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.
Maxwell tensor evaluation problem in a hyperelastic material
Posted Apr 9, 2014, 6:51 a.m. EDT Low-Frequency Electromagnetics, MEMS & Nanotechnology, MEMS & Piezoelectric Devices, Structural Mechanics Version 4.3, Version 5.2 5 Replies
Please login with a confirmed email address before reporting spam
Hello,
I am using Comsol V4.3 and below I will try to define my problems:
To simplify, I am modelling in 2D. The general outline of the setup: a piece of hyperelastic material is sandwiched between two compliant electrodes. One electrode is a terminal with a potential of 2000 V and the other one is grounded. I want to calculate the force on the dielectric and the deformation of the dielectric.
I have tried to set up two models. One with solely the Electromechanics (emi) module and another model consisting of the Movings Mesh (ale), Solid Mechanics (solid) and Electrostatics (es) modules.
In the emi-module I am able to get to a good solution when I am using a linear elastic material. But when I change the material to a hyperelastic material using the Ogden-equations, I get an error that Comsol is not able to evaluate the variable (mod1.)emi.nTsx. To my knowledge this is the x-component of the Maxwell stress tensor.
I have checked that my mesh is symmetrical (mapped mesh with distribution) according to the advice of Ivar Kjelberg. In these two models I have left out the surrounding air since I do not think this is essential to solve the problem.
My next guess is to work with mesh displacements to seperate the displacement and the Maxwell tensor calculation, but I have not yet found out how.
I have done a lot of searching and reading in order to find a solution myself but I haven't found anything that helped me past this point. I am glad to provide the model files I have if anyone is interested.
Oh and could somebody also tell me how to show the boundary numbers in the Graphics Window?
Thanks!
I am using Comsol V4.3 and below I will try to define my problems:
To simplify, I am modelling in 2D. The general outline of the setup: a piece of hyperelastic material is sandwiched between two compliant electrodes. One electrode is a terminal with a potential of 2000 V and the other one is grounded. I want to calculate the force on the dielectric and the deformation of the dielectric.
I have tried to set up two models. One with solely the Electromechanics (emi) module and another model consisting of the Movings Mesh (ale), Solid Mechanics (solid) and Electrostatics (es) modules.
In the emi-module I am able to get to a good solution when I am using a linear elastic material. But when I change the material to a hyperelastic material using the Ogden-equations, I get an error that Comsol is not able to evaluate the variable (mod1.)emi.nTsx. To my knowledge this is the x-component of the Maxwell stress tensor.
I have checked that my mesh is symmetrical (mapped mesh with distribution) according to the advice of Ivar Kjelberg. In these two models I have left out the surrounding air since I do not think this is essential to solve the problem.
My next guess is to work with mesh displacements to seperate the displacement and the Maxwell tensor calculation, but I have not yet found out how.
I have done a lot of searching and reading in order to find a solution myself but I haven't found anything that helped me past this point. I am glad to provide the model files I have if anyone is interested.
Oh and could somebody also tell me how to show the boundary numbers in the Graphics Window?
Thanks!
5 Replies Last Post Mar 13, 2017, 3:45 p.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
This seems to be a bug in the software. If so, we will fix this for the next version of COMSOL. In the meantime could you send the model you are working on into support (support@comsol.com) and mention that I asked for the case to be assigned to me in the forum? Then I will be able to get back to you with a workaround.
In answer to your other question:
"Could somebody also tell me how to show the boundary numbers in the Graphics Window?"
You cannot directly display the boundary numbers in the graphics window, however to identify numbers with boundaries in COMSOL 4.3 you can select a boundary physics feature in the model tree and then choose 'View>Selection List' to pull up a list of numbers associated with the boundaries. When you click on the individual numbers then the relevant boundaries will be highlighted in the user interface.
In answer to your other question:
"Could somebody also tell me how to show the boundary numbers in the Graphics Window?"
You cannot directly display the boundary numbers in the graphics window, however to identify numbers with boundaries in COMSOL 4.3 you can select a boundary physics feature in the model tree and then choose 'View>Selection List' to pull up a list of numbers associated with the boundaries. When you click on the individual numbers then the relevant boundaries will be highlighted in the user interface.
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Mr. Ransely,
I will send the model to the support as you asked. Another way to simulate the bug is to use the example model: "MEMS modul -> actuators -> electrostatically_actuated_cantilever" When I replace the linear elastic material with a hyperelastic material I get the same error.
In the meanwhile I have been able to converge to a solution in my second model. (with the solid mechanincs and electrostatics module)
To further discuss the visualisation of the boundary numbers: I have opened some files that users have put on the discussion forum and when I opened those files the numbers would be presented on the graphics screen. Was this made possible in another version and has it been removen maybe in V4.3? I found it visually very aidful.
Thanks for your time and help,
DHam.
I will send the model to the support as you asked. Another way to simulate the bug is to use the example model: "MEMS modul -> actuators -> electrostatically_actuated_cantilever" When I replace the linear elastic material with a hyperelastic material I get the same error.
In the meanwhile I have been able to converge to a solution in my second model. (with the solid mechanincs and electrostatics module)
To further discuss the visualisation of the boundary numbers: I have opened some files that users have put on the discussion forum and when I opened those files the numbers would be presented on the graphics screen. Was this made possible in another version and has it been removen maybe in V4.3? I found it visually very aidful.
Thanks for your time and help,
DHam.
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi DHam,
To display boundary numbers (and domain, edge and point numbers), make sure to check the "Show geometry labels" checkbox in the View node under Definitions.
Jeff
To display boundary numbers (and domain, edge and point numbers), make sure to check the "Show geometry labels" checkbox in the View node under Definitions.
Jeff
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Thanks Jeff and DHam - it seems like there is always something new to learn about COMSOL, even for those of us who work here!
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
Hello,
I wish to model a hyperelastic strain sensor and have run into the same error above as DHam. The model will simulate using a linear elastic domain, but moving to a hyperelastic nonlinear domain results in an undefined Maxwell stress tensor. In the application library models, changing the material from a linear elastic material to a non-linear material, such as in the capacitive pressure sensor, also results in the error. However, I am using Comsol 5.2.
Can you confirm that this bug was fixed in version 5.2? If so, I will continue to check my boundary conditions, simplify my model, etc.
Thank you,
Kristen
I wish to model a hyperelastic strain sensor and have run into the same error above as DHam. The model will simulate using a linear elastic domain, but moving to a hyperelastic nonlinear domain results in an undefined Maxwell stress tensor. In the application library models, changing the material from a linear elastic material to a non-linear material, such as in the capacitive pressure sensor, also results in the error. However, I am using Comsol 5.2.
Can you confirm that this bug was fixed in version 5.2? If so, I will continue to check my boundary conditions, simplify my model, etc.
Thank you,
Kristen