Tracking down model geometry errors on failed analysis

[andrewdroth]

After a simulation fails it can be tricky to find what part of the model caused the error. It is particularly cumbersome when the issues lies within a large part with thousands of lines.

Currently we look at the log and check which part it stumbled on and note the element #.

We then do a Check Model, and search for that element numer using the Specify feture in the Element Information dialog. That returns some node numbers

We then use the Current Results dialog to specify the node number, and with that we can track down the location of one of the nodes listed as it is now highlighted.

Is there an easier way to track down problem areas?

[AstroJohnPE · Accepted solution]

Hi Andrew,

You should be able to eliminate a couple of your steps.

1. When you specify the element ("Results Inquire: Element Information: Specify"), the element is selected and highlighted. There is no need to then inquire on the nodes (although the nodes also provide the coordinate, so you have a better idea of where to look).
2. If you do want the nodes, use "Selection > Subentities > Nodes". No need to do any typing.
3. Since the analysis was started, the model should already be "checked". You should be able to click on the "Results" tab instead of doing the "Analysis > Analysis > Check Model".

[andrewdroth]

Thanks John,

That does speed up the process.

For some reason I didn't think the results tab was availible for selection of a failed analysis. I'll have to try that next time.

Andrew