• Registered: 2013-01-22
  • Posts: 10

Topic: Convergence Issues

Hi Atena Users/Admin.,

I will give an insight on how my model is setup to start out. I am modeling a beam that has 3 voids running through the length of the web and each void has 16 sides (making a shape similar to a circle). I am applying an area load to 7 sides of the 16 sides, the lowest 7 faces, for the middle void first then moving to the top void, both in different loading cases (the bottom void does not have any loading). After those two loading cases, I am applying another area load to all 16 faces on the middle void and the following case consists of area load for all 16 faces on top void. The analysis steps in summary are as follows:

1: body force (dead load) and supports applied
2: -0.4455 kip/in line load at each end of the beam
3: -0.307 ksi area load on bottom 7 faces middle void
4: -0.110 ksi area load on bottom 7 faces top void
5: -0.125 ksi area load on ALL 16 faces middle void
6: take load step 5 off
7: -0.125 ksi area load on ALL 16 faces top void
8: take load step 7 off

The beam is supported in 4 locations (fixed all directions) all 1ft. off of center in two directions of the beam (making a square) and are geometrically uniform. I have gotten the same model to run but where load steps 3 and 4 were being applied at the same time. Since I have updated it to where they are being applied at different times, I have been having convergence issues during load step 4.

I have changed my mesh to see if that would help and have created tried a Modified Newton-Raphson solution parameters. The Mod. NR gets me further in the analysis but still does not converge. I have increased the number of steps to help as well but that has not.

Any help would be great. I am not sure why it will not converge anymore. Hopefully I was clear on explaining the situation but don't hesitate if clarification is needed. Please and thank you for the help!

Quote Post 1

  • dpryl
  • CC software developer
  • Offline
  • Registered: 2007-07-13
  • Posts: 1,107

Re: Convergence Issues

Hello,
please see ATENA Troubleshooting, 2.2.6 Floating point problems - Division by zero, Loss of >7 digits on the matrix diagonal, Warning: Sign of diagonal changed when reducing equation, Floating point exception: Multiple floating points traps, Zero or negative jacobian, InvertA: zero determinant, CCMaterials Extended AExc: CC5ParamYield F:project_stress_on_f_division by zero, and similar and 2.2.2 I get the message "The execution is killed due to violation of stop iteration criteria", what does it mean? The latest version of the document is available from our web under Products - ATENA - Documentation.

If you still can not find the problem source, send us your model etc. following the instructions in 2.1.1 I have a problem not listed here.

dpryl's Website

Quote Post 2

3 Reply by baallenba (edited by baallenba 2014-05-12 20:07:22)

  • Registered: 2013-01-22
  • Posts: 10

Re: Convergence Issues

I have figured out the convergence issue, that is all solved. Some loading was wrong. Now that the model is ran all the way through, I am in the post-processor trying to look at different steps to see how the structures reaction changed through the loading steps and when I am going to a different step, I get a popup that says "Catastrophic failure." The structure disappears and doesn't allow me to see anything. If I click back on the last step, the structure reappears. This has happened with a previous model as well but I figured it might just be isolated to that model. Why am I getting this catastrophic failure?

I am using Atena3D as well. It is updated to the current version.

Thank you.

Quote Post 3

  • dpryl
  • CC software developer
  • Offline
  • Registered: 2007-07-13
  • Posts: 1,107

Re: Convergence Issues

This sounds like hitting the 2GB memory limit for 32-bit Windows applications. You can send us your model such that we can try to reproduce and investigate the problem source (see Troubleshooting 2.1.1).

dpryl's Website

Quote Post 4