negative element volume problem in CFX/icem-cfd

adam2008 · July 30, 2009, 15:50

hi,
Got an error as below in CFX. I am using Icem-CFD to generate mesh. But in Icem-cfd, i already did smoothing and all determinant values in icem-cfd are above zero. I am wondering, how can I find this mesh problem in icem-cfd? Which mesh quality parameter should i check in icem-cfd? Because each time, cfx takes a long time to read in icem-cfd mesh due to large size, it is really waste of time when this error comes out. So if I can know this problem in icem-cfd, that would be very helpful. thanks a lot

ERROR #002100012 has occurred in subroutine cVolSec. |
| Message: |
| A negative ELEMENT volume has been detected. This is a fatal |
| error and execution will be terminated. The location of the first |
| negative volume is reported below. |
| Volume : -0.5637E-12 |
| Location : ( 0.10718E+01, 0.61251E-02, -0.61110E-01)

ghorrocks · July 30, 2009, 22:18

Hi,

Usually this occurs in a deforming mesh simulation when an element turns itself inside out but in my experience it is possible in ICEM to use poor smoothing and also generate negative volume elements. You have to be careful with smoothing in ICEM, it can generate negative volume elements using the default settings. The problems are usually at the interface between prism layers and tet elements in tricky bits of the geometry.

Make sure you do a element volume check in ICEM, and don't do too much smoothing.

Glenn Horrocks

Josh · April 14, 2010, 19:25

Glenn, you've saved me before, hopefully you can again.

Despite that I have used minimal smoothing in ICEM (I just smoothed for orthogonality, being that it is a hexahedral grid), my grid appears to have negative volume elements, as shown below.

As shown on the left in blue writing and in the histogram, there are almost 63 negative volume elements (8 + 55), with the smallest being -18.7586 mm^3. I need to get rid of these, but am not sure how.

Any ideas?

mvoss · April 16, 2010, 10:52

just a guess,

reassociate the edges/lines AND then try to move the verticies on the line. If done right you´ll see the verticies snapping on the edges. i my case while encountering such problems, i did a edge/line association, so everything was looking good but the verticies haven´t been moved to the linelocation ending up in meshverticies out of the surface -->(distorted elements-->neg. element)

neewbie

Josh · April 16, 2010, 17:23

Hi Neewbie -

Thanks for the reply.

I did re-associate the edges/lines before moving the vertices. The vertices did snap on the edges, as shown with my blocking here:

ghorrocks · April 18, 2010, 08:22

In these situations then I do not smooth in ICEM. Just mesh and proceed. Smoothing can be more harm than good sometimes.

Josh · April 18, 2010, 17:04

Thanks for the reply, Glenn.

Oddly enough, I ran another test case without any smoothing, and STILL had negative volume elements.

More details about this situation can be found in this thread:
http://www.cfd-online.com/Forums/ans...tml#post255073

ghorrocks · April 18, 2010, 19:26

Don't give up yet! Meshing is a complicated skill.

Although this post is best discussed on the thread you link to as it is really a meshing question, not a CFX question.

July 30, 2009, 15:50	negative element volume problem in CFX/icem-cfd	#1
adam2008 New Member Join Date: Jul 2009 Posts: 25 Rep Power: 17	hi, Got an error as below in CFX. I am using Icem-CFD to generate mesh. But in Icem-cfd, i already did smoothing and all determinant values in icem-cfd are above zero. I am wondering, how can I find this mesh problem in icem-cfd? Which mesh quality parameter should i check in icem-cfd? Because each time, cfx takes a long time to read in icem-cfd mesh due to large size, it is really waste of time when this error comes out. So if I can know this problem in icem-cfd, that would be very helpful. thanks a lot ERROR #002100012 has occurred in subroutine cVolSec. \| \| Message: \| \| A negative ELEMENT volume has been detected. This is a fatal \| \| error and execution will be terminated. The location of the first \| \| negative volume is reported below. \| \| Volume : -0.5637E-12 \| \| Location : ( 0.10718E+01, 0.61251E-02, -0.61110E-01)

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[blockMesh] BlockMesh FOAM warning	gaottino	OpenFOAM Meshing & Mesh Conversion	7	July 19, 2010 15:11
negative volume with sphere	jafarizade	FLUENT	2	May 21, 2009 04:28
negative solid volume error	Luis	CFX	3	October 30, 2007 19:04
Negative volume in mesh motion	skarp	CFX	5	July 12, 2007 00:51
FSI (negative volume element)	Flying pigs	CFX	2	April 23, 2007 08:07

July 30, 2009, 22:18		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,871 Rep Power: 144	Hi, Usually this occurs in a deforming mesh simulation when an element turns itself inside out but in my experience it is possible in ICEM to use poor smoothing and also generate negative volume elements. You have to be careful with smoothing in ICEM, it can generate negative volume elements using the default settings. The problems are usually at the interface between prism layers and tet elements in tricky bits of the geometry. Make sure you do a element volume check in ICEM, and don't do too much smoothing. Glenn Horrocks

April 14, 2010, 19:25		#3
Josh Senior Member Joshua Counsil Join Date: Jul 2009 Location: Halifax, Nova Scotia, Canada Posts: 366 Rep Power: 18	Glenn, you've saved me before, hopefully you can again. Despite that I have used minimal smoothing in ICEM (I just smoothed for orthogonality, being that it is a hexahedral grid), my grid appears to have negative volume elements, as shown below. As shown on the left in blue writing and in the histogram, there are almost 63 negative volume elements (8 + 55), with the smallest being -18.7586 mm^3. I need to get rid of these, but am not sure how. Any ideas?

April 16, 2010, 10:52		#4
mvoss Senior Member Matthias Voß Join Date: Mar 2009 Location: Berlin, Germany Posts: 449 Rep Power: 20	just a guess, reassociate the edges/lines AND then try to move the verticies on the line. If done right you´ll see the verticies snapping on the edges. i my case while encountering such problems, i did a edge/line association, so everything was looking good but the verticies haven´t been moved to the linelocation ending up in meshverticies out of the surface -->(distorted elements-->neg. element) neewbie

April 16, 2010, 17:23		#5
Josh Senior Member Joshua Counsil Join Date: Jul 2009 Location: Halifax, Nova Scotia, Canada Posts: 366 Rep Power: 18	Hi Neewbie - Thanks for the reply. I did re-associate the edges/lines before moving the vertices. The vertices did snap on the edges, as shown with my blocking here:

April 18, 2010, 08:22		#6
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,871 Rep Power: 144	In these situations then I do not smooth in ICEM. Just mesh and proceed. Smoothing can be more harm than good sometimes.

April 18, 2010, 17:04		#7
Josh Senior Member Joshua Counsil Join Date: Jul 2009 Location: Halifax, Nova Scotia, Canada Posts: 366 Rep Power: 18	Thanks for the reply, Glenn. Oddly enough, I ran another test case without any smoothing, and STILL had negative volume elements. More details about this situation can be found in this thread: http://www.cfd-online.com/Forums/ans...tml#post255073

April 18, 2010, 19:26		#8
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,871 Rep Power: 144	Don't give up yet! Meshing is a complicated skill. Although this post is best discussed on the thread you link to as it is really a meshing question, not a CFX question.