Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Internal Error In Geometry Composition

Please login with a confirmed email address before reporting spam

The below post is related to an archived discussion


Similar to what was said in the linked post above (which hase been closed), Lately I am spending around 70% of my time fighting the software because when building geometries the boolean operation (the last stage in building a geometry) fails to work.

The error occurs even when using only simple shapes that intersect at straigh angles (nothing complex like the intersection of doubly curved surfaces). The error message produced is not helpfull and its occurance is very inconsistent. This is the biggest issue I have had with the software, and the problem is still here after years and years! We have tried building geometries in an extarnal CAD program and importing, but this will lead to similar behaviour with unextpected/inconsistent errors popping up when the time comes to do the boolean operation (and finaly start the actual work).

This is a call to the software team to give this issue TOP PRIORITY. In my opinion more time should be given to fixing this issues instead of adding more modules, more features or changing them.

Please respond if you share this opinion, so the extend of the issue can be made clear to the developers. I know fellow workers share the same experience (and I suspect that there are more like us out there).

Kind Regards, Marten Nijhof


2 Replies Last Post Aug 3, 2018, 4:12 p.m. EDT
Walter Frei COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 6 years ago Aug 3, 2018, 12:02 p.m. EDT

Dear Marten,

Our development team is indeed constantly working on improving the geometry algorithms, and we can say that these types of errors are getting fewer and fewer with each software release.

What would be particularly helpful here is for you to send us samples of the geometries that are causing you issues. If you want these cases to be held in confidence, please send them to support@comsol.com. Without seeing specific cases that are causing issues, it is quite difficult to address otherwise. Also, our support team will usually also be able to suggest workarounds.

Best Regards, Walter Frei COMSOL Support

Dear Marten, Our development team is indeed constantly working on improving the geometry algorithms, and we can say that these types of errors are getting fewer and fewer with each software release. What would be particularly helpful here is for you to send us samples of the geometries that are causing you issues. If you want these cases to be held in confidence, please send them to support@comsol.com. Without seeing specific cases that are causing issues, it is quite difficult to address otherwise. Also, our support team will usually also be able to suggest workarounds. Best Regards, Walter Frei COMSOL Support

Please login with a confirmed email address before reporting spam

Posted: 6 years ago Aug 3, 2018, 4:12 p.m. EDT

Hi,

I understand your situation. I have had numerous such errors over the years because I was trying to load about 1000+ spheres with an outer thin layer around each sphere in a representative volume element (to solve diffusion-reaction problem). The problem happend when three or more concentric spheres spatially intersect, which led to degenerate geometry and I could not proceed any further, no matter what. I normally seek help from COMSOL, but in this instance, I decided not to because I shifted my strategy to particle based methods. You might wonder why I had to include a thin layer explicitly around each spehere rather than treating them as implicit layer using tangential derivatives. However, my goal was to understand some phenomenon within the layers and moreover for some spheres the aspect ratio between the concentric spheres were not so big and thus could introduce inaccuracies.

If I had not changed my strategy, I would have certainly sought help from COMSOL team, who have provided me excellent support over the last 10 years or so. I hope you will be able to resolve your problem, but it will be nice to post the solution for our benefit.

Suresh

Hi, I understand your situation. I have had numerous such errors over the years because I was trying to load about 1000+ spheres with an outer thin layer around each sphere in a representative volume element (to solve diffusion-reaction problem). The problem happend when three or more concentric spheres spatially intersect, which led to degenerate geometry and I could not proceed any further, no matter what. I normally seek help from COMSOL, but in this instance, I decided not to because I shifted my strategy to particle based methods. You might wonder why I had to include a thin layer explicitly around each spehere rather than treating them as implicit layer using tangential derivatives. However, my goal was to understand some phenomenon within the layers and moreover for some spheres the aspect ratio between the concentric spheres were not so big and thus could introduce inaccuracies. If I had not changed my strategy, I would have certainly sought help from COMSOL team, who have provided me excellent support over the last 10 years or so. I hope you will be able to resolve your problem, but it will be nice to post the solution for our benefit. Suresh

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.