Negative tetrahedral volumes

[sitanshu iyengar]

Hi all,

I am doing a research on the bone fracture mechanisms of trabecular bone.

I have a series of micro-CT images that need to be meshed so that i can simulate the volumes on ABAQUS.

i meshed the volume using cgalmesh function and on visual inspection found nothing wrong, however, ABAQUS says that the mesh has negative or very small tetrahedral elements.

any help on how to get over this problem will be greatly appreciated.

Thank you.

Sincerely

Sitanshu

[Fang, Qianqian]

For some reason, my below reply did not get posted on Google Groups. Resending it again.

From: Fang, Qianqian
Sent: Tuesday, September 8, 2020 12:34 PM
To: iso2mes...@googlegroups.com; sitanshu iyengar
Subject: Re: [iso2mesh-users] Negative tetrahedral volumes

hi Sitanshu,

check two things:

First, check to see if you have exported the element label (5th column of elem, if present) correctly - often times, if the label was treated as node indices, it can confuses the validator.

Next, call meshreorient.m to ensure all elements have positive volume, like

newelem=meshreorient(node(:,1:3),elem(:,1:4));

see

https://github.com/fangq/iso2mesh/blob/master/meshreorient.m

let me know if this can fix the problem.

Qianqian

 

Thank you.

Sincerely

Sitanshu

--
You received this message because you are subscribed to the Google Groups "iso2mesh-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to iso2mesh-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/iso2mesh-users/df1a44ab-f9ec-4f86-95a7-cee144f4e0e7n%40googlegroups.com.

 

 

[filip]

I encountered the same problem and found a solution. Before opening elem in Abaqus, I swapped the 4th and 3rd columns. Mr. Fang was right; it turned out to be a bookkeeping issue as he mentioned in his earlier posts about the same problem.

elem(:,[1,2,3,4])=elem(:,[1,2,4,3]);

Filip

[Fang, Qianqian]

That’s a helpful comment, thanks for sharing your successful conversion.

 

From your comment, looks like Abaqus assumes the node orientation of the tet is different from iso2mesh – iso2mesh assumes the node orientation so that it’s volume is positive, looks like Abaqus assumes the opposite (negative volume). Flipping any two columns will flip the sign of the volume.

To view this discussion on the web visit https://groups.google.com/d/msgid/iso2mesh-users/785dd7cf-9342-4b2b-9525-8f520db9a9a7n%40googlegroups.com.