Seismosoft | Earthquake Engineering Solutions

Hi,
I'm actually modelling a 12 storey frame-wall 3D structure and try to pass dynamic time-histories.
right now I'm testing my model by applying multiple static analysis and keep having fbd_inv problem.
I read the help system many times, but apart of saying that the matrice couldn't be inverse, I don't know hoe to solve this problem. The eigenvalue analyse runs perfectly.
Probably that my convergence criteria and iterative strategies aren't the best one, but for a big model with multiple nodes constraints, I'm not sur what would be the best values.
I've also put some performance criteria check and the program isn't able to apply the entire load even if the rebars never yield.

So how do you solve fbd_inv problems??

At the same time, is it possible to have models that can run for 15-20 hours??

Thanks

Hi Benoit,

1) I'm sure there are past posts on this matter. Please, search the Forum for “convergence problems”.
2) Usually dynamic time-history analyses of large complex 3D models may last several hours. Please, refer to the Help System of the program (-> Running an analysis and processing the results -> Processor).

Federica