BandArpackSolver: Maximum Iteration Reached

Araj · Post by **Araj** » Mon Jan 31, 2005 8:19 am

Hello,

I am trying to perform an eigen value analysis, but I am getting this message.

"BandArpackSolver: Maximum Iteration Reached"

Also while performing Transient analysis, it does not converge.

Any suggestions regarding this and the type of error with which it is related to.

Thank you,

With regards,
Aloke

silvia · Post by **silvia** » Mon Jan 31, 2005 9:28 am

is the first a fatal error or just a warning?
is a static pushover analysis of your structure working, before you do a transient analyisis? where is it not converging?
do you want to post your input file?

Araj · Post by **Araj** » Mon Jan 31, 2005 10:19 am

Thank you Silvia for your concern,

"BandArpackSolver: Maximum Iteration Reached"

It does not mention whether it is a fatal error or warning. But it just comes like that. I get highly unlikely values as eigenvectors (-6.23 * 10^66).

The Static Pushover analysis works.

The transient analysis fails at 0.01 seconds which is my first step. But Warning appears that CTestDispIncr. failed to converge after 10 iterations.

I am analyzing a 3 bay 2-D frame of 2 storey.

Thank you,

Rgds,
Aloke

silvia · Post by **silvia** » Mon Jan 31, 2005 10:42 am

how about posting your input file???
do you have the masses defined correctly???

Araj · Post by **Araj** » Mon Jan 31, 2005 10:53 am

It consists of many files linked together and I guess cannot be posted. Sorry for this.

I have verified the Gravity Load Analysis and I am sure there is nothing wrong till this part.

I guess its only the part of Eigenvalue analysis which is giving trouble. I guess, my definition of masses are correct as I have used element mass in "element nonlinearBeamColumn".

I shall be glad if you can suggest what does this error actually mean and what parameters cause this error.

Thanking you,

With Regards,
Aloke

silvia · Post by **silvia** » Mon Jan 31, 2005 11:20 am

the gravity load analysis has really nothing to do with masses and eigenvalues.
you might want to try nodal masses before you do element masses to make sure you have everything working, as some people are experiencing problems with element mass.
if you think it might help, you can post the part of the input that pertains to this.