second order analysis (no convergence)

[admamouhiba]

Hello,

I have read all the threads mentioning the P-delta analysis and second order effects. However I can't find a solution to the convergence problem in my model, I have changed the tolerance and the increments number too. But I always get a bunch of instabilities and the "no convergence" error.

please note that I did not have any instabilities with the buckling analysis, and I discretized all the model elements for more precision in the buckling results. I find that for the first mode the critical factor<3, which means my structure is "sway" and that I need to take into account the non linearity . I have created a seperate load case "second ordre" which is the same as "ELU+" , the  load case I am interested in. and I set the first mode I obtain from the buckiling analysis with "ELU+" to be the initial deflection for the non linear analysis.

I don't know why I cannot attach my robot file, but here are some screen shots. I apprecaiate any help.

Thanks in advance.

@Pawel.Pulak @mustafahesenow 

[Krzysztof_Wasik]

Hi @admamouhiba 

Share model for verification. Save it as RTD without results then Zip before sharing.

[admamouhiba]

Thanks for the reply @Krzysztof_Wasik

here is the file.

[Krzysztof_Wasik]

Hi @admamouhiba 

Shared file does not contain any model. Try onece again

[admamouhiba]

@Krzysztof_Wasik 

I hope it works this time.

[Krzysztof_Wasik · Accepted solution]

HI @admamouhiba 

In this case I would recommend switch ing off DSC option in Job preferences. Keep in mind that forces applied in case 10 are bigger than buckling load

[admamouhiba]

I am afraid that does not solve the problem @Krzysztof_Wasik . if you look at attached screenshot, critical factor is 1.28, which means load case 8 needs to be multiplied by 1.28(I understand it's very small) to reach the buckling load, meaning it's not bigger than buckling load.

one more question, why do I get instabilies after running the P-delta anlysis and what does the "no convergence" mean? and what happens if the load case is bigger than the buckling load?

[Krzysztof_Wasik]

Hi @admamouhiba 

In your model second order analysis is set for case 11. which has loads much bigger than in load case 8

When DSC is off case 11 converges, so results can be obtained for all load cases.

Could you explain problem with more details.

[admamouhiba]

I see thank you so much. But shouldn't DSC be on because  I have releases in my model?

[Krzysztof_Wasik · Accepted solution]

Hi @admamouhiba 

When in releases fixed/released setting is used (no elasticity and nonlinear behavior are defined) DSC might become deactivated while analysis.

[admamouhiba]

Hello @Krzysztof_Wasik ,

Why is it that when I define the buckling deflection L/200, I get instabilities and the calculations don't converge? Please note that I corrected the load for the non linear analysis to be the same as for the buckling analysis.

Thanks a lot.