Coupling electrostatics (dynamics) with vibroacoustics

General discussion about Elmer
Post Reply
SqFKYo
Posts: 3
Joined: 08 Dec 2009, 14:17

Coupling electrostatics (dynamics) with vibroacoustics

Post by SqFKYo »

http://www.csc.fi/english/pages/elmer/e ... ctricforce shows that Elmer can be used in calculating electrostatic force. On another hand the Elmer poster says that it can also handle vibroacoustics. Now would it be possible to combine these two by using AC current, and calculate electrostatic force at different points of the cycle, and then combine it with vibroacoustics to see what kind of sound the bending elastic material would produce?

I'm mostly looking for yes/no answer here so that I know whether I need to find another FEM-simulator, or should I just try to dig into the manual to learn how to make this happen (have tried reading it, but it's going slowly, and if the answer is "no"...). Thanks in advance for any responses, and sorry if the answer is obvious.
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: Coupling electrostatics (dynamics) with vibroacoustics

Post by raback »

Basically yes, but this does not work out of the box. Also, I would not recommend combining transient and harmonic equations.

If you choose time domain you could basically use transient elasticity equations coupled with electrostatics and even Navier-Stokes. This will end to a quite heavy computation.

If you choose frequency domain it will be rather difficult to get consistant solution i.e. it is difficult to take into account the electrostatic and acoustic forces (pressure from Helmholtz) when you compute the displacements. In simple geometries you may perhaps express them in terms of spring and damping coefficients but generally this will be difficult. Basically this means approximation where inertial forces dominate. Also frequency domain will always be limited to small displacements. Currently also the electrostatics does not provide a harmonic version so you might need to use some other similar equation that does, or involve a little coding.

To conclude, yes this field could be studied (and also has been studied) with Elmer but there are currently still some tampering particularly in the coupled frequency domain analyses while transient may not be computationally feasible.

-Peter
Post Reply