Adding new roughness boundary conditions for SST #2573

marcosch27 · 2025-09-17T14:54:45Z

Proposed Changes

Hello everyone,

This PR is one of two contributions resulting from my Master’s Thesis.

In this PR, I’ve added new roughness boundary conditions for the SST turbulence model, based on the approach described in this paper by Aupoix.

The attached PDF includes documentation as well as validation results for a fully turbulent flat plate test case. If needed, I can add a validation test case to the SU2 suite.

Please let me know if any modifications are needed.

Related Work

PR Checklist

I am submitting my contribution to the develop branch.
My contribution generates no new compiler warnings (try with --warnlevel=3 when using meson).
My contribution is commented and consistent with SU2 style (https://su2code.github.io/docs_v7/Style-Guide/).
I used the pre-commit hook to prevent dirty commits and used pre-commit run --all to format old commits.
I have added a test case that demonstrates my contribution, if necessary.
I have updated appropriate documentation (Tutorials, Docs Page, config_template.cpp), if necessary.

bigfooted · 2025-09-17T20:36:19Z

Thanks for the contribution! Can you add the new config options to config_template.cfg and add the flat plate setup? You can add the .cfg file to this PR, and the mesh and a converged restart file to a PR for the TestCases repository.

pcarruscag

Add a testcase please, something of verification quality, are there known results from literature for flatplates?

pcarruscag · 2025-09-21T23:06:14Z

Common/include/option_structure.hpp

+  NONE,                 /*!< \brief No option / default. */
+  WILCOX1998,           /*!< \brief Wilcox 1998 boundary conditions for rough walls  / default. */


Both are default?

I will fix this issue. When surface roughness is used, the Wilcox boundary condition should be the default. To align with the current implementation, I will set WILCOX2006 as the default boundary condition.

pcarruscag · 2025-09-21T23:07:48Z

Common/include/option_structure.hpp

+  ROUGHSSTParsedOptions.wilcox2006 = IsPresent(ROUGHSST_OPTIONS::WILCOX2006);
+  ROUGHSSTParsedOptions.limiter_knopp = IsPresent(ROUGHSST_OPTIONS::LIMITER_KNOPP);
+  ROUGHSSTParsedOptions.limiter_aupoix = IsPresent(ROUGHSST_OPTIONS::LIMITER_AUPOIX);


you don't check for 1998? Is it valid if all of these are true?

pcarruscag · 2025-09-21T23:10:06Z

SU2_CFD/src/solvers/CTurbSSTSolver.cpp

+        /*--- Aupoix eddy viscosity limiter ---*/ 
+        else if (roughsstParsedOptions.limiter_aupoix) {
+
+          su2double k0Plus = ( 1.0 /sqrt( constants[6])) * tanh(((log((kPlus +EPS ) / 30.0) / log(10.0)) + 1.0 - 1.0*tanh( (kPlus + EPS) / 125.0))*tanh((kPlus + EPS) / 125.0));


why log / log(10) instead of log10(...)?

In the implementation, the current expression is retained to remain consistent with the formulation in the reference paper, even though it would be mathematically equivalent to using a base-10 logarithm.

Put a comment if you like, but this is log10, thanks.

pcarruscag · 2025-09-21T23:13:19Z

SU2_CFD/src/solvers/CTurbSSTSolver.cpp

+          solution[1] = FrictionVel*FrictionVel*S_R/(laminar_viscosity/density);
+        } 
+        /*--- Knopp eddy viscosity limiter ---*/
+        else if (roughsstParsedOptions.limiter_knopp) {


The options are mutually exclusive so allowing a list of enums in the config is not the most appropriate.
Better to change to a single enum option?
If it ever becomes possible to specify a list, the config syntax will still be compatible.

pcarruscag · 2025-09-21T23:14:20Z

SU2_CFD/src/solvers/CTurbSSTSolver.cpp

+            S_R = (200/(kPlus+EPS))*(200/(kPlus+EPS));
+          else
+            S_R = 100/(kPlus+EPS) + ((200/(kPlus+EPS))*(200/(kPlus+EPS)) - 100/(kPlus+EPS))*exp(5-kPlus);


Use the pow function please

Suggested change

S_R = (200/(kPlus+EPS))*(200/(kPlus+EPS));

else

S_R = 100/(kPlus+EPS) + ((200/(kPlus+EPS))*(200/(kPlus+EPS)) - 100/(kPlus+EPS))*exp(5-kPlus);

S_R = pow(200/(kPlus+EPS), 2);

else

S_R = 100/(kPlus+EPS) + (pow(200/(kPlus+EPS), 2) - 100/(kPlus+EPS))*exp(5-kPlus);

pcarruscag · 2025-09-21T23:14:58Z

SU2_CFD/src/solvers/CTurbSSTSolver.cpp

+
+        su2double solution[2];


Suggested change

su2double solution[2];

su2double solution[2] = {};

marcosch27 · 2025-10-09T16:34:27Z

Thanks for the contribution! Can you add the new config options to config_template.cfg and add the flat plate setup? You can add the .cfg file to this PR, and the mesh and a converged restart file to a PR for the TestCases repository.

Thanks for the reply. I added just one entry to select the boundary condition at the wall, which is "ROUGHSST_OPTIONS", since surface roughness is specified using the existing entry "WALL_ROUGHNESS". As shown in the document attached to the PR, the present test case is the NASA ZPG Flat Plate test case with the addition of surface roughness to verify the effects of different boundary conditions for rough walls for the SST model. I will upload the test case in a PR for the TestCases repository.

marcosch27 · 2025-10-09T16:43:04Z

Add a testcase please, something of verification quality, are there known results from literature for flatplates?

I will add the test case, and in this paper, it is possible to find an overview of the effects of the boundary conditions for rough walls for the SST model on a fully turbulent flat plate test case, including the boundary conditions that I have implemented.

marcosch27 · 2025-10-15T08:58:27Z

Hi @pcarruscag and @bigfooted,

As part of this pull request, I’ve a validation test case for the new roughness boundary condition in the SST model.
I’d like to open a pull request to add this test case to the su2code/TestCases repository, but I currently don’t have permission to push a new branch there.

Could you please grant me temporary access or let me know the preferred procedure for contributing this test case?
Alternatively, I can open a pull request from my fork if that’s the standard approach.

Thanks a lot for your help!

pcarruscag · 2025-10-15T13:29:43Z

I've invited you to the SU2 organization

marcosch27 · 2025-10-29T09:22:55Z

Hi!

I’ve opened a pull request at su2code/TestCases#178, where I’ve uploaded the validation test case for the newly implemented boundary conditions.

The results obtained with the different boundary conditions are presented in terms of the log-law profiles and velocity shift, confirming the findings reported in the literature.

pcarruscag · 2025-11-01T20:00:36Z

Great, can you add regression tests to *_regression.py scripts?
You can modify regression.yaml to target your new branch with the mesh and restart files.
The configs you just add to the code repo (so this PR) because that's is easier to maintain when options change.

pcarruscag · 2025-11-03T22:14:52Z

SU2_CFD/src/variables/CTurbSSTVariable.cpp

+  su2double limiter = a1*Solution(iPoint,1) / (F2(iPoint)*vorticity_mag + EPS);
+  limiter = min(1.0, limiter);
+  SST_Limiter(iPoint) = limiter;


@marcosch27 can you leave this for another PR please?
We already use a lot of memory, especially for adjoint, so I'm not keen on adding more storage just to output values.

Ok, that's clear, I've removed all the changes related to the SST limiter output.

SU2_CFD/src/output/CFlowOutput.cpp

pcarruscag

Thank you for the tests, I think it's good to go after these minor comments.

pcarruscag · 2025-11-19T23:43:30Z

TestCases/rans/flatplate/roughness/bc_wilcox2006/turb_SST_flatplate_roughBCWilcox2006.cfg

+% 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)


Can you cleanup the unused/default options please? Take a look at our other config files to see what I mean.

Suggested change

% 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)

MUSCL_TURB= NO

pcarruscag · 2025-11-19T23:44:22Z

TestCases/rans/flatplate/roughness/bc_wilcox2006/turb_SST_flatplate_roughBCWilcox2006.cfg

+% Slope limiter (NONE, VENKATAKRISHNAN, VENKATAKRISHNAN_WANG,
+%                BARTH_JESPERSEN, VAN_ALBADA_EDGE)


We no longer put the lists of options in all configs because then if we remove or add an option everything is out of date.

Suggested change

% Slope limiter (NONE, VENKATAKRISHNAN, VENKATAKRISHNAN_WANG,

% BARTH_JESPERSEN, VAN_ALBADA_EDGE)

pcarruscag · 2025-11-19T23:44:47Z

TestCases/rans/flatplate/roughness/bc_wilcox2006/turb_SST_flatplate_roughBCWilcox2006.cfg

+% Multi-Grid Levels (0 = no multi-grid)
+MGLEVEL= 0
+%
+% Multi-grid cycle (V_CYCLE, W_CYCLE, FULLMG_CYCLE)
+MGCYCLE= V_CYCLE
+%
+% Multi-grid pre-smoothing level
+MG_PRE_SMOOTH= ( 1, 2, 3, 3 )
+%
+% Multi-grid post-smoothing 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


None of this is used if the level is 0, etc.

pcarruscag · 2025-11-20T00:04:42Z

TestCases/rans/flatplate/roughness/bc_wilcox2006/turb_SST_flatplate_roughBCWilcox2006.cfg

+% 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"                                                %


You can probably grab and modify our most recent config for a flatplate to make it easier to cleanup the files

pcarruscag · 2025-11-20T00:05:02Z

SU2_CFD/src/solvers/CTurbSSTSolver.cpp

        nodes->SetSolution(iPoint,solution);
        LinSysRes.SetBlock_Zero(iPoint);
-
+        


Suggested change

Common/include/option_structure.hpp

pcarruscag · 2025-11-20T00:07:56Z

TestCases/serial_regression.py

+    # FLAT PLATE, ROUGHNESS BC KNOPP SST
+    turb_flatplate_sst_roughBCKnopp           = TestCase('turb_sst_flatplate_roughBCKnopp')
+    turb_flatplate_sst_roughBCKnopp.cfg_dir   = "rans/flatplate/roughness/bc_knopp"
+    turb_flatplate_sst_roughBCKnopp.cfg_file  = "turb_SST_flatplate_roughBCKnopp.cfg"
+    turb_flatplate_sst_roughBCKnopp.test_iter = 10
+    turb_flatplate_sst_roughBCKnopp.test_vals = [10.000000, 0.053020, -3.454853, -0.684543, -0.886080, 2.140376, 1.043068, 4.808919, -0.203494, 0.053645]
+    test_list.append(turb_flatplate_sst_roughBCKnopp)
+
+    # FLAT PLATE, ROUGHNESS BC AUPOIX SST
+    turb_flatplate_sst_roughBCAupoix           = TestCase('turb_sst_flatplate_roughBCAupoix')
+    turb_flatplate_sst_roughBCAupoix.cfg_dir   = "rans/flatplate/roughness/bc_aupoix"
+    turb_flatplate_sst_roughBCAupoix.cfg_file  = "turb_SST_flatplate_roughBCAupoix.cfg"
+    turb_flatplate_sst_roughBCAupoix.test_iter = 10
+    turb_flatplate_sst_roughBCAupoix.test_vals = [10.000000, 0.053252, -3.575414, -0.761810, -0.998912, 2.003238, 0.907276, 4.807309, -0.197354, 0.051349]
+    test_list.append(turb_flatplate_sst_roughBCAupoix)


Move a couple of these tests to parallel_regression, please, so that all actions take more or less the same time.

…egression

pcarruscag · 2025-11-24T15:30:17Z

TestCases/rans/flatplate/roughness/bc_wilcox2006/turb_SST_flatplate_roughBCWilcox2006.cfg

+WALL_ROUGHNESS = (wall, 400e-6)
+KIND_ROUGHSST_MODEL = WILCOX2006
+MATH_PROBLEM= DIRECT
+RESTART_SOL= NO


You are adding restart files to the testcases repo, so the cases should probably be restarted? (which is good to track converged values instead of just a few iterations)

Suggested change

RESTART_SOL= NO

RESTART_SOL= YES

pcarruscag · 2025-11-24T15:30:37Z

.github/workflows/regression.yml

        with:
          # -t <Tutorials-branch> -c <Testcases-branch>
-          args: -b ${{github.ref}} -t develop -c develop -s ${{matrix.testscript}}
+          args: -b ${{github.ref}} -t develop -c feature_turb_flatplate_SST_RoughBCs -s ${{matrix.testscript}}


Remember to revert this to develop

bigfooted added the changelog:feature label Sep 17, 2025

pcarruscag reviewed Sep 21, 2025

View reviewed changes

marcosch27 mentioned this pull request Oct 27, 2025

Add SST rough wall boundary condition validation test case (flat plate) su2code/TestCases#178

Open

pcarruscag reviewed Nov 3, 2025

View reviewed changes

github-advanced-security bot found potential problems Nov 3, 2025

View reviewed changes

SU2_CFD/src/output/CFlowOutput.cpp Fixed Show fixed Hide fixed

marcosch27 added 10 commits November 14, 2025 15:14

Boundary conditions for SST for rough walls

b6c4c13

Fix formatting

e517e4d

Fix issues about rough SST BCs formulations

8f633e5

Add SST Limiter Output

0d3eaad

Revert SST Limiter Output changes

3b7b0ff

Switch to regression branch

92f1940

Fix rough BCs options issues

65d8ea5

Added rough flatplate test cases

129f7db

Fixed values regression test rough flat plate

292af31

Fix regression test roughness

26ce36f

marcosch27 force-pushed the feature_SST_RoughBCs branch from 39274fc to 26ce36f Compare November 17, 2025 16:04

pcarruscag reviewed Nov 20, 2025

View reviewed changes

Cleaned regression tests for rough flat plate and moved to parallel r…

1d14e38

…egression

pcarruscag reviewed Nov 24, 2025

View reviewed changes

pcarruscag approved these changes Nov 24, 2025

View reviewed changes

		NONE, /!< \brief No option / default. /
		WILCOX1998, /!< \brief Wilcox 1998 boundary conditions for rough walls / default. /

		% Slope limiter (NONE, VENKATAKRISHNAN, VENKATAKRISHNAN_WANG,
		% BARTH_JESPERSEN, VAN_ALBADA_EDGE)

		nodes->SetSolution(iPoint,solution);
		LinSysRes.SetBlock_Zero(iPoint);

Adding new roughness boundary conditions for SST #2573

Are you sure you want to change the base?

Adding new roughness boundary conditions for SST #2573

Conversation

marcosch27 commented Sep 17, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Proposed Changes

Related Work

PR Checklist

Uh oh!

bigfooted commented Sep 17, 2025

Uh oh!

pcarruscag left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

marcosch27 commented Oct 9, 2025

Uh oh!

marcosch27 commented Oct 9, 2025

Uh oh!

marcosch27 commented Oct 15, 2025

Uh oh!

pcarruscag commented Oct 15, 2025

Uh oh!

marcosch27 commented Oct 29, 2025

Uh oh!

pcarruscag commented Nov 1, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

pcarruscag left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

marcosch27 commented Sep 17, 2025 •

edited

Loading