Learning Forum

[ahg]

Hello,

I am trying to run a simulation where a 'rocket' (triangular section) moves laterally within an enclosure due to the surrounding fluid flow pushing it forward.

[Rob]

Have a look in the Help system for Fluent. There are two videos that go through moving mesh. In your case I think you've tried to smooth the mesh rather than remesh. n

[ahg]

Hi Rob,nThanks for your advice. Regarding the smoothing, I think that i just a poor screenshot on my part, the re-meshing option (deselecting the smoothing made no difference):nI haven't been able to find the videos you are referring to, seems like the help only includes descriptions, but I will investigate further and see whether they help. I've tried to follow videos off YouTube, but I can't spot the error.nThanks again for your help,nAtreyn

[Rob]

Check the cell size in your model compared to the remesh criterion too: the defaults aren't always appropriate. n

[ahg]

Hi Rob,nThanks again.nI believe the cell size in the model matches the remesh criteria (~max cell size 5.8e-4). A maximum cell skewness is also set of 0.7, which I think should trigger the remeshing, before negative cells occur.nBest,nAtreyn

[Rob]

I should have spotted this, check 11.6.2.3 in the Fluent User's Guide, Solution mode, 2020R2 documentation. n

[ahg]

Hi Rob,nJust to clarify, are you saying that using a triangular mesh will resolve this? I have tried this with the local cell remeshing and a triangular mesh, and am now receive a floating point error.nI am still messing around with the settings to hopefully try resolve this as it does appear to be a different problem.nThanks a bunch.nAtreynnEDIT: It appears the floating point error occured because I had switched to a pressure based solver. If I was to switch back to a density solver the divergence occurs, and negative cell volume occurs again.n

[Rob]

Is the solution remeshing, and then failing?n

[ahg]

Hi Rob,nThe solution looks to just be failing. No remeshing appears to be taking place - is there any way to check in the logs or similar?nThanks,nAtreyn

[Rob]

Logs tend to tell you what happened, what you want is images at each mesh update. If you display cell size how does that compare to the remesh settings? n

[ahg]

Hi Rob,nThanks for your advice. I was unable to get this dynamic mesh to work, so have opted for a different approach. As such, am happy to close this thread. nThanks,nAtreyn