Why is my large harmonic magnetic analysis aborting?


This problem may be related to defect 64752, which was related to having a local array that was dynamically allocated on the stack. In ANSYS, we must always allocate memory using the ANSYS heap memory manager because arrays can quickly grow to be GBs in size, and considering that stack memory is only 16 MB or 32 MB, it is by far not enough to hold anything that related to the number of nodes, elements, or DOF in the model, for example.

Unfortunately, the only possible workarounds would be to change solvers, change platforms, or decrease the problem size until the bug is avoided. Since this area of the Sparse solver code was completely rewritten in v12, this bug is inherently removed in that version.





Show Form
No comments yet. Be the first to add a comment!