X Bracing for retrofitting

03-Analytical/modelling capabilities
Ralazem
Posts: 63
Joined: 08 Feb 2019, 02:38

X Bracing for retrofitting

Post by Ralazem »

I have done a static pushover analysis of an old RC building that has been badly designed. My analysis had fully converged for the first model.

For the second model, I am applying some x bracing in the spans where local element failures occur, however, I cannot get the strengthened model to converge. How is that possible? I tried modelling the bracing as elastic elements and as truss elements and neither models are converging. What might cause this and how can I get it to converge?
Ralazem
Posts: 63
Joined: 08 Feb 2019, 02:38

Re: X Bracing for retrofitting

Post by Ralazem »

Update: Changing the brace parameters, I can get it to converge but it does not seem to improve the structure's behavior. It makes more elements reach the code checks compared to the strengthened one. What needs to be done here?

PS. I've gone through the forum and validation files and sample models and all that.
User avatar
seismosoft
Posts: 1184
Joined: 06 Jul 2007, 04:55

Re: X Bracing for retrofitting

Post by seismosoft »

Hello Ralazem,
The best way to see why your model diverges is to go to the post-processor, to the Convergence tab and see which elements or which regions of the structure have convergence problems.
As for the fact that with the strengthening, you get more failures, it could be related to asymmetries (in plan and/or in elevation) in the strength and stiffness distribution of the new members.
Regards,
Seismosoft Support
Ralazem
Posts: 63
Joined: 08 Feb 2019, 02:38

Re: X Bracing for retrofitting

Post by Ralazem »

Thank you, I will try looking at the failed elements and re-running the analysis. But, in the case where I cant change the material properties, and for FBPH elements, does changing the number of fibers lead to better convergence?

Also, on another note, for my un-strengthened building, I changed a minor value in steel ( small change) and now the whole model wont converge.. returned the value to the original one where it was converged but now it wont converge. It seems that convergence is very sensitive to small changes..
User avatar
seismosoft
Posts: 1184
Joined: 06 Jul 2007, 04:55

Re: X Bracing for retrofitting

Post by seismosoft »

Convergence is sensitive to small changes only in very unstable models. This instability is due to either bad modelling practices, or structures that are close to collapse.
Note also that there is no way that you return to a previous state and you get a different behaviour. Probably you must have changed something else, and the analysis cannot converge.
Seismosoft Support
User avatar
seismosoft
Posts: 1184
Joined: 06 Jul 2007, 04:55

Re: X Bracing for retrofitting

Post by seismosoft »

Regarding the number of fibres, the tendency is that the more fibres you have the better the convergence is. However, this is a parameters that generally does not play a significant role in the convergence of the analysis.
Seismosoft Support
Ralazem
Posts: 63
Joined: 08 Feb 2019, 02:38

Re: X Bracing for retrofitting

Post by Ralazem »

Can I please send you my model for you to see if I have applied any bad modelling practices? I know the model is close to collapse, but since i managed to get it to converge before, I don't know what the issue is now.

https://www.dropbox.com/s/hrp6pfct7um1p ... H.spf?dl=0
User avatar
seismosoft
Posts: 1184
Joined: 06 Jul 2007, 04:55

Re: X Bracing for retrofitting

Post by seismosoft »

Hello Ralazem,
The problem was caused by the fact that you had very small plastic hinge lengths. If you change it back to the default value (i.e. 16.67%), which is a very reasonable value, the analysis will run without problems, even if some of the elements are totally deactivated at some stage, due to failures that you have defined in the Performance Criteria (btw a 20% residual strength might be a better approach. Moreover, a 0.8% concrete strain is often not an extreme value that causes complete failure of the member (this is the case for only old badly detailed structures)
Seismosoft Support
Ralazem
Posts: 63
Joined: 08 Feb 2019, 02:38

Re: X Bracing for retrofitting

Post by Ralazem »

I am following an equation by Priestly from his book "displacement based seismic design of structures" to calculate plastic hinge lengths. The work i am doing is for a PhD paper, so I will need to justify any plastic hinge length I use, so if I do use the 16.67%, I would have to justify it. What is the basis for the recommendation of this value?

Also, wouldn't a 20% residual strength cause even more convergence issues, as the columns will fail sooner?? Also, I took it as 80% residual based on Australian practices.Overall, the structure I am analyzing is in fact an old badly designed building, and the purpose of this analysis is that I apply some retrofitting options and study the difference in behavior. So for such structure, would my model, as it is currently, be considered properly modeled, or do you notice bad modelling practices?
Ralazem
Posts: 63
Joined: 08 Feb 2019, 02:38

Re: X Bracing for retrofitting

Post by Ralazem »

Also it seems that the fundamental time period is sensitive to changes in PH, but shouldn't it be based on elastic properties? When compared the time period of this model with another software, I would have to use a PH of 1-2% to get a matching value, but ofcourse that value is unreasonable.
Post Reply

Return to “03-Analytical/modelling capabilities”