ISSUE RIGID DIAPHRAGM MASONRY SHELL

[mmc]

Dear forum members,

I am modelling masonry arcades using shell elements. I have used a rigid diaphragm constraint to simulate the effect of the roof on top of the arcades being rigid. I have constrained the master node by locking Uz, Rx and Ry. After that, I set up the modal analysis steps, requesting 6 modes of vibration.

However, when I submit the analysis, a warning message appears indicating that 'MumpsParallelSolver::solve(void) - Error -3 returned in substitution dmumps()'. However, the results I obtained only show one local mode, which does not make sense. I have tried other solvers and elastic-plastic materials, and either the analysis does not run or the results are incorrect.

I would be grateful if someone could point out where the error is. Please find attached a simplified version of the model that I am trying to run.

I look forward to receiving your feedback and thank you in advance.

[STKO Team]

I don't see any error, try to update your opensees and/or STKO.

modes_mason.png (264.79 KiB) Viewed 569 times

[mmc]

Hi STKO Team,

Thanks for answering so fast.

Thats weird. When I run the analysis, I get the Mumps error and the analysis continues until I get the results. However, as you can see in the screenshot, there is definetly a problem with the solver. Also, the period is super high.

I have STKO 4.1.0 and Opensees 3.7.2 downloaded from prof Nicola Tarque's.

[STKO Team]

That's weird... I tested it with those exect versions and it works fine on my side.
1) Make sure to you have the correct latest versions: just re-download them to be sure
2) maker sure the file you sent here to the forum is the one you are testing

[mmc]

I have reinstalled both STKO and OpenSees using the 3.7.2 file from here: https://www.researchgate.net/publicatio ... ltext=true.
I have run the same file uploaded.
I get the same results...
I am not sure what it might be happening here.
I run the file from here using shell elements and rigid diap and I dont have any problems: viewtopic.php?p=10658&hilit=rigid+diaf#p10658
I have changed the element property from ASDShellT3 to ShellNLDKGT and I get your moral shapes.

[STKO Team]

I just tested with those exact versions and I don't see any problem

[mmc]

Sorry, I have tried to run the model in 3 different computers and it does not run correctly. We have checked that both STKO and OpenSees are updated.

[STKO Team]

But from the output of your commandline it seems you are running an eigenvalue analysis with more than 1 processor.
Is it true?
In that case, OpenSees eigen command can only run with 1 processor

[mmc]

The model (uploaded at the beginning of this thread) has one partition. I run the analysis with openseesmp. I havent had problems with this. Actually, if I change the type of element to ShellDKGT (for example), the model works.

[STKO Team]

It's really strange because I cannot reproduce the same error on my side.
Can you show the version of OpenSees you are using? (at the beginning of the command line when you run opensees).

if you want, contact info@asdeasoft.com, and we can arrange a meeting to see what the problem is