Force plot does not converge CFD

I am currently working on a university project involving a Computational Fluid Dynamics (CFD) analysis of a train part, specifically the ‘wippe’ component.

In the provided screenshot, noticeable extreme fluctuations can be observed despite implementing measures such as region and local refinements, as well as using inflate boundary layers. Additionally, I’ve experimented with various solutions from online forums, including employing 2 to 3 Cartesian boxes around the part, utilizing a finer mesh, increasing local refinement, adjusting relaxation factors, and more. Unfortunately, these attempts have not yielded the desired results.

My hypothesis is as follows: The wind direction is along the z-axis, and the ‘wippe’ part, located at the top, exhibits symmetry concerning the x-axis. Consequently, I suspect that after the wind interacts with the initial part of the top, dead wake regions are formed. The subsequent interaction of the second part of the “wippe” with these chaotic wake regions may be the root cause of the observed fluctuations.

I am seeking advice on potential solutions to address this issue. Any insights or suggestions would be greatly appreciated.



Hi there,
Thank you for posting on the forum!
Can you please share the project link?
Thank you

In my experience geometry that cause significant unattached turbulent flow usually has oscillatory spikes in convergence even when it is considered to be converged. I believe in that case a moving average of the result can be used for determining whether convergence has been achieved. You can do that with the ORSI feature of the yPlusHistogram program, which you can find by searching this forum.