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 The information provided may be out of date.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.

Problems with displacement in 2D axis-symmetric model

Please login with a confirmed email address before reporting spam

I'm having several issues trying to calculate the change in volume of a circular chamber (surrounded by an elastomeric material) due to pressure:

1. When I try to visualize the deformation (Results > 2D plot group > Surface [solid.disp] > Deformation), the cross-section of the chamber appears non-deformed. I'm plotting the spatial plane and my scale factor is 1.

2. I'm also trying to calculate the volume change of the chamber. I defined an integration model coupling (Definitions > Model couplings > Integration) corresponding to the domain of interest and set the frame to spatial. In results, I created a global evaluation (Derived values > Global evaluation) and entered "intop1(1)" under expression. The calculated volume of the chamber remains the same despite the pressure applied to the chamber.

file attached


1 Reply Last Post 2012年1月22日 GMT-5 10:58
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 2012年1月22日 GMT-5 10:58
Hi

first of all you are in 2D-axi, so if you integrate the section area, without the 2*pi*r you will not get a volume as I expect you wanted, or you can check the "compute integrand in revolved geometry" to have COMSOL looking after the 2*pi*r for you (for that specific integral, only)

And there is a little unit bug with a "Defined Integration" in 2D-axi and the computed integrand in revolved geometry checked) there is a "m" missing in the default units. COMSOL has promised me they will correct this in one of the next releases

You are better off with a Derived Variable Integration surface + "Compute Volume integral" checked, here the units are correct. In any case the numerical values are correct for both integration operators.

Then compare integration of the spatial frame and of the material frame, the latter should be constant , ut the spatial should change

But this does not explain why your integration are not changing with the deformation ;)

Got it:
It is linked with the Stationary - Geometry non-linearities check box, that you have not selected, and its linked to the fact that you integrate a volume that is not part of the solid, hence you do not have the spatial frame driven as in a "solid".
Try to make 2 integrands, spatial and material for domain 1 and for domain 2
Then solve with or without a) solid linear elastic material force linear check/uncheck, and Stationary Non-linear geometry checked and non checked

This seems to be something that has changed with v4.2a, I do neither not fully understand it any longer ;)

I have rechecked, including the release notes: you need to turn on the "solver - geometric non linearity" to get spatial deformation values correct with integrations, this way changed in 4.2a and was probably because too many users got confused while referring to the spatial frame and linking back to the model they made the model non linear.

Unfortunately those who understood are now confused by this new change in definition. So even if the spatial frame is distinguished from the material x,y,z and X,Y,Z respectively, its driven by the deformation ONLY when the non-linearity is turned ON, and in the same time COMSOL proposes Green-Lagrange strain tensor, see the doc (then x=X+u again) the integration on the spatial frame do not differ from the material frame until we turn specifically ON the non-linearities, now in the study node.

So try to turn on the non linearities in the solver. Then for your case you need to include the domain 2 in the solid, give it a material data of the type E=1Pa, u=0, rho=1 and it should do

--
Good luck
Ivar
Hi first of all you are in 2D-axi, so if you integrate the section area, without the 2*pi*r you will not get a volume as I expect you wanted, or you can check the "compute integrand in revolved geometry" to have COMSOL looking after the 2*pi*r for you (for that specific integral, only) And there is a little unit bug with a "Defined Integration" in 2D-axi and the computed integrand in revolved geometry checked) there is a "m" missing in the default units. COMSOL has promised me they will correct this in one of the next releases You are better off with a Derived Variable Integration surface + "Compute Volume integral" checked, here the units are correct. In any case the numerical values are correct for both integration operators. Then compare integration of the spatial frame and of the material frame, the latter should be constant , ut the spatial should change But this does not explain why your integration are not changing with the deformation ;) Got it: It is linked with the Stationary - Geometry non-linearities check box, that you have not selected, and its linked to the fact that you integrate a volume that is not part of the solid, hence you do not have the spatial frame driven as in a "solid". Try to make 2 integrands, spatial and material for domain 1 and for domain 2 Then solve with or without a) solid linear elastic material force linear check/uncheck, and Stationary Non-linear geometry checked and non checked This seems to be something that has changed with v4.2a, I do neither not fully understand it any longer ;) I have rechecked, including the release notes: you need to turn on the "solver - geometric non linearity" to get spatial deformation values correct with integrations, this way changed in 4.2a and was probably because too many users got confused while referring to the spatial frame and linking back to the model they made the model non linear. Unfortunately those who understood are now confused by this new change in definition. So even if the spatial frame is distinguished from the material x,y,z and X,Y,Z respectively, its driven by the deformation ONLY when the non-linearity is turned ON, and in the same time COMSOL proposes Green-Lagrange strain tensor, see the doc (then x=X+u again) the integration on the spatial frame do not differ from the material frame until we turn specifically ON the non-linearities, now in the study node. So try to turn on the non linearities in the solver. Then for your case you need to include the domain 2 in the solid, give it a material data of the type E=1Pa, u=0, rho=1 and it should do -- 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.