COMSOL HELP (Eigenfrequency Analysis) Micro-rotor

[bilz]

Hello Everyone,

I'm new with comsol and would appreciate some help from anyone willing to help me out.

Ok so here's my question, I'm trying to perform an Eigenfrequency Analysis for an electrostatic micro-rotor. The problem is I want to take in consideration the effects of the surrounding air as well, so does anyway know how can I set up the ambient surrounding the rotor to be AIR.

I'm really thankful, in advance, to anyone who responds.

 

[bilz]

Ok maybe I should formulate the question all over again. It doesn't matter what kind of device it is, in my case it is a micro-rotor but it can be a cantilever or anything. The question is how to set up the AIR around the device while performing an Eigenfrequency analysis using COMSOL...

 

[Greg Freeman]

I may be speaking from ignorance of what exactly you are trying to accomplish, but including aerodynamic/fluid affects on a vibrating object is a very complex problem that I doubt you could just plug into a computer and expect a reasonable answer. The best I think you would be able to accomplish would probably be an approximation with the air acting as a damper.

The problem lies in the fact that the boundaries for the coupled CFD problem will be moving.

I'm wondering though if someone else has more experience in this matter.

 

[bilz]

I am not sure of what exactly you were talking about either. But yes i want to take into account the effects of air damping while performing the eigenfrequency analysis...

Anyone??

 

[bilz]

No one? please I am really in trouble over here... I really need to know how to perform the eigenfrequency analysis in air. because if I perform the normal eigenfrequency analysis in comsol it would be like in vacuum... i need the surrounding atmosphere to be air...

PLEASE HELP!

 

[Mech_Engineer]

As was already stated, your best bet is probably to try doing a multiphysics analysis coupling CFD with the frequency response.

Your only other possibility would be to try and come up with an approximation for the part's damping coefficient in air and plugging that into the harmonic response simulation. The real question is really whether you need to get numbers, or you need to get the correct numbers...

 

[bilz]

As was already stated, your best bet is probably to try doing a multiphysics analysis coupling CFD with the frequency response.

Your only other possibility would be to try and come up with an approximation for the part's damping coefficient in air and plugging that into the harmonic response simulation. The real question is really whether you need to get numbers, or you need to get the correct numbers...

I saw your comments for the question about the cantilever beam, in which someone was trying to obtain the eigenfrequencies of the cantilever, as i understand these eigenfrequencies and eigenmodes oof the cantilever in vacuum right? i want to do exactly the same thing (which i already did) but this time considering the air not in vacuum...

I'm sorry but what do you exactly mean by coupling CFD with the frequency response?
Thanks for your replies I truly appreciate it.

 

[Mech_Engineer]

CFD is computational fluid dynamics- a.k.a. finite element anlysis for fluids. CFD can be used to solve for fluid flows for temperature, pressure, velocity, etc.

I don't personally use COMSOL (I did see it demoed), my company uses ANSYS Multiphysics.

If you are using COMSOL Multiphysics (not COMSOL mechanical or structural, whatever the regular mechanical package is called), it's biggest strength is that it can "couple" different physical phenomenea. For a CFD example, you could solve for the temperature of a part with a flow of hot air blowing over it (CFD-Thermal) or how much a part deflects in a high-velocity stream of air (CFD-Structural). Applications can go into other realms as well. For example modeling the inductive heating of a metal bar (EM-Thermal) can be done, as well as things like thermal expansion/deflection (Thermal-Structural).

In your case, I would think you can couple a harmonic-response problem with CFD to see no only what happens to the part, but what happens to the air around it. COMSOL will essentially solve the two separate problems like a set of simultaneous equations with continuity conditions. Just so you know, it may take a long time to solve, depending on your available computing horsepower.

It's important to note, CFD is not an easy thing to implement. It takes a good amount of knowledge in fluid dynamics as well as the specific software package's nuances. If you're not sure what you're doing, the numbers you get may not be anything close to reality, and the software has no way to know that.

 

[bilz]

Thanks a lot for your reply, comments and suggestions. I understand it is not an easy task, especially to couple the harmonic response with CFD and that it will take a lot of computing resources... I was thinking just out of the top of my head, what if I just draw a volume surrounding the rotor (or any structure) and assign to it the properties of air and solve for the eigenfrequencies... what do you think about that? what would i get?

 

[Mech_Engineer]

...what if I just draw a volume surrounding the rotor (or any structure) and assign to it the properties of air and solve for the eigenfrequencies... what do you think about that? what would i get?

As in treat the air as a soft solid surrounding the part? No, that's won't work. Now if you're talking about drawing a volume and then using a CFD analysis on it, that's exactly what you will have to do.