Error: Negative volume and Creating empty surface

Faby · June 5, 2012, 06:59

Hello everyone,

I have two interface walls defined as dynamic mesh, that move as a rigid body according to an udf in a flow domain. The problem is in transient time.
When I try to compute the solution, the simulation stops with this error:

"Warning: no positive-volume exist.
Error: update-dynamic-mesh failed. Negative cell volume detected
Note: zone-surface: cannot create surface from sliding interface zone.
Creating empty surface."

If I see the mesh surrounding dynamic mesh zones (interface walls) it's deformed a lot. But, I don't know how to fix it and I can't understand the "Note".

Could anyone help me?

Thank you in advance.

Far · June 5, 2012, 07:17

It is producing negative volume due to dynamic mesh setup. Either check the mesh settings (use tetra ) or dynamic mesh settings.

Faby · June 6, 2012, 11:15

Quote:

Originally Posted by Far

It is producing negative volume due to dynamic mesh setup. Either check the mesh settings (use tetra ) or dynamic mesh settings.

I'm already using tetra mesh...
I tried with declaration of fluid zone surrounding the particle as deforming zone...but it doesn't work...

sadjad.s · October 4, 2012, 16:45

i've got your problem either.
it seems that problem is within using interface walls in remeshing method

Faby · October 5, 2012, 08:41

At the end I did't use interface walls , but a mesh around the moving body that connects the wall of moving body and the mesh of external fluid domain.

sadjad.s · October 6, 2012, 08:04

at last i found the answer (:
actually i was modelling a simple 3d model of falling sphere in water.
i 've successfully done it with interface walls.
my problem was that i didn't take buoyancy into account; sphere had much less density than water so it went up in about 0.01 sec. (density of sphere is calculated by fluent which is mass divided by volume )
don't forget to choose both interfaces as "rigid body" and "passive" in dynamic mesh options.

Faby · October 8, 2012, 05:45

Quote:

Originally Posted by sadjad.s

at last i found the answer (:
actually i was modelling a simple 3d model of falling sphere in water.
i 've successfully done it with interface walls.
my problem was that i didn't take buoyancy into account; sphere had much less density than water so it went up in about 0.01 sec. (density of sphere is calculated by fluent which is mass divided by volume )
don't forget to choose both interfaces as "rigid body" and "passive" in dynamic mesh options.

How did you take buoyancy into account?

nkme2007 · October 10, 2012, 06:48

Hello All,

I want to do analysis of heat transfer from water flowing through pipes submerged inside concrete. I am modelling in GAMBIT and wish to analyse it on Ansys FLUENT.

Can anybody help me out, how to model and simulate?

Does any tutorials exist?

sadjad.s · October 10, 2012, 18:05

hi mate,
actually with 6DOF option in dynamic mesh, fluent automatically considers buoyancy force between solid(ie. box) and fluid(water).
my problem was that i considered rather little mass for box (in the main problem it was sphere) so that it went up!
in this example(2d version), i made a structred grid around box and let it move with the same speed as box.(as you can see in picture).
and used remeshing method for unstructured grid.
i use fluent v14.0 64 bit, if you want, post me your email so i send you case & data and udf.
test.jpg

sadjad.s · October 10, 2012, 18:13

hi mate,
modelling "water through pipe which is surrounded by concrete", it appears to be a rather simple task.
how much are you familiar with gambit & fluent?
you wanna model it 2d or 3d?

nkme2007 · October 11, 2012, 01:08

Quote:

Originally Posted by sadjad.s

hi mate,
modelling "water through pipe which is surrounded by concrete", it appears to be a rather simple task.
how much are you familiar with gambit & fluent?
you wanna model it 2d or 3d?

Sadjad,

I am a beginner of GAMBIT & FLUENT. My aim is to analyse the heat transfer and to analyse the temperature distribution among the concrete. I think 2D would suffice me, can you help me out?

sadjad.s · October 11, 2012, 04:32

your analysis seems easy if you set all problem's specifications correct.
but you need first to learn meshing with Gambit.
by doing these tutorials at this site(Cornell University), you'll have a rather sufficient view of Gambit&Fluent.
https://confluence.cornell.edu/displ...arning+Modules
Then you can start simulating what you want.

nkme2007 · October 11, 2012, 05:25

Quote:

Originally Posted by sadjad.s

your analysis seems easy if you set all problem's specifications correct.
but you need first to learn meshing with Gambit.
by doing these tutorials at this site(Cornell University), you'll have a rather sufficient view of Gambit&Fluent.
https://confluence.cornell.edu/displ...arning+Modules
Then you can start simulating what you want.

But, I didn't find any tutorial that is nearer to my work.

sadjad.s · October 11, 2012, 12:18

you are right but if you are a beginner, you need to learn these two softwares from base.
by doing those tutorials you will get familiar with meshing via Gambit and some basic CFD modellings with Fluent.

nkme2007 · October 11, 2012, 12:38

Quote:

Originally Posted by sadjad.s

you are right but if you are a beginner, you need to learn these two softwares from base.
by doing those tutorials you will get familiar with meshing via Gambit and some basic CFD modellings with Fluent.

Thank you sadjad!

subha_meter · October 23, 2012, 05:09

Hi Sadiad,

I had the same experience of the solid sphere bouncing off the liquid surface instead of submerging. In 2D case, how did you specify the volume of the sphere. I believe in the 6DOF property UDF, you are giving mass and moment of inertia as the required inputs.

Regards,

Quote:

Originally Posted by sadjad.s

hi mate,
actually with 6DOF option in dynamic mesh, fluent automatically considers buoyancy force between solid(ie. box) and fluid(water).
my problem was that i considered rather little mass for box (in the main problem it was sphere) so that it went up!
in this example(2d version), i made a structred grid around box and let it move with the same speed as box.(as you can see in picture).
and used remeshing method for unstructured grid.
i use fluent v14.0 64 bit, if you want, post me your email so i send you case & data and udf.
Attachment 16128

sadjad.s · October 23, 2012, 14:56

dear mate,
no need to specify volume, because fluent compute it by geometry.
you just need to enter mass of sphere via udf and density will be computed by "density=mass/volume".
a circle in 2d is actually a cylinder which has one meter depth so in order to compute volume just multiply section are by one.
to solve your problem just make sure that density of sphere is higher than fluid.
if there is still problem, send your email address to send you case&data&udf.

subha_meter · October 23, 2012, 19:55

Hi Sadjad,

Thanks for your reply.

Earlier, I calculated the mass based on a sphere which is actually a cylinder in 2D. This caused the problem since cylinder volume considered by FLUENT is higher than the sphere and consequently density of solid became lower than the fluid.

I have fixed the problem now.

Thanks again!

Regards,

Quote:

Originally Posted by sadjad.s

dear mate,
no need to specify volume, because fluent compute it by geometry.
you just need to enter mass of sphere via udf and density will be computed by "density=mass/volume".
a circle in 2d is actually a cylinder which has one meter depth so in order to compute volume just multiply section are by one.
to solve your problem just make sure that density of sphere is higher than fluid.
if there is still problem, send your email address to send you case&data&udf.

subha_meter · November 6, 2012, 18:01

Hi Sadjad,

Although the dynamic mesh model worked for heavier particle (density > liquid). for lighter particle (particle density < liquid density), it seems there's some problem. The particle bounces off the interface instead of floating on the liquid. Any suggestion?

Regards,

Quote:

Originally Posted by sadjad.s

dear mate,
no need to specify volume, because fluent compute it by geometry.
you just need to enter mass of sphere via udf and density will be computed by "density=mass/volume".
a circle in 2d is actually a cylinder which has one meter depth so in order to compute volume just multiply section are by one.
to solve your problem just make sure that density of sphere is higher than fluid.
if there is still problem, send your email address to send you case&data&udf.

sadjad.s · November 7, 2012, 14:36

hi mate,
if you are sure that particle must go up (i.e. density of particle is less than liquid density), then you must use very little time step.(even in order of e-7)
as it is obvious, particle will throw up very quickly, so in order to catch motion, use little time step.
as a suggestion, the order of time step should be in a way that in each time step, your body moves less than cell height.

June 5, 2012, 06:59	Error: Negative volume and Creating empty surface	#1
Faby New Member Join Date: Apr 2012 Posts: 17 Rep Power: 14	Hello everyone, I have two interface walls defined as dynamic mesh, that move as a rigid body according to an udf in a flow domain. The problem is in transient time. When I try to compute the solution, the simulation stops with this error: "Warning: no positive-volume exist. Error: update-dynamic-mesh failed. Negative cell volume detected Note: zone-surface: cannot create surface from sliding interface zone. Creating empty surface." If I see the mesh surrounding dynamic mesh zones (interface walls) it's deformed a lot. But, I don't know how to fix it and I can't understand the "Note". Could anyone help me? Thank you in advance.

October 10, 2012, 18:05		#9
sadjad.s Member sadjad Join Date: Jan 2012 Posts: 72 Rep Power: 14	hi mate, actually with 6DOF option in dynamic mesh, fluent automatically considers buoyancy force between solid(ie. box) and fluid(water). my problem was that i considered rather little mass for box (in the main problem it was sphere) so that it went up! in this example(2d version), i made a structred grid around box and let it move with the same speed as box.(as you can see in picture). and used remeshing method for unstructured grid. i use fluent v14.0 64 bit, if you want, post me your email so i send you case & data and udf. test.jpg Last edited by sadjad.s; October 10, 2012 at 18:21.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[Other] mesh airfoil NACA0012	anand_30	OpenFOAM Meshing & Mesh Conversion	13	March 7, 2022 18:22
[blockMesh] error message with modeling a cube with a hold at the center	hsingtzu	OpenFOAM Meshing & Mesh Conversion	2	March 14, 2012 10:56
channelFoam for a 3D pipe	AlmostSurelyRob	OpenFOAM	3	June 24, 2011 14:06
[blockMesh] BlockMesh FOAM warning	gaottino	OpenFOAM Meshing & Mesh Conversion	7	July 19, 2010 15:11
[blockMesh] Axisymmetrical mesh	Rasmus Gjesing (Gjesing)	OpenFOAM Meshing & Mesh Conversion	10	April 2, 2007 15:00

June 5, 2012, 07:17		#2
Far Senior Member Sijal Join Date: Mar 2009 Location: Islamabad Posts: 4,558 Blog Entries: 6 Rep Power: 54	It is producing negative volume due to dynamic mesh setup. Either check the mesh settings (use tetra ) or dynamic mesh settings.

October 4, 2012, 16:45		#4
sadjad.s Member sadjad Join Date: Jan 2012 Posts: 72 Rep Power: 14	i've got your problem either. it seems that problem is within using interface walls in remeshing method

October 5, 2012, 08:41		#5
Faby New Member Join Date: Apr 2012 Posts: 17 Rep Power: 14	At the end I did't use interface walls , but a mesh around the moving body that connects the wall of moving body and the mesh of external fluid domain.

October 6, 2012, 08:04		#6
sadjad.s Member sadjad Join Date: Jan 2012 Posts: 72 Rep Power: 14	at last i found the answer (: actually i was modelling a simple 3d model of falling sphere in water. i 've successfully done it with interface walls. my problem was that i didn't take buoyancy into account; sphere had much less density than water so it went up in about 0.01 sec. (density of sphere is calculated by fluent which is mass divided by volume ) don't forget to choose both interfaces as "rigid body" and "passive" in dynamic mesh options.

October 10, 2012, 06:48		#8
nkme2007 Member Tamil Nadu Join Date: Oct 2012 Posts: 44 Rep Power: 14	Hello All, I want to do analysis of heat transfer from water flowing through pipes submerged inside concrete. I am modelling in GAMBIT and wish to analyse it on Ansys FLUENT. Can anybody help me out, how to model and simulate? Does any tutorials exist?

October 10, 2012, 18:13		#10
sadjad.s Member sadjad Join Date: Jan 2012 Posts: 72 Rep Power: 14	hi mate, modelling "water through pipe which is surrounded by concrete", it appears to be a rather simple task. how much are you familiar with gambit & fluent? you wanna model it 2d or 3d?

October 11, 2012, 04:32		#12
sadjad.s Member sadjad Join Date: Jan 2012 Posts: 72 Rep Power: 14	your analysis seems easy if you set all problem's specifications correct. but you need first to learn meshing with Gambit. by doing these tutorials at this site(Cornell University), you'll have a rather sufficient view of Gambit&Fluent. https://confluence.cornell.edu/displ...arning+Modules Then you can start simulating what you want.

October 11, 2012, 12:18		#14
sadjad.s Member sadjad Join Date: Jan 2012 Posts: 72 Rep Power: 14	you are right but if you are a beginner, you need to learn these two softwares from base. by doing those tutorials you will get familiar with meshing via Gambit and some basic CFD modellings with Fluent.

October 23, 2012, 14:56		#17
sadjad.s Member sadjad Join Date: Jan 2012 Posts: 72 Rep Power: 14	dear mate, no need to specify volume, because fluent compute it by geometry. you just need to enter mass of sphere via udf and density will be computed by "density=mass/volume". a circle in 2d is actually a cylinder which has one meter depth so in order to compute volume just multiply section are by one. to solve your problem just make sure that density of sphere is higher than fluid. if there is still problem, send your email address to send you case&data&udf.

November 7, 2012, 14:36		#20
sadjad.s Member sadjad Join Date: Jan 2012 Posts: 72 Rep Power: 14	hi mate, if you are sure that particle must go up (i.e. density of particle is less than liquid density), then you must use very little time step.(even in order of e-7) as it is obvious, particle will throw up very quickly, so in order to catch motion, use little time step. as a suggestion, the order of time step should be in a way that in each time step, your body moves less than cell height.