Unable to see the element number for the negative determinant

Hello,

Im trying to simulate a dam breach along a river. I was able to get it to run and work how I wanted to with a rough mesh of 3m. However, when I make the fine smaller to get better results (1m) I am getting the error "Negative determinant. In the telemac console the error shown as follows: “GEOELT: ELEMENT ****** : NEGATIVE DETERMINANT”. I have read the other posts about the Negative determinants. I have analysed my mesh and did not find any elements with 0 area or clockwise elements.

Any suggestions as on what might be wrong with the mesh or the setup of the telemac files.

PS: I have attached the relevant files

Leimbach_nth-2.cas (3.34 KB)
DGM_Leimbach.cli (749 KB)

No files :expressionless:

Fine mesh means more precision…

Did you make you analysis on the slf file?

Sometimes rounding truncation could leads to this problem

Hope this helps

Sorry for the lack of files, i get an error that the size of files is too large even though it is under 100mb. I’m try to upload them ASAP.

yes, in the cas file i did use the analyse mesh option but there was no information provided.

I’ll check the truncation. Thank you for the suggestion.

EDIT: The slf file is 200mb and too large to be uploaded and I get the same error message when I try to uplaod the ZIP file (67Mb). Idk if this would help, but im attaching the ss of the mesh analysis from BK


probabily the IELEM with negative det has a value larger than 999999:

in sources\utils\bief\geoelt.f

...

        IF(DET.LT.1.D-20) THEN

          WRITE(LU,99) IELEM

99        FORMAT(1X,'GEOELT: ELEMENT ',1I6,' : NEGATIVE DETERMINANT')

          CALL PLANTE(1)

          STOP

        ENDIF



...

just change 1I6 to 1I9 in line “99” and recompile. You will see the element number with DET.LT.1.D-20.

Hello [name removed],

FYI your suggestion was pushed to main and will be available in next release.

Thanks for helping users.

[name removed]