FEMM needed

Matlab w/ PDE toolbox - used to be about $160 for the student edition. It's not really built in functionality to solve acoustics problems, although they do give an example of sorts... It would also be pretty difficult to do any kind of fully coupled model. It's basically useful for looking at wave propagation given a unit input. You can also predict polar patterns, etc. Check out my webpage for some stuff I've done with it.

I think someone here was also developing stand-alone code to do the same sort of thing, so you might try doing a search.
This probably isn't particularly useful to anyone, but I use NASTRAN at work for structural modeling. It's far from free. There's also an acoustics solver bundled in with that. I've used that some to model my room. Calculating room modes was a piece of cake. Anything else got a bit more challenging. I did some frequency response runs and eventually got some reasonable looking results. You mesh the air. You can define acoustic absorber elements with frequency dependent absorbtion (technically acoustic impedance). For sources you can define an acoustic source or apply an enforced pressure.

I have a model of the LAB horn built, but haven't got anything to successfully run with it.

After sitting at a computer working with FEM (for a living) for 50+ hours a week, I don't have a huge motivation to sit there even longer in my time off to do more. :xeye:

I'm sure I'll do some more with it sometime in the future. I want to do some sound absorber modeling (rigid fiberglass stuff) before spending money on some.

I posted a bit before.
link to batdorf rambling

- Robert