[dradenkovic]

Hello.

I am comparing results of SIMPLE and PIMPLE pseudo - transient approach in channel, k-epsilon model in both cases.

I have noticed that value of yPlus is different in these two cases, for the same mesh.

For comparation, when SIMPLE is applied, yPlus at lower wall is around 10. For the same mesh, also k-epsilon, when PIMPLE is applied, yPlus at lower wall is around 70.

Any help what is the reason? I think that reason exists, since in new OpenFoam (I have dev version, compiled few weeks ago), yPlus is dependent from solver, for example, now command is

simpleFoam -postProcess -func yPlus


Regards,
Darko

[dradenkovic]

I tried both methods (and SIMPLE and PIMPLE) on pitzDaily example; both methods give the same result. And yPlus is the same, independently from method.

But in channel, I see different solutions when I apply these methods.
Velocity profile, pressure, turbulent kinetic energy - everything is different. Starting solution for transient simulation is converged k-epsilon solution, but that doesn't help. No wonder why yPlus is different in these cases.

So, to conclude, yPlus does not have any relationship with the chosen method (SIMPLE or PIMPLE); but new question is the reason for the above mentioned behavior ... Steady and transient solution after enough time should be the same in this case, right? What could be the reason for this discrepancy?


Regards,
Darko

[wyldckat]

Quick answer: I got your PM. Since you didn't specify the exact solvers you have used and you didn't specify the exact case set-up that you are using, then I'm guessing here that you are mixing a compressible result with an incompressible result; or maybe you're not waiting for the flow to stabilize in the transient solution provided by PIMPLE. Keep in mind that with PIMPLE, it could take anything between 10 seconds or 10 days inside simulated time, for a simulation to reach steady-state.

The other possibility is that the time accuracy is revealing details that the steady-state doesn't reveal - edit: for example, vortices being released near walls.

Please provide more details, as requested here: http://www.cfd-online.com/Forums/ope...-get-help.html

[dradenkovic]

I forgot to check kinematic viscosity - it was different for these two cases. Days passed until I noticed this, unfortunately.

At the end, result for SIMPLE and PIMPLE was the same, as other parameters.

Thank you!

Regards,
Darko