 biplane if you will.
On the left is the flow pattern that is obtained on a very-well resolved grid. The shock waves and their interactions are all captured quite well, with very sharp resolution of the shocks. On the top right is a solution adapted grid where the pressure gradient was the feature chosen as the adaptation criterion. The final grid contains nearly 38,000 points, and the computed drag coefficient on the lower airfoil drag is 767 counts.
On the lower right is the grid that results when the error-based grid adaptation method is used.
With only 3800 nodes – ten times fewer than the pressure based adaptation – the computed drag coefficient on the lower airfoil is 766 counts.
Notice that the error-based method has not resolved many of the flow interactions that one might think are absolutely necessary to resolve for an accurate prediction of the drag in this supersonic flow. In fact, apart from the leading and trailing edges, the only readily discernable feature that has been adapted to is the part of the bow shock from the upper airfoil that impinges on the lower airfoil.
The relatively simple example on the previous slide suggests that in cases of complex flow interactions, intuitive, ad-hoc adaptation schemes may be quite wasteful of grid points, leading to needlessly long computation times.
The remarkable savings in grid points has been seen in many other cases, including viscous flows. However, we should be careful not to oversell this methodology at this point. It is still evolving, and some significant difficulties lie ahead. For viscous flows, much development work still needs to occur in the adaptation mechanics for highly anisotropic cells. Unsteady flows, which may be quite important in S&C applications, still need theoretical development. The method as developed to date relies on a convergent solution to a steady state. Finally, it should be noted that the development of the adjoint solver is very labor intensive.
We believe that now is the time to promote a more aggressive use of CFD for S&C. There will be failures – they are to be expected – but that is the only way to make progress.
We feel that a coordinated effort between experiment and CFD is required. In addition to the usual force and moment data that the comes out of experiments geared toward S&C, more detailed information, including flow visualization, pressure data and velocity distributions are needed for CFD calibration. Such coordinated studies should include fundamental studies on simple configurations – to allow for careful grid and time step convergence studies – as well as studies on complete configurations of interest to industry in order to maintain relevance.
Finally, we believe that a computing workshop along the lines of the recent Drag Prediction Workshops should be held for a problem of interest to the S&C community. In these workshops, multiple codes are applied to the specified configuration (using supplied grids and/or grids of the participant’s making), with comparisons made to experiment for assessing accuracy.
Over the past 25 years the field of Computational Fluid Dynamics has made tremendous strides.
Wings can be designed and tested with complete confidence that the results will be as expected.
Their reliability and range of applicability has grown. Now is the time to explore what can be done with CFD at the corners of the flight envelope.
My presentation is broken down into three areas: 1. A description of the kinds of “surprises” that Boeing Commercial Airplanes has experienced in the last 25 years; 2. A sampling of the kinds of CFD modeling that we are doing in support of stability & control and loads issues 3. A brief discussion as to how we, as an government/industry/academia team might operate.
The 777 had many miss-predictions that were discovered during flight testing. None of them major, but things that had we known about them during the design phase we would have designed a different airplane. An example of this is the fact that the airplane is nearly .01Mach faster than planned. Had we known this we could have reduced the sweep of the wing to improve low speed performance and reduce weight or thickened the wing to reduce weight and increase fuel volume.
The 737NG family had several “surprises of its own”.
The flaps-up stall characteristics turned out to be unacceptable. The airplane exhibited too much stick lightening during stall (pitch up. The dark black line on the left of the plot show the pitch characteristics for the rollout configuration.
The addition of a stall strip on the inboard wing leading edge and three vortilons on the outboard slat improved the stall characteristics.
These drawings show the size and location of the inboard stall strip These drawings show the size and location of the vortilons on the outboard slats.
Another issue for the 737NG was a lateral trim issue. The story goes like this: If the trailing edge flaps aren’t rigged quite right they can separate. The figure shows the left hand flaps separating. This separation flows back and blankets one side of the vertical tail. The velocity difference across the tail causes the tail to yaw the airplane nose left. As the airplane yaws the left wing increases in sweep and the right wing decreases in sweep. This causes a decrease/increase in their lift curve slopes thus creating more lift on the right wing and less on the left. This lift imbalance causes the airplane to roll left wing down.
Here are flight test photos showing normal flow on the flap and side of body in the pictures on the left, and separated flow on the right.
This problem has caused us to have to rerig and refly predelivery flight tests until the problem has been resolved. This can cause delay in delivery (and revenue) as well as the additional costs of the extra flight tests An asymmetric stall on the 767-400 at flaps 15 was resolved with the use of vortilons across several of the outboard slat segments So what are the emerging capabilities in CFD that may help us to avoid these kinds of surprises on future airplanes?
There is no question as to the economic impact the successful application of CFD throughout the flight envelope will have. We have gone from building and testing 77 wings on the 767 to 11 for the 737NG. This is a huge savings in both money and time. The areas of high lift design, stability and control validation and simulator database development, and loads database development will all reap huge rewards from the intelligent application of CFD.
Unstructured grid technology makes it now possible to model configurations having a great degree of geometric complexity.
Examples showing the nature of the unstructured grid used to model the 777.
Using automated gridding procedures it is now possible to get converged solutions for high lift configurations in a matter of a few days – instead of weeks or months. This solution was obtained using AFLR3 and CFD++ No, it’s not perfect yet, by a long shot. But if we are to do any investigations of low speed stability and control issues we must first be able to compute the basic wing/body characteristics.
Solution adaptive techniques, like the one shown here, will help to mitigate the ever-increasing need for larger and larger computers. This will allow grid to be placed only where it is needed and minimize it everywhere else.
Two modeling efforts are shown here. In the bottom left we have used CFD to model the effect of changing Reynolds number. The elevator on the left, operating at atmospheric wind tunnel conditions, is separated. As the Reynolds number is increased toward flight the separation is seen to disappear, thus increasing the effectiveness of the elevator.
The other three figures (top left, top right, bottom right) show the effect of modeling the wing- mounted vortex generators. Roughly 23 million grid points were used in the block structured code to model the 777 with 16 VG’s Here are two videos comparing the wing with and without vortex generators. Notice how much longer the wing stays attached when the VGs are present.
We have discovered that in order to more accurately predict the effectiveness of spoilers we must model the wing mounted VGs ahead of them. Here the right wing has the spoilers down, the let wing has them deployed.
Here you can see the affect of Reynolds number on the flow field ahead of the spoilers.
Here you can see the lift curve, drag polar, and pitch characteristics. Clearly the CFD has done a good job of modeling the airplane.
A comparison of wing pressures with CFD calculations shows stunning results.
Here you can see the section Cls and span load. The circle represents the airplane fuselage – imagine it coming out of the page. The right wing is carrying its normal load – the spoilers are deployed on the right hand wing. The left wing shows the dramatic lift loss due to the spoilers.
Here you can see the effect of increasing Reynolds number of the aft loading, pitch characteristics, and lift at zero angle of attack.
So how do we go about doing all this work? Well, to be in COMSAC you should have to bring something to the table. In my view that means airplane configurations, wind tunnel data, flight test data and the people to do the analysis work. This is the “cost of admission” – oh yeah, you have to be willing to share it with the other team members.
I don’t think we need a great many people – just the right ones. And they need access to best computing resources that we can make available. If NASA could provide the computing then industry could provide the people.
Security is a huge issue. Can we create an environment like a classified program where the participants see everything and the outside world sees nothing? It will make it easier to open up if we’re protected from interested, “prying” eyes.
NASA – can you make this happen?
I have secured the permission of Boeing management to allow the release of 777 geometry, wind tunnel, flight test and CFD results within the COMSAC arena – provided the proper security measures are in place. We need protection from Airbus, Embraer, and others.
Thank you for your time. I am looking forward to working with each of you as we work to make the COMSAC idea a reality. I believe there is much to be gained for engaging in this activity and will work diligently to make it happen. I hope you will, too.
Excellent agreement up to 20 degrees. Pitch-up predicted. Max lift over estimated by 5 %.
Discrepancy between CFD and wind tunnel test increases with angle of attack. CFD provides flow physics insight to help size and locate control surfaces.
The loss in pitch stability due to removal of a distorted tail cone was under predicted using CFD.
CFD provides flow physics insight for individual tail fins.
Testing in the stall angle-of attack region must be done using small increments in angle of attack or sideslip (one degree or less) or the true stall characteristics may not be identified In the stall region and beyond expect aerodynamic nonlinearities from flow separation due to shock waves, vortex breakdown, and wake flow.
Aerodynamic hysteresis effects can occur in sideslip data as well as in pitch data. Are there limitations in current CFD to predict hysteresis?
Wind tunnel accurately predicted non-zero yawing moment caused by asymmetric vortices, but what success rate do we have with CFD?
Final modifications made to the F-5F.
The Shark nose eliminated the forebody vortex asymmetry.
Significantly improved directional stability to high angles. Change to LEX delayed wing stall.
More elliptical forebody shape produced stable restoring moment.
Good agreement between CFD result and wind tunnel up to 25 degrees. Good agreement between CFD result and simulation based on flight testing up to 20 degrees.
This presentation will discuss Computational Fluid Dynamics as it is used for tactical aircraft and weapons development. Primary emphasis will be products designed, developed and built in St.
Louis reflecting the presenter’s background and experience, though it is believed that similar issues will be found wherever tactical vehicles are developed.
Over the past several years, Boeing has used Computational Fluid Dynamics extensively for a variety of development and analysis tasks. In configuration development, CFD is used to direct design effort for both internal and external flow to maximize performance. Wind Tunnel distortion effects are accurately estimated using CFD. Sting and distortion testing, if it is conducted at all, is pushed of to a later part of the program. Blended configurations with embedded engines, like most of the configurations we work with, have extensive aero propulsion interaction and CFD has proven a valuable method for investigating and quantifying these interactions. Again, both internal and external flow must be modeled.
CFD has proven valuable in modeling flow around sensors, pods, stores, weapons, and weapons bays for a variety of uses including stability and control, sensor performance, and structural loads. CFD is a valuable tool for investigating weapon separation issues, estimating stability and control of a small vehicle embedded in the flowfield of a much larger air vehicle.
Aerodynamic investigations might include control power issues, separated flow investigations, estimating effects of external changes, and so on. All of these are very amenable to CFD analysis.
There is constant pressure from program management, both customer and company, to shorten development time and reduce costs. Wind tunnel testing is both costly and time consuming and may prove inaccurate due to geometry inaccuracies and Reynolds Number scaling effects that are hard to predict. Yet it is critical to fully investigate a configuration, eliminating flaws prior to design freeze because it is terribly expensive to change anything later in the program. Once the design is frozen, design loads are required to begin detail design. In a highly maneuverable vehicle maneuvering inertia loads are often as high or higher than air loads. Therefore, a fairly detailed concept for the flight control system, necessitating a detailed and accurate aero database, are required to develop credible loads. CFD should play an increasing role in this development.
Accuracy is important because of the expense of redesign late in the program.
The presenter anticipates others will speak of need for analysis at cruise conditions and for low speed, high lift configurations. That is important for tactical vehicle too, so chalk up another “yes” vote for these capabilities. However, many critical design points and most of our problem areas involve massive separation on the air vehicle. We need to predict these flow characteristics.
For instance, both F-15 and F-18 aircraft have design points at or near max sustained load factor at which the wing shows significant separation. Max instantaneous turn rate, at maximum lift, is also a significant parameter in close in combat. An important point for UCAV is nose down pitch capability at high angle of attack. Weapons typically turn at very high g’s and have significant separation. A max range JDAM trajectory puts it at maximum lift for most of the flight. All of these conditions are legitimate design points and will require that accurate CFD solutions be available rapidly during the design phase through operational phase of these programs.
Damping derivatives at all angles of attack are important. Low angle of attack and high speed often set the control system capabilities. Cross wind landing in the high lift configuration depends on reasonable estimates of damping in all three axes. Departure and spin resistance as well as spin recovery estimation depends upon both forces and moments as well as damping coefficients at very high angles of attack and at high rates.
While tactical vehicles generally are designed with very strong structures, the also fly at very high speeds. Flexibility effects at high dynamic pressure severely reduces control power. We need to be able to estimate these effects far better than we can today. In particular, if we are to make use of the flexibility, as we are in the Advanced Aeroelastic Wing program, it is imperative that we be able to combine CFD with structural analysis codes to predict and then to design in favorable aeroelastic behavior.
Weapons influence on the parent airframe and that airframe influence on weapon separation are also extremely important for tactical vehicles. After all, the main job of the aircraft is to get the warheads and sensors out to the battle where they can do their job. Safe and effective separation cannot be overemphasized and it should be designed in from the beginning of the program.
Much has been said over the years and probably en this symposium about the first four items on this lest. I Probably can’t add anything to the discussion that has not been said.
Time dependent solutions are possible and will become more necessary as they get more possible with increased speed of solutions. There are many applications waiting, dynamic lift effects, transient internal and external flow phenomena, and weapon separation to name but a few.
Finally, verification and validation of CFD codes and applications is extremely important. The codes themselves should be verified and validated, but perhaps it is even more important to verify and validate the application of the code. Improper preparation, faulty assumptions, and schedule pressure cause perfectly good codes to give the wrong answer with great precision. As an industry we cannot afford to be led astray by our trusted tools.
Getting the vertical tail size correct on the first try has been a challenge for over 50 years. Most of the Century Series fighters suffered from roll coupling brought on by a combination of high rates of roll and insufficient stability to counter the resultant buildup in alpha and beta.
Enlarging the vertical tails was the most common and successful solution.
Early tests of the “Have Blue” stealth prototype with its inward canted vertical tails proved unsatisfactory, so outward canted tails were selected for the F-117A. Even so, on the first flight of the F-117A (6/18/81), directional stability and control were both found to be far less than predicted, so the tails were increased in size by 50%. Dick Abrams, head of flight test at Lockheed, stated that “the same mistake wasn’t made on the YF-22”. On the YF-22, several methods were used to predict the required vertical tail size and the largest value was used with an additional safety factor included. No problems were encountered during flight test and the tail size for the proposed production configuration was actually decreased in size from the prototype.
In the early 1970’s, the F-15 and YF-17 suffered from reduced control effectiveness due to aeroelastic effects. In the case of the YF-17, the effect had been predicted but not the magnitude, for the F-15, the effect had not been predicted.
Early flights of the YF-17 showed unsatisfactory roll performance at high dynamic pressure conditions, a problem that was traced to excessive wing flexibility which reduced aileron effectiveness. Aileron and differential tail authority were increased but both were hinge moment limited so the final roll performance was still less than desired. The wing of the F-18 was made significantly stiffer to avoid a similar problem.
Maximum load factor attained in high-g pullups of the F-15 was found to be less than predicted at transonic speeds. In addition, the short period damping was found to be underpredicted although the short period frequency matched predictions. An aeroelastic analysis showed tha t aft fuselage bending reduced the tail effectiveness while a dynamic time lag between the fuselage and build-up in lift of the tail reduced the short period damping. Due to its abundance of control authority, neither problem degraded operational effectiveness.
During departure tests of the F-16, deep-stall trim conditions were encountered at + and – 55 degrees angle of attack. Although the possibility of a deep-stall had been considered, none was expected at the c.g. locations where the flights were conducted. Analysis of the data indicated a major pitching moment discrepancy between the wind tunnel predictions and flight. High angle of attack tests had been conducted in four facilities. A detailed study of the results from these and other tests indicated that Reynolds number effects, model support interference, and the engine nozzle position were the major contributors to the discrepancy. This study was conducted in the early 1980’s and CFD was not used as a diagnostic tool.
A wind tunnel test of the F-15 STOL and Maneuver Technology Demonstrator (SMTD) in ground effect was conducted at the McDonnell Low Speed Wind Tunnel. A fixed ground board with no boundary layer removal system was used. Cold jet thrust reversers were used. The results showed a large plume induced pitchup which increased with sideslip. The control laws were modified to prevent a tail strike during landing. A later test with a model was conducted at the Langley Vortex Research facility. This test indicated virtually no plume induced pitching moment change. Excessive nose down moments were apparent in early flight tests that almost bottomed the nose gear. Analysis of the flight test data indicated a plume induced nose down moment, different from both tunnel tests but closer to the Langley results. The flight control laws were modified and future tests were completed without incident. At the time of this problem (early 1990’s) CFD was too immature to tackle this type of analysis.
Almost 30,000 hours of wind tunnel testing for aerodynamics/S&C were conducted on the Space Shuttle. The vast majority of this data is for Mach<8. At higher speeds, the limited data available were corrected for viscous interaction and aeroelastic effects analytically. No corrections were made for real gas effects. Early flight showed large discrepancies in the predicted trim settings of the body flap and elevons. Analysis of the data indicated an error in predicted center of pressure of approximately 1% body length (2% m.a.c.) for M>10. Extensive calculations using CFD were conducted to study the discrepancy. These indicated that the primary contributor was real gas effects, with Mach number effects and viscous effects also playing a role.
The F/A-22 and V-22 both suffered from tail buffet caused by unsteady vortex effects emanating from the LEX (F/A-22) and forebody (V-22). In both cases, extensive flight tests and CFD analyses were conducted. Actively controlled rudders were briefly considered as solutions in both cases. For the F/A-22, the problem was solved with an internal structural modification. For the V-22, a large fixed strake was added above each door.
The Orbital Sciences Pegasus booster was designed without wind tunnel tests. Engineering level codes and a limited amount of CFD (see AIAA paper 91-0190) were used for aerodynamics and stability and control. Good agreement between predictions and flight test were found (see AIAA paper 93-0520). A larger vehicle, Pegasus XL, was designed with a longer fuselage, higher mounted wing and modified tails. The first two flights of the XL version (June 94, June 95) ended in failure after the booster had to be destroyed. For the first time, wind tunnel tests were conducted, along with CFD analyses. These showed that lateral stability was not well predicted for XL version. The flight control system was modified and successful flights resumed in March 1996.
Except for the F-16 deep stall, all of the flight test problems discussed in this presentation occurred in the heart of the flight envelope. This is also true for other problems discussed in this conference such as the F-18 “wing drop”.
Resolving differences in wind tunnel results can be very important. This is definitely an area where CFD can contribute.
Typical S&C analyses done by AFRL or ASC fall in three general categories. Examples are shown which occurred over the past 12 months. In many of these cases, short time lines were involved.
CFD was used to a limited extent for the first two items only (DC-10, B-1B). In all other cases, engineering level codes or wind tunnel results were used. In many cases, use of CFD would be overkill. AFRL is somewhat atypical in that most programs involving any S&C analysis are at the early conceptual or preliminary design stage. Most analysis of fielded systems is done within ASC.
Engineering codes are defined here as Datcom-type semi-empirical codes, vortex lattice codes and inviscid 3-D panel codes. These are still the standard in both industry and government for generation of S&C data bases. There have been no major theoretical developments for this class of codes for the past decade. Most improvements have come in the form of graphical user interfaces and including the codes within larger design synthesis tools. While this latter “improvement” definitely increases productivity, it also increases the chance for GIGO as these tools become more “black box” in nature.
While engineering codes typically give good results at low angles of attack for many parameters, others such as pitching moment at zero angle of attack are very difficult to predict. Accurate values of Cmo are also difficult to get from a wind tunnel. Many of the yawing moment parameters are also difficult to predict (primarily for vertical tailless vehicles). In addition, these codes are of little or no use for separation based control devices or aero-propulsive interference effects. Ground effect are also difficult to predict.
Most CFD calculations within AFRL and ASC are geared towards problems requiring highly detailed flowfields. Problems shown here are: optical distortion of the airborne laser due to the forebody flow of the aircraft; hose oscillation behind the KC-135R multi-point refueling system pod; and plume impingement from a Stinger missile on the tail of a Predator UAV. Lift and drag calculations are also done using CFD. Problems shown here are: laminar flow investigation on the Global Hawk; transonic performance on the ALCM; and an X-45A computation that was done in concert with a wind tunnel program.
CFD is rarely used for S&C calculations within AFRL or ASC. For the DC-10 WASP problem discussed previously, contractor CFD calculations were performed. For the B-1B WCMD problem, prior CFD calculations from AEDC for the B-1B with weapons bay doors open were used to generate a data base of local flow conditions to put into a store separation simulation.
One impediment to use of CFD for S&C on future programs is the trend away from large development programs (F-22/V-22/JSF) towards low budget demonstrators with potential production at a future date. These programs typically operate with tight time constraints and very low budgets.
The X-40A program is typical. It is a subsonic Space Maneuver Vehicle technology demonstrator that was airdropped from a helicopter and autonomously flown to the landing site.
It was completed in 36 months for less than $20M. The aero/S&C analysis consisted of an engineering code (APAS) and two wind tunnel entries, a configuration development test in a very small facility and a data base test on the final configuration in an 8x12 ft low speed facility.
No CFD was done at low speeds for aero or S&C (some CFD was done to assess hypersonic characteristics).
This chart shows the major impediments to CFD use in S&C at WPAFB. Turnaround time has always been a problem but has shown steady improvement with unstructured grids and advances in computational power. The ASC capability shown represents a complete configuration analysis using the Navier Stokes solver COBALT. Continually changing configurations during design also hampers the utility of CFD.
The second issue is somewhat of an organizational problem. The “CFD” branch within AFRL is an isolated entity which typically support projects that are externally funded. In a 1999 downsizing, AFRL management declared “stability and control is a solved problem” and abolished the stability and control group. CFD related S&C efforts (high angle of attack dynamic derivative computations) ceased and the eight remaining “S&C” engineers now reside within five different branches.
There are many problems that are difficult and/or very expensive to wind tunnel test, for which engineering level codes are completely inadequate. Two are shown here. These should be areas ripe for CFD.
Propeller effects are once again of interest. Slipstream effects for both STOL and VTOL configurations are both difficult to predict and expensive to measure. Although semi-span tests are efficient from a cost point of view, they only give half of the answer. CFD developments in the rotary wing community that should be of some use here. There has also been a loss of corporate memory both in government and industry regarding propellers (there was a Propeller Laboratory at WPAFB in the 1950’s).
Ground effects for both STOL and VTOL vehicles have always been difficult to predict.
Boundary layer removal has always been a key problem, moving ground belts are notoriously difficult. CFD work that has been done in this area with impinging jets has demonstrated that half-span analyses give incorrect results due to significant mixing and that turbulence modeling is critical.
Raytheon manufactures a range of business aircraft. These are sold under the Beechcraft and Hawker badges.
At the bottom end there are the piston-powered Bonanza and Baron. They are followed by the King Air products; C90A, 200, & 350 with 4, 6, & 8-passenger cabins respectively.
In the jet market, the base product is the Beech Premier 1 followed by the Hawker 400XP (previously the Beechjet), the Hawker 800XP, and finally the Hawker Horizon which is in flight test now.
Beech and Hawker also have a long tradition of producing special-mission aircraft. A few examples are shown here.
The S&C engineer needs tools to rapidly and accurately address the technical issues in all phases of the airplane design and production. Some of the requirements are hard and fast such as those found in the certification regulations. Other issues demand the use of engineering judgment, often without adequate data for making the decisions. That is where CFD can provide the needed information and insight into the physics of the flow situation.
There is a wide range of CFD tools available to an aerodynamicist. The aerodynamics engineer should be familiar with the features and limitations of each of these tools in order to make the best decision of how to proceed.
Understanding the flow physics of a configuration is of paramount importance in solving the problems on a project. Is it an inviscid problem? Is it a viscous problem? What level of CFD analysis is appropriate to generate the needed answer?
Ideally, the CFD modeling needs to be done by a member of the project team so that there is good communication. In addition, it is important that the CFD analysis be timely. The time to respond can easily make or break the usefulness of CFD to the project.
What are the generic areas of interest that repeatedly appear in the various projects?
Empirical methods and closed form solutions are useful for linear coefficients, but the interesting problems often involve non-linear coefficients.
It is too costly to evaluate a range of configurations in the wind tunnel during development.
Using CAD and modern gridding software, it is possible to evaluate many configuration changes for a wide range of flight conditions by using multiple CFD codes before the loft lines for the first wind-tunnel model are set.
Rate, or rotary, derivatives are also difficult to estimate and even harder to measure in a wind tunnel. Accurate rate derivatives are needed for the simulations which are becoming more and more a part of the S&C toolbox.
There are many different elements that can influence the handling qualities and performance of an airplane once the basic configuration has been defined. It would be useful to have proven CFD techniques to investigate these elements and help with their design and implementation.
A completely new tail isn’t always the best answer for the financial success of the project.
Sometimes adding a few devices will do the trick.
A larger vertical tail adds weight. Small appendages can provide the extra area. These taillets actually reduce some of the critical loads.
Separated flows due to high adverse pressure gradients are hard to analyze and lead to many undesirable flight characteristics. Mach induced separations can lead to control surface or flap buzz. The effects of ice accretions are becoming increasingly importance as regulations are tightening up. Highly swept wings are becoming more common and require extra care in the design.
Subtle leading edge strips are used to control the stall (and yes those are VSAERO-generated streamlines with the local pressure distribution superimposed by the changing color of the stripes).
This useful pre-flight cup holder also provides very tame stall characteristics and excellent spin prevention and recovery. The design method was “cut and try”.
Vortilons are powerful flow control devices. They are normally added during developmental flight testing.
No, that’s not a manufacturing glitch. This small leading-edge discontinuity is intentional and generates a lower stall speed. Not all vortex devices are big and ugly. ( If anything, maybe it should have been slightly bigger so that there would have been fewer comments about how manufacturing must have goofed up.)
These VG’s were used to eliminate an uncommanded, high-altitude, long-period, pitch oscillation. Probably could do it with fewer, but that would have required more testing.
Would an NS calculation now allow us to check for this before flight?
Control systems are no longer limited to a single surface. Using spoilers in conjunction with ailerons is common and almost necessary given the current certification regulations. Multiple spoiler panels are used for safety and can also double as speed brakes. Spoiler effectiveness with flaps deflection in always an interesting study where CFD methods are heavily relied upon.
Predicting control surface hinge moments is very difficult. Wind tunnel tests can sometimes lead to incorrect conclusions due to limitations of Reynolds number and instrumentation sensitivity.
Re-design is sometimes required late in the program. Accurate CFD tools could be very valuable.
This is a very small wedge in front of an aileron, but has a very large effect on the hinge moment. Sometimes it just doesn’t take much.
Rudder lock can sometimes be remedied by employing flow tripping devices. This stall bar on the rudder “sucks” the rudder back from a possible over boosted condition.
In meeting the minimum control speed in a multi-engine airplane, the rudder must provide large control forces for relatively small pilot effort. A bulge on the trailing edge can help balance the control force as a function of sideslip angle as opposed to that due to surface deflection. This provides freedom from rudder lock while keeping pedal forces low for the engine out condition.
Turboprop airplanes require consideration of slipstream effects. They are often also capable of very high angles of attack due to the ability of the airplane to “hang on the prop” at high power settings and low airspeeds. While powered wind-tunnel tests are common for these aircraft, the dynamic pressure and Reynolds number for these tests are very low. The use of validated CFD methods for these flight conditions would be of great benefit.
Manufacturing always struggles to build airplanes exactly the way we design them. Tolerances are exceeded, parts are dropped and damaged, and repairs are made. The aerodynamicist must decide whether to approve or disapprove these out-of-tolerance conditions. Usually there is little or no data to support the engineering decision. CFD could help.
There are other disciplines on the project team that also depend on S&C results.
Simulations are used during development, but also used through the life of the product for training.
Loads are calculated for a conditions throughout the entire operating envelope of the aircraft. In order to accurately trim the plane or fly a specific maneuver, the S&C coefficients need to accurate. Some loads may be pilot-effort limited. Without good hinge moment estimates, extra conservatism will needed to insure a safe design.
Here are some basic considerations to develop a CFD model.
Can the model be simplified? Do I have to model the entire aircraft or will a smaller model suffice? In fact, how small can I make it in order to answer the question reliably and quickly?
Match the physics of the problem with the CFD tool to be used: - Is there a feature of the configuration that will require special care or a special technique to model? What level of CFD is required.
- I may still need transonic capability to obtain accurate leading-edge suction at stall.
- Is the flow attached or separated? If separated, is it from the trailing edge or is it a leading- edge vortex? The CFD engineer should use this to help select the method of analysis.
Unless the engineer is familiar with the CFD tool, the result will be neither quick nor calibrated.
When the problem focuses on the Aircraft, we are interested in gross effects.
When the problem is at the configuration element level, we need to have the appropriate detail in our CFD models.
Here is an example of an Aircraft-Focused modeling from many years ago. This type of modeling is still appropriate for many cases and is used regularly in loads analysis. Although the results on the next page are inviscid, it is also easy to add viscosity to these models now.
These plots were used to show the correlation of analysis with experiment for the aircraft loads.
This simple model provided an excellent match to the tightly-controlled, wind-tunnel model and good agreement to the 85% Proof-of-Concept (POC) Demonstrator during a steady 2-g turn.
(See Applied Computational Aerodynamics, Ed. Press Henne, Progress in Astronautics and Aeronautics, Volume 125 Published by AIAA, 1990, Chapter 16) Typical lift and pitching moment behavior are shown here for a business-jet configuration. The solid circles denote estimated Clmax levels using the trailing-edge, pressure-difference rule.
Note that for the flaps 30 case, the experimental data matches an effective flap deflection that is 3 degrees less than the geometric angle for the model as designed. This tends to indicate that the bracket tolerance and aeroelastics of the flap installation combine to reduce the effective flap deflection.
Medium and small-size business aircraft typically have unpowered control surfaces. This example shows how CFD codes can be used to study the complex interactions of various aileron shapes with the wing and cove. These results can be used to evaluate these shapes and choose ones for further evaluation in a wind tunnel or in flight.
This is an example from over 20 years ago that shows even simple-lifting surface methods have their place in the preliminary assessment of a configuration.
This simple modeling accurately matched the spanload and force and moment curves in the linear regions.
More recently (see AIAA paper # 98-2739) panel method geometries have been used to effectively simulate deflected spoilers with VSAERO and Tranair.
The buffet boundary for a business jet is an important limit to understand. This plot shows how the boundary-layer behavior from Tranair results can be used to make a useful estimate of this limit (see AIAA paper # 2000-0380).
The use of Navier-Stokes codes now could possibly make the estimate even better.
It is important for an aircraft program to be able to accurately estimate the behavior of the entire aircraft and also model details of the flow past specific configuration elements. There will likely need to be more than one CFD model to efficiently get the needed results.
A CFD plan for a program should utilize appropriate tools for each phase.
This presentation offers a perspective on the NASA COMSAC initiative from the vantage point of aerospace industry .
In this centennial year of Wright Brothers’ “controlled” powered flight, it is rather fitting that we hold a symposium devoted exclusively to the problem of more effectively predicting stability and control characteristics of flight vehicles!
Although the symposium is limited in scope to only computational methods, resulting improvements in our ability to computationally predict S&C characteristics can have far- reaching implications.
The presentation starts out by highlighting the rationale and urgency for advancing the state of the art in computational methods for S&C.
A couple of examples are included to illustrate the capabilities and deficiencies of the current suite of methods.
This is followed by a brief discussion of the challenges for the COMSAC initiative.
A summary chart concludes this presentation.
This chart outlines the primary motivation behind the need to advance the state of the art in computational methods for S&C.
The aerospace industry is deeply engaged in transitioning to a Simulation Based Acquisition strategy that the Department of Defense has adopted. Lockheed Martin is committed to implementing this strategy to meet customer expectations.
If successful, substantial benefits will accrue in quality and affordability of future flight vehicles for meeting national needs. However, realizing the benefits of this strategy depends on our ability to generate data in a timely and affordable manner to meet the demands of credible modeling and simulations.
A wide variety of data is required to support the Modeling and Simulation needs of a modern flight vehicle development effort.
Stability & Control data is absolutely crucial because it has significant influence over decisions about numerous aspects of flight vehicle development, some of which are listed in this chart.
Computational methods are the key enablers for meeting the demands of S&C data.
Since heretofore wind tunnels have been the primary means of generating S&C data for flight vehicle design, why not further improve the testing capabilities? Why should we look for computational methods as an option? This chart addresses these questions.
Although wind tunnels offer a proven and mature capability, they are deficient in supporting the demands of a M&S environment for successfully implementing the SBA strategy.
Reynolds number scaling is a fundamental limitation that plagues almost all wind-tunnel testing.
Accurately predicting full-scale model characteristics using the measured S&C characteristics of a sub-scale model continues to be major challenge. This limitation, in principle, can be overcome by computational methods.
This chart summarizes the nearly four decades of focused development in producing ever more capable computational methods. The available methods can be generally grouped in four categories. As one moves up from the linear potential methods (Level I) to the Navier-Stokes methods (Level IV), there is a significant increase in capabilities resulting from the use of a more complete model of physics. However, the increased capability is generally accompanied by the penalties of much longer turnaround times and higher computer and labor resources.
Consequently, all levels of methods are not equally effective for supporting the M&S demands.
One way to assess the effectiveness of computational methods is to consider Effectiveness as a product of Quality and Acceptance factors (as shown on the right hand side of this chart).
Accuracy and credibility of computational results are the primary Quality factors, and timeliness and affordability are the key Acceptance factors.
The next three charts illustrate the effectiveness of the linear potential and Euler methods that are presently in widespread use.
The linear potential methods used by Lockheed Martin fall in the category of vortex-lattice methods (using planar surface to represent geometry) and panel methods (using quadrilateral patches to approximate the actual geometric shape).
These methods are capable of generating S&C data including force and moment derivatives.
Space shuttle data generated using our panel code, QUADPAN, is shown on the right hand side—excerpted from a SAE paper published almost two decades ago.
Due to the simplified nature of their flow physics model, the linear potential methods are not valid for analyzing nonlinear aerodynamic effects associated with shocks (transonic Mach numbers) and vortical flows (high angles of attack).
These methods afford rapid turnaround and require relatively low levels of labor and computer resources. Therefore, their Acceptance factors are high but the overall effectiveness is low except for limited regions of the flight envelope where the flows do not contain shocks or regions of separated flow.
As the Euler methods began to mature by the late 1980s, Lockheed Martin performed several in- house and NASA-sponsored studies during the 1990s to assess their capabilities and deficiencies for preliminary design applications. Part of the effort was devoted to examining the control effects. An extensive study was conducted for the ICE configuration using the adaptive Cartesian-grid SPLITFLOW code. The aim was to assess the code’s ability to predict aerodynamic characteristics due to the deployment of leading-edge flaps as well as trailing-edge elevons and spoilers.
In general, the code captured the significant changes in flow characteristics and the trends of the force and moment increments. This chart illustrates the correlation of force and moment o increments for the 60 deflected spoiler case. Predictions for some angles of attack agree well with data while others do not. The side force and lateral-directional coefficients for this asymmetric deflection case show good correlation with data trends, except for the highest angle of attack of 25 degrees. One of the key benefits of computational methods is their ability to provide details of the off-body flow interaction which might have contributed to unanticipated changes in control effectiveness. Computational methods can provide advance warning of potential problems.
This chart shows results for the attached flow case for symmetrically deflected trailing-edge elevon on the ICE vehicle. The normal force (C ), axial force (C ) and pitching moment (C ) N A m o increments due to deflection match quite well with the measured increments except at 18 angle of attack, where the code overpredicts the normal force and nose-down moment. On further investigation, it was found that for some angles of attack such as this, the solution had difficulty in converging and the force and moment predictions settled in at levels that did not correlate well with data. At other angles of attack both above and below the troublesome one, the solution proceeded much better in terms of convergence and predictions correlated well with test data.
Based on the results of the various studies, it could be concluded that unstructured-grid Euler methods can be applied to predict control effects to support preliminary design of combat vehicles with sharp, swept leading edges. For configuration trade studies in early design stages, Euler methods are efficient given the substantial improvement in turnaround time and cost over viscous Navier-Stokes analysis by a factor of 3 to 5. However, the effectiveness levels are far from being satisfactory, and efforts must continue to render Euler and Navier-Stokes methods fully effective for preliminary design applications in an IPPD environment.
NASA’s COMSAC initiative offers the right approach at the right time for advancing our capabilities in predicting S&C data using computational methods.
Ongoing advances in computers and algorithms may allow sufficiently rapid turnaround to produce a complete database in a timely manner.
Such a capability is essential to supporting the Modeling & Simulation needs of the SBA strategy that DoD and aerospace industry have adopted for future flight vehicle development.
However, one of the principal challenges facing the COMSAC suite of codes is to guarantee that the predictions faithfully represent reality. The traditional approach of validating computational results with wind-tunnel data defeats the purpose—it adds time and cost! The COMSAC initiative must provide methods and techniques with a high level of Effectiveness.
This chart further outlines the challenges associated with achieving high levels of effectiveness.
A few key Quality and Acceptance factors are listed here that must be simultaneously enhanced to reach the desired Effectiveness targets.
This chart lists several considerations for the COMSAC program plan to ensure that the program is able to deliver what it promises.
In this presentation, a rationale was provided for the urgent need to develop computational methods for S&C.
The current state of the art was briefly discussed, and challenges for the COMSAC initiative were highlighted.
In moving forward, careful planning is critically important for maximizing the return on investment and to deliver what is promised.
I will review the trends for wind tunnel testing since the Wright Brothers. Review data requirements for S&C. Show a few examples of CFD in challenging flow fields. Compare capabilities and time requirements for CFD and WT. And recommend the biggest payoff areas for applying CFD to S&C.
It would be nice to have a subscale model that could do everything that the full scale vehicle can do, except for Rn, but no model or tunnel can do all that. So we have to break up the terms into smaller pieces and test them individually and build them back up to represent the full scale aircraft at operating conditions. So we have models for basic, unpowered, static effects, then add inlet effects, jet effects, support system corrections, dynamic effects, etc. We measure component loads and hinge moments on yet a different model.
We have learned lessons from many generations of wind tunnel testing and the art has evolved to better represent the full scale aircraft. Many of these standard practices evolved from finding surprises in flight, and going back and figuring out why the surprises occurred. CFD doesn’t have this experience yet in S&C applications.
I found the graphic on the left at a couple of web sites. It is wind tunnel hours on a log scale vs time since the Wright brothers. It indicates that WT test hours are growing exponentially. I made the plot on the right on a linear scale to illustrate the trend. But it is based on incomplete data and includes some special cases.
When I add more data points and focus on the range that doesn’t include space flight, it shows that the trend is not exponential, but rather has a broad range, even for aircraft in the category of supersonic fighters.
It would be interesting to talk more detail about some of the examples, but we don’t have time.
A typical program today might take about 15,000 hours of wind tunnel time for one supersonic fighter aircraft.
This table is to compare the data requirements between performance and S&C.
Some aero data is easy to predict, others are challenging. The most challenging is combinations of challenging conditions, such as Here is an example of CFD predictions at Mach 0.9 for low AOA lift and pitching moment.
Inviscid CFD showed good trends, but an offset. Viscous CFD got closer to the correct absolute answer.
The next step would be to refine the grid and possibly the turbulence models to get a better answer.
I think that CFD can match wind tunnel data at low AOA given time and given the WT data.
But without the WT data, you can’t be sure you have the right answer.
Once you calibrate the CFD to the WT data, you then have a set of parameters for the specific CFD tool that can be used to look for incremental changes to the vehicle and to understand the flow physics.
I will show CFD to predict transonic characteristics at elevated AOA and beta. This is a particularly challenging condition because it includes a mixture of strong shocks, strong vortices, regions of separation, and unsteady flow. And the situation is even more complex because this flow also varies with Mach, aoa, beta, tef, ht, stores, probably rates, probably altitude due to aeroelasticity. This challenges both WT and CFD.
Our objective was to improve lateral stability.
We started off with 6M grid cells but did not match well enough. We double the cells and got better correlation, but still had opposite trends with TEF. We tripled the cells to 18M and got improved correlation but still not matched. Some coefficients were still substantially off, even for incremental effects.
So this illustrates my point that CFD can try various levers to improve correlation with WT and may get to a match, given time and given the WT data, but can’t be relied on to get the right answer without the WT data.
It is still worthwhile to do since it provides understanding of flow physics, and provides a tool to look at effects of small changes.
Here is an example of CFD at high AOA at low speed. Our objective was to find nose down pitching moment at 55 deg AOA. CFD showed forebody changes that could provide nose down Cm or -0.046. We tested several changes similar to the CFD recommended geometry. Some of them provided nose down at 55 aoa, but all of them also had nose up at lower AOA, which moved the critical point to a lower aoa and provided no benefit. If we had time to work this further, we might have been able to find a successful change, but time and money ran out.
My experience with CFD process is: So to summarize my examples, there are some flow conditions of great interest that are challenges to CFD.
But CFD is a tremendous tool. Future benefits to S&C could be as significant if we get to work using it, learning the capabilities, finding out how to get increased confidence. It has been used successfully for: Requirements vary by program phase.
This is my experience in the use of CFD and wind tunnel for performance and S&C by program phase. We mostly use the wind tunnel, but also use some predictive methods in all phases.
The biggest payoff for CFD would be in conceptual and preliminary design, where the most important configuration decisions are made.
To summarize the benefits of applying CFD to S&C.
Here is the trend of computing power and turnaround time. Moore’s Law illustrates the exponential growth of computing power, which is phenomenal. This has allowed much finer grids and greater number of iterations, as illustrated in lower graph. What we can do now in 2 days would have taken 600 days in 1988.
But the wind tunnel is still far faster. To do a 15,000 hour wind tunnel program takes about 1.7 years of wind tunnel time, but would take over 1200 years of CFD. So wind tunnel testing is still required for S&C databases.
CFD has great potential for helping in S&C, but we will have to work it now to make that happen.
My name is Lawrence Green. I work in the NASA Langley Multidisciplinary Optimization Branch. There, I sit both literally and figuratively between the high-fidelity CFD gurus and the S&C practitioners here at NASA Langley. The talk title is now somewhat different than was published in your program, but now reflects the intended focus of the talk. I'd like to thank my co-authors, Angela Spence and Patrick Murphy for their help, and the conference organizers for allowing me to speak today.
This is an outline of my presentation. First, I'll give some S&C background and then describe some CFD background. I'll state my research objectives for the past year. I'll then describe several CFD methods and results for static, steady rate, and dynamic S&C derivatives. Finally, I'll summarize and draw some conclusions from the work.
Aircraft designers, flight control designers, and the flight dynamics community in general must account for the variation of forces and moments under maneuvering conditions. Maneuvering F&M can be largely different than static F&M and may exhibit nonlinear unsteady behaviors. To help designers a large knowledge base has been developed using S&C parameters to characterize aircraft dynamics. Dynamic stability parameters (derivatives), in particular, provide information about the “stiffness” and “damping” of the dynamic system. This discussion will focus on issues associated with how the “damping” derivatives, that characterize F&M variation with respect to angular rates, are modeled and measured.
nd Aircraft equations of motion (EOM) begin with Newton’s 2 Law. Application of this law requires an inertial reference frame but for convenience the equations are translated to the rotating aircraft body-axis system. This produces the nonlinear inertial terms shown above.
Equations for translation, rotation, and kinematic relationships produce a system of nonlinear ordinary differential equations. These equations are written with several other assumptions to make the discussion more tractable: 1) Earth is inertial reference frame with no curvature; 2) Airplane is rigid with lateral symmetry; 3) Thrust acts along fuselage and through c.g.; 4) Still atmosphere, i.e., no winds or gusts; 5) Constant mass with no internal mass movements, constant inertia; 6) Body axis system is fixed to aircraft.
The Faero & Maero terms are the aerodynamic forces and moments acting on the aircraft and these must be defined in order to solve the system of ODEs. This step is where S&C issues for modeling and measuring the “damping” derivatives occur. Defining the aerodynamic model is a substantial step and a key area where CFD can make significant contributions to solving S&C problems.
To see how dynamic stability derivatives can arise, the conventional aero model for pitching moment is shown. The conventional approach to defining the aero F&M is to assume that these functions can be expanded in a linear series with constant coefficients. Nonlinear, high-order, frequency-dependent, or time-dependent terms are assumed to be zero. Additional simplifying assumptions that can be applied as appropriate are: 1) the aircraft is a rigid-body (no aero-elastic responses in structure or controls); 2) no sharp discontinuities in aero; and 3) no stochastic processes.
As an example under these assumptions, the non-dimensional pitching moment equation can be written as shown. The key S&C parameters for this discussion are the two derivatives with respect to pitch rate and angle-of-attack rate. These two angular rates can be quite different in flight.
A conventional technique, used for many years to estimate damping derivatives such as Cmq, is to perform 1-dof, planar, forced-oscillation (FO) tests. For this test the model is placed at various angles of attack in a wind tunnel and allowed to undergo forced sinusoidal oscillations at different frequencies and usually relatively small amplitudes. Oscillations are usually done about pitch, roll, and yaw axes. This method of testing produces the so-called “in-phase” and “out-of- phase” derivatives usually designated by a bar over the derivative (see equation). These coefficients are basically the Fourier coefficients obtained from harmonic analysis of the FO measurements.
A long history of S&C testing has produced many methods of testing to obtain various stability and control derivatives, however, the FO test is the primary method and virtually the only method generally available today. Past decisions made under the general belief that other test methods were not needed lead to very few facilities in the world today having advanced capabilities or previously available capabilities such as plunging to obtain angle of attack rate derivatives, directly.
Historically, the conventional aerodynamic model has been used very successfully for developing aircraft and for predicting flight responses in certain parts of the flight envelope.
However, nonlinear or unsteady behaviors can be observed in different flight regimes such as during transonic maneuvering, rapid maneuvers, or at high angle of attack. Flight predictions can quickly deteriorate at higher angles of attack and for cases where the assumptions previously stated are not satisfied. Also results can be very configuration dependent.
The first problem S&C engineers must address is that wind tunnel measurements do not support the conventional model assumption of constant coefficients and in particular constant damping derivatives. This is shown by the large frequency dependence of the in-phase and out-of-phase derivatives at high angles of attack. The second problem is that the conventional FO test technique for measuring damping produces a combination of derivatives rather than separate values, as required by the series expansion of the aero F&M in the conventional model. This occurs because the angle-of-attack rate and pitch rate are kinematically constrained to be equal in single dof FO testing. Both of these problems can be ameliorated with the use of CFD.
Impacts of poor models for design are clear, especially increased research, development, and certification costs. Most of the flows of interest to the S&C community involve nonlinear, unsteady, and/or separated flows.
CFD brings unique capabilities to the S&C table. The code user can choose from a variety of code and grid fidelities to solve their problems. Many codes offer the capabilities to simulate both conventional and advanced dynamic testing techniques, as well as augmenting and enhancing current testing techniques. In addition, several derivative extraction techniques can be brought to bear on the problems including the traditional method of finite differences, symbolic manipulation, and complex mathematics. I’ve chosen to use a technique called automatic differentiation, which was developed by mathematicians. The technique can provide both f forward and reverse modes, applicable to both conventional S&C needs and those of morphing vehicles. There are a variety of tools available. I’ve used the ADIFOR tool, developed by Rice University and Argonne National Labs, to extract exact derivatives from legacy FORTRAN codes via repeated application of the chain rule.
Some previous work in which this technique was used include the application of ADIFOR to the TLNS3D code in order to extract longitudinal derivatives with respect to Mach, alpha, Reynolds number, and geometric parameters. Later, as the technical monitor for a GWU Master’s student, we again applied ADIFOR to both the PMARC and CFL3D codes for the explicit purpose of computing static and steady rate S&C derivatives. With the PMARC code, we also performed a control placement effectiveness study for a morphing vehicle. In the CFL3D code, a steady state solution method for steady rate derivatives was implemented to improve the computational speed significantly over true time dependent calculations. I have also been involved in the nd development of techniques for the efficient calculation of 2 derivatives and uncertainty propagation which allow me to place uncertainty bounds on my computed forces and moments and their associated S&C derivatives.
The previous work demonstrated the potential for using CFD to compute S&C derivatives but only scratched the surface of what could be done. As a result of the previous work I have continued to advocate for NASA to apply CFD to S&C problems. An area that I have personally advocated for the inclusion within COMSAC is the investigation of various code fidelities ranging from digital DATCOM to DES, to assess and quantify the accuracy and computational resource requirements of the various code fidelities. My branch intends to develop semi- automated means to identify and use the right code and grid fidelities code for a given application, based upon user requirements. I have an interest in bringing bounded S&C analyses into the early design process. I also returned to this area of research to extend, develop, and demonstrate CFD methods for dynamic S&C derivatives within forced oscillation and pure rotary motions. I was also interested in demonstrating the computational separation of lumped dynamic S&C derivatives, which was proposed in our 1999 paper, but never developed or demonstrated.
This plot shows several examples of computed forces and moments from both the PMARC and CFL3D codes. The upper left had figure shows CN as a function of angle of attack in degrees.
The upper lright figure shows CS, the lower left figure shows Cm, and the lower right figure shows the rolling moment, Cl. The solid line with circles is the measured data, the dashed lines with pluses are the PMARC data, the dashed lines with diamonds are the CFL3D Euler data.
Convergence problems were encountered with the Euler solutions, leading us to begin to compute Navier-Stokes solutions, as shown in the next figure.
This figure shows the Navier Stokes computations for Cm as a function of angle of attack. Three flow regimes were observed: attached flow up to about 6 degrees alpha, vortical flow up to about 15 degrees alpha, and vortically bursting flows beyond 15 degrees alpha. Wind tunnel measurements were taken at 1 degree increments, whereas the computations were performed only at 5 degree increments, which unfortunately, misses some key features of the measured data. The comparisons for CN and CA were in good agreement across the angle of attack range.
As in the picture on the ASCOT slide at the start of this talk, looking downstream, the vehicle surface is colored with contours of the pressure coefficient. Vortical flow structures are shown over the wing upper surface. The comparisons of CN and CA with WT were deemed to be excellent or good and are not shown.
It should be noted that there were some unexplained effects observed in the raw wind tunnel data which suggest that the model may have been slightly mis-aligned and geometrically unsymmetric upon installation.
This is the same plot with the inclusion of ADIFOR-CFL3D calculations. The two green dash- dot lines are the Euler mode of ADIFOR-CFL3D. The diamonds are the medium grid and the stars are the fine grid. The fine grid viscous ADIFOR-CLF3D Spalart turbulence model results are shown with red dotted lines and triangles. You will note that at 7.5 deg alpha the fine grid Euler converged to a significantly different value that the medium grid. The grids are not converging in a second order fashion as the grids are sequenced. It is therefore believed that different flow physics are being modeled in the two grids. Because of this reservation, the Euler results will be removed for clarity from the plots of the other two longitudinal derivatives.
The left graph depicts the derivative of axial force coefficient with respect to angle of attack on the vertical axis. The right graph depicts the derivative of pitching moment coefficient with respect to angle of attack on the vertical axis. The turbulent Navier-Stokes solution had difficulty matching the pitching moment derivative exactly, but did show a similar shape to the data. Pitching moment is one of the more difficult quantities to predict with CFD, therefore, a derivative of pitching moment would be even more challenging.
To move on to the lateral derivatives.
The derivatives of side force coefficient with respect to angle of sideslip or beta is shown on the vertical axis. Only the inviscid ADIFOR-CLF3D Euler is shown because the viscous cases were performed with a half span model and therefore cannot show the lateral derivatives. The Euler code does not nail the derivatives exactly, but does detect the break in the data above 5 deg angle of attack.
The left graph depicts the derivative of rolling moment coefficient with respect to angle of sideslip on the vertical axis. The right graph depicts the derivative of yawing moment coefficient with respect to angle of sideslip on the vertical axis. The ADIFOR-CFL3D maybe seeing derivative values that are underestimated by the large step in the wind tunnel central differencing calculation or the asymmetry in the wind tunnel data.
This figure illustrates our attempts to compute roll rate derivatives and pressure increments to compare with rotary balance data with a Navier-Stokes code. The agreement is generally good, but the anomalous results for delta CM in the rotary balance comparison have not been unexplained.
The types of comparisons done, and the resulting accuracy to which derivatives were obtained are pointed out. Note that even “poor” comparisons were better than other previous linear aerodynamic approximations. Note the time required to obtain solutions and the memory required.
This figure illustrates the lift and moment coefficients time histories in response to an imposed small amplitude (5 degree) oscillation of the angle of attack, as measured in the 12 Foot Low Speed Wind Tunnel. The moment response, which is directly related to the surface pressure distribution, illustrates the rapidly changing characteristics of the vortical and bursting flows at 30.8 and 50.8 degrees mean value of alpha. PMARC time histories for the lift were similar, but the moment differed considerably, and required further study.
The traditional wind tunnel technique for the separation of lumped derivatives is to use both a curved flow wind tunnel to identify pitch rate effects and plunge motions to identify angle of attack effects, neither of which are common experimental techniques today. In 1999, the Park/Green paper proposed that CFD offers the potential to computationally separate lumped dynamic S&C derivatives, but the technique was neither developed, nor demonstrated, at that time. Recently, I have proposed three techniques to computationally separate lumped dynamic S&C derivatives. The first technique applies the usual data processing techniques for wind tunnel data to the CFD time histories. If static and pure rotary data are available, an algebraic manipulation can be used to extract force and moment derivatives with respect to alpha-dot and q-dot. The second and third derivative separation techniques involve the solution of two simultaneous equations for these same alpha-dot and q-dot derivatives.
This figure shows the typical so-called in-phase and out-of-phase lumped dynamic derivative response for the lift force during forced pitch oscillations at various k rates.
This figure shows the typical so-called in-phase and out-of-phase lumped dynamic derivative response for the pitching moment during forced pitch oscillations at various k rates.
This is an example of computational separation of lumped dynamic derivatives using separation technique 1. Measured data was processed as usual. Then algebraic manipulation was used to extract the q-dot derivatives. The derivative CL-q-dot is shown in the left hand figure for various k rates; the derivative CM-q-dot is shown in the right hand figure.
To summarize, conventional S&C modeling and experimental techniques are generally good for steady, low to moderate vehicle orientations, and perhaps mild separation, but have notable limitations in transonic, or fully separated flows. CFD offers capabilities complimentary to the conventional S&C techniques. Several methods for the application of CFD to S&C have been discussed, including the computation of forced oscillation motions, the computation of lumped dynamic S&C derivatives, and several computational separation techniques for use with lumped dynamic S&C derivatives. All of these steps are important to establish the credibility of CFD in the S&C arena. Additional detail on all these topics will be presented in the two pending papers which have been accepted for presentation at the Reno 2004 meetings.
Form Approved REPORT DOCUMENTATION PAGE OMB No. 0704-0188 The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number.
PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS.
1. REPORT DATE (DD-MM-YYYY) 2. REPORT TYPE 3. DATES COVERED (From - To) 04 - 2004 01- Conference Publication 4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER COMSAC: Computational Methods for Stability and Control 5b. GRANT NUMBER 5c. PROGRAM ELEMENT NUMBER 6. AUTHOR(S) 5d. PROJECT NUMBER Fremaux, C. Michael and Hall, Robert M. (Compilers) 5e. TASK NUMBER 5f. WORK UNIT NUMBER 23-762-45-AF 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION REPORT NUMBER NASA Langley Research Center Hampton, VA 23681-2199 L-18378A 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR'S ACRONYM(S) National Aeronautics and Space Administration NASA Washington, DC 20546-0001 11. SPONSOR/MONITOR'S REPORT NUMBER(S) NASA/CP-2004-213028/PT1 12. DISTRIBUTION/AVAILABILITY STATEMENT Unclassified - Unlimited Subject Category 08 Availability: NASA CASI (301) 621-0390 Distribution: Nonstandard 13. SUPPLEMENTARY NOTES An electronic version can be found at http://techreports.larc.nasa.gov/ltrs/ or http://ntrs.nasa.gov 14. ABSTRACT The unprecedented advances being made in computational fluid dynamic (CFD) technology have demonstrated the powerful capabilities of codes in applications to civil and military aircraft. Used in conjunction with wind-tunnel and flight investigations, many codes are now routinely used by designers in diverse applications such as aerodynamic performance predictions and propulsion integration. Typically, these codes are most reliable for attached, steady, and predominantly turbulent flows. As a result of increasing reliability and confidence in CFD, wind-tunnel testing for some new configurations has been substantially reduced in key areas, such as wing trade studies for mission performance guarantees. Interest is now growing in the application of computational methods to other critical design challenges. One of the most important disciplinary elements for civil and military aircraft is prediction of stability and control characteristics. CFD offers the potential for significantly increasing the basic understanding, prediction, and control of flow phenomena associated with requirements for satisfactory aircraft handling characteristics.
15. SUBJECT TERMS Stability and Control; Computational Fluid Dynamics; Stability Derivatives; Aerodynamics 19a. NAME OF RESPONSIBLE PERSON 18. NUMBER 17. LIMITATION OF 16. SECURITY CLASSIFICATION OF: OF ABSTRACT STI Help Desk (email: help@sti.nasa.gov) a. REPORT c. THIS PAGE b. ABSTRACT PAGES 19b. TELEPHONE NUMBER (Include area code) (301) 621-0390 U U U UU 3 85 Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std. Z39.18