The Analysis has stopped because the Solver has exited unexpectedly

[h.schlichter]

I have a very simple model for computing natural convection in an air volume. After calculating thermal radiation the analysis fails with "The Analysis has stopped because the Solver has exited unexpectedly". I have no idea about tha reason for this error. Attached you find my model.

Regards

H. Schlichter

[marwan_azzam]

Please verify the direction of Gravity in the Solve dialog Physics tab (nan = not a number).

Also, verify the dimensions of your model.  A 2 Km piece of glass is quite impressive!

[h.schlichter]

Thank yo for your answer.

• Some times ago I already detected the problem of "disturbing" the gravity direction. At the start of the analysis I definitely set the direction to 0,0,-1. It must have been changed by the faulty analysis.
• Admittedly I did not control the dimensions of my CFD model, because I always switch of the measuring tape. But I do not understand how the measures in th CFD model occured. I exported the model directly from SimStudio Tools, and there the glass wall is 20 meters high (and this is intended). I attach the SimStudio model.

By the way, is there any possibility to measure coordinates or distances in the CFD model?

Regards

H. Schlichter

[h.schlichter]

I found this post:

https://forums.autodesk.com/t5/cfd-forum/cfd-2017-update-1-model-dimensions-in-mm-incorrect/m-p/6784...

Obviously it is a bug in SimStudio/CFD. Is there any fix available?

[Amal.C]

Hi @h.schlichter,

Regarding the value of the gravity coordinates, please use the comma next to the "M" letter on the German Keyboard. I saw that this problem happens when the user sets these coordinates using the comma of the numerical pad. 

As far as the Simstudio Tools issue, we currently recommend that you export your CAD Model into SAT or STEP and then import it into CFD. We apologize for this inconvenience. 

I hope this helps.

Thanks,

Amal

[h.schlichter]

Thank you for your response, but I never use the numeric keypad. In this case I use backspace to erase the default "0" and then "-1", but I do not use the numeric keypad.

Regards

H. Schlichter

[h.schlichter]

I exported my model from SimStudio as iges and imported to CFD. Now it seems to work. But what if I want to actualize an existing design study? The bug should be fixed as soon as possible.

Regards

H. Schlichter

[Amal.C · Accepted solution]

Hi @h.schlichter,

Please delete everything you have in the gravity field and set manually the numbers as well as the comma (next to the M letter). It should work in this way 🙂

I just tested it on your model and was able to launch the simulation for some iterations without solver crash.

In case you want to actualize an existing study, importing the geometry should be the same:

Add/Update Design > Browse > Choose the file extension > Select it >  Update selected Design

I would like to inform you that we are currently working on it and it is indeed set as high priority 🙂 

Thanks,

Amal

[h.schlichter]

Thank you for your response. For information: the only possibility to launch CFD from SimStudio without scaling error is to construct the model in cm!

Regards

H. Schlichter

[Amal.C]

Hi @h.schlichter,

Thanks for sharing this 🙂 

Regarding the issue with the gravity vector, I would like to check if you were finally able to launch your simulation without the solver crash. 

Please let me know if you still need help.

Thanks,

Amal

[h.schlichter]

Hi Amal,

I think the problem with the gravity vector was caused by the crashed simulation. After the crash the gravity vector was nan,nan,nan.

I could start and finish my simulation successfully by one of the two methods:

1. Exporting the model from SimStudio as 'iges' and reimporting to CFD
2. Constructing the model in centimeter and lauching CFD (see my last post)

But it's a workaround, not a solution.

Regards

H. Schlichter

[Amal.C]

Hi @h.schlichter,

Thanks for your feedback.

We are dealing here with two different issues:

1. Clicking on the Solve button leads to changing the coordinates of the gravity vector to nan,nan,nan: This was causing the crash of the simulation. I am glad to hear that it was finally resolved and that you were able to run your simulation
2. Launching a geometry into CFD directly from Simstudio Tools displays incorrect scaling in CFD: It is a known issue which has no relationship with the simulation crash. I agree that what we currently have is a workaround. The Development Team is working to fix it as soon as possible.

I hope this helps 🙂

Thanks,

Amal

[h.schlichter]

I must cancel my last statement. I constructed a model in centimeter in SimStudio and launched CFD directly. The model was scaled by a factor of 100. The only way is exporting as 'iges' and reimporting.

[h.schlichter · Accepted solution]

The crash was caused by a scaling error in CFD 2017. In CFD 2018 it did not occur up to now. Case can be closed.

H. Schlichter