
[Sponsors] 
Issue with turbulent viscosity contour  SU2 v7.0.6 

LinkBack  Thread Tools  Search this Thread  Display Modes 
May 31, 2021, 05:23 
Issue with turbulent viscosity contour  SU2 v7.0.6

#1 
New Member
Join Date: May 2021
Posts: 5
Rep Power: 5 
Hi,
I am currently computing the flow on the CERAS aircraft using SU2. The flow is computed in RANS (SA model) at a Mach number of 0.78 and a target lift coefficient CL = 0,6. I have achieved a first computation using a “coarse” mesh close to the wall as illustrated in the figure Coarse_Mesh.png and considering SU v7.0.6. I have defined 2 boundary conditions : “Farfield” and “Heat Flux wall”. The numerical parameters are very close to the turbulent flat plate tutorial except for the spatial discretization scheme, which is JamesonSchmidtTurkel (JST) instead of ROE. The computation did converge and loose 4 order of magnitude on the rho residual. However, I was surprised by the conical shape of the viscous contour in the symmetry plane, as illustrated in the figure (Viscosity_v706.png). The contour shown is the ratio of Eddy_Viscosity over Laminar_Viscosity. After this first computation, I also achieved a simulation considering SU2 v6.2.0 with the same mesh. I did plot the viscous ratio contour in the symmetry plane, and this time it appears to be confined close to the wall, as illustrated in figure (Viscosity_v620.png). However, the convergence is much slower and the residuals do not decrease as much as with SU2 v7.0.6. Finally, I did increase the number of cells close to the wall, as illustrated in figure (Fine_Mesh.png), and I ran the simulation with SU2 v7.0.6 until 9 orders of magnitude in the residuals are lost. This time the viscous ratio contour is confined as expected to the wall and does not develop following the conical shape shown before (Viscosity_v706_FineMesh.png). However, the value of the viscous ratio in the boundary layer seems to be very high (several thousands) while I was expecting several hundred at most. With this additional computation, I was able to compare the profiles of the boundary layer between v6.2.0 and v7.0.6 along the fuselage (illustrated in figures BL_X_5d0m.png to BL_X_25d0m.png in the following message). As it is shown, the turbulent content is much higher in the profiles taken from v7.6.0 while the boundary layer appears to be “closer to a laminar profile”. (I also ran the turbulent flat plate with both versions of SU2 and plotted the profiles of the boundary layer, which where exactly the same) Could you help me with these simulations. Maybe I do not have a look at the right quantities or the numerical parameters are not suitable for this simulation? (I did also enclosed the configuration file I did use for the v7.6.0 simulation) Thank you in advance! 

May 31, 2021, 05:26 

#2 
New Member
Join Date: May 2021
Posts: 5
Rep Power: 5 
Here are the boundary layer profiles I could not attach the the preceding message.
I also copy here the configuration file I used for the computation with SU2 v7.0.6. Thank you in advance  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % % % SU2 configuration file % % Case description: Turbulent flow over flat plate with zero pressure gradient % % Author: Thomas D. Economon % % Institution: Stanford University % % Date: 2011.11.10 % % File Version 5.0.0 "Raven" % % % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %  DIRECT, ADJOINT, AND LINEARIZED PROBLEM DEFINITION % % % Physical governing equations (EULER, NAVIER_STOKES, % WAVE_EQUATION, HEAT_EQUATION, FEM_ELASTICITY, % POISSON_EQUATION) SOLVER= RANS % % If NavierStokes, kind of turbulent model (NONE, SA, SA_NEG, SST) KIND_TURB_MODEL= SA % % Mathematical problem (DIRECT, CONTINUOUS_ADJOINT) MATH_PROBLEM= DIRECT % % Restart solution (NO, YES) RESTART_SOL= YES %  COMPRESSIBLE AND INCOMPRESSIBLE FREESTREAM DEFINITION % % % Mach number (nondimensional, based on the freestream values) MACH_NUMBER= 0.78 % % Angle of attack (degrees) AOA= 0.0 % % Sideslip angle (degrees) SIDESLIP_ANGLE= 0.0 % % Freestream temperature (288.15 K by default) FREESTREAM_TEMPERATURE= 288.15 % % Reynolds number (nondimensional, based on the freestream values) REYNOLDS_NUMBER= 6.500000e+06 % % Reynolds length (in meters) REYNOLDS_LENGTH= 1.0 % %  REFERENCE VALUE DEFINITION % % % Reference origin for moment computation REF_ORIGIN_MOMENT_X = 0.00 REF_ORIGIN_MOMENT_Y = 0.00 REF_ORIGIN_MOMENT_Z = 0.00 % % Reference length for pitching, rolling, and yawing nondimensional moment REF_LENGTH= 34.07 % % Reference area for force coefficients (0 implies automatic calculation) REF_AREA= 122.41 %  CL DRIVER DEFINITION % % % Activate fixed lift mode (specify a CL instead of AoA, NO/YES) FIXED_CL_MODE= YES % % Target coefficient of lift for fixed lift mode (0.80 by default) TARGET_CL= 0.600000 % % Estimation of dCL/dAlpha (0.2 per degree by default) %DCL_DALPHA= 0.2 % % Maximum number of iterations between AoA updates UPDATE_AOA_ITER_LIMIT= 250 % % Number of iterations to evaluate dCL_dAlpha by using finite differences (500 by default) ITER_DCL_DALPHA= 200 %  BOUNDARY CONDITION DEFINITION % % % NavierStokes wall boundary marker(s) (NONE = no marker) MARKER_HEATFLUX= ( Wingbodytail, 0.0 ) % % Symmetry boundary marker(s) (NONE = no marker) MARKER_SYM= ( NONE ) % % Farfield boundary marker(s) (NONE = no marker) MARKER_FAR= ( Farfield ) % % Marker(s) of the surface to be plotted or designed MARKER_PLOTTING= ( Wingbodytail ) % % Marker(s) of the surface where the functional (Cd, Cl, etc.) will be evaluated MARKER_MONITORING= ( Wingbodytail ) %  COMMON PARAMETERS DEFINING THE NUMERICAL METHOD % % % Numerical method for spatial gradients (GREEN_GAUSS, LEAST_SQUARES, % WEIGHTED_LEAST_SQUARES) NUM_METHOD_GRAD= GREEN_GAUSS % % CourantFriedrichsLewy condition of the finest grid CFL_NUMBER= 10.0 % % Adaptive CFL number (NO, YES) CFL_ADAPT= NO % % Parameters of the adaptive CFL number (factor down, factor up, CFL min value, % CFL max value ) CFL_ADAPT_PARAM= ( 0.1, 2.0, 100.0, 1e5 ) % % RungeKutta alpha coefficients RK_ALPHA_COEFF= ( 0.66667, 0.66667, 1.000000 ) % % Number of total iterations ITER= 10000 %  SLOPE LIMITER DEFINITION % % % Coefficient for the limiter VENKAT_LIMITER_COEFF= 0.1 % % Coefficient for the sharp edges limiter ADJ_SHARP_LIMITER_COEFF= 3.0 % % Reference coefficient (sensitivity) for detecting sharp edges. REF_SHARP_EDGES= 3.0 % % Remove sharp edges from the sensitivity evaluation (NO, YES) SENS_REMOVE_SHARP= NO %  MULTIGRID PARAMETERS % % % MultiGrid Levels (0 = no multigrid) MGLEVEL= 3 % % Multigrid cycle (V_CYCLE, W_CYCLE, FULLMG_CYCLE) MGCYCLE= V_CYCLE % % Multigrid presmoothing level MG_PRE_SMOOTH= ( 1, 2, 3, 3 ) % % Multigrid postsmoothing level MG_POST_SMOOTH= ( 2, 2, 2, 2) % % Jacobi implicit smoothing of the correction MG_CORRECTION_SMOOTH= ( 0, 0, 0, 0 ) % % Damping factor for the residual restriction MG_DAMP_RESTRICTION= 0.8 % % Damping factor for the correction prolongation MG_DAMP_PROLONGATION= 0.8 %  FLOW NUMERICAL METHOD DEFINITION % % % Convective numerical method (JST, LAXFRIEDRICH, CUSP, ROE, AUSM, HLLC, % TURKEL_PREC, MSW) CONV_NUM_METHOD_FLOW= JST % % Monotonic Upwind Scheme for Conservation Laws (TVD) in the flow equations. % Required for 2nd order upwind schemes (NO, YES) MUSCL_FLOW= YES % % Slope limiter (NONE, VENKATAKRISHNAN, VENKATAKRISHNAN_WANG, % BARTH_JESPERSEN, VAN_ALBADA_EDGE) SLOPE_LIMITER_FLOW= NONE % % 2nd and 4th order artificial dissipation coefficients JST_SENSOR_COEFF= ( 0.5, 0.02 ) % % Time discretization (RUNGEKUTTA_EXPLICIT, EULER_IMPLICIT, EULER_EXPLICIT) TIME_DISCRE_FLOW= EULER_IMPLICIT %  TURBULENT NUMERICAL METHOD DEFINITION % % % Convective numerical method (SCALAR_UPWIND) CONV_NUM_METHOD_TURB= SCALAR_UPWIND % % Monotonic Upwind Scheme for Conservation Laws (TVD) in the turbulence equations. % Required for 2nd order upwind schemes (NO, YES) MUSCL_TURB= NO % % Slope limiter (VENKATAKRISHNAN, MINMOD) SLOPE_LIMITER_TURB= VENKATAKRISHNAN % % Time discretization (EULER_IMPLICIT) TIME_DISCRE_TURB= EULER_IMPLICIT %  CONVERGENCE PARAMETERS % % % Convergence criteria (CAUCHY, RESIDUAL) CONV_FIELD= RMS_DENSITY % % Min value of the residual (log10 of the residual) CONV_RESIDUAL_MINVAL= 14 % % Start convergence criteria at iteration number CONV_STARTITER= 10 % % Number of elements to apply the criteria CONV_CAUCHY_ELEMS= 100 % % Epsilon to control the series convergence CONV_CAUCHY_EPS= 1E6 % %  INPUT/OUTPUT INFORMATION % % % Mesh input file MESH_FILENAME= Test_raffinementV7.su2 % % Mesh input file format (SU2, CGNS, NETCDF_ASCII) MESH_FORMAT= SU2 % % Mesh output file MESH_OUT_FILENAME= pyCAPS_SU2.su2 % % Restart flow input file SOLUTION_FILENAME= solution_flow.dat % % Restart adjoint input file SOLUTION_ADJ_FILENAME= solution_adj.dat % % Output file format (PARAVIEW, TECPLOT, SLT) TABULAR_FORMAT= TECPLOT % % Output file convergence history (w/o extension) CONV_FILENAME= history % % Output file restart flow RESTART_FILENAME= restart_flow_pyCAPS_SU2.dat % % Output file restart adjoint RESTART_ADJ_FILENAME= restart_adj.dat % % Output file flow (w/o extension) variables VOLUME_FILENAME= flow_pyCAPS_SU2_Point5 % % Output file adjoint (w/o extension) variables VOLUME_ADJ_FILENAME= adjoint % % Output objective function gradient (using continuous adjoint) GRAD_OBJFUNC_FILENAME= of_grad.dat % % Output file surface flow coefficient (w/o extension) SURFACE_FILENAME= surface_flow_pyCAPS_SU2_Point5 % % Output file surface adjoint coefficient (w/o extension) SURFACE_ADJ_FILENAME= surface_adjoint % % Writing solution file frequency OUTPUT_WRT_FREQ= 1000 % % % Screen output SCREEN_OUTPUT= (INNER_ITER, WALL_TIME, RMS_DENSITY, RMS_NU_TILDE, LIFT, DRAG) HISTORY_OUTPUT=(ITER,REL_RMS_RES,RMS_RES, AERO_COEFF) 

May 31, 2021, 17:05 

#3 
Senior Member
Pedro Gomes
Join Date: Dec 2017
Posts: 466
Rep Power: 13 
Hi,
We seem to have an ongoing problem with JST at symmetry planes (but not other schemes) possibly because of a slight inconsistency between the discretization of internal fluxes and boundary fluxes. https://github.com/su2code/SU2/issues/1125 You are missing out on many fixes, improvements, and features by not using the most recent version of the code. 

June 1, 2021, 03:44 

#4 
New Member
Join Date: May 2021
Posts: 5
Rep Power: 5 
Hi,
Thank you for your answer, I did ask the IT support for an installation of the most recent version of SU2. But to clarify on my ongoing computations, I did compute the whole aircraft, not only the half (sorry, my statement was not clear in the above messages). Anyway, I will give a try to the computation with the most recent version of SU2. I did extract the quantities in the symetry plane but it is not a boundary condition. I did also try to run the simulation with the ROE scheme but it diverges really quickly, leading to CL and CD of 10**6. I hope it helps to understand my issue, anyway I come back to you when I am able to run the simulation with the latest version. Thank you again 

June 28, 2021, 09:11 

#5 
New Member
Join Date: May 2021
Posts: 5
Rep Power: 5 
Hi,
I am now able to run SU2 v7.1.1 simulations and I tried to achieve the simulation I presented before, consisting of a full aircraft configuration. In the following, the figures are extracted in the symetry plane during the postprocessing of the calculations. Firstly, I did achieve the simulation with the fine mesh I showed you before and considering the JST scheme. The visosity ratio (show in figure "Viscosity_Fine_JST.png") presents the same characteristics as before. The value of the ratio of viscosity is still very high (several thousands close to the boundary layer). Secondly, I did run the simulation on the coarse mesh I showed in the first message. However, with this mesh, the conical shape of the viscosity ratio appears again, as illustrated in "Viscosity_Coarse_JST.png". Finally, I did try to use the Roe scheme in combination with the coarse mesh. I was forced to initialize the simulation from the result obtained after a JST computation. But, as shown in the onvergence plot in figure "Convergence_Roe.png", the calculation diverges instead of converging and crashed afte 200 iterations. I also join after a copy of the configuration file for the Roe simulation. Could you help me with these calculations? I was hoping to find a viscosity contour close to the one obtained with SU2 v6.2.0 Thank you in advance %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % % % SU2 configuration file % % Case description: Turbulent flow over flat plate with zero pressure gradient % % Author: Thomas D. Economon % % Institution: Stanford University % % Date: 2011.11.10 % % File Version 5.0.0 "Raven" % % % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %  DIRECT, ADJOINT, AND LINEARIZED PROBLEM DEFINITION % % % Physical governing equations (EULER, NAVIER_STOKES, % WAVE_EQUATION, HEAT_EQUATION, FEM_ELASTICITY, % POISSON_EQUATION) SOLVER= RANS % % If NavierStokes, kind of turbulent model (NONE, SA, SA_NEG, SST) KIND_TURB_MODEL= SA % % Mathematical problem (DIRECT, CONTINUOUS_ADJOINT) MATH_PROBLEM= DIRECT % % Restart solution (NO, YES) RESTART_SOL= YES %  COMPRESSIBLE AND INCOMPRESSIBLE FREESTREAM DEFINITION % % % Mach number (nondimensional, based on the freestream values) MACH_NUMBER= 0.78 % % Angle of attack (degrees) AOA= 0.0 % % Sideslip angle (degrees) SIDESLIP_ANGLE= 0.0 % % Freestream temperature (288.15 K by default) FREESTREAM_TEMPERATURE= 288.15 % % Reynolds number (nondimensional, based on the freestream values) REYNOLDS_NUMBER= 6.500000e+06 % % Reynolds length (in meters) REYNOLDS_LENGTH= 1.0 % %  REFERENCE VALUE DEFINITION % % % Reference origin for moment computation REF_ORIGIN_MOMENT_X = 0.00 REF_ORIGIN_MOMENT_Y = 0.00 REF_ORIGIN_MOMENT_Z = 0.00 % % Reference length for pitching, rolling, and yawing nondimensional moment REF_LENGTH= 34.07 % % Reference area for force coefficients (0 implies automatic calculation) REF_AREA= 122.41 %  CL DRIVER DEFINITION % % % Activate fixed lift mode (specify a CL instead of AoA, NO/YES) FIXED_CL_MODE= YES % % Target coefficient of lift for fixed lift mode (0.80 by default) TARGET_CL= 0.600000 % % Estimation of dCL/dAlpha (0.2 per degree by default) %DCL_DALPHA= 0.2 % % Maximum number of iterations between AoA updates UPDATE_AOA_ITER_LIMIT= 250 % % Number of iterations to evaluate dCL_dAlpha by using finite differences (500 by default) ITER_DCL_DALPHA= 200 %  BOUNDARY CONDITION DEFINITION % % % NavierStokes wall boundary marker(s) (NONE = no marker) MARKER_HEATFLUX= ( 1, 0.0 ) % % Symmetry boundary marker(s) (NONE = no marker) MARKER_SYM= ( NONE ) % % Farfield boundary marker(s) (NONE = no marker) MARKER_FAR= ( 2 ) % % Marker(s) of the surface to be plotted or designed MARKER_PLOTTING= ( 1 ) % % Marker(s) of the surface where the functional (Cd, Cl, etc.) will be evaluated MARKER_MONITORING= ( 1 ) %  COMMON PARAMETERS DEFINING THE NUMERICAL METHOD % % % Numerical method for spatial gradients (GREEN_GAUSS, LEAST_SQUARES, % WEIGHTED_LEAST_SQUARES) NUM_METHOD_GRAD= GREEN_GAUSS % % CourantFriedrichsLewy condition of the finest grid CFL_NUMBER= 10.0 % % Adaptive CFL number (NO, YES) CFL_ADAPT= NO % % Parameters of the adaptive CFL number (factor down, factor up, CFL min value, % CFL max value ) CFL_ADAPT_PARAM= ( 0.1, 2.0, 100.0, 1e5 ) % % RungeKutta alpha coefficients RK_ALPHA_COEFF= ( 0.66667, 0.66667, 1.000000 ) % % Number of total iterations ITER= 10000 %  SLOPE LIMITER DEFINITION % % % Coefficient for the limiter VENKAT_LIMITER_COEFF= 0.1 % % Coefficient for the sharp edges limiter ADJ_SHARP_LIMITER_COEFF= 3.0 % % Reference coefficient (sensitivity) for detecting sharp edges. REF_SHARP_EDGES= 3.0 % % Remove sharp edges from the sensitivity evaluation (NO, YES) SENS_REMOVE_SHARP= NO %  MULTIGRID PARAMETERS % % % MultiGrid Levels (0 = no multigrid) MGLEVEL= 3 % % Multigrid cycle (V_CYCLE, W_CYCLE, FULLMG_CYCLE) MGCYCLE= V_CYCLE % % Multigrid presmoothing level MG_PRE_SMOOTH= ( 1, 2, 3, 3 ) % % Multigrid postsmoothing level MG_POST_SMOOTH= ( 2, 2, 2, 2) % % Jacobi implicit smoothing of the correction MG_CORRECTION_SMOOTH= ( 0, 0, 0, 0 ) % % Damping factor for the residual restriction MG_DAMP_RESTRICTION= 0.8 % % Damping factor for the correction prolongation MG_DAMP_PROLONGATION= 0.8 %  FLOW NUMERICAL METHOD DEFINITION % % % Convective numerical method (JST, LAXFRIEDRICH, CUSP, ROE, AUSM, HLLC, % TURKEL_PREC, MSW) CONV_NUM_METHOD_FLOW= ROE % % Monotonic Upwind Scheme for Conservation Laws (TVD) in the flow equations. % Required for 2nd order upwind schemes (NO, YES) MUSCL_FLOW= YES % % Slope limiter (NONE, VENKATAKRISHNAN, VENKATAKRISHNAN_WANG, % BARTH_JESPERSEN, VAN_ALBADA_EDGE) SLOPE_LIMITER_FLOW= NONE % % 2nd and 4th order artificial dissipation coefficients JST_SENSOR_COEFF= ( 0.5, 0.02 ) % % Time discretization (RUNGEKUTTA_EXPLICIT, EULER_IMPLICIT, EULER_EXPLICIT) TIME_DISCRE_FLOW= EULER_IMPLICIT %  TURBULENT NUMERICAL METHOD DEFINITION % % % Convective numerical method (SCALAR_UPWIND) CONV_NUM_METHOD_TURB= SCALAR_UPWIND % % Monotonic Upwind Scheme for Conservation Laws (TVD) in the turbulence equations. % Required for 2nd order upwind schemes (NO, YES) MUSCL_TURB= NO % % Slope limiter (VENKATAKRISHNAN, MINMOD) SLOPE_LIMITER_TURB= VENKATAKRISHNAN % % Time discretization (EULER_IMPLICIT) TIME_DISCRE_TURB= EULER_IMPLICIT %  CONVERGENCE PARAMETERS % % % Convergence criteria (CAUCHY, RESIDUAL) CONV_FIELD= RMS_DENSITY % % Min value of the residual (log10 of the residual) CONV_RESIDUAL_MINVAL= 14 % % Start convergence criteria at iteration number CONV_STARTITER= 10 % % Number of elements to apply the criteria CONV_CAUCHY_ELEMS= 100 % % Epsilon to control the series convergence CONV_CAUCHY_EPS= 1E6 % %  INPUT/OUTPUT INFORMATION % % % Mesh input file MESH_FILENAME= pyCAPS_SU2.su2 % % Mesh input file format (SU2, CGNS, NETCDF_ASCII) MESH_FORMAT= SU2 % % Mesh output file MESH_OUT_FILENAME= pyCAPS_SU2.su2 % % Restart flow input file SOLUTION_FILENAME= solution_flow.dat % % Restart adjoint input file SOLUTION_ADJ_FILENAME= solution_adj.dat % % Output file format (PARAVIEW, TECPLOT, SLT) TABULAR_FORMAT= TECPLOT % % Output file convergence history (w/o extension) CONV_FILENAME= history % % Output file restart flow RESTART_FILENAME= restart_flow_pyCAPS_SU2.dat % % Output file restart adjoint RESTART_ADJ_FILENAME= restart_adj.dat % % Output file flow (w/o extension) variables VOLUME_FILENAME= flow_pyCAPS_SU2_Point5 % % Output file adjoint (w/o extension) variables VOLUME_ADJ_FILENAME= adjoint % % Output objective function gradient (using continuous adjoint) GRAD_OBJFUNC_FILENAME= of_grad.dat % % Output file surface flow coefficient (w/o extension) SURFACE_FILENAME= surface_flow_pyCAPS_SU2_Point5 % % Output file surface adjoint coefficient (w/o extension) SURFACE_ADJ_FILENAME= surface_adjoint % % Writing solution file frequency OUTPUT_WRT_FREQ= 1000 % % % Screen output SCREEN_OUTPUT= (INNER_ITER, WALL_TIME, RMS_DENSITY, RMS_NU_TILDE, LIFT, DRAG) HISTORY_OUTPUT=(ITER,REL_RMS_RES,RMS_RES, AERO_COEFF) 

Thread Tools  Search this Thread 
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Problem with divergence  TDK  FLUENT  13  December 14, 2018 06:00 
Turbulent Viscosity Ratio...(I know, an old issue)  Freeman  FLUENT  7  March 2, 2009 16:28 
Turbulent Viscosity  Floating Point Error  Johnny B  FLUENT  1  November 26, 2003 08:42 
On limiting to turbulent viscosity ratio!  varghese  FLUENT  2  November 15, 2003 08:56 
Turbulent viscosity in a riser  ap  FLUENT  8  April 19, 2003 08:00 