BeamSolver3D: wave propagation

Numerical methods and mathematical models of Elmer
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: BeamSolver3D: wave propagation

Post by raback »

Hi

Great, added that as default so no need to put it in the future.

This relates to at which stage mass matrix is added to the linear system. Historically (when there was less memory) it was added (with given timestepping) to the matrix at element level. The current recommended way is to maintain the mass matrix and add it in global level. In fact, I had no clue that it would solve the issue but I just wanted to separate mass and stiffness matrix so I could narrow down the issue...

-Peter
palyvar
Posts: 3
Joined: 24 Jan 2023, 21:57
Antispam: Yes

Re: BeamSolver3D: wave propagation

Post by palyvar »

Thank you both very much, now the solution looks realistic and the problems with NaN Norm disappeared (at least in what I have tried so far).
Post Reply