cfl3d and fun3d analysis of hiliftpw-2 workshop casesfun3d.larc.nasa.gov/papers/fun3d_hlw.pdfcfl3d...

1http://fun3d.larc.nasa.govHiLiftPW-2 June 21-22, 2013

CFL3D and FUN3D Analysis of

HiLiftPW-2 Workshop Cases

Elizabeth M. Lee-Rausch & Christopher L. Rumsey

NASA Langley Research Center

2nd AIAA CFD High Lift Prediction Workshop (HiLiftPW-2)

Sponsored by the Applied Aerodynamics TC

San Diego, CA

June 22-23, 2013


CFD Code Descriptions

• CFL3D Structured-Grid Code

– Parallel 3D compressible cell-centered finite-volume RANS

– Full Navier-Stokes equations

– Upwind Roe flux-difference splitting for inviscid fluxes

– Spalart-Allmaras and SST turbulence models, fully-turbulent

– SST-Gamma-Re-Theta (GRET), transition

– Implicit local time-stepping using AF relaxation for linear system

– Restart computations from lower angles of attack

• FUN3D Unstructured-Grid Code

– Parallel 3D compressible finite-volume RANS for mixed-element meshes

– Full Navier-Stokes equations-node centered

– UMUSCL 0.5 scheme (Averaging + Upwind) for inviscid fluxes

– Spalart-Allmaras turbulence model, fully-turbulent

– Implicit local time-stepping using multi-color point Gauss-Seidel relaxation for

linear system

– Restart computations from lower angles of attack

3http://fun3d.larc.nasa.gov

Cases and Grids

HiLiftPW-2 June 21-22, 2013

Struct.

1 to 1 – A

(Boeing)

Unst.

Nodal - D

(UWYO/Cessna)

FUN3DSA

(Mixed)

CFL3D

(Case 1 only)

SA

SST (incomplete)

GRET (incomplete)

• Required Case 1 and Case 2

• Optional Case 3 not done


Coarse Medium Fine

Struct.

1-to-1 A

Boeing

11M 34M 105M

Mixed ver.

Unst. D

UWYO/Cessna

10M 31M 76M

Comparison of Grid Sizes


Case 1 – FUN3D Grid Convergence, SA


Lift Drag Pitching Moment

ALPHA

CL

0 10 20 301

1.5

2

2.5

3

3.5

002.1 case1 FT coarse no tracks

002.1 case1 FT medium no tracks002.1 case1 FT fine no tracksExp. ETW

CD

CL

0.1 0.2 0.3 0.4 0.5 0.61

1.5

2

2.5

3

3.5

CM, CM25

CL

-0.6 -0.4 -0.2 01

1.5

2

2.5

3

3.5


Case 1 – CFL3D Grid Convergence, SA



ALPHA

CL

0 10 20 301

1.5

2

2.5

3

3.5

002.2 case1 FT coarse no tracks002.2 case1 FT medium no tracks

002.2 case1 FT fine no tracksExp. ETW

CD

CL

0.1 0.2 0.3 0.4 0.5 0.61

1.5

2

2.5

3

3.5

CM, CM25

CL

-0.6 -0.4 -0.2 01

1.5

2

2.5

3

3.5


Case 1 – CFL3D results, other models



Medium Grids


Case 1 – Grid Convergence, SA




Case 1 – Grid Convergence, AoA=7



Note: Grid D (fine) was discovered to have

incorrect min spacing – so not of similar

family to C and M.







family to C and M.


Case2 – Reynolds Number Study



ALPHA

CL

0 10 20 301

1.5

2

2.5

3

3.5

002.1 case2a (low Re) FT medium w tracks002.1 case2b (hi Re) FT medium w tracksExp. LSTW low Re

Exp. ETW hi Re

CD

CL

0.1 0.2 0.3 0.4 0.5 0.61

1.5

2

2.5

3

3.5

CM, CM25

CL

-0.6 -0.4 -0.2 01

1.5

2

2.5

3

3.5


Case 1 Surface Restricted Streamlines


AOA 16deg

Medium Grids

FUN3D CFL3D


FUN3D Surface Restricted Streamlines


AOA 16deg

Medium Grid

Case 1 Case 2b




AOA 20deg

Medium Grids

FUN3D CFL3D




AOA 20deg

Medium Grid

Case 1 Case 2b




AOA 18.5deg

Medium Grid

Exp. Oil FlowCase 2a


FUN3D Pressure Distributions


Case 1 Medium Grid

Eta = 0.45

Slat Main Flap

x

Cp

1640 1660 1680 1700 1720

-9

-8

-7

-6

-5

-4

-3

-2

-1

0

1

2

x

Cp

1400 1500 1600

-9

-8

-7

-6

-5

-4

-3

-2

-1

0

1

2

x

Cp

1360 1370 1380 1390 1400

-15

-10

-5

0

Black - 21 deg

Red - 7degBlue - 16degGreen - 18.5deg




Case 2b

Eta = 0.45

Slat Main Flap

x

Cp

1640 1660 1680 1700 1720

-9

-8

-7

-6

-5

-4

-3

-2

-1

0

1

2

x

Cp

1400 1500 1600

-9

-8

-7

-6

-5

-4

-3

-2

-1

0

1

2

x

Cp

1360 1370 1380 1390 1400

-15

-10

-5

0

Black - 21 deg





Case 1 Medium Grid

Eta = 0.75

Slat Main Flap

x

Cp

1840 1860 1880

-9

-8

-7

-6

-5

-4

-3

-2

-1

0

1

2

x

Cp

1650 1700 1750 1800 1850

-9

-8

-7

-6

-5

-4

-3

-2

-1

0

1

2

x

Cp

1630 1640 1650 1660 1670

-15

-10

-5

0

Black - 21 deg





Case 2b

Eta = 0.75

Slat Main Flap

x

Cp

1840 1860 1880

-9

-8

-7

-6

-5

-4

-3

-2

-1

0

1

2

x

Cp

1650 1700 1750 1800 1850

-9

-8

-7

-6

-5

-4

-3

-2

-1

0

1

2

x

Cp

1630 1640 1650 1660 1670

-15

-10

-5

0

Black - 21 deg



Velocity contours over flap, AoA=7


Plane 2


Velocity profiles over flap, AoA=7


Note: exp data at low Re, shown for reference only

Plane 2

Case 1 Medium/Fine Grids


Effect of Re on velocity profiles, AoA=7


Plane 1-window B Plane 1-window D Plane 2-window D

Plane 2-window E Plane 3-window E Plane 3-window E


Summary (1)

• Grid convergence study (no brackets)¯ Structured grid and unstructured grid results with same

turbulence model (SA) compare reasonably well in forces/moments

¯ There is flap separation at all AoAs

¯ FUN3D on unstructured D grid shows more separation than CFL3D on structured A grid

¯ Maximum lift coefficient is over-predicted

¯ At higher alphas, there are larger differences between codes (esp. in the moment coefficient)

¯ CFL3D on structured A grid series captures wake profiles better than FUN3D on unstructured D grid series



Summary (2)

• Bracket study¯ Including brackets improves results (forces, moments,

surface pressures)

¯ General trends captured

¯ CL,max not predicted well due to early main wing separation

¯ Slat wakes not resolved on D grids

• Reynolds number study¯ General trends captured, but CL,max not predicted well

¯ Velocity profiles compare with experiment only qualitatively



N-2/3

CL

0 1E-05 2E-05

2.6

2.8

3

3.2

3.4

16 deg

Black - Workshop Grids

Red - 4th Adapted Grids

N-2/3

CD

0 1E-05 2E-05

0.26

0.28

0.3

0.32

0.34

16 deg

N-2/3

CM

0 1E-05 2E-05-0.7

-0.6

-0.5

-0.4

16 deg

Next Steps• Adjoint-based adaption : output functional L/D

– Off-body adaption

– C1 coarse and medium grids (SA model)

– AOA 16deg

– Four adaption cycles (4 additional adjoint and flow solutions)


11M16M31M

62M


C1 Medium Grid Adaption η=70%


Original Grid – 31M nodes




Adapted 4 – 62M nodes


C1 Medium Grid Adaption Flap Cp


Orig. Grid – 31M nodes : Adapted 4 – 62M nodes

η=72% η=82% η=89%

x

cp

,c

p_

ET

W_R

UN

23

8

1800 1820 1840 1860 1880

-4

-2

0

2

Symbols - Exp.Black - C1 Med. GridRed - Adapted Grid

x

cp

,c

p_

ET

W_R

UN

23

8

1880 1900 1920

-4

-2

0

2


x

cp

,c

p_

ET

W_R

UN

23

8

1920 1940 1960

-4

-2

0

2



Effect of Turb. Model on Flap Cp


SADM – SA model with Dacles-Mariani correction

η=89%

x

cp

,cp

_E

TW

_R

UN

23

8

1920 1940 1960

-4

-2

0

2

Red - 4th Adapted Grid SA

Blue - 4th Adapted Grid SADM


Backup Slides



FUN3D Residual Convergence


Iteration

R_

1,R

_6

C_

L

C_

D

0 20000 40000 6000010

-10

10-8

10-6

10-4

10-2

100

102

104

1

1.5

2

2.5

3

3.5

4

4.5

5

0

0.1

0.2

0.3

0.4

0.5

HiLiftPW2 DLR-F11 Case1 Med. SA 22.4 deg

7deg

16deg18.5deg

20deg 21deg 22.4deg

C1 Medium Grid Polar

R6

R1


ITER

RE

SID

M-1

CL

CD

0 50000 100000 15000010

-10

10-8

10-6

10-4

10-2

100

102

104

1

1.5

2

2.5

3

3.5

4

4.5

5

0

0.1

0.2

0.3

0.4

0.5

7deg

16deg

18.5deg20deg

21deg22.4deg

CFL3D Residual Convergence


C1 Medium Grid Polar


Objectives– Assess the numerical prediction capability (meshing, numerics, turbulence modeling,

high-performance computing requirements, etc.) of current-generation CFD

technology/codes for swept, medium-to-high-aspect ratio wings for landing/take-off

(high-lift) configurations.

– Develop practical modeling guidelines for CFD prediction of high-lift flow fields.

– Determine the elements of high-lift flow physics that are critical for modeling to enable

the development of more accurate prediction methods and tools.

– Enhance CFD prediction capability for practical high-lift aerodynamic design and

optimization.

Focus– The HiLiftPW-2 test cases are based on the European High Lift Programme

(EUROLIFT) DLR F11 high-lift configuration.

• Representative of a commercial wide-body twin-jet high-lift configuration

• 3 elements : full-span slat, main and full-span flap

• Landing configuration : slat 26.5 degs and flap 32 degs

– A significant amount of high-quality surface and flow field data are available, including

data for an assessment of Reynolds number scale effects.

• Force and moment (ETW and B-LSWT)

• Surface pressures (ETW and B-LSWT)

• PIV and oil flow (B-LSWT)

HiLiftPW-2 - Rehearsal Background


HiLiftPW-2 – DLR F11


PIV PlanesPressure Tap Stations

η=17.6%

η=70%

η=87.4%

η=44.9%___

η=75.1% ___


• Case 1 : Grid convergence study (required)

– Mach = 0.175, free-air

– Angles-of-attack to be computed (deg) = 7 and 16 (18.5, 20, 21 and 22.4)

– ReMAC = 15.1 M, fully turbulent

– DLR F11 "Config 2" - Slat 26.5 deg, Flap 32 deg (Wing/Body/HL system + SOB Flap Seal)

– Coarse, medium, fine and extra-fine

• Case 2a and 2b: Reynolds number study (required)

– Mach = 0.175, free-air

– Angles-of-attack to be computed (deg) = 0, 7, 12, 16, 18.5, 19, 20 and 21

– ReMAC = 1.35M (Case2a) and 15.1 M(Case2b), fully turbulent

– Medium density mesh from Grid Convergence Study

– DLR F11 "Config 4" - Slat 26.5 deg, Flap 32 deg (Config 2 + Slat Tracks and Flap Track Fairings)

• Case 2c (OPTIONAL) - Low Reynolds Number Condition with Transition

• Case 3 (OPTIONAL) - Full Configuration Study

• Case 4 (OPTIONAL) - Turbulence Model Grid-Convergence Verification Study

HiLiftPW-2 - Rehearsal Background




AOA 7deg

Medium Grids

FUN3D CFL3D




AOA 7deg

Medium Grid

Case 1 Case 2b







family to C and M.

cfl3d and fun3d analysis of hiliftpw-2 workshop casesfun3d.larc.nasa.gov/papers/fun3d_hlw.pdfcfl3d...

Documents