I was wondering how you expect the response would look like if you perform a typical cyclic test on the element? Can you please provide a quick sketch?

Dear Thanh Do,

To model one story RC MRF with non-ductile effect, I use the beam-column joint element parameters based on the A.B Liel journal papers related to Seismic Collapse Safety of Reinforced Concrete Buildings:II. Comparative Assessment of Non-Ductile and Ductile Moment Frames. Then I used the elastic beam column element with lumped plasticity model at the end of beam and column elements. I use the modified Ibarra bilinear material models at the end of the elastic elements. Then, when I plot the moment curvature response, the plot terminated after reaching peak moments with our any post-capping response. Similarly, the pushover curve shows very high shear force and drops suddenly after reaching peak base shear. The peak base shear is very high compared the base shear at the cracked point or first change of stiffness point.

I am looking forward getting some modeling advice and selection appropriate material model.

Hi

Use fibre-based approach !

By this way you can consider an unconfined stress-strain relationship, I mean spread plasticity is recommended. I recommend you to use OpenSEES.

Hi Sadjad,

I am  pleased you provide me a good advice.

I would like to inform you that I use the OpenSees to model beam and column elements with the spread plasticity as you recommended for one story simple model. Then I employed Joint2d element to represent beam column joint. However, I have some issued regarding material properties, time period, and push over curve. First, I used the modified Ibarra material at the end of beam and column with different values, but I got the same time period for all different initial stiffness parameters. Next, I got reasonable response up to the first yield point (i.e. Baser shear (V) =0.2W) but very high base shear (V =1.0W) in post-yielding region up to peak response). Then, the response drops suddenly and vertically almost zero or some residual base shear with variation of parameters. I hope I am expecting loss of stiffness after peak, but I got such undesired response.

Or, do I have to use two different inelastic springs using the modified Ibarra bilinear unixial material due to non-ductile joint: one at the end of beams and columns, the four inelastic springs for beam column joints? In fact, I am using only one spring in my model. 

Dear Sadjad,

I model one story RC frame with fiber based approach for nonlinear beam and column elements, and joint2d elements for beam-column joint with four springs based on the modified Ibarra bilinear models and elastic shear spring. Although I got some improvement in the response, the selection of bilinear parameters for these four springs or ignoring some other modeling considerations in the model. In deed, These bilinear parameters are selected from dissertation or journal papers by Dr. Curt Haselton. The result shows very high stiffness or very high shear after first crack point, which is undesirable.