Non Convergence of non-linear problem

Anonymous · ‎10-09-2017

I'm having problems with my existing model while doing the P-Delta analysis. I tried going through the forum for a potential solution but did not find one. I did select "Matrix Update after each iteration" since I saw this as a solution in most of the previous treads. Unfortunately this did not help.

Please Help and thank you! Attached I have included the model.

Rafacascudo · ‎10-09-2017

this is the model of the other thread with mesh problems , isn´t it??

Rafael Medeiros
Did you find this post helpful? Feel free to Like this post.
Did your question get successfully answered? Then click on the ACCEPT SOLUTION button.

Anonymous · ‎10-09-2017

Yes, correct.

Rafacascudo · ‎10-09-2017

I mean , this one you attached now is the same still with those mesh problems

Rafael Medeiros
Did you find this post helpful? Feel free to Like this post.
Did your question get successfully answered? Then click on the ACCEPT SOLUTION button.

Anonymous · ‎10-09-2017

No, this is an updated model form the last one. I did fix the mesh issues with your proposed solution.

Rafacascudo · ‎10-09-2017

this is what I see when I open the attached model

Rafael Medeiros
Did you find this post helpful? Feel free to Like this post.
Did your question get successfully answered? Then click on the ACCEPT SOLUTION button.

Anonymous · ‎10-09-2017

That does not look right... I have attached the model again. It should be completely meshed as well as wind loading in both directions. I believe the previous model did not have any wind loads. Thanks again for your help.

Rafacascudo · ‎10-10-2017

You have unsolved instabilities type 1 in your model. Types 1 and 2 cannot be ignored. Model will not converge while there´s still unsloved instabilities.

You should watch webinar about instabilities on

Build your Robot Structural Analysis IQ - YouTube

also on Job preferences /meshing you should mark only "iterative adjustement of FE mesh". Also , when you reach a satisfying mesh , you must freeze them all. It will save you a lot of analysis time.

On the same url above you can watch webinars 1 and 2 about meshing

Rafael Medeiros
Did you find this post helpful? Feel free to Like this post.
Did your question get successfully answered? Then click on the ACCEPT SOLUTION button.

Anonymous · ‎10-11-2017

Hi Rafa

Thanks again for your response. I was able to eliminate the majority of the inabilities type 1. The location of these instabilities were at nodes I added at the center of mass of every level to apply the wind loads. I eliminated this nodes and instead added the loads to the columns and walls. This eliminated the original instabilities but after running a non P-Delta analysis I did get a instability type 1 at node 798. The problem is I can't locate this node. It doesn't appear in the node table as shown on the attached image. Also, this model does not converge on just a gravity load combination 1.2DL + 1.6LL. I don't understand how this can happen, maybe I modeled a wall panel incorrectly? Can you please take a look?

Thank You!

Rafacascudo · ‎10-11-2017

I couldn´t find it either . Probably 798 is a calculation node(internal ,not created by the user). But anyway, even with that instability warning type 1, nodal displacements seem to be in a normal range for that kind of structure.

I tried changing the Solver from automatic(which is recommended) to Skyline , and to my surprise there was an error warning.

those nodes have horizontal supports in the same plane of the slab diaphragm. That´s the reason. I deleted the support assignment , but it is still giving an error execution problem 1901 and the same instability warnings.

So , certainly official support will know what is going on with your model

Other than that you should fix some spots in your mesh. There are some like this below

if you look closer and turn the contour points on ,you can see that there are 2 less than 1mm apart.

Rafael Medeiros
Did you find this post helpful? Feel free to Like this post.
Did your question get successfully answered? Then click on the ACCEPT SOLUTION button.

Artur.Kosakowski · ‎10-12-2017

Hi @Anonymous

Try these settings:

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

Artur Kosakowski

Anonymous · ‎10-12-2017

Thanks again Rafa for your swift response. I did fix the Mesh issues but still can't get my model to work.

Anonymous · ‎10-12-2017

Artur

I still can't get a convergence with these settings. Please Help!

Artur.Kosakowski · ‎10-12-2017

Hi @Anonymous

The model with the results for the combination 15.

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

Artur Kosakowski

Anonymous · ‎10-12-2017

Hi Artur

Can you elaborate and what you did to the model to make it work?

Thank You!

Artur.Kosakowski · ‎10-13-2017

Hi @Anonymous

The only changes were the nonlinear analysis parameters which I already posted picture for and switching on DSC algorithm in Job Preferences.

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

Artur Kosakowski

Anonymous · ‎10-16-2017

Unfortunately these parameters don't work for the other load combinations that include Wind forces. Any other ideas?

Artur.Kosakowski · ‎10-17-2017

Hi @Anonymous

Check this file (I used the correction to structural axes and slightly increased the model generation tolerance).

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

Artur Kosakowski

Community

Non Convergence of non-linear problem

Non Convergence of non-linear problem

Non Convergence of non-linear problem