Document
www.nasa.gov June 5-9, 2017 Denver, CO
Andy Broeren
Frédéric Moens
NASA Glenn Research Center ONERA—The French Aerospace Lab
Brian Woodard and Jeff Diebold
University of Illinois at Urbana-Champaign ASE Conference
of an 8.9% Scale Semispan Swept
th
Wing for Assessment of Icing Effects
Low-Reynolds Number Aerodynamics
AIAA 9 National Aeronautics and Space Administration www.nasa.gov
Outline
Introduction Objectives and Approach Experimental Methodology Model Mounting Evaluation Clean Wing Aerodynamics CFD Simulation Comparison Ice Roughness Simulation Comparison Summary Acknowledgements
• • • • • • • • •
National Aeronautics and Space Administration www.nasa.gov
Introduction
Development and use of 3D icing simulation tools. Lack of ice accretion and aerodynamic data for large- scale, swept wing geometries. Aerodynamic understanding important for evaluating efficacy of 3D icing simulation tools. Multi-faceted research effort called SUNSET II.
• • • •
National Aeronautics and Space Administration www.nasa.gov
)
12×10
) aerodynamic test
≈
Re
2.4×10
≤
Re
Introduction
High fidelity and low fidelity
−
Low-Reynolds number ( campaigns. The artificial ice shapes were developed based upon a series of ice-accretion tests in the NASA Icing Research Tunnel. Higher-Reynolds number (up to aerodynamic test campaigns.
Aerodynamic understanding important for evaluating efficacy of 3D icing simulation tools. • • • National Aeronautics and Space Administration www.nasa.gov
Objectives and Approach
Perform experimental and computational assessment of clean-wing aerodynamics, model installation and simulation of small ice roughness. Perform aerodynamic testing with 8.9% scale semispan swept wing model of CRM65 at low-Reynolds number. Perform 3D RANS simulations of clean wing fully turbulent and with free transition. Parametric study of model-mounting configurations. Investigate techniques for simulating small ice roughness.
Objectives • Approach • • • •
National Aeronautics and Space Administration www.nasa.gov
Common Research Model (CRM)
Commercial transport class configuration. Contemporary transonic supercritical wing design. Publically available and otherwise unrestricted for world-wide distribution. A 65% scale CRM was selected as the full-scale, reference swept-wing geometry for this research. CRM65 size airplane is comparable to Boeing 757.
• • • • •
National Aeronautics and Space Administration www.nasa.gov .
P C .
M C ,
Experimental Methodology
D C , L C − Aerodynamic testing performed at Wichita State University Beech Wind Tunnel. Test section size 7-ft x 10-ft. 8.9%-scale semispan model of CRM65 geometry. Reynolds numbers = 0.8, 1.6 and 2.4×10 Corresponding Mach numbers = 0.09, 0.18 and 0.27. Measure integrated aerodynamic performance with force balance Measure surface pressure - Mini-tuft and surface-oil flow visualization.
• • • • • • • • National Aeronautics and Space Administration www.nasa.gov
Model Mounting Configurations
National Aeronautics and Space Administration 0.04 www.nasa.gov 0.03 D C 0.02 0.01 Circular Splitter, Circular Shroud Circular Splitter, Streamlined Shroud Rectangular Splitter, Streamlined Shroud No Splitter, No Shroud 0.00 0.0 0.2 0.4 0.6 0.8 -0.4 -0.2 L C M C -0.04 0.00 0.04 0.08 0.12 0.16 0.20 0.24 0.28 = 0.27.
M , 8 10 12 14 16 18 (deg.)
= 2.4×10
Model Mounting Configurations
Effect of model mounting on aerodynamic performance at Re -6 -4 -2 • 0.0 0.2 0.4 0.6 0.8 1.0 1.2 -0.4 -0.2 L C National Aeronautics and Space Administration www.nasa.gov = 13.2 deg., α = 0.44, 28 Circular Splitter, Circular Shroud Circular Splitter, Streamlined Shroud Rectangular Splitter, Streamlined Shroud No Splitter, No Shroud y/b (in.)
x = 0.27.
M , = 2.4×10
Model Mounting Configurations 21
Surface pressure distribution at Re • 0.0 1.0 -7.0 -6.0 -5.0 -4.0 -3.0 -2.0 -1.0 p
C
National Aeronautics and Space Administration 0.04 www.nasa.gov = 0.09 = 0.18 = 0.27 M M M 0.03 , , , 6 6 6 10 10 10 D C 0.02 = 0.8 = 1.6 = 2.4 Re Re Re 0.01 0.00 0.0 0.2 0.4 0.6 0.8 -0.4 -0.2 L C M C -0.04 0.00 0.04 0.08 0.12 0.16 0.20 0.24 0.28 8 10 12 14 16 18 (deg.)
Clean Model Aerodynamics
Effect of Reynolds and Mach number on clean wing configuration.
-6 -4 -2 • 0.0 0.2 0.4 0.6 0.8 1.0 1.2 -0.4 -0.2 L C National Aeronautics and Space Administration p 60.0 C -0.2 -0.6 -1.0 -1.4 -1.8 -2.2 -2.6 -3.0 50.0 www.nasa.gov 40.0 (in.)
30.0 x = 0.18.
20.0 M , = 11.1 deg.
10.0 0.0 = 1.6×10 0.0 10.0 20.0 30.0 40.0 50.0 60.0 y (in.)
Re p C 60.0 -0.2 -0.6 -1.0 -1.4 -1.8 -2.2 -2.6 -3.0 50.0 40.0 (in.)
30.0 x
Clean Model Aerodynamics
20.0 = 9.0 deg.
10.0 Surface pressure distribution at 0.0 • 0.0 10.0 20.0 30.0 40.0 50.0 60.0 y (in.)
National Aeronautics and Space Administration www.nasa.gov = 11.1 deg., and α = 0.18.
M ,
Clean Model Aerodynamics
= 1.6×10 Mini-tuft and surface-oil flow visualization at Re • National Aeronautics and Space Administration www.nasa.gov = 0.18.
M , p 60.0 C -0.2 -0.6 -1.0 -1.4 -1.8 -2.2 -2.6 -3.0 50.0 = 1.6×10 40.0 Re (in.)
30.0 x 20.0
Clean Model Aerodynamics
= 13.6 deg.
10.0 = 13.6 deg., and Surface-pressure distribution and mini-tuft flow visualization at α 0.0 0.0 • 10.0 20.0 30.0 40.0 50.0 60.0 y (in.)
National Aeronautics and Space Administration www.nasa.gov = 0.18.
M , p C 60.0 -0.2 -0.6 -1.0 -1.4 -1.8 -2.2 -2.6 -3.0 50.0 = 1.6×10 40.0 Re (in.)
30.0 x 20.0
Clean Model Aerodynamics
= 14.1 deg.
10.0 = 14.1 deg., and Surface-pressure distribution and mini-tuft flow visualization at α 0.0 0.0 • 10.0 20.0 30.0 40.0 50.0 60.0 y (in.)
National Aeronautics and Space Administration www.nasa.gov = 0.18.
M , = 1.6×10 Re
Clean Model Aerodynamics
Surface-pressure distribution animation at • National Aeronautics and Space Administration www.nasa.gov cells cells cells = 8) corresponding to T N
CFD Simulation Methodology
Test-section floor included as symmetry plane. Wing: ~9.4×10 Splitter: ~6.5×10 Collar grid: ~0.65×10 − − − − CFD simulation included the wing and splitter plate, no shroud. Chimera overset grid based upon ONERA methodology. ONERA elsA solver for 3D compressible RANS equations. One equation Spalart-Allmaras turbulence model. Free-transition model criteria based upon free-stream turbulence intensity of 0.11% ( WSU wind tunnel.
• • • • • National Aeronautics and Space Administration 0.15 www.nasa.gov 0.10 M C 0.05 0.00 -0.05 0.0 0.2 0.4 0.6 0.8 1.0 L = 0.18.
-0.4 -0.2 C = 8) M 0.15 , N 0.10 D C = 1.6×10 0.05 WSU Experiment CFD Turbulent CFD Transition ( Re 0.00 0.0 0.2 0.4 0.6 0.8 1.0 L -0.4 -0.2 C (deg.)
CFD Simulation Comparison
eo g -4 Clean wing performance at • L 0.0 0.2 0.4 0.6 0.8 1.0 -0.4 -0.2 C National Aeronautics and Space Administration www.nasa.gov = 0 deg.
α = 0.18.
M , = 1.6×10
CFD Simulation Comparison
Re Surface oil flow visualization and transition location at and • National Aeronautics and Space Administration 1.0 , www.nasa.gov 0.8 = 0.84 0.6 = 1.6×10 x/c y/b Re 0.4 Local 0.2 0.0 = 13.1 deg. and 0.0 1.0 -6.0 -5.0 -4.0 -3.0 -2.0 -1.0 p α C 1.0 =8) N 0.8 = 0.28 0.6 x/c WSU Experiment CFD Turbulent CFD Transition ( y/b 0.4 Local
CFD Simulation Comparison
0.2 = 0.18.
0.0 Surface pressure distribution at M • 0.0 1.0 -6.0 -5.0 -4.0 -3.0 -2.0 -1.0 p C National Aeronautics and Space Administration www.nasa.gov
Roughness Simulation Methodology
Full-span artificial ice shapes were bolted to the wing leading edge. Artificial ice shapes were made using rapid-prototype manufacturing (RPM). Small ice roughness was simulated with regular pattern of hemispheres in the RPM shape. Aerodynamic results were compared to carborundum grit of equivalent size applied to the clean leading edge.
• • • • National Aeronautics and Space Administration 0.04 www.nasa.gov 0.03 D C 0.02 = 0.010 in. roughness = 0.18.
k 0.01 M , = 0.010 in. Grit roughness with gaps = 0.010 in. Grit roughness no gaps Clean RPM, k k 0.00 0.0 0.2 0.4 0.6 0.8 L -0.4 -0.2 C = 1.6×10 M C Re -0.04 0.00 0.04 0.08 0.12 0.16 0.20 0.24 0.28 8 10 12 14 16 18 (deg.)
Roughness Simulation Comparison
Aerodynamic performance at -6 -4 -2 0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 • L -0.4 -0.2 C National Aeronautics and Space Administration www.nasa.gov = 0.09, 0.18 and M and
Summary
= 0.8, 1.6 and 2.4×10 conditions, the flow over the outboard sections of the Re M and Re Circular splitter plate and streamlined shroud selected for further work. For all wing separated as the wing stalled with the inboard sections near the root maintaining attached flow. This behavior was captured for 3D RANS CFD simulations with free transition model, with opposite results for fully turbulent simulations. Size of RPM-based hemispherical roughness limited to height = 0.010 inches due to manufacturing limitations.
– – – – Experimental and computational study of 8.9% scale CRM65 semispan wing at 0.27. Four different model mounting configurations were investigated. A detailed study of clean wing aerodynamics was performed: Artificial ice roughness simulated with hemispherical patterns in RPM shapes generated aerodynamic effects equivalent to similar size carborundum grit roughness.
• • • • National Aeronautics and Space Administration www.nasa.gov :
Acknowledgements
.
Special thanks also to the WSU Beech Wind Tunnel staff and to NASA—Advanced Air Transport Technology Project FAA ONERA Boeing University of Illinois University of Virginia University of Washington Sponsor Organizations • • • Supporting Organizations: • • • • *** William Yoshida at Univ. of Illinois for developing surface pressure contour plots National Aeronautics and Space Administration