Dear Baikuntha,

I had involved in a project with same questions and I am going to tell you my experiences which there is no proof to be totally correct.

1- I have used  Hysteretic material of opensees to define shear versus strain behavior of panel zone with transformed moment-deformation behavior.

2- I faced with some issues on using equalDOFs command and using Joint2D material.

3- I presume it would be more rational to use empirical modifications in beam hinges behavior to capture effects of joints or panel zones on seismic response. I know there are some references but at the moment I don't remember anything.

I hope it helps

Bests,

Reza 

Dear Reza,

Thank you for sending important issues for the modeling of panel zone.

As my understanding, it is better to used hinges at the end of beams and columns using Modified Ibarra-Medina-Krawinkler Deterioration Model with Bilinear Hysteretic Response (Bilin Material). Then, there is no need to use Joint2D element to capture effect of panel zones.

Some researches have been using both beam with hinges and joint2D element to capture different levels of damages or effects of beam and joints separately. I hope we can capture most of significant effects only using elastic beam with end hinges and there is no need to use Joint2D element.

In OpenSees, the BeamColumnJoint element is also available. Have you think of using this element instead of Joint2D element or improving models with relevant parameters for Joint2D element too.

Dear Baikuntha,

I was wondering if you could possible elaborate on your statement that you got "unexpected results" using the modified Ibarra-Medina-Krawinkler deterioration model? What is the limitation of the model?

Thank you very much.

Dear Thanh Do,

When I used the modified Ibarra-Medina-Krawinkler model at four nodes of the beam column elements using joint2d element. The parameters of these springs are determined by the equations provided by Dr. C. B Haselton dissertation and his journal paper for ductile moment resisting frames. In my one story model, the elastic beamcolumn element are used for beams and columns. Then, the modified Ibarra bilinear material is introduces as mentioned above.

The moment and chord rotation curve has terminated at or near peak moment, and  it does not show post-capping response. Similarly, when I plot the pushover curve, the curve drops suddenly after reaching peak base shear, which is almost 4 or 5 times higher than the base shear at the first cracked or change of stiffness point. After sudden change of base shear, the residual base shear goes up with increasing stiffness, instead of flattening after sudden drop. These results indicates some major issues either in material  properties, element properties, or other issues.

Please advice me what might the reason behind these unexpected results. I am looking forward getting some suggestion to improve reinforced concrete beam column joint.

Thank you for your description. I think it would be easier to debug if you test the with a simpler model, maybe just the joint itself and see what response it generates. The problem might be: 

(1) You haven't used "proper" parameters. By playing with different parameters, you might have a better sense. If the problem goes away, then maybe it is the reason. If not, then it is something else.

(2) Numerical issue with the Ibarra-Krawinkler model. Have you detected any convergence problem, i.e. no quadratic convergence or even no convergence at all? The model might not be robust in your tough case. 

I am no expert in joint modelling, but here are some thoughts on numerical simulations in general. It is easier to debug working with a smaller model. Good luck and let me know your discoveries!

Dear Mr. Than Do,

Thank you very much for your prompt response.

Actually, I use a simple one story model with the some parameters based on equations by Dr.  C. B Haselton for ductile RC frames. First, the time period was excessively high even though I change initial stiffness value. It means that the use of joint2d element does not provide the real time period or due to modeling issues. Although I changed parameters with different values, the pushover curve and Moment-curvature shows almost similar trend. Second,there is some numerical issues in pushover analysis.

It sounds like the joint2d with the Ibarra-Krawinkler has some accuracy and numerical limitations in your case. Why don't you try other models/software and compare the results? Good luck

Good question.. Please share me the best answer might you trust...

Regards…

Baikuntha Silwal

When modelling the beams and columns as elasticbeamcolumn elements which properties do we need to consider if the section is RC?

Hi Nallagatla,

Could you please share the purpose and expected result from this modeling? Why do you need this model instead of simplified model?

Please see the link below:

http://opensees.berkeley.edu/OpenSees/doc/Altoontash_Dissertation.pdf

I believe this may help for your research.

Thanks,

Baikuntha Silwal

Baikuntha Silwal

Hi Baikuntha,

I need to do Incremental Dynamic Analysis on three storied - three bay structure as a part of my master's project.Currently I am learning OpenSees,I found in one research paper they had done incremental dynamic analysis on Lumped plasticity model.Can you please tell me if it is possible simplified model other than this lumped plasticity model.

Thanks,

Nallagatla Bhagya sree

Cristian V. Miculas

If your code failing in Gravity analysis itself then I suggest checking the Eigenvalues before doing analysis, If your eigenvalues are reasonable then you can proceed to change the analysis parameters, I neither used Penalty constraint handler nor Joint2D, but I can say that selection of alphaM and alphaS matters a lot.

The below note is from the OpenSees website :

The degree to which the constraints are enforced is dependent on the penalty values chosen. Problems can arise if these values are too small (constraint not enforced strongly enough) or too large (problems associated with conditioning of the system of equations).

You can check the OpenSees User forum.