Analysis

[ehsanala]

hello

I made the model wih damper.the model to run with algorithm linear and Model will converge . If the model without damper to analyze ,with algorithmNewton will converge. what is problem ?
the comparison of the two models is correct?

If dt of earthquake is equal to 0.02.and to reduce dt of analysis equal 0.001.analysis is correct?

thanks

[fmk]

you need to rephrase the question.

[ehsanala]

hi

I built a structural model with damper. Algorithm is only linear convergence is analyzed. But when Newton algorithm to analyze the structure without damper works. Comparing the two models is correct?

thanks

[mhscott]

Look at the time history response using the two algorithms ... you should see some differences. The linear algorithm should only be used for very small time steps because it is explicit.

[nhemafukiganesh]

I've a 3 story RC building model for pushover analysis which was carried out in Opensees 2.4 version, 32bit Windows 7 computer. The curve Base Shear vs top story displacement was plotted.
Now I've new computer with Window 8, 64bit computer. And I run the same model using latest version OpenSees 2.4.2. The result Base Shear vs top story displacement different than previous analysis.
Further,
1) When I analyzed the Pushover analysis before in 1st computer the analysis stop within 300mm of control node disp, but now it does not stop rather it runs more than 500mm.

2) The Pushover curve is so smooth and high base shear in 2nd analysis than in previous.
Is it due to newer version of OpenSees or due to 64 bit computer? How these different curve generated? I'm confused which analysis should I consider for my research work? Your help would be appreciated. Thank you.

[brag006]

are you sure you didn't change anything? Can you post the code please

[nhemafukiganesh]

brag006 wrote:
> are you sure you didn't change anything? Can you post the code please
The files are same but just run in Opensees 2.4 and Opensees 2.4.2. My model is not in single file, it contain 6 files. How to upload file here? I don't know. If I just paste the code of each file it's very large and even it does not run in Opensees. If any body interested in solving this problem I can provide the files. Thank you.

[brag006]

send your files to my email. My email is my forum user name at aucklanduni.ac.nz

[nhemafukiganesh]

What is your email address? The following is error in yahoo when I try to send email in your address, (aucklanduni.ac.nz).
Invalid Address
We cannot find a match for this email address:

aucklanduni.ac.nz

Please use an email address that looks like user@example.com or a nickname or a distribution list name from your Yahoo7 contacts.

[brag006]

yea before aucklanduni.ac.nz add my username - brag006

[civil]

dear frank
please sau us about this problem. why the results are different in opensees 2.4.0 & 2.4.2?

[fmk]

the problem may have been due to the fact that someone modified the default integration rule associated with the dispBeam and forceBeam (nonlinearBeam) column elements. While the change to the default beamColumn was necesseary, given a typical use was to replace a single forceBeamColumn elements with many dispbeamColumn elements , it was not intended to also be the default for the forceBeamColumn.

Version 2.4.3 which I just put out to remedy this, returns the default behaviour to the forceBeamColumn. If you are using the dispBeamColumn and what the old behaviour, you must use the -integration Lobotto option.

[mhscott]

Use the 'print ele $eleTag' command to make sure you have the right integration rule. Should be Legendre for dispBeamColumn and Lobatto for forceBeamColumn (nonlinearBeamColumn)

[civil]

dear mhscott and other friends

I use dispBeamColumn element in opensees version 2.4 and when i use print ele $eleTag command, it is Lobatto.

1) Did you mean Lobatto and dispBeamColumn element is wrong?
2) Are my analysis results wrong in this case?
3) how can we define Lobatto or Legendre for elements?
I dont change anything in the commamd and use e.g.
element dispBeamColumn 1 1 2 10 4 1

[civil]

civil wrote:
> dear mhscott and other friends
>
> I use dispBeamColumn element in opensees version 2.4 and when i use print
> ele $eleTag command, it is Lobatto.
>
> 1) Did you mean Lobatto and dispBeamColumn element is wrong?
> 2) Are my analysis results wrong in this case?
> 3) how can we define Lobatto or Legendre for elements?
> I dont change anything in the commamd and use e.g.
> element dispBeamColumn 1 1 2 10 4 1

please answer this question
thanks in advanse