Learning Forum

[floatingstones]

Hello all,

I am using the student version of Ansys 2021 R1, which limits the number of nodes and elements to 125k each.

I've worked out my mesh to be under 125k (more precisely 110k elements and 90k nodes). Tried a simple simulation with gravity only and obtained results.

However, when applying hydrostatic pressure, the following error appears:

Your product license has numerical problem size limits, you have exceeded these problem size limits and the solver cannot proceed.

My assumption is that the Pressure object creates SURF154 elements above the shell elements in order to apply the load. And these new elements add to the count, surpassing the limit. Is this correct? If so, is the only way to solve it making the mesh more coarse?

any insight is helpful! thanks!

[Rob]

Very likely - contact also adds cells to the count that don't show up in the mesh. The other solution is to use one of the Uni Research licences.
There is an option to renumber the nodes. The check is made against the highest node ID, not the total count of nodes. Renumbering means you don't have "gaps" in the sequence so reducing the highest node ID.

[floatingstones]

Hey thank you for your answer! Could you guide me into the University Research licence? What's the condition for using it?
Gonna check the renumbering option out, although I think the problem is only with the element count.

[Rob]

You'll need to talk to your supervisor re the Research licence as the University pay for those. Those licences have some limitations re doing commercial work but remove the cell count limit and work with the campus parallel licences.

[floatingstones]

Hey so my University does not have a Research license on Ansys, this is why I'm using the free Student version (https://www.ansys.com/academic/students/ansys-student).
As I see in the website, the Student License only has limits on nodes/elements (Problem Size Limits), this is why I induced my error was on the extra elements created by pressure/contacts.
But now my mesh is under 64k (half of the limit) and I am still getting the error. Even if the pressure created an extra of the totality of the elements, that wouldn't explain the error. Do you have any idea what might cause numerical problem size limits in this Student license?

[floatingstones]

Update: Reduced the node number and it worked, so the problem still was the node/element count.

[Rob]

Glad to hear it's working. Now you can go and lean on your University to get Research so you can run bigger models! ;)

[floatingstones]

Update 2: Weak Springs also add to the element count, I think those were the problems actually!
I'll advertise!

[Rob]

:)
One of the benefits of CFD, we have 512k cells available and I don't think our virtual cells add to the count.