Caedium crashes when attempting to use Force Monitors on a *.stp file

Submitted by malteahrens on January 8, 2011 - 11:27

Problems

Over the past days I have been trying to run CAE simulations on a file generated in CATIA. I have been exporting the file from CATIA as a *.stp file and then importing it into Caedium. Applying the Velocity (U) vector field to the object works, but when seeking to apply a Vector Variable Monitor like Force (F), Caedium crashes (see https://cid-a45fe702de180b23.office.live.com/browse.aspx/Public/Engineer... for a screenshot, Windows' Debug output, and a sample *.stp file generated by CATIA that fails in Caedium).

Here's what I did in Caedium (brief):
1. Import file
2. Apply the volume substance (air) to the outer box / volume
3. Change all the fluid conditions as in the "Flow over a rotating wheel..." tutorial
4. Apply the wall condition to all the faces of volume
5. Add an inlet & outlet (on opposite sides) to 2 of the faces of volume
6. Add a linear velocity to the "ground" face of volume
7. Apply Accuracy:High to the faces of the inner box / volume_1
8. Attempt to add a Force Monitory of "Vector Variables":F

This only seems to apply to CATIA's exported files, as, for example, I am able to complete the "Flow Over a Rotating Wheel with Moving Ground" tutorial, using its *.brep file, without any problems (this includes Force Monitors).

Any help would be appreciated.

Malte

Edit: This is fixed in Caedium v3.0.2.

Confirmed bug, but you don't have a single volume

Submitted by symscape on January 8, 2011 - 12:49.

Thanks for the files and the bug report with which I've been able to reproduce the crash you are seeing. The next version of Caedium will include a fix for this problem.

However, the inner box is not part of your simulation. You need to subtract the inner box from the outer box to form a single flow volume. As things are you are asking for the fluid force on a surface that isn't part of a simulation and that's causing the crash.

Now have a single volume (working), but getting RANS Flow failed

Submitted by malteahrens on January 9, 2011 - 12:19.

Ah, that makes sense. Thanks for the help. Updating that part in CATIA and bringing it back into Caedium worked!

So, I have begun moving to "slightly more advanced" shapes, trying to emulate the 'wheel in a box' shape from the "Flow over a rotating wheel..." tutorial. I am able to insert it (my CATIA version of it, see: https://cid-a45fe702de180b23.office.live.com/self.aspx/Public/Engineerin... ) into Caedium, and run both Vector Variables and Vector Fields.

However, when I apply a 'angular velocity' onto the 'wheel', as in the tutorial, and then run / 'play' the simluation, I get the following error:

Warning:
RANS Flow failed
Assertion Failed.
Expression: is(UP_TO_DATE)
File: ..\..\..\trunk\src\of\ofcase.cpp
Line: 575

Possible Workaround

Submitted by symscape on January 9, 2011 - 12:55.

Are you running the latest version of Caedium v2.4.2? To check - display the About Dialog using Help->About.

I tried setting up a case with your geometry and it seemed to work fine, then I applied the Angular Velocity and I didn't see the error.

As a workaround, try saving the project (.sym) and reverting (rereading it) using File->Revert. If you still see the error can you share this project so I can take a look?

"Just" seems to work

Submitted by malteahrens on January 9, 2011 - 20:49.

Yes, I am running Caedium v.2.4.2. I did a revert on the solution and that did not work. However, when I created a new solution and did everything again, it just seemed to work and do the Angular Velocity fine without an error.

This seems a bit strange, considering that I repeated (i.e. created a new solution with the same part and same settings to see if that would work) the simlulation yestereday at least 3 times during which it always returned that error, and now it is working, with no noticable difference in how I am creating the solution. But, at least it works!

Thank you for your help!

Still lIke to take a look at project file

Submitted by symscape on January 10, 2011 - 08:18.

Great to hear you are up and running, but if you have the Caedium project file (.sym) for the case with the error I'd still like to take a look - it sounds like a bug to me.

The link to the .sym file

Submitted by malteahrens on January 10, 2011 - 11:55.

Sure, here is the link: https://cid-a45fe702de180b23.office.live.com/self.aspx/Public/Engineerin...

Animated Axes Causing Error

Submitted by symscape on January 10, 2011 - 14:03.

Great thanks for that. I took a look and I see the problem. It's not really a bug, more of a unexpected side-effect that I've not explained very well.

If you perform a transform when the Simulation Time isn't at the beginning (Time = 0) you are actually animating (moving with respect to time) the entity you transformed. So in your case you animated the rotation axes, so it moved as the simulation progressed and this caused a condition update to trigger and throw off the RANS Flow solver resulting in the error you reported.

If you don't want to animate the axes, which I presume you don't, and you are mid-way through a simulation (i.e., Time > 0) you will need to:

Click the Lock button on the Home toolbar - to disconnect the solver from the simulation time control
Click the Start button on the toolbar - to return the simulation to t = 0
Create and transform your geometry or axes
Toggle the Lock button on the toolbar - to return to your original simulation time
Resume the simulation

I know it's the opposite of what you want, but it might help to see an animation example to help clarify why this feature is present.

Of course! It makes sense now.

Submitted by malteahrens on January 12, 2011 - 04:31.

That is very logical, thanks for the explanation (the animation helped!).

Symscape