Pushdown analysis in steel moment frames using SAP2000 1

[theosia]

Does anyone knows how to do pushdown analysis using SAP2000?

My analysis is to investigate the progressive collapse in a steel moment frame using sap2000
The procedure is to remove for example the corner column and apply the load combination that is provided by the GSA2003 guidelines with amplification factor 2(D.L+0.25L.L) and i have to get as a result the pushdown curve with load factor against displacement.
I have to increase the load gradually and push with load control the joint of the column that i had remove!
The analysis is nonlinear static analysis,and i had problem when i defined the analysis. I can't get the results

 

[ishvaaag]

Please consider that if your structure and considered cases do not converge to solution to the extent of the load grows, it is not finding solution and hence won't be showing results.

 

[theosia]

thanks for you reply, the problem is I don't do the appropriate procedure to give the results, if you have any idea how to do the procedure that will be very useful
Im confused with the way to do the pushdown analaysis and get the curve. Im not sure if applied the correct load cases with the correct amplification factor
thanks

 

[ishvaaag]

The one thing that SAP2000 won't be making is automatic removal of members whilst the structure deteriorates.

In your question maybe (I have not tried) one could proceed this way:

1 Subdivide heavily in segments your structure, to the extent that the analyzed segments are stub-like and unlikely to buckle themselves.

2 Define customized nonlinear material to huge strains showing almost nil stiffness there

3 Set your loads in such way that the progression of loadings can be made apparent. Maybe one way can be staged construction or loading, where you can state the loads added at each stage, and see the results per stage

4 Ensure the analysis is nonlinear and P-Delta

In this way you will get to the equilibrium if feasible without collapse with maybe some members dangling buckling or something so whilst other standing members are showing stiffness enough to get equilibrium and keep the thing working. As soon as no equilibrium is met somewhere SAP2000 won't be providing results.

Hence quite likely as an engineer you need to outsmart the code by dimensioning the members in such a way -as stiff and in such disposition- that the mandatory cases enforced by the code to be met by your structure do so with your foreseen secondary structural mechanisms, there available for the cases to check. Then, analyses like those proposed above will show the suppressed parts as redundant for the wanted behavior when such thing happens, since the program will be showing equilibrium. By staging the loads, the program will be revealing then the stability you will have been willingly introducing with your mind.

This last paragraph refers mainly to my suspicion of that what simply is happening in your case is that your damaged structure shows not to have such secondary mechanism in place, and then, the model lacking convergence to solution, the program does not show results.

 

[dik]

Very good explanatiov.

Dik

 

[theosia]

Thanks for you procedure that you have noticed me to follow.
Basic the problem is where i defined my load case, i dont know if I increment my applied load correct
I have attached two files to see what i have done
I hope these information will help better to discuss my problem
Thanks

 

[ishvaaag]

I have perused your added info. One first observation that comes to my mind is that depending on the state when the column is suppressed, hinges may form firs atop (it may not need to be necessary your case). It is usual (or at least can be that) the central column shorten more than outer, what can lead to relative lower reserve capacity at top corners than elsewhere, but of course that depends on members and loads standing.

https://wiki.csiberkeley.com/display/kb/Progressive+collapse+analysis

Above a reference of CSI on pushdown.

Will be gathering info and continue investigating your issue.

 

[theosia]

Thank you so much for your response and your comments!its very helpful! I will check what u have sent me and i will inform you about my progress!
its very useful!
thanks

 

[ishvaaag]

I also can think an scenario on why when setting the limit of the analysis in terms of a limit load you may not be finding solution whereas when setting a limit solution in terms of a limit displacement you find. Simply, the solution pertaining to the limit load may need levels of displacement superior to those stated in your alternative analysis on progressive displacements to some limit. So you may not find a solution for the limit load sinc the structure not able to get equilibrium for the limit load, whereas yes for the limit displacement.

On the chart, I see the one proposed is not a direct product of SAP2000, or if it is, it seems to have been at least relabeled. In any case, one should be able to link the laod factor to the load being applied at the moment and then make the plot...