Jump to content
GMS, SMS, and WMS User Forum
hpai

rma2: Problem inside COEFS routine

Recommended Posts

Hi,

new to this forum. i'm using rma2 in sms9.2 to model a river confluence.

below is what appears in the output.txt prompt after 213 lines:

=== Problem inside COEFS routine. Do 180.

Element= 535 Gauss pt= 1 out of NGP= 16

Divide by DETJ= -0.0229

J11-J22-J12-J21= 1.1526 0.38006 0.74030 0.62257

stop in coefs

guessing the integration is having some issues at the element location at a specific gauss point in the quadrilateral. any suggestions on how to proceed? (sorry for vague question... not exactly sure what is causing).

thx!

h

Share this post


Link to post
Share on other sites

h,

A similar post appears in the old forum. I have included it below.

- John

problem in COEFS

Hi I am facing this problem while trying to run RMA4.5. the output is presented here. Can anybody please tell me why this things is happening. Any comments would be apriciated. Thanks.

-----------------------------------------------------------------------

=== Problem inside COEFS routine. Do 180.

COEFS

Element= 13821 Gauss pt= 1 out of NGP= 16

Divide by DETJ= -0.3970

J11-J22-J12-J21= -1.6550 0.23989 0.0000 -1.1725

stop in coefs

------------------------------------------------------------------------

Re: problem in COEFS

Hi

i have found very interesting thing. My original geo file shows these value for particular nodes.

GNN 40343 340388.7711 5937444.7697 610.2145

GNN 40344 340388.5769 5937445.1376 610.2379

GNN 40345 340387.7779 5937445.1406 610.2586

GNN 40346 340387.9721 5937444.7728 610.2348

but after running the gfgen45.exe ot1 file shows this truncated X,Y and Z for those same nodes. This things happened when my mesh elements are of very small (.5 m). This truncated X,y values are creating problem. Making my area unusual, i.e. negative or zero.

40343 340388.78 5937445.00 610.21

40344 340388.56 5937445.00 610.24

40345 340387.78 5937445.00 610.26

40346 340387.97 5937445.00 610.23

Is it problem of RMA2? Did any body face this problem? Did anybody successsfully run for very small mesh size?

Any comments would be apriciable.

Thanks.

Mostafa

Re: problem in COEFS

From the message written to your output file, it appears that the Jacobian determinant resulting from the mapping between the parent and global element is negative at one of the integration points. This is a very bad situation, but will not necessarily destroy your solution. The negative determinant is probably caused by a strangely formed element. Check the element, and perhaps rebuild the mesh in the vicinity of that element. The program ought to warn of the existence of negative Jacobians at integration points before a simulation. Did you ask SMS to carry out a "mesh check" before your run?

David Froehlich

Re: problem in COEFS

RMA2 has used real type variable for all floating point numbers, so if your x and y are big enougth then rma2 will truncate the number and then it will use the wrong number to calculate the result.

For example, if x1=38453195.4678, x2=38453195.4778, rma2 maybe truncate them to the same value, then error appears.

Share this post


Link to post
Share on other sites

Thanks!

So, not exactly sure what "mesh check" meant; I do run model checks before actually running.

I did "relax" the mesh which worked though.

glad to see i wasn't the only dummy having these troubles =)!

h

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...