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 The information provided may be out of date.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.

Postprocessing in cyclindrical coordinates


I am doing a plane strain analysis of a pipe with a crack in it.

I would like to get the stress components in the cyclindridcal coordinate system (radial, hoop stress) rather than in the cartesian coordinate system (x & y stress components).

There seems to be do obvcious way of doing this.

Is there anyone who can suggest something?

Thanks,
--
Arthur Rupel

12 Replies Last Post Nov 22, 2016, 12:22 AM EST
Henrik Sönnerlind COMSOL Employee
Posted: 5 years ago Feb 7, 2013, 4:08 AM EST
Hi,

This is the way to do it:

1. Add a cylindrical coordinate system with suitable axis directions under Definitions.
2. Select that system in the Linear Elastic feature.
3. Solve
4. Plot local stress tensor components ("Stress tensor, local coordinate system")

Regards,
Henrik
Hi, This is the way to do it: 1. Add a cylindrical coordinate system with suitable axis directions under Definitions. 2. Select that system in the Linear Elastic feature. 3. Solve 4. Plot local stress tensor components ("Stress tensor, local coordinate system") Regards, Henrik

Posted: 5 years ago Feb 21, 2013, 2:49 PM EST
I am using plastic deformation-isotropic hardening-von mises stress, tangent modulus.--
There is no problem in defining the cyc coord in definitions. However I cannot find anyway of stating this with the material properties that I am using. This is not a linear elastic material.

Thanks
Arthur Rupel
I am using plastic deformation-isotropic hardening-von mises stress, tangent modulus.-- There is no problem in defining the cyc coord in definitions. However I cannot find anyway of stating this with the material properties that I am using. This is not a linear elastic material. Thanks Arthur Rupel

Henrik Sönnerlind COMSOL Employee
Posted: 5 years ago Feb 21, 2013, 5:05 PM EST
Hi,

You still have a Linear Elastic node (where you set Young's Modulus etc.) to which Plasticity is a sub-node in the model tree. In Linear Elastic there is a coordinate system setting even if the plasticity is isotropic.

Regards,
Henrik
Hi, You still have a Linear Elastic node (where you set Young's Modulus etc.) to which Plasticity is a sub-node in the model tree. In Linear Elastic there is a coordinate system setting even if the plasticity is isotropic. Regards, Henrik

Posted: 5 years ago Mar 5, 2013, 4:57 PM EST
I defined the cyclindrical coordinate system in definitions.

In the linear elastic option, the only coordinate system allowed in the global. There does not seem to be a way of changing this.

When I try to plot r, theta stresses in the cyclindrical system, all I get are the x & y stresses-even in the changed local system.

There has to be a way for results to see that I want cyclindrical stresses, but I cannot find it.

Thanks,

--
Arthur Rupel
I defined the cyclindrical coordinate system in definitions. In the linear elastic option, the only coordinate system allowed in the global. There does not seem to be a way of changing this. When I try to plot r, theta stresses in the cyclindrical system, all I get are the x & y stresses-even in the changed local system. There has to be a way for results to see that I want cyclindrical stresses, but I cannot find it. Thanks, -- Arthur Rupel

Posted: 5 years ago Mar 6, 2013, 1:28 AM EST
Hi

When you define user coordinate systems, COMSOL defines the transformation matrix, and its inverse, so you can use these also directly (see the doc) COMSOL remains in the Cartesian x,y,z coordinate system, so you need to apply these transforms manually to the Results section, f you need to map back

--
Good luck
Ivar
Hi When you define user coordinate systems, COMSOL defines the transformation matrix, and its inverse, so you can use these also directly (see the doc) COMSOL remains in the Cartesian x,y,z coordinate system, so you need to apply these transforms manually to the Results section, f you need to map back -- Good luck Ivar

Henrik Sönnerlind COMSOL Employee
Posted: 5 years ago Mar 6, 2013, 7:21 AM EST

I defined the cyclindrical coordinate system in definitions.

In the linear elastic option, the only coordinate system allowed in the global. There does not seem to be a way of changing this.




The reason is that only coordinate systems which belong to the material frame can be selected for the material definition. So if you change the setting in the coordinate system from Spatial to Material, then it will show up.

Regards,
Henrik
[QUOTE] I defined the cyclindrical coordinate system in definitions. In the linear elastic option, the only coordinate system allowed in the global. There does not seem to be a way of changing this. [/QUOTE] The reason is that only coordinate systems which belong to the material frame can be selected for the material definition. So if you change the setting in the coordinate system from Spatial to Material, then it will show up. Regards, Henrik

Posted: 5 years ago Mar 6, 2013, 10:49 AM EST
Thanks! It worked.
This may be something other people should note.

--
Arthur Rupel
Thanks! It worked. This may be something other people should note. -- Arthur Rupel

Posted: 4 years ago Nov 27, 2013, 8:33 AM EST

Hi,

This is the way to do it:

1. Add a cylindrical coordinate system with suitable axis directions under Definitions.
2. Select that system in the Linear Elastic feature.
3. Solve
4. Plot local stress tensor components ("Stress tensor, local coordinate system")

Regards,
Henrik



About the fourth step "Plot local stress tensor components ("Stress tensor, local coordinate system")", I got a question: when I choose the "Stress tensor(Spatial)", I got the same plots compare with the "Stress tensor, local coordinate system". But "Stress tensor(Spatial)" seems to be defined according to the Cartesian coordinate system. How can I get the stress tensor of Cylindrical system component?

Thanks,
Fan shuyu
[QUOTE] Hi, This is the way to do it: 1. Add a cylindrical coordinate system with suitable axis directions under Definitions. 2. Select that system in the Linear Elastic feature. 3. Solve 4. Plot local stress tensor components ("Stress tensor, local coordinate system") Regards, Henrik [/QUOTE] About the fourth step "Plot local stress tensor components ("Stress tensor, local coordinate system")", I got a question: when I choose the "Stress tensor(Spatial)", I got the same plots compare with the "Stress tensor, local coordinate system". But "Stress tensor(Spatial)" seems to be defined according to the Cartesian coordinate system. How can I get the stress tensor of Cylindrical system component? Thanks, Fan shuyu

Henrik Sönnerlind COMSOL Employee
Posted: 4 years ago Nov 27, 2013, 10:05 AM EST


Hi,

This is the way to do it:

1. Add a cylindrical coordinate system with suitable axis directions under Definitions.
2. Select that system in the Linear Elastic feature.
3. Solve
4. Plot local stress tensor components ("Stress tensor, local coordinate system")

Regards,
Henrik



About the fourth step "Plot local stress tensor components ("Stress tensor, local coordinate system")", I got a question: when I choose the "Stress tensor(Spatial)", I got the same plots compare with the "Stress tensor, local coordinate system". But "Stress tensor(Spatial)" seems to be defined according to the Cartesian coordinate system. How can I get the stress tensor of Cylindrical system component?

Thanks,
Fan shuyu


Hi,

The "Stress tensor(Spatial)" is as you say oriented along the global Cartesian coordinate system, whereas the "Stress tensor, local coordinate system" has orientations determined by the coordinate system selection in the Linear Elastic node.

Note that "Step 3: Solve" is not strictly necessary if the material is isotropic, and you just want to change the display directions for "Stress tensor, local coordinate system". It is then sufficient to instead do an "Update solution" for the study.

Regards,
Henrik
[QUOTE] [QUOTE] Hi, This is the way to do it: 1. Add a cylindrical coordinate system with suitable axis directions under Definitions. 2. Select that system in the Linear Elastic feature. 3. Solve 4. Plot local stress tensor components ("Stress tensor, local coordinate system") Regards, Henrik [/QUOTE] About the fourth step "Plot local stress tensor components ("Stress tensor, local coordinate system")", I got a question: when I choose the "Stress tensor(Spatial)", I got the same plots compare with the "Stress tensor, local coordinate system". But "Stress tensor(Spatial)" seems to be defined according to the Cartesian coordinate system. How can I get the stress tensor of Cylindrical system component? Thanks, Fan shuyu [/QUOTE] Hi, The "Stress tensor(Spatial)" is as you say oriented along the global Cartesian coordinate system, whereas the "Stress tensor, local coordinate system" has orientations determined by the coordinate system selection in the Linear Elastic node. Note that "Step 3: Solve" is not strictly necessary if the material is isotropic, and you just want to change the display directions for "Stress tensor, local coordinate system". It is then sufficient to instead do an "Update solution" for the study. Regards, Henrik

Posted: 4 years ago Nov 28, 2013, 12:33 AM EST
Thanks very much !

Fan shuyu
Thanks very much ! Fan shuyu

Posted: 11 months ago Nov 22, 2016, 12:21 AM EST
Thanks for local coordinate selection tips . Its very useful !
Thanks for local coordinate selection tips . Its very useful !

Posted: 11 months ago Nov 22, 2016, 12:22 AM EST
It works ! Many thanks!
It works ! Many thanks!

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.