Problem with DTHA

04-Unexpected behaviour/errors
Post Reply
Maher
Posts: 5
Joined: 27 Mar 2020, 03:27

Problem with DTHA

Post by Maher »

Hi

I am facing issues with Dynamic Time History Analysis it researches a certain step and starts to give Max_tol, prd_lte, and terminates. I changed the converging criteria and element tolerance. This helped to apply the whole record, however, after looking at the strains in the elements they seem to be unrealistic and very low.

I am not sure what parameters could affect the strain.
I am trying to get some reasonable results from my analysis

User avatar
seismosoft
Posts: 743
Joined: 06 Jul 2007, 04:55

Re: Problem with DTHA

Post by seismosoft »

There are may posts on convergence problems in this forum. Furthermore, refer to the part in the manual that describes how to overcome convergence difficulities. Note that, when adopting a very loose convergence criterion, in order to make the analysis converge, this can affect the accuracy of the results.
SeismoSoft Support

Maher
Posts: 5
Joined: 27 Mar 2020, 03:27

Re: Problem with DTHA

Post by Maher »

Hi
I went through the sesimo-struct manual and the seismo-struct convergence solving video as well, and searched here for a solution to my problem
despite all that, I couldn't find a solution to my problem and because of that I posted my problem here.

I already did response spectrum analysis and nonlinear pushover analysis on the model, however, the issue raised when I tried to perform NTHA.
I need your advise and help in solving this issue

User avatar
seismosoft
Posts: 743
Joined: 06 Jul 2007, 04:55

Re: Problem with DTHA

Post by seismosoft »

Can it be that simply the structure cannot sustain the applied earthquake loading? What is the max. pga of the records?
Seismosoft Support

Maher
Posts: 5
Joined: 27 Mar 2020, 03:27

Re: Problem with DTHA

Post by Maher »

Hi ,

It is a bridge, I performed pushover analysis to it and Response analysis as well, the record I am using has a 0.4 PGA .
In pushover, the structure withstands 0.5 m in displacement and showed reasonable strain values in the elements (as I predefined strains for certain damage states, however in the NHTA it just went up to 3 seconds of the records with almost no displacement, OR when I changed the convergence criteria it performed the whole analysis but having unrealistic low strain and reaching a displacement of 1.5 m and did not even reach slight damage in comparing to the pushover analysis, I am using an elastic element for the deck as I am doing spine model. Most of the unbalance i am facing is in the nodes connecting the elastic elements and I think that this could be the issue, however, I don't know how to overcome it

User avatar
seismosoft
Posts: 743
Joined: 06 Jul 2007, 04:55

Re: Problem with DTHA

Post by seismosoft »

We do not know the details of your model, however If the analysis reaches deformations of 1.50m it is reasonable that you cannot make it converge (especially if you have geometric nonlinearities activates). A pga of 0.4g is by no means small, but please make sure that you are applying a correct conversion to the program units (i.e. factor of 9.81 in the case of kN and m).
Secondly, indeed a very loose convergence criterion might give very inaccurate results.
One final suggestion would be to decrease the time step of the dynamic analysis
Seismosoft Support

Post Reply

Return to “04-Unexpected behaviour/errors”