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.
Material domain changes duringe parameter sweep
Posted Mar 2, 2016, 8:14 a.m. EST Geometry, Mesh, Structural Mechanics Version 5.2, Version 5.3 2 Replies
Please login with a confirmed email address before reporting spam
Hello,
we have some issues with a rather simple problem. The Model is a piezoelectric cylinder with a crack. The cylinder is exposed to stress between both planes and the crack itself has a triangular profile. The stress is alligned to close the crack. When the crack is deeper than half of the cylinder, we observed that the crack opened in our simulation, which means that the displacement was positive instead of negative.
We avoided this problem by defining a second material within the crack. However during a parametric sweep where the angle of the crack was changed betwe 10 and 25 degree, we observed that the material of the crack changed automatically to the material of the cylinder.
Does anybody know, why this happens and how it can be avoided? Thank you in advance,
Ulf Zimmermann
we have some issues with a rather simple problem. The Model is a piezoelectric cylinder with a crack. The cylinder is exposed to stress between both planes and the crack itself has a triangular profile. The stress is alligned to close the crack. When the crack is deeper than half of the cylinder, we observed that the crack opened in our simulation, which means that the displacement was positive instead of negative.
We avoided this problem by defining a second material within the crack. However during a parametric sweep where the angle of the crack was changed betwe 10 and 25 degree, we observed that the material of the crack changed automatically to the material of the cylinder.
Does anybody know, why this happens and how it can be avoided? Thank you in advance,
Ulf Zimmermann
2 Replies Last Post Jun 7, 2017, 12:05 p.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
Hi
for me that is strange, I do not see why, and have never experienced "material" change in COMSOL.
However, I have a comment on stress in a cylinder with PZT materials, the cylindrical coordinate gives a different behavior than a linear block, and one can be confused by what appears to be a "bizarre" stress and deformation behavior in cylindrical coordinates, while it can be explained once one develop completely the strain and stress tensor.
Now this might not be your case, but I have burned my finger before on that one.
Structural requires a fully 3D model (like EM) due to the non diagonal cross coupling of the stiffness tensor, or in other words the presence of a non-zero Poisson modulus.
--
Good luck
Ivar
for me that is strange, I do not see why, and have never experienced "material" change in COMSOL.
However, I have a comment on stress in a cylinder with PZT materials, the cylindrical coordinate gives a different behavior than a linear block, and one can be confused by what appears to be a "bizarre" stress and deformation behavior in cylindrical coordinates, while it can be explained once one develop completely the strain and stress tensor.
Now this might not be your case, but I have burned my finger before on that one.
Structural requires a fully 3D model (like EM) due to the non diagonal cross coupling of the stiffness tensor, or in other words the presence of a non-zero Poisson modulus.
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
7 years ago
Hello, some time later down the line.
I'm experiencing exactly the same issue, the problem in my case occurs when the parametric sweep causes some domains to disappear (because the increment in size cause by the sweep) then suddenly Comsol decides to change the material of the remaining domains in a fairly arbitrary way and damages the model, so I have to close it and open it again prior changing that or entirely re-assigning the materials to the domains in the original way.
Is there a setting to force Comsol to stop doing that?
Thanks
I'm experiencing exactly the same issue, the problem in my case occurs when the parametric sweep causes some domains to disappear (because the increment in size cause by the sweep) then suddenly Comsol decides to change the material of the remaining domains in a fairly arbitrary way and damages the model, so I have to close it and open it again prior changing that or entirely re-assigning the materials to the domains in the original way.
Is there a setting to force Comsol to stop doing that?
Thanks