CFD Online Discussion Forums - How to speed up reconstructPar ?

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)

- OpenFOAM (https://www.cfd-online.com/Forums/openfoam/)

- - How to speed up reconstructPar ? (https://www.cfd-online.com/Forums/openfoam/78056-how-speed-up-reconstructpar.html)

Hi Bruno,

you're right, my residual plot is similar to yours. I know it's just a tutorial case but I have a similar convergence behavior in a real problem (case is very similar to the motorbike case).

I don't consider this as converged. The pressure residual is at 0.01
Sure, you can get a good first impression of the flow but nothing more.
The values for pressure and velocity should be pretty inaccurate or what would you say?

Regards,
Toni

Hi Toni,

Allow me to quote myself :)

Quote:

Originally Posted by wyldckat (Post 281101)

But I'm not a CFD specialist, so it's only a hunch ;)

I also said that there are holes in the motorbike that shouldn't exist. Those holes might be damaging the solution, therefore it won't converge in 500 iterations.

AFAIK, the motorBike case was first introduced when snappyHexMesh was included in OpenFOAM. And at that time, it was a tutorial solely for snappyHexMesh, with no set-up for solver execution. So I think it's only natural that the tutorial has not yet been validated with a real experimental data.

But if you want to aim for convergence, try AdMesh for plugging the holes in the motorBike STL and increase the number of iterations to 1000-1500. At least it's what I would do, if I really needed it to work...

Best regards,
Bruno