"Nick Müller" wrote in message
...

Have you considered a solenoid pump for your model? I don't have any
reason
to believe it would be the answer, but I'm curious.

Solenoid not, but something like a piezo might be an option.
But as soon as you have a plunger, you have trouble.

Since you're into it this deep, I'd like to suggest something you may want
to consider: magnetostrictive metals. I have some (indirect) experience with
them. They produce much more force than piezo, and they're good for (IIRC)
20 kHz. They activate very simply, with a solenoid-type coil, but they do
have some mechanical hysteresis.

They're used in some mechanical applications now. The technology was
developed for sonar transducers. Small bars of the material are reasonably
priced.

Have fun. It sounds like you've developed your work into something close to
laboratory research.

--
Ed Huntress