Accurate simulation?

Hi I just created a new simulation, I was wondering if this project is accurate or close to accurate? I think it might be more accurate than the past, because one, I used region refinments to an extent, two I used symmetry on this simulation, three I made the mesh better quality than the last version. It also took less time to come up with a flat out result please take a look, here is the link https://www.simscale.com/workbench/?pid=450324700395851699&mi=spec%3A3abe0a78-89f7-4fd8-b17b-481e0e76f3d7%2Cservice%3ASIMULATION%2Cstrategy%3A17

Thank you for your time and help,
Frank lucci

Hey Frank,

To begin with, you can check whether the simulation reached a converged state according to this guide: How to Check Convergence of a CFD Simulation? | SimScale
The boundary conditions seem just fine :slight_smile: Make sure to check the y plus value in the solution fields as well.
How to Control and Analyze Y+ in Meshing Simulation | SimScale

Best regards,
Fillia

Hello again, I have decided to run the simulation a bit more to make the convergence a bit more perfect. I can’t get my head around the y plus value. Is there a general rule of thumb I can use for the y plus value?

Thank you again,
Frank
Lucci

Check here: There are 5 levels of region refinements, how much should each one be - #3 by jairogut

This should be helpful :slight_smile:

Best regards,
Fillia

So the y+ is 30 to 100 in the finest spot of the mesh or the first and most refined region, what should be the y+ of the 2nd region refinment and the flow region? I’d don’t understand.

Sorry for the trouble,
Frank
Lucci

Hi Frank!

Y+ is only computed for the walls, especially those with wall modeling boundary conditions (function or full resolution). Please refer to this article for some background:

So do you have to have the y+ correct for an accurate simulation? Or can it be close to accurate and the simulation will still be accurate?

The correct y+ guarantees that the wall boundary condition is giving valid results.

If you are not interested in the accuracy of the flow fields (velocity and pressure) at the walls, but for example in regions far from them, then I guess you do not have to worry about perfect y+ values. Otherwise, you do.

So if it is a symmetrical type simulation do you have to have good y+ values or can you have ok y+ values and get accurate results?

Please read the reference and analyse with care!

Y+ is only important in no-slip walls

If the y+ is correct according to the selected wall model, then this point is checked.

The next step to assure the accuracy of the simulation would be to perform a mesh independence analysis.

Hi again I looked every were and I couldn’t find it, could you describe what it is and how to do it?
Thank you,
Frank
Lucci

Hey Frank, make sure to add the ‘Turbulence’ under the ‘Field calculations’, so you can check the y+ values in the post-processor after the run.

Result Control | Simulation Setup | SimScale.