Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

volume change

Please login with a confirmed email address before reporting spam

I have a simple question,

I want to calculate the change in the volume of a deformed body . I am trying to get it using the boundary integration of the total displacement in the post processing option. but i couldn`t be sure whether it is really representing the change of the volume or not. or if there is any other way, please inform me.
tnx in advance

23 Replies Last Post 12.05.2012, 11:55 GMT-4

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24.06.2010, 16:14 GMT-4
why not just calculate the volume and subtract? (by integrating "1" in 3D)
why not just calculate the volume and subtract? (by integrating "1" in 3D)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24.06.2010, 16:20 GMT-4
Danial Tz
thanks!
can you tell me how the integration(the steps) can be caculated in the post processing?
Best regards,
Danial Tz thanks! can you tell me how the integration(the steps) can be caculated in the post processing? Best regards,

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 25.06.2010, 02:52 GMT-4
You just integrate (subdomain integration) 1.

int (1) dv = V
frame: spatial
You just integrate (subdomain integration) 1. int (1) dv = V frame: spatial

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 25.06.2010, 03:09 GMT-4
Hi

the inital volume is by integrating over the the spatial frame, and the final is over the material in V4 or ref frame in V3.5a but you might have to "turn the ref on" in the physics settings) no ?

Ivar
Hi the inital volume is by integrating over the the spatial frame, and the final is over the material in V4 or ref frame in V3.5a but you might have to "turn the ref on" in the physics settings) no ? Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 25.06.2010, 05:48 GMT-4

Hi

the inital volume is by integrating over the the spatial frame, and the final is over the material in V4 or ref frame in V3.5a but you might have to "turn the ref on" in the physics settings) no ?

Ivar


Why the other way around?

e.g. this is the plot of the surface (because it is 2D) of fsi example in Comsol modeling. Blue is the ref integral and red is the spatial (ALE).

cl.ly/8e860437f5a4a986d442

Or am I misunderstanding something?
[QUOTE] Hi the inital volume is by integrating over the the spatial frame, and the final is over the material in V4 or ref frame in V3.5a but you might have to "turn the ref on" in the physics settings) no ? Ivar [/QUOTE] Why the other way around? e.g. this is the plot of the surface (because it is 2D) of fsi example in Comsol modeling. Blue is the ref integral and red is the spatial (ALE). http://cl.ly/8e860437f5a4a986d442 Or am I misunderstanding something?

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 25.06.2010, 09:11 GMT-4
Hi

You are probably right, I mixing up things too.

In V4 COMSOL has renamed and ordered it better, I have understood, for v4: spatial = Eulerian (X,Y,Z...), material = Lagrangian (x,y,z...) and mesh = on the mesh (Xm,Ym,Zm..), but mesh does not necesarily move as the material, nor as the fluid.

It's the Lagrangian or Euler way of seing things, (the one who likes to swim with the flow and the one only looking on things only from the coast :)

see "Frames" p401 V4 doc: mph_ug.pdf

so from v3.5a its the "reference" frame that has now become "material" frame in V4, I'm i right now ?

the V4 doc is not fully free of references to the "reference frame" yet

I agree this is important to get straight
Ivar
Hi You are probably right, I mixing up things too. In V4 COMSOL has renamed and ordered it better, I have understood, for v4: spatial = Eulerian (X,Y,Z...), material = Lagrangian (x,y,z...) and mesh = on the mesh (Xm,Ym,Zm..), but mesh does not necesarily move as the material, nor as the fluid. It's the Lagrangian or Euler way of seing things, (the one who likes to swim with the flow and the one only looking on things only from the coast :) see "Frames" p401 V4 doc: mph_ug.pdf so from v3.5a its the "reference" frame that has now become "material" frame in V4, I'm i right now ? the V4 doc is not fully free of references to the "reference frame" yet I agree this is important to get straight Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 25.06.2010, 09:35 GMT-4
Hi Ivar,

In FSI cases, when I apply material to the plot, nothing seems has moved (also mesh). Then when I say Spatial it looks the way it should; with all the deformation. So practical understanding of it was

3.5 -------- 4.0
reference --> material
ALE ----> Spatial
mesh? no use so far
Hi Ivar, In FSI cases, when I apply material to the plot, nothing seems has moved (also mesh). Then when I say Spatial it looks the way it should; with all the deformation. So practical understanding of it was 3.5 -------- 4.0 reference --> material ALE ----> Spatial mesh? no use so far

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 25.06.2010, 09:58 GMT-4
Hi Danial

And as I'm mostly in structural, hence Lagrangian, that is why I'm getting it up side down w.r.t. to you ;)
Ivar
Hi Danial And as I'm mostly in structural, hence Lagrangian, that is why I'm getting it up side down w.r.t. to you ;) Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 29.12.2010, 13:00 GMT-5
Hello Danial,

I'm working with 3.5. Could you pls help me with following question. I have a 3D subdomain 1 with a hole in the middle, which expands as a result of the pressure on the hole boundaries. I should calculate a growing hole volume at each timestep. If I ntergrate a subdomain 1 over the spatial frame (int(1)dV) at each timestep, it will not include the hole volume. thanks
Hello Danial, I'm working with 3.5. Could you pls help me with following question. I have a 3D subdomain 1 with a hole in the middle, which expands as a result of the pressure on the hole boundaries. I should calculate a growing hole volume at each timestep. If I ntergrate a subdomain 1 over the spatial frame (int(1)dV) at each timestep, it will not include the hole volume. thanks

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 30.12.2010, 04:53 GMT-5
Hi

are you saying you have one domain (a square with a hole) and no mesh in the hole, but pressure boundary condition on the hole, hence hole deformation with time, and you want the hole volume ?

if so, the hole being total volume of square (outside boundary assumed fixed) minus (-) domain "1", and the volume of domain "1" is the integration of operand "1" over the volume (or the surface * thickness in 2D) and in Spatial frame if ST in v4

--
Good luck
Ivar
Hi are you saying you have one domain (a square with a hole) and no mesh in the hole, but pressure boundary condition on the hole, hence hole deformation with time, and you want the hole volume ? if so, the hole being total volume of square (outside boundary assumed fixed) minus (-) domain "1", and the volume of domain "1" is the integration of operand "1" over the volume (or the surface * thickness in 2D) and in Spatial frame if ST in v4 -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 31.12.2010, 13:51 GMT-5
Thanks a lot! The problem is that the cube boundaries are not fixed. I probably should integrate the displacements over the external cube boundaries and subtract integration of operand "1". Should all these be in the ALE (spatial) coordinates? Could you please also refer me to the place in COMSOL documentation where I can read about integration of operand "1" in v.3.5.
Thanks a lot,
Regina
Thanks a lot! The problem is that the cube boundaries are not fixed. I probably should integrate the displacements over the external cube boundaries and subtract integration of operand "1". Should all these be in the ALE (spatial) coordinates? Could you please also refer me to the place in COMSOL documentation where I can read about integration of operand "1" in v.3.5. Thanks a lot, Regina

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 01.01.2011, 05:09 GMT-5
Hi

if you are using the ALE, and you have no particular physics inside the domain of the hole, you can still mesh it in the ALE section and integrate thereafter witha coupling integration variable of operand "1" and get the "hole" volume to the mesh precision.

One thing that took a loooong time for me to get used to is the implicit COMSOL notation:
you select some domains or boundaries and then you write a formula in GUI field, often with "unusual units" this should be read mathematically, simply (once you know):

SUM_over_selected_Domains_or_Boundaries( int( int( int( Your_formula_for_an_element, dx ), dy ), dz )

In carthesian, if you select some other coordinates mnually you must alo correct the integrationoperands, the COMSOl coordinate transforms, applyes these to getr back to carthesian coordinates.

If you are in 3D structural (smsld in 3.5 or solid in v4) and have a gravity load on two domains/volumes you select the two domains, and type in the GUI field:

rho_smsld[kg/m^3]*9.81[m/s^2]

this means implicitely in COMSOL notations

F[N] = SUM_over_the_two_Domains( int( int( int( rho_smsld_of_each_domain_i [kg/m^3]*9.81[m/s^2] , dx ) , dy ) , dz )


(note the rho_smsld of each domain !) and if you check the units now it is logical ;)

Up to you to correct for the thickness or te loop length 2*pi*r for 2D , 2-Axi respectively

--
Good luck
Ivar
Hi if you are using the ALE, and you have no particular physics inside the domain of the hole, you can still mesh it in the ALE section and integrate thereafter witha coupling integration variable of operand "1" and get the "hole" volume to the mesh precision. One thing that took a loooong time for me to get used to is the implicit COMSOL notation: you select some domains or boundaries and then you write a formula in GUI field, often with "unusual units" this should be read mathematically, simply (once you know): SUM_over_selected_Domains_or_Boundaries( int( int( int( Your_formula_for_an_element, dx ), dy ), dz ) In carthesian, if you select some other coordinates mnually you must alo correct the integrationoperands, the COMSOl coordinate transforms, applyes these to getr back to carthesian coordinates. If you are in 3D structural (smsld in 3.5 or solid in v4) and have a gravity load on two domains/volumes you select the two domains, and type in the GUI field: rho_smsld[kg/m^3]*9.81[m/s^2] this means implicitely in COMSOL notations F[N] = SUM_over_the_two_Domains( int( int( int( rho_smsld_of_each_domain_i [kg/m^3]*9.81[m/s^2] , dx ) , dy ) , dz ) (note the rho_smsld of each domain !) and if you check the units now it is logical ;) Up to you to correct for the thickness or te loop length 2*pi*r for 2D , 2-Axi respectively -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 01.01.2011, 05:35 GMT-5
Great, thanks a lot!
Regina
Great, thanks a lot! Regina

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 12.05.2011, 17:54 GMT-4
Ok, just not following discussion on how to calculate volume change.

What I am trying to determine is the volume change of a tank with internal pressure.

How do I go about doing this?

Thanks,

Roy Martin
Ok, just not following discussion on how to calculate volume change. What I am trying to determine is the volume change of a tank with internal pressure. How do I go about doing this? Thanks, Roy Martin

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 13.05.2011, 01:06 GMT-4
Hi

the volumes are calculated on a mesh, so if your tank is "hollow" and you apply the pressure on the walls it would not work like that ;)

Workaround: you add the internal volume, as a "vacuum" no force item and then calculate the volume

--
Good luck
Ivar
Hi the volumes are calculated on a mesh, so if your tank is "hollow" and you apply the pressure on the walls it would not work like that ;) Workaround: you add the internal volume, as a "vacuum" no force item and then calculate the volume -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 13.05.2011, 11:37 GMT-4

Ok, I have an internal volume and set up the volume integration for the domain representing the internal volume to evaluate an expression of 1.

Now, is the calculated volume the deformed volume or the original volume?
Is there a way to calculate both in comsol so that I can get the difference? Rather than try to compare a comsol mesh volume to a solidworks solid volume...

Thanks,

Roy
Ok, I have an internal volume and set up the volume integration for the domain representing the internal volume to evaluate an expression of 1. Now, is the calculated volume the deformed volume or the original volume? Is there a way to calculate both in comsol so that I can get the difference? Rather than try to compare a comsol mesh volume to a solidworks solid volume... Thanks, Roy

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 13.05.2011, 16:36 GMT-4
Hi

in v4.1 and solid physics, you have now by default the "deformed mesh on" wich means that the spatial frame (defomed one) and the material fram (original ones) are different. Make 2 operators, one for the spatial frame, one on the reference frame, and integrate the same domain (or boundary) after having solved, or at least updated an existing solution. Then you shoudl get the different volumes, before - after

--
Good luck
Ivar
Hi in v4.1 and solid physics, you have now by default the "deformed mesh on" wich means that the spatial frame (defomed one) and the material fram (original ones) are different. Make 2 operators, one for the spatial frame, one on the reference frame, and integrate the same domain (or boundary) after having solved, or at least updated an existing solution. Then you shoudl get the different volumes, before - after -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 13.05.2011, 17:14 GMT-4
Ahhh, sucess.

What I did:

created a volume integration of the domain of interest with an expression of 1. this calculates the volume of the domain. The data set is the solution of interest.

Then, in the solution to which the date set points, change the Frame from Spatial to Material or Mesh. With spatial being the deformed geometry and material/mess being the reference (original) geometry.

Solve the integration for both frames, take the difference, and I get the volume change.

correct me if I stated anything wrong.

Another related question. In trying different things early, I came accros the variable for volumetric strain. If this variable is used as the expression in the volume integration, the result is very close to the difference calculated above if the Frame used is Spatial (deformed) and almost the same if the Frame used is Material/Mesh. Is this coincendence, or what should happen?

Thanks for all the help,

Roy
Ahhh, sucess. What I did: created a volume integration of the domain of interest with an expression of 1. this calculates the volume of the domain. The data set is the solution of interest. Then, in the solution to which the date set points, change the Frame from Spatial to Material or Mesh. With spatial being the deformed geometry and material/mess being the reference (original) geometry. Solve the integration for both frames, take the difference, and I get the volume change. correct me if I stated anything wrong. Another related question. In trying different things early, I came accros the variable for volumetric strain. If this variable is used as the expression in the volume integration, the result is very close to the difference calculated above if the Frame used is Spatial (deformed) and almost the same if the Frame used is Material/Mesh. Is this coincendence, or what should happen? Thanks for all the help, Roy

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 14.05.2011, 03:19 GMT-4
Hi

well done, welcome the the frame game, this is more usuall in fluidics, buit solid stress strain, is in ome way "flow" with specific conditions. So now yu know eveything about Euler and Lagrange ?

By the way in V3.5 one must turn on the deformed frame in the Physics tab for solid structural.

For the rest of your coment I believe so, but I would have to take a look at the equations underneath (turn on the equations in COMSOL, I find it very useful) so you could as well look it up too ?

--
Good luck
Ivar
Hi well done, welcome the the frame game, this is more usuall in fluidics, buit solid stress strain, is in ome way "flow" with specific conditions. So now yu know eveything about Euler and Lagrange ? By the way in V3.5 one must turn on the deformed frame in the Physics tab for solid structural. For the rest of your coment I believe so, but I would have to take a look at the equations underneath (turn on the equations in COMSOL, I find it very useful) so you could as well look it up too ? -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 03.06.2011, 19:36 GMT-4
Hi,

Just trying to understand regarding volume .

I am implementing as follows:
operator name intop1
source selection boundary, selection "inner_surface"
variable
name: vol, expression intop1(1)


Does this mean vol represent the volume enclosed by "inner_surface"?

Regards,
Susant
Hi, Just trying to understand regarding volume . I am implementing as follows: operator name intop1 source selection boundary, selection "inner_surface" variable name: vol, expression intop1(1) Does this mean vol represent the volume enclosed by "inner_surface"? Regards, Susant

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 06.06.2011, 10:51 GMT-4
Hi

1) In which frame ? you have also a frame field in the operator definition spatial or Materil (or mesh ?)
2) volume are on domains, not on boundaries, there is something I'm not following, or not understand

--
Good luck
Ivar
Hi 1) In which frame ? you have also a frame field in the operator definition spatial or Materil (or mesh ?) 2) volume are on domains, not on boundaries, there is something I'm not following, or not understand -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 12.05.2012, 08:56 GMT-4
Hi there,

I am using 4.2a and I am trying to calculate the volume of my deformed solutions. I have tried the things suggested in this thread, but I cannot get it to work. My model uses the Solid Mechanics module and the PDE module to model poroelasticity in the brain.

I can see the deformation in my plots, and even though I have tried all frame types in my volume integration, the number I get for the volume remains the same as the non-deformed geometry no matter what I do. I am trying to see how volume change as a result of prescribed displacement on the surface affects pressure, but for this I need to know the magnitude of volume change.

I am wondering if, perhaps, there is some check box I need to check or uncheck to make this work. Does it make a difference that I use an imported mesh versus one generated from a geometry?

Thanks,
Carolyn.
Hi there, I am using 4.2a and I am trying to calculate the volume of my deformed solutions. I have tried the things suggested in this thread, but I cannot get it to work. My model uses the Solid Mechanics module and the PDE module to model poroelasticity in the brain. I can see the deformation in my plots, and even though I have tried all frame types in my volume integration, the number I get for the volume remains the same as the non-deformed geometry no matter what I do. I am trying to see how volume change as a result of prescribed displacement on the surface affects pressure, but for this I need to know the magnitude of volume change. I am wondering if, perhaps, there is some check box I need to check or uncheck to make this work. Does it make a difference that I use an imported mesh versus one generated from a geometry? Thanks, Carolyn.

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 12.05.2012, 11:55 GMT-4
Hi

I believe that is because things has canged in 4.2a w.r.t. earlier, due to too many people being confused by the frames.

If you want to get the deformed volume, try turning ON the non-linear geometry tick in the SOLVER node, resolve sour moel then integrate the domain with the integrand of "1" over the SPATIAL frame x,y,z (not the Material frame X,Y,Z)

I believe this shold do it in v4.2a

the best is to try it out on a simple model i.e. a "soft" volume in 3D solid with a gravity body load agains a fixed boundary

--
Good luck
Ivar
Hi I believe that is because things has canged in 4.2a w.r.t. earlier, due to too many people being confused by the frames. If you want to get the deformed volume, try turning ON the non-linear geometry tick in the SOLVER node, resolve sour moel then integrate the domain with the integrand of "1" over the SPATIAL frame x,y,z (not the Material frame X,Y,Z) I believe this shold do it in v4.2a the best is to try it out on a simple model i.e. a "soft" volume in 3D solid with a gravity body load agains a fixed boundary -- Good luck Ivar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.