h5m faceting errors

[Lorenzo Chierici]

Dear all, 

During a .step to .h5m faceting conversion through DAGMC I got several error warning , though the execution runs till the end. 

Finally I got the following faceting summary information:

----Curve fail information -----

There were 57 curves that could not be faceted

---- All surface faceted correctly -----

How serious is that? How would I be able to detect the defective curves in the geometry? I can convert the h5m newly generated file into stl or vtk formats through mbconvert, but I'm not actually able to detect any potential problems. 

Any insights or suggestions highly appreciated. 

Thanks, 

Lorenzo 

[Andrew Davis]

Hi Lorenzo

I presume you're using Cubit? Following the workflow here or something like it - https://svalinn.github.io/DAGMC/usersguide/cubit_workflow.html?

Failing curve faceting will potentially impact your water tightness of faceting. Have you run make/check_watertight? If you're geometry is reported as watertight you're ok

Thanks

Andy

--
You received this message because you are subscribed to the Google Groups "DAGMC Users and Collaborators" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dagmc-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dagmc-users/837b71a5-2160-4a1e-ac35-dc265e7df8e6n%40googlegroups.com.

[Lorenzo Chierici]

Hi Andy,

Thank you for your answer. Yes I'm using Cubit and openmc for simulation.

When I run check_watertight after make_watertight, that's what I get:

geometry check
number of surfaces=22941
number of volumes=6

84/4072004 (0.00206287%) unmatched edges
2/22941 (0.00871802%) unsealed surfaces
2/6 (33.3333%) unsealed volumes
leaky surface ids=18440 18454  
leaky volume ids=1 5  
45.958 seconds

The strange thing is that the more I reduce the faceting_tolerance the higher the unsealed surfaces and leakages.

Thank you,

Lorenzo

To view this discussion on the web visit https://groups.google.com/d/msgid/dagmc-users/CAGNcaP_zbsYkzHBj0iNLvdzBqfL9w_R0qGGYjLWr7M9p_G_kvA%40mail.gmail.com.

[tim....@wisc.edu]

Hi Lorenzo,

You need to determine the cause of the leaky volumes reported by check_watertight.  I suggest carefully examining the geometry in Cubit after importing the step file.  Besides the leaky volumes, remember that you need to examine the surfaces, curves, and vertices that form those volumes.  Do not try to imprint and merge until the geometry is "clean".  Imprinting and merging a geometry that is not "clean" will most likely result in poorly formed models.  Most of this is described in the link Andy sent:

https://svalinn.github.io/DAGMC/usersguide/cubit_workflow.html

If you can share your step file, then we may be able to take a look at it to see if we see anything wrong. 

Tim

[Lorenzo Chierici]

Dear Tim

Thank you for your reply. I've managed to reduce the number of curves that could not be faceted to 3 and after I run check_watertight I got zero leakages and zero unsealed surfaces.

Should I consider my geometry ok then?

If I reduce the faceting_tolerance from 1e-2 to 1e-4 I get several failures during the meshing process saying "failed to fix inverted triangles in surface " but when I run check_watertight I get the same zero values. 

Any suggestions are appreciated.

Thanks,

Lorenzo

--
You received this message because you are subscribed to a topic in the Google Groups "DAGMC Users and Collaborators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/dagmc-users/mg7OGLi9YDs/unsubscribe.
To unsubscribe from this group and all its topics, send an email to dagmc-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dagmc-users/db5f1002-5c11-4f09-866d-596d5837b0d9n%40googlegroups.com.

[tim....@wisc.edu]

Lorenzo,

I think it might be helpful for Patrick to weigh in on this as he is the expert on the make_watertight process.  I personally would not feel confident in this geometry. 

Is it possible to share your original stp file so we could see what might be happening in the whole workflow process?  In the past, for the complex CAD files I worked on for ITER, I always had difficulty trying to import stp files.  The importer mostly worked but often there would be a very small number of CAD errors that were difficult/impossible to fix.  I would expect the stp importer in Cubit to be better now but I am not familiar with the details of the current version of the importer.

Tim

[Lorenzo Chierici]

Hi Tim,

If you could take a look at the step files and give an opinion that would be great. I attached the 3 bare minimum components.

Thank you very much for your support.

Regards,

Lorenzo

You received this message because you are subscribed to the Google Groups "DAGMC Users and Collaborators" group.

To unsubscribe from this group and stop receiving emails from it, send an email to dagmc-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dagmc-users/4d08ab16-0106-4a50-80ad-7ea999d744f0n%40googlegroups.com.