[shotta287]

Hi all,
Please see the attached screenshot of my geometry. I am modelling the extraction of a blood clot. Blood primary phase, blood clot secondary. VOF multiphase model. The problem is when running the solution the velocity field keeps dirverging and error comes up saying global courant number greater than 250. I think it is due to the quality of my mesh. At the moment my minimum orthogonal quality is 0.24. It could be because there is a change of diameter in my geometry. I tried the solution with a tube with constant diameter and it worked (same settings). So what would be the best way to mesh my geometry? Any assistance would be much appreciated

[shotta287]

Anyone? It is needed for my university project which is due in next week! Help would be much appreciated.

[scipy]

Since your mesh only has 26k elements, why not try and increase that by an order of magnitude? Add some edge sizing refinement at the location of the diameter change maybe...

[shotta287]

Originally Posted by scipy

Since your mesh only has 26k elements, why not try and increase that by an order of magnitude? Add some edge sizing refinement at the location of the diameter change maybe...

Hi thank you for replying. Would relevance increase the order of magnitude? Also, I am solving via a non-iterative time advancement to reduce calculation time. The solution doesn't solve this way, only by hybrid initialization. Is a different type of mesh required for non-iterative?

[Far]

Originally Posted by shotta287

Hi thank you for replying. Would relevance increase the order of magnitude? Also, I am solving via a non-iterative time advancement to reduce calculation time. The solution doesn't solve this way, only by hybrid initialization. Is a different type of mesh required for non-iterative?

order of magnitude means multiplication by 10. So increase the mes to 0.26 million and check your solution. And definitely edge sizing, relevance etc will refine the mesh.

If possible, do the hexa mesh and I am sure you will find pretty good tutorials for exactly same geometry in ICEM CFD hexa.

No, different type of mesh is not required. It is more or less depends on the options chosen in solver.

[shotta287]

Originally Posted by Far

order of magnitude means multiplication by 10. So increase the mes to 0.26 million and check your solution. And definitely edge sizing, relevance etc will refine the mesh.

If possible, do the hexa mesh and I am sure you will find pretty good tutorials for exactly same geometry in ICEM CFD hexa.

No, different type of mesh is not required. It is more or less depends on the options chosen in solver.

Please see the attached picture. This geometry also has a default mesh and has a minimum orthogonal quality of 0.62. I am able to run non iterative solution with this.
In my first post, the mesh shown there has a orthogonal quality of 0.24 and for some reason does not run non-iterative advancement. I am keeping the solver settings the same for both meshes. What could be going wrong?