Unphysical pressure field in LES PitzDaily tutorial case
 

Hi Openfoamers,

I was looking into the LES PitzDaily tutorial case(tutorials/incompressible/pisoFoam/les/pitzDaily) and found the pressure result is a terrible disaster: the whole pressure field is fluctuating rapidly and apparently that's not what we expect to have.
I did a raw search and found some threads also had similar confusions. Since a few years have passed, I am posting a new thread to see if there are any solution to this problem in the newest version of Openfoam?

https://www.cfd-online.com/Forums/op...d-problem.html
https://www.cfd-online.com/Forums/op...-pisofoam.html
https://www.cfd-online.com/Forums/op...ofoam-les.html

Thanks a lot!
Jeremy

I did not have such effects with simulating the Pitz/Daily geometry, but had similar effects with other cases.

One word before: The pressure is harder to calculate than the velocity because it't calculation is indirect. Therefore the accuracy of the presser is often a problem and with it all values derived form it.

Strong pressure fluctuations point to inducing to much turbulence in the systems. This may be the case if you use a turbulence model which has a tendency to do so, combined with a too large time step / a too large maximal Courant number. The pimpleFoam algorithm is especial prone to that because it tries to predict a solution using a large time step and correct it afterwards. It may be necessary to use real small time steps / maxCo here. The wall time for getting a solution may decrease dramatic if you lessen the time step. And the accuracy gets better,

Jeremy, post some results, a comparison with other results and a description of the spatiotemporal schemes you used. Otherwise is very difficult to help!

Also the LES configuration is very important

Sent from my Lenovo A5000 using CFD Online Forum mobile app

Hello guys!


Any update?
It seems that the reason is due to the inlet condition, turbulentInlet.



Is there anyway to fix them with a postprocessing?