Seismosoft | Earthquake Engineering Solutions

I have modeled four girders of bridge sections and connected them with dummies. The bridge girders are in the X-Y plane. I am applying truckload in -ve Z direction and doing a pushover analysis. I had a similar model in SAP2000. I have done the calibration for linear analysis and have almost similar deflections to the original model.

For non-linear analysis, I am performing a pushover analysis of truck loads. I have three modes of failure for my model considering failure
whichever is reached first: 1) Crushing of concrete when compression strain reaches -0.0035, 2) Failure in reinforcement when strain reaches 10%, and 3) When deflection at any node reaches 0.0075 times the span length i.e. here 0.09m as my span length is 12m.

For pushover analysis, I have applied loads of 500 kN and divided them into 500 steps i.e. I am increasing the load by 1kN at each step. The failure in
concrete is obtained at the 347th step. However, the deflections are more than they should be.

I am using the Load Control loading phase with 1.0 target load factor and 500 steps. It keeps on running until failure in the material is reached. Also, I have many nodes whose deflection I need to keep in check in order to consider the failure.

When I try to have response control at one node the other nodes have deflection more than 0.09m in -ve Z-direction.
Can anyone please help me to define the failure deflection for multiple nodes or stop analysis when it is observed?

When the response is highly nonlinear very small differences in the load lead to very large differences in the deformations. For such cases the Response Control, whereby the defrormations are checked, is usually a better option.
Not sure what you mean with "When I try to have response control at one node the other nodes have deflection more than 0.09m in -ve Z-direction." however you need to control the node with the larger deformation increase at the critical steps.
Seismosoft Support

Thank you for your reply.

As I am having four girders in the bridge. I have four control nodes at the middle of the bridge for each girder.

I am doing the pushover analysis with different load cases in which I am moving trucks in the transverse direction of the bridge. So when I change the load position I am not sure which of my girder will first reach the deflection limit.

I have tried with the response control phase defining it for all the four nodes in the middle of girders however it ends up doing a pull-push analysis instead of stopping the analysis after any of the four control nodes has reached the deflection limit first.

I wanted to know is there any way to keep in check the deflection of all the four girders in a way that when any of them reaches its deflection the SeismoStruct stops the analysis instead of doing a pull-push analysis.

Possibly, you could use SeismoStruct's performance criteria with the Stop or Pause type of notification.
Seismosoft Support

I have defined two performance criteria for failure in steel and concrete.
I am not sure how to define performance criteria to control deflection in girder.
Could you please elaborate on how to define performance criteria to control the deflection.

Unfortunately the current version of SeismoStruct does not have drift based performance criteria. You could possibly check these indirectly through the chord rotation of the piers.
Seismosoft Support

Can the pushover curve from a static pushover analysis be downloaded as an excel file?

Dear jessie_7,

You may copy all the results available in table format and paste them to an excel file.

Thank you

Thank you very much for your reply.

I have calibrated my bridge model for static analysis with the same bridge model which is modeled as grillage in SAP2000. The results are almost similar with only a 4 % difference. However, when I perform the Static Pushover analysis I am getting the failure at much higher loads as compared to the SAP2000 model, and when I tried to check the deflection at the load step for which the SAP2000 model gives me failure it is 11 times higher deflection in SeismoStruct.

Could you please suggest to me any solutions because I have rechecked my model many times but I have no idea what to do?

Thank you in advance.
Dipesh