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.

3D Fluid Structure Interaction Problem Will Not Converge

Please login with a confirmed email address before reporting spam

I would like to model a thin iron cantilever beam bending due to constant water flow normal to the beam. I am having trouble getting the solution to converge for high aspect ratio beams such as the one in the attached file. If I take the aspect ratio significantly lower (less than 5 or 10), then the solution converges; however, I would like to have a solution for a high aspect ratio wire.

Did anyone else run into problems with fluid-structure interaction high aspect ratio models?

Thanks!


3 Replies Last Post 03.06.2010, 04:32 GMT-4

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 02.06.2010, 17:04 GMT-4
Any inverted mesh problem due to higher deflection? otherwise I can't think of why a longer beam would cause problem and not a shorter one (except numerical errors, ...).
Any inverted mesh problem due to higher deflection? otherwise I can't think of why a longer beam would cause problem and not a shorter one (except numerical errors, ...).

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 02.06.2010, 17:13 GMT-4
I do occasionally get inverted mesh problems when I draw similar models, but for this particular model, I do not get inverted mesh problems.

I believe the length of the beam in this model is 50 mm and it does not converge; however, when I run the same exact model with a 5 mm length, it does converge.
I do occasionally get inverted mesh problems when I draw similar models, but for this particular model, I do not get inverted mesh problems. I believe the length of the beam in this model is 50 mm and it does not converge; however, when I run the same exact model with a 5 mm length, it does converge.

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 03.06.2010, 04:32 GMT-4
Hi

I'm no fluidics expert, but I would say start in 2D its faster to debug.

When I try fluidics I often run into convergence problems, like you.

One thing I'm always fighting with is the gauge pressure and absolute pressure, for me, in my mind, p=0 means rho=0 hence no mass to act on the structure, but this is not so in fluidics.
Nevertheless, and I do not know really why, when I use p0=1 for initial pressure (and outlet pressure) my models goes further, or converges better, for some time, (your model too) but often like here never ends. In fact the slightest mesh change also changes the convergence, so we are missing some physics "knob" to turn correctly here.

I hope that the CFD module will be better tailored, but fluidics has always been a special domain for it gurus ;)

Have fun Comsoling
Ivar
Hi I'm no fluidics expert, but I would say start in 2D its faster to debug. When I try fluidics I often run into convergence problems, like you. One thing I'm always fighting with is the gauge pressure and absolute pressure, for me, in my mind, p=0 means rho=0 hence no mass to act on the structure, but this is not so in fluidics. Nevertheless, and I do not know really why, when I use p0=1 for initial pressure (and outlet pressure) my models goes further, or converges better, for some time, (your model too) but often like here never ends. In fact the slightest mesh change also changes the convergence, so we are missing some physics "knob" to turn correctly here. I hope that the CFD module will be better tailored, but fluidics has always been a special domain for it gurus ;) Have fun Comsoling Ivar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.