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.
FSI Module Boundary Conditions
Posted 3 giu 2016, 14:11 GMT-4 Studies & Solvers Version 5.1 1 Reply
Please login with a confirmed email address before reporting spam
Hi All,
I'm using the FSI module (Fluid Structure Interaction) to simulate a sphere in a fluid. My problem is that I want to simulate an elastic sphere rotating in a viscous fluid. So, I've created my geometry as a region of fluid, with a solid sphere in it. Normally I would use the 'sliding wall' boundary condition on the sphere, but it is causing the 'fluid solid interface' node to be overwritten. This is an undesirable effect, so I went in to the equation view and modified the VxWall, VyWall, and VzWall nodes to reflect their individual components of velocity at the surface of the sphere. I defined a spherical coordinate system for the sphere and solved for the individual components of velocity as:
u = -sysU.r*Omega*sin(sysU.phi*sin(sysU.theta)
v = sysU.r*Omega*sin(sysU.theta)*cos(sysU.phi)
w = 0
I had success with a similar approach when I was only trying to get the elastic sphere to translate about one axis (z), but cannot seem to get this current simulation to work. The error that I am getting is "Undefined value found in the equation residual vector. There are 476 degrees of freedom giving NaN/Inf in the vector for the variable x". Is there a better way to do this, or do you have any suggestions?
Thank you,
Adam
I'm using the FSI module (Fluid Structure Interaction) to simulate a sphere in a fluid. My problem is that I want to simulate an elastic sphere rotating in a viscous fluid. So, I've created my geometry as a region of fluid, with a solid sphere in it. Normally I would use the 'sliding wall' boundary condition on the sphere, but it is causing the 'fluid solid interface' node to be overwritten. This is an undesirable effect, so I went in to the equation view and modified the VxWall, VyWall, and VzWall nodes to reflect their individual components of velocity at the surface of the sphere. I defined a spherical coordinate system for the sphere and solved for the individual components of velocity as:
u = -sysU.r*Omega*sin(sysU.phi*sin(sysU.theta)
v = sysU.r*Omega*sin(sysU.theta)*cos(sysU.phi)
w = 0
I had success with a similar approach when I was only trying to get the elastic sphere to translate about one axis (z), but cannot seem to get this current simulation to work. The error that I am getting is "Undefined value found in the equation residual vector. There are 476 degrees of freedom giving NaN/Inf in the vector for the variable x". Is there a better way to do this, or do you have any suggestions?
Thank you,
Adam
1 Reply Last Post 10 giu 2016, 12:24 GMT-4
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
I was able to impose the rotation BC in the equation view, but the solver will not converge and says there's an "Undefined Value Found in the Equation Residual Vector", unless I fix every node on the rotating sphere which I believe is throwing off my results. Any ideas?
Thanks,
Adam
Thanks,
Adam