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.
How can I remove a prescribed displacement/
Posted Apr 5, 2010, 10:35 p.m. EDT 2 Replies
Please login with a confirmed email address before reporting spam
Hi,
My model has a polymer film over a fluid layer using a fluid structure interaction mode.
I want to calculate a reaction force when the film is pressed and released, and a recovery time to the original shape.
In other words, on the press boundary, Rz = delta for t<=t1, and after t1, Rz should be removed, not Rz = 0.
But I have some difficulty applying such bounday conditions properly.
Could you let me know how to set the BC for the recovery, or any other technique to run the same situation?
My model has a polymer film over a fluid layer using a fluid structure interaction mode.
I want to calculate a reaction force when the film is pressed and released, and a recovery time to the original shape.
In other words, on the press boundary, Rz = delta for t<=t1, and after t1, Rz should be removed, not Rz = 0.
But I have some difficulty applying such bounday conditions properly.
Could you let me know how to set the BC for the recovery, or any other technique to run the same situation?
2 Replies Last Post Apr 6, 2010, 1:37 a.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
If it's just one time step like that, you might be able to run the simulation up to t1, then save the result and use it as the initial condition for a second simulation where you've removed the BC in question.
Does that get you anywhere? Maybe it wouldn't work if it would interrupt inertia effects or something like that.
-Jeff
Does that get you anywhere? Maybe it wouldn't work if it would interrupt inertia effects or something like that.
-Jeff
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
You can use some "brute force method", at least it works for me in structural.
When you apply a constraint COMSOL adapts the equations in "Physics - Equation Systems - Boundary setting - Weak tab" variables "constr" and "constrf", you can add here a time bolean function of the type
(Ry_smpn-v)*(t<0.001) if the constraint is in y to apply your initial condition during the first 1 msec, or just *(t==0) in for these TWO variables multiplying the corrsponding elements therein.
Note that this will issu some warnings as you are changing the boundary conditions when you solve.
Furthermore you are tampering with the equationsin in such a way that COMSOL does not understand it anymore, therefor it "locks" the equations (you have always the "reset equation" to come back to COMSOL nomal understanding).
This is not enough, as when you apply suddenly a restriction of this type, in transient, you system starts (if you leve the initial values as default 0) in a stressfree state, so you must first run a static analysis with constant t=0 to get the starting values distributed, then save this and restart (by removing constant t=0) and switch to transient mode and then run your analysis.
Try it out carefully with a simple canteliver, it might have some other side effects, so always check carefully your results
Have fun Comsoling
Ivar
You can use some "brute force method", at least it works for me in structural.
When you apply a constraint COMSOL adapts the equations in "Physics - Equation Systems - Boundary setting - Weak tab" variables "constr" and "constrf", you can add here a time bolean function of the type
(Ry_smpn-v)*(t<0.001) if the constraint is in y to apply your initial condition during the first 1 msec, or just *(t==0) in for these TWO variables multiplying the corrsponding elements therein.
Note that this will issu some warnings as you are changing the boundary conditions when you solve.
Furthermore you are tampering with the equationsin in such a way that COMSOL does not understand it anymore, therefor it "locks" the equations (you have always the "reset equation" to come back to COMSOL nomal understanding).
This is not enough, as when you apply suddenly a restriction of this type, in transient, you system starts (if you leve the initial values as default 0) in a stressfree state, so you must first run a static analysis with constant t=0 to get the starting values distributed, then save this and restart (by removing constant t=0) and switch to transient mode and then run your analysis.
Try it out carefully with a simple canteliver, it might have some other side effects, so always check carefully your results
Have fun Comsoling
Ivar