mass scaling on quasi static problems using explicit solvers

[assafwei]

Hi,

I am using explicit solver to solve highly nonlinear problems that are impossible to solve using implicit solver, I dump the dynamic oscilations and use mass scaling to reduce solve time.
My question is this - since I am not intersted in the dynamic phenomena (and I dump them) is there still a limit for the mass scaling? Why shouldnt I use as much possible mass scaling as long as I am solving quasi static problems?

Thanks.

 

[Dave442]

Hi,

There is nothing stopping you doing this. If you are not interested in dynamic effects and are only modelling quasi-static explicit, you should reduce your simulation time as much as possible by increasing your loading rate and/or using mass scaling. You just need to ensure that kinetic energy remains a small fraction of internal energy (~5%) thoughout the duration of of the analysis. I try to avoid using excessive damping and mass scaling.

Good luck,
Dave

 

[assafwei]

Thanks Dave,

Is there a rule of thumb for how much damping is OK? Needless to say that it has a huge effect on the results, but I am really having a hard time judging how much damping is too much...

Thanks.

 

[Dave442]

Hi again,

Unfortunately I do not have a rule of thumb for selecting damping parameters. It will depend on what you are trying to model and exactly how you are applying damping. I would suggest reading the documentation and reviewing the literature. Also, if you provide more details of your analysis maybe someone can help you.

Dave