eigenvalues analysis
-
elenavaseva
- Posts: 7
- Joined: 05 Mar 2010, 05:24
eigenvalues analysis
_Thank I will check the units -
-
elenavaseva
- Posts: 7
- Joined: 05 Mar 2010, 05:24
Re: eigenvalues analysis
I try to make simple linear analysis -eigen value of two story one bay frame with concetrate mass I have used
section - rss
section mat - unc
matarial - el_mat
element - effrm
and mass conctrate in all nod-column - beam
in my case plane frame - 4 node
And after run -
warning could not re-orthogonalise all
I chak my model alot of time - no results
before i used real r/c element and the results were the same
When I remove concterate mass i have received results in case of r/c element with reinf using again el-mat but only the section are rc no rss.
section - rss
section mat - unc
matarial - el_mat
element - effrm
and mass conctrate in all nod-column - beam
in my case plane frame - 4 node
And after run -
warning could not re-orthogonalise all
I chak my model alot of time - no results
before i used real r/c element and the results were the same
When I remove concterate mass i have received results in case of r/c element with reinf using again el-mat but only the section are rc no rss.
Re: eigenvalues analysis
Hi Elena,
From what i could understand, your structure is very simple with few DOF. This way, only few eigenvalues can be found and, probably, as your higher modes increase, you period decrease to very low values. The error that you report indicates that the program can't find all the eigenvalues that you define in the settings - expected situation based in what i mentioned before.
I suggest you to look at the period values and mode shape found and verify if they are reasonable. If not, a possible mistake could be on the units definition of your model.
Hope this can help you...
Romain
From what i could understand, your structure is very simple with few DOF. This way, only few eigenvalues can be found and, probably, as your higher modes increase, you period decrease to very low values. The error that you report indicates that the program can't find all the eigenvalues that you define in the settings - expected situation based in what i mentioned before.
I suggest you to look at the period values and mode shape found and verify if they are reasonable. If not, a possible mistake could be on the units definition of your model.
Hope this can help you...
Romain
-
elenavaseva
- Posts: 7
- Joined: 05 Mar 2010, 05:24
Re: eigenvalues analysis
I check all information but I can not find some reason for this very simple model Please do i have to snet you this small model
Re: eigenvalues analysis
Hi,
You can send me the file.
Maybe i can help you...
Cheers,
Romain
You can send me the file.
Maybe i can help you...
Cheers,
Romain
-
seismosoft
- Posts: 1263
- Joined: 06 Jul 2007, 04:55
Re: eigenvalues analysis
Dear Elena,
We can confirm that your model is correct, and that it is instead the eigenvalue solver that is not behaving as it should - we will correct this apparent bug in the eigensolver in a future release of SeismoStruct.
In the meantime, and as a workaround to make the analysis run, we advise you to define your mass in the 'z' direction as well. That should do the trick.
Thank you for bringing this problem to our attention.
Seismosoft Support
We can confirm that your model is correct, and that it is instead the eigenvalue solver that is not behaving as it should - we will correct this apparent bug in the eigensolver in a future release of SeismoStruct.
In the meantime, and as a workaround to make the analysis run, we advise you to define your mass in the 'z' direction as well. That should do the trick.
Thank you for bringing this problem to our attention.
Seismosoft Support