OpenSees 2.3.0 won't run with ReinforcingSteel

[tazarv]

Dear All,
I have tried to run a SDOF RC column in OpenSees 2.3.0 under an earthquake. That can not proceed while this file is OK in OpenSees 2.2.2c. Also, when I change the rebars to Steel02, OpenSees 2.3.0 works perfectly.
What is your idea about this?
Please reply to this post or alternatively to http://opensees.berkeley.edu/community/ ... =2&t=32867

Thanks,
Mostafa

[fmk]

i am curious as to why they give different results .. none of the steel materials was modified in the last revision .. what concrete material are you using?

[tazarv]

Hi Frank,

I have found something more interesting for you. I have checked my file in 6 different systems. Surprisingly, all 4-PC with Intel Processors could analyze the model with ReinforcingSteel perfectly. However, 2 systems with AMD Processors got failed after 42 seconds (when earthquake is getting stronger). I don't have more systems to check it now. I have checked two different models first with concrete02 and second with Concrete01WithSITC. The issue exists in both model when I use ReinforcingSteel but with steel02 it's fine.

What's you idea?

[hlayssi]

Hello,

I have been facing with the same problem in OpenSees 2.3.1 and the ReinforcingSteel Material.
Is there any achievement on the problem?
The Opertaing System is Win 7- 64Bit.

Thanks very much.
Hamed

[vesna]

Are you using 32Bit version of Tcl?

[tazarv]

I have tried both 32 and 64. I think the problem is what I mentioned before, AMD processors. I had to change my CPU to Intel and it's working now.

[hlayssi]

Hello back,

How can I find the 32-64 version of the TCL. As far as I know, mine is the one downloaded from the OpenSees website.

thanks,
hamed

[vesna]

In that case you are using 32-bit version of Tcl.

Another comment: I used ReinforcingSteel extensively as well as Steel02 and I noticed that Steel02 is more stable then ReinforcingSteel. For some earthquakes I would not have any problems while using Steel02 but I would not be able to achieve the convergence with ReinforcingSteel. The problem may be machine that you use as Tarzev was mentioning but I also think that there is something inherent in ReinforcingSteel material that triggers numerical instability that leads to convergence problems.

[hlayssi]

Hello,

Thanks for your comment on Steel02 vs. Reinforcing steel.
Well, at the moment, with a displacement controlled integrator, I am doing quite well with both of them.
However, I have faced with the same difficulties in the convergence when using Reinforcing Steel Material (in certain loading protocols - I am doing reversed cyclic tests).
I don't think it is absolutely from the machine, I think there is something weird in the material coding as you have noted.


Thanks again,
Hamed

[jackywww]

what concrete material are you using?