FSI - Latest Discussions - COMSOL Forums

Airfoil in Air using FSI

Wed, 16 May 2012 22:31:14 +0000

Hi,

I am trying to compute an FSI simulation of an airfoil with a constant wind flowing from the right to the left with slip wall condition. The model will converge if the Re number is low (less than 10^5). As I increase my Re, the model will not converge. In additional, my inlet ramp flow velocity will need to be increase very slowly before it reaches constant. I am wondering if there is something wrong in my setup? Or, can FSI be used for high Re.

I am wondering if anybody can point me to the right direction.

Thanks a million in advance!

Ting

Moving Interior Wall in cdf

Tue, 15 May 2012 14:25:58 +0000

Hi,

I am studying pipe flow, currently in 2D.

I have placed a rigid interior wall, making a certain angle with the direction of flow, inside the pipe. Ideally, I would like it to rotate 'freely' around its first edge (see picture, the red point would be the axis). As of now, I am only using the Laminar Flow module.

By freely, I mean subjected to the flow. How can I achieve this?

1) the freely rotating axis
2) make the mesh follow the interior wall

I am reading this http://www.comsol.com/support/knowledgebase/970/ and looking for examples.

Thank you

Inflation of a cavity filled with a compressible fluid

Fri, 04 May 2012 05:28:20 +0000

Hi,

I'm trying to model a cavity being filled with a compressible fluid. The cavity is pressurized, due theoretically to the intrusion of new fluid, which causes deformation of the elastic domain surrounding it.

I've tried to model this using Structural mechanics and applying a load on the boundary of a sphere (embedded within another bigger domain) but my solution doesn't seem to account for the compressibility of the sphere. Any idea of what I'm doing wrong ?? Should I use fsi instead of Structural Mechanics ??

Thanks a lot !

Karen.

FSI+Rotating mechinery

Sat, 28 Apr 2012 06:28:01 +0000

hello, I am ready to simulate the strength of strired tank of baffle. I used the Rotating Mechinery model and the plan strain model. when I initial the model, the comsol 3.5 gives error:Messages:
Failed to evaluate expression
- Expression: cos(-2*pi*rpm*t)*X-sin(-2*pi*rpm*t)*Y

Failed to evaluate variable
- Variable: t
- Geometry: 1
- Boundaries: 14,15,16,17,18,19,20,21,22,23, ...

Can anyone help me fix this error?

Thank you!

two values at a single node

Fri, 27 Apr 2012 22:27:43 +0000

I have an FSI model.
When I try to output the WSS on a particular edge, I notice that there are atleast 20 points where I have two values for WSS i.e., at each of these 20 points, there are two values for WSS.

I am not sure why this is happening. It looks like COMSOL is counting each point twice and I cannot understand why its happening. I am using the union option to create the geometry. So at two intersecting edges, only one point should be accounted for.

I appreciate if someone can clarify what's going on.

Thanks,
Sirisha

converting COMSOL code to the MATLAB ???

Thu, 26 Apr 2012 11:21:43 +0000

Hello,

I'm working with COMSOL to simulate a chemical experience, but it takes too much time to display the results.
So i want to find a link between comsol and matlab if i can convert the code of comsol to matlab that will be faster than using comsol alone !!!
So am asking if i can do this and how ????

watting your answer
thank you very much :)

Postprocessing of Fluid structure interaction result

Wed, 18 Apr 2012 00:18:58 +0000

Hi, dear all,

I am trying out the FSI module, and my model somewhat get converge.
I apply the inlet a pulse wave inlet uniform velocity, with outlet pressure=0, things goes well, but at 10.5s, the sin wave do not looks like what it should look, as 1st attachment.

I want to know what is going wrong so l use laminar flow module, all things go well as attach,a wave occur. as in second attachment
both use time dependent solver and exactly the same solver setting.

so i suspect the error comes from the solid mechanics during fluid structure interaction.

is anyone share experience with me?may i know how i can fix the small "dip" at the last second?

Meshing Using Geometry Difference Feature

Sun, 15 Apr 2012 21:15:47 +0000

I'm doing a project for an aerospace class in which I'm completing a fluid-structure analysis of a balloon design. In order to generate the balloon structure and interior helium I used a difference in geometry (See the attached COMSOL model). When I go to mesh the model setup I recieve an error that domain 2, the balloon domain, doesn't mesh properly. I figure there may be an issue with the geometry difference feature. Can anyone provide any insight on my issue? I wish to generate a mesh of the entire model ASAP.

Thank you in advance!

FSI - stationary problem - parametric vs single studies

Thu, 05 Apr 2012 21:10:41 +0000

I am working on a model that uses the COMSOL in-built FSI physics.

I am running a stationary study .However my inlet and outlet boundary conditions are dependent on time "t". So I set "t" as a parameter and define its value to run the stationary study.

Now, the issue I am facing is, I want to solve the stationary problem for several "t" t=0:0.1:1.
I can set this up by using the "continuation parameter" and use the parametric solver for the solution. But when I do this, the problem fails to converge for any "t" other than t=0. It fails due to inverted mesh elements. In order to get rid of the inverted mesh elements, I tried using linear shape order, I tried several different meshes but it still does not work for fine meshes (it works for a couple of coarse meshes though but that mesh density is not good enough for my study).

However, the very same stationary problem works if I set up 10 different studies (Each study corresponding to each t i.e., t=0 is one study, t=0.1 is another study ....t=1 is another study). I don't understand why this is happening and I am hoping that someone has an explanation or a suggestion to troubleshoot.

Thanks for the help.
Sirisha

Viscous Fluid damping

Thu, 05 Apr 2012 06:06:37 +0000

Hello,

I am trying to calculate the Q-factor for micro-mechanical resonators in fluids. To do this, I have to use FSI interface (navier stokes equation), doing a transient simulation and then calculate Q-factor.

However, The problem is that time domain simulations take a lot of time. So, could I use thermo-acoustic model in COMSOL which is based on linearized NS for viscous damping calculation ?

Also, what is the difference between using FSI (NS) and thermoacoustic solid interaction model (linearized NS)?
Is it all about the amplitude of solid displacement ?

Waiting for your reply

Regards,
Mohamed

Need some help in meshing, inverted mesh elements warnings

Tue, 27 Mar 2012 07:23:37 +0000

I have an FSI problem, by moving a membrane I would like to see the fluid move.

When I setup the simulation, I've played around with the settings, looked at how other models were setup, but I can't get rid of the 'inverted mesh elements' warnings.

Does anyone have any advice on this model?

pulse response of FSI problem

Thu, 22 Mar 2012 11:10:03 +0000

I'm investigating an FSI problem.
I'm interested in the frequency response of a membrane of a reservoir with water but I have found that a frequency analysis with FSI can be a bit tricky.

Instead, I have done a pulse response of the membrane. By applying a force pulse using a smoothed step function on the top of the membrane. I then decrease the time steps and smoothness of the pulse to find the point at which the membrane will 'ring'. From this ringing, I can deduce the first resonant frequency.

The ringing can be seen by plotting the centre deflection (fsi.disp) as a function of time. I've included this plot, as a reference, the applied step function is plotted as well, (scaled so you can compare it).

Currently, I'm using a 2D model with 'coarser' mesh.

For the solver, I clear the update scaled absolute tolerance in the time stepping field.
I find I need to switch to 'fully coupled' solver at shorter time intervals for the simulation to work.

I would like to expand this simulation to a 3D problem. Currently I'm solving for: 29730 dof and the solution time is 344 s. (5 minutes, 44 seconds). (core i3, winXP 64-bit, 8GB of ram)

When I switch to 3D the computing time is going to increase a lot.

The first question is: Is this a valid method of simulating the device, to get an idea of the resonance frequency?

Secondly: what can I do to decrease the computation time, keeping in mind that in 3D the mesh will be much larger.

Gear Pump Modeling

Wed, 21 Mar 2012 14:41:43 +0000

I am looking for anyone to share their experiences in modeling a gear pump in COMSOL. I am looking for feedback on modules that were used, setup of problem incl. boundary conditions, as well as solver configurations.

Thank you in advance,
Andreas.

FSI problem: wall tension plot

Wed, 21 Mar 2012 11:41:43 +0000

Hi,

I have a FSI model (balloon filled with water) which solved successfully, however I don't know if I can plot wall tension of the balloon. Anyone know how to plot wall tension (or wall stress) in COMSOL?

Looking forward for your help. Thanks!

Simulating fluid structural interaction, without the structural part?

Wed, 29 Feb 2012 13:22:42 +0000

The ultimate goal is to simulate the workings of a membrane pump.

In order to learn more about the pump, and also comsol, I want to first forego simulating the mechanical part.

What I would like to do is: model a reservoir with two open boundarys and see how the fluid flows if I change the shape of one wall of the reservoir. This wall is the membrane. At this moment, I am not interested in the structural properties of he membrane. I want a prescribed movement of that wall. which will be in the form a squared sine function. So I want the shape of the wall to change over time. (vibrate) and see how this influences the liquid in the reservoir.

It is not the same as doing a parametric sweep and changing the model shape. This just reruns the same simulation with a different shape, it doesn't 'remember' the previous condition with respect to conservation of mass.

An alternative is to model the membrane wall as an inlet and applying a varying inflow velocity. This is okay, but it doesn't show in the graphs that I'm actuall changing the shape of the membrane.

Any suggestions on how to simulate this? Will I need to use FSI?

Many thanks for any suggestions.

-Jaap

boundary conditions

Tue, 28 Feb 2012 10:28:13 +0000

Dear all,

I have an incompressible fluid flow problem. If i have only inlet but no outlet in the model (the mass conservation is fulfilled by expanding volume), which boundary condition (velocity or pressure) is best to be implemented?

In fluid flow (laminar flow) model, is it a reference pressure must be defined somewhere in the model? If yes, is that mean I hv to find a way to define reference pressure if i use velocity as inlet condition?

What if FSI physics - is there any reference pressure needed?

solver configuration setting

Mon, 09 Jan 2012 16:01:16 +0000

Hi:

Would some one teach me what the solver configuration means? When I follow the inkjet model, it says set "dependent variables 2, mod1_u, scale" as 10, and set "mod1_p" as1e4. What does that mean?
Thanks a lot.

Asynchronous Machine time dependent startup, singularity error

Thu, 21 Jul 2011 16:00:12 +0000

Hello,

I am trying to simulate a ASM in the rotating machines physics, time dependent.
The Stator slots are coupled to the electrical circuit physics for a Voltage excitation
The rotor slots are also coupled there to define the short circuit resistants.
I am using the ODE Interface for a rotation velocity equation.

When simulating for 20ms, with a timestep of 1ms(or less), everything is fine (rotor slowly starts rotating).
But when I want to calculate for a longer time, the computation stops somewhen in the middle, saying something like:

"Repeated error test failures. May have reached a singularity.
Last time step is not converged."

I choose the mesh as extremely fine and tried a step size of 55us.
As "Maximum Step" and "Initial Step" I'm using the same value as for "Step Size".
But it does not help...

Does anybody have any advice?

Thanks in advance

Automatic Remeshing for Moving Meshes

Tue, 28 Jun 2011 13:05:25 +0000

According to the release notes, version 4.2 is supposed to have a new automatic remeshing feature for moving meshes:

"For a simulation that includes a moving mesh, a new automatic remeshing feature makes more extreme deformation states possible. When the mesh is deformed beyond a user-defined mesh quality threshold, the automatic remeshing activates, and the simulation is automatically continued, starting from the new mesh."

However, I am working with a fluid-structure interaction model, and I can't seem to get this feature to work nor find any mention of it in the documentation or help files. Has anybody tried this out?

Thanks,
John

in 2D fsi mode , i cant define "contact pair" ?

Sun, 19 Jun 2011 17:00:12 +0000

Dear Dr. Ivar,

I am trying to simulate natural heart valve in 2D fsi mode. when leaflets contact an error occurred.

Repeated error test failures. May have reached a singularity.
Time : 0.002592892726514386
Last time step is not converged.

When I define “contact pair” in definition and add “contact 1” in fsi node , I get error:

Failed to find consistent initial values.
Last time step is not converged.

How can I use contact pair? Or is there another way to modeling this example?

Please help me

Thank you so much!