Eigenvalue analysis

04-Unexpected behaviour/errors
Post Reply
rezasheikhi
Posts: 10
Joined: 17 Sep 2017, 20:46

Eigenvalue analysis

Post by rezasheikhi » 21 Dec 2018, 11:03

Hello,

When I run eigenvalue analysis I get this warning:
Could not re-orthogonalise all Lanczos vectors
https://www.dropbox.com/s/y47z0t8ui8ved ... e.spf?dl=0

Thank you,
Reza Sheikhi

huffte
Posts: 827
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

Re: Eigenvalue analysis

Post by huffte » 21 Dec 2018, 15:01

Did you investigate the help system? It is very clear on this issue. Directly from the help text:

"When running an eigenvalue analysis using Lanczos algorithm, user may be presented with a message stating: "could not re-orthogonalise all Lanczos vectors", meaning that the Lanczos algorithm could not calculate all or some of the vibration modes of the structure. . . . This message typically appears when too many modes are sought, e.g. when 30 modes are asked in a 24 DOF model, or when the eigensolver cannot simply find so many modes (even if DOFs > modes)."

If you change you requested number of eigenvalues from 10 to 6, you will no longer see the error. This seems reasonable given the model has only two lumped masses. You should investigate the participation factors to ensure you capture a sufficient percentage of the total mass with 6 modes.
Tim Huff

rezasheikhi
Posts: 10
Joined: 17 Sep 2017, 20:46

Re: Eigenvalue analysis

Post by rezasheikhi » 21 Dec 2018, 16:38

Thank you for your quick answer, I requested 6 number of eigenvalue now I get "Imaginary or null eigen-solution" for modes 2 to 6

huffte
Posts: 827
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

Re: Eigenvalue analysis

Post by huffte » 21 Dec 2018, 18:11

Is there a particular reason you are using Lanczos eigensolver? If you specify 'Automatic' you will get virtually 100% participation in the first 6 modes, none of which are null. You give up much too easily.
Tim Huff

rezasheikhi
Posts: 10
Joined: 17 Sep 2017, 20:46

Re: Eigenvalue analysis

Post by rezasheikhi » 21 Dec 2018, 21:41

How many Ritz vectors in directions should I request in this case?

huffte
Posts: 827
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

Re: Eigenvalue analysis

Post by huffte » 22 Dec 2018, 01:52

If I specify 2 Ritz vectors in each direction I get 6 modes and 100% mass participation in all three directions.
Tim Huff

Post Reply