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.
Moving Mesh issues in v4
Posted 2010年8月10日 GMT-4 10:27 1 Reply
Please login with a confirmed email address before reporting spam
I am attempting to port a model from v3.5 to v4 which uses the moving mesh(ale) interface. The model was working properly in v3.5 but I have had some issues getting the expected behavior in v4. I am attaching 2 simplified versions of what I am trying to do which demonstrate the issues I am running into. To my knowledge the 2 attached versions are identical except for the fact that the physics interfaces were added to the model in different orders.
a) ES_ALE_v1.mph – the electrostatic interface was added first then the moving mesh interface was added
b) ALE_ES_v1.mph – the moving mesh interface was added first then the electrostatic interface was added
I think there maybe be a bug which is causing the difference in behavior, do you agree?
Regardless, neither version is behaving as I'd like. Version a) is graphically displaying what I think the simulation is actually calculating though.
In Version b) – 2D Plot Group 1 – select time 0.6. As you can see block #4 has moved up as expected and is also a solid deep red. This is expected also since a fixed voltage of 10 is being applied to it's boundaries. However clicking on the solid red block within the block #3 region shows that the voltage is not 10v there. Instead it is something less.
In Version a) one can observe the movement of block #4 by going to 2D Plot Group 1 – select time 0.6 and Frame: Spatial. In this case you see that the mesh has moved as expected but the voltage does not follow the boundaries of the moving mesh.
I need to move block #4 and have it remain at 10v all the time. The rest of the model should respond accordingly based on the movement of block #4. Any suggestions you could provide are welcome.
a) ES_ALE_v1.mph – the electrostatic interface was added first then the moving mesh interface was added
b) ALE_ES_v1.mph – the moving mesh interface was added first then the electrostatic interface was added
I think there maybe be a bug which is causing the difference in behavior, do you agree?
Regardless, neither version is behaving as I'd like. Version a) is graphically displaying what I think the simulation is actually calculating though.
In Version b) – 2D Plot Group 1 – select time 0.6. As you can see block #4 has moved up as expected and is also a solid deep red. This is expected also since a fixed voltage of 10 is being applied to it's boundaries. However clicking on the solid red block within the block #3 region shows that the voltage is not 10v there. Instead it is something less.
In Version a) one can observe the movement of block #4 by going to 2D Plot Group 1 – select time 0.6 and Frame: Spatial. In this case you see that the mesh has moved as expected but the voltage does not follow the boundaries of the moving mesh.
I need to move block #4 and have it remain at 10v all the time. The rest of the model should respond accordingly based on the movement of block #4. Any suggestions you could provide are welcome.
Attachments:
1 Reply Last Post 2010年8月20日 GMT-4 13:53
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
COMSOL support helped with this issue, I will summarize the fix below so that other users will not make the same mistake.
Essentially the primary stumbling block was that I had not allowed for Free Deformation in my model in the area(“air”) around my oscillating plate(block/domain #4). When I do this in either of the models I attached in the above post, the plate now oscillates as expected and is held at 10V while the area around the plate is appropriately coupled to it and responds as expected.
I am attaching the file that COMSOL provided as an example of how to set up this kind of model.
Essentially the primary stumbling block was that I had not allowed for Free Deformation in my model in the area(“air”) around my oscillating plate(block/domain #4). When I do this in either of the models I attached in the above post, the plate now oscillates as expected and is held at 10V while the area around the plate is appropriately coupled to it and responds as expected.
I am attaching the file that COMSOL provided as an example of how to set up this kind of model.
Attachments: