Negative Volume Problem

mueen · February 29, 2016, 04:07

I'm having problem of negative cell volume detected! in a quad mesh of two concentric cylinders in Fluent for 2D dynamic meshing case. I did meshing of cylinders in Gambit and then export it to Fluent solver [Gambit]. The JPEG images are attached which shows the volume of dynamic mesh before and after the error detected..
Can u plz help me in this regard.

Thank u!

-mAx- · February 29, 2016, 04:24

I assume, if you were able to export your mesh in Fluent, then you didn't have any cell with negative volume (else the export would fail).
Then your problem lies on your dynamic mesh setting.
If you are handling quad mesh, then I suppose you are using cell layering.
If everything is well set, you should try to reduce your dt.
this tutorial may help you:
http://aerojet.engr.ucdavis.edu/flue...tg/node181.htm

mueen · February 29, 2016, 05:40

Sir I'm using smoothing in dynamic meshing, I reduced time step size from 0.01 to 0.0001 but again after some time same error occurred.

-mAx- · February 29, 2016, 06:19

no idea with smoothing, sorry.

mueen · March 1, 2016, 01:17

Sir I'm using UDF in along with dyanamic meshing. Is it possible that Udf is incorrect?

-mAx- · March 1, 2016, 04:15

then check your motion without udf.
If you don't have cells with negative cells, then it should lie on your udf

mueen · March 1, 2016, 12:58

Sir, I've used UDF for different grid sizes along with different domain size but with same UDF I got results for only one of those grid sizes with no error & other grid sizes produce error.

edd313 · March 1, 2016, 19:56

Which smoothing tecnique are you using?
Say you are using spring-based, you may try to increase the "number of iterations". It will be more time consuming to update the mesh, but you will get better quality and eventually no negative volume.

mueen · March 2, 2016, 01:14

yes sir I'm using spring based smoothing but after a few number of iterations it shows error of negative volume!!!
Should I use different smoothing technique???

CeesH · March 2, 2016, 02:27

are all parts of your domain moving properly?
for example, I long ago did a falling ball simulation with 6DOF, and although the ball was moving properly, the cells in front and before the ball did not remesh along, meaning the cells in front of the ball would soon overlap with the ball itself - leading to negative cell volumes.

You can see in 'preview mesh movement' where it goes wrong, typically.

mueen · March 2, 2016, 03:34

Sir, can I use TUI for smoothing of quad elements in 2d, as suggested by one of the member of CFD forum!
if yes; can u guide me how to write TUI.

February 29, 2016, 04:24		#2
-mAx- Super Moderator Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41	I assume, if you were able to export your mesh in Fluent, then you didn't have any cell with negative volume (else the export would fail). Then your problem lies on your dynamic mesh setting. If you are handling quad mesh, then I suppose you are using cell layering. If everything is well set, you should try to reduce your dt. this tutorial may help you: http://aerojet.engr.ucdavis.edu/flue...tg/node181.htm mueen likes this. __________________ In memory of my friend Hervé: CFD engineer & freerider

February 29, 2016, 06:19		#4
-mAx- Super Moderator Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41	no idea with smoothing, sorry. __________________ In memory of my friend Hervé: CFD engineer & freerider

March 1, 2016, 04:15		#6
-mAx- Super Moderator Maxime Perelli Join Date: Mar 2009 Location: Switzerland Posts: 3,297 Rep Power: 41	then check your motion without udf. If you don't have cells with negative cells, then it should lie on your udf __________________ In memory of my friend Hervé: CFD engineer & freerider Last edited by -mAx-; March 2, 2016 at 01:17.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
multiphaseEulerFoam (OF2.3.0) : Courant number explodes when running in parallel	Mehrez	OpenFOAM Running, Solving & CFD	10	May 18, 2016 11:44
Negative Volume during Mesh Motion Analysis	giov_ingr	FLUENT	2	December 13, 2013 06:09
Problem of negative volume	Anam	ANSYS	0	September 11, 2011 04:27
[ICEM] negative mesh volume problem (icem-cfd/cfx)	adam2008	ANSYS Meshing & Geometry	5	April 16, 2010 12:21
Negative volume problem in Two-way FSI coupling	fred	CFX	3	August 16, 2006 10:03

February 29, 2016, 05:40		#3
mueen New Member Pakistan/Punjab Join Date: Dec 2015 Posts: 17 Rep Power: 10	Sir I'm using smoothing in dynamic meshing, I reduced time step size from 0.01 to 0.0001 but again after some time same error occurred.

March 1, 2016, 01:17		#5
mueen New Member Pakistan/Punjab Join Date: Dec 2015 Posts: 17 Rep Power: 10	Sir I'm using UDF in along with dyanamic meshing. Is it possible that Udf is incorrect?

March 1, 2016, 12:58		#7
mueen New Member Pakistan/Punjab Join Date: Dec 2015 Posts: 17 Rep Power: 10	Sir, I've used UDF for different grid sizes along with different domain size but with same UDF I got results for only one of those grid sizes with no error & other grid sizes produce error.

March 1, 2016, 19:56		#8
edd313 New Member Edoardo Join Date: Nov 2015 Posts: 6 Rep Power: 11	Which smoothing tecnique are you using? Say you are using spring-based, you may try to increase the "number of iterations". It will be more time consuming to update the mesh, but you will get better quality and eventually no negative volume.

March 2, 2016, 01:14		#9
mueen New Member Pakistan/Punjab Join Date: Dec 2015 Posts: 17 Rep Power: 10	yes sir I'm using spring based smoothing but after a few number of iterations it shows error of negative volume!!! Should I use different smoothing technique???

March 2, 2016, 02:27		#10
CeesH Senior Member Cees Haringa Join Date: May 2013 Location: Delft Posts: 607 Rep Power: 0	are all parts of your domain moving properly? for example, I long ago did a falling ball simulation with 6DOF, and although the ball was moving properly, the cells in front and before the ball did not remesh along, meaning the cells in front of the ball would soon overlap with the ball itself - leading to negative cell volumes. You can see in 'preview mesh movement' where it goes wrong, typically.

March 2, 2016, 03:34		#11
mueen New Member Pakistan/Punjab Join Date: Dec 2015 Posts: 17 Rep Power: 10	Sir, can I use TUI for smoothing of quad elements in 2d, as suggested by one of the member of CFD forum! if yes; can u guide me how to write TUI.