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 The information provided may be out of date.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.
Boundary conditions on inlets have big effect
Posted 5 apr 2012, 10:25 GMT-4 Fluid & Heat, Studies & Solvers Version 4.2a 1 Reply
Please login with a confirmed email address before reporting spam
I have a simple 2D model of a reservoir with a membrane connected to an inlet and outlet.
It is filled with water.
I want to simulate the flow throught the inlets as function of membrane movemement. For that I use an FSI model.
I try to simulate the transient behaviour when a smoothed force step function is applied to the membrane from outside.
I can simulate this reasonably well if I choose the boundary conditions for the fluid inlets as: open-boundary, no viscous stress.
However, when I want to simulate more complicated transient behaviour, this boundary condition results in a non-stable oscilation that over time will crash the simulation. It works for a simple step function, but for periodic behaviour the unstable oscillation will get larger and larger.
When I set the boundary conditions of the open-boundarys to 'normal stress' f0 = 0 N/m^2, then the simulation will work initially, but over time the time steps will get to small and the result will not converge.
By plotting during the simulation I can see that the new boundary conditions make more sense than the old, but I don't understand why it runs into a wall.
Could somebody help with this?
I've included the model using the old boundary conditions, so you can run a test (5 minutes on Core i3). But I would like to change the boundary conditions of the open boundary 1 to 'zero normal stress'.
It is filled with water.
I want to simulate the flow throught the inlets as function of membrane movemement. For that I use an FSI model.
I try to simulate the transient behaviour when a smoothed force step function is applied to the membrane from outside.
I can simulate this reasonably well if I choose the boundary conditions for the fluid inlets as: open-boundary, no viscous stress.
However, when I want to simulate more complicated transient behaviour, this boundary condition results in a non-stable oscilation that over time will crash the simulation. It works for a simple step function, but for periodic behaviour the unstable oscillation will get larger and larger.
When I set the boundary conditions of the open-boundarys to 'normal stress' f0 = 0 N/m^2, then the simulation will work initially, but over time the time steps will get to small and the result will not converge.
By plotting during the simulation I can see that the new boundary conditions make more sense than the old, but I don't understand why it runs into a wall.
Could somebody help with this?
I've included the model using the old boundary conditions, so you can run a test (5 minutes on Core i3). But I would like to change the boundary conditions of the open boundary 1 to 'zero normal stress'.
Attachments:
1 Reply Last Post 13 apr 2012, 09:24 GMT-4
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
The problem seems to go away when I reduce the applied force.
When the force is reduced by a factor of 2, the simulation finishes correctly and gives results as expected.
When the force is reduced by a factor of 2, the simulation finishes correctly and gives results as expected.