I have finished constructing a balanced microphone preamplifier. The circuit that I used is here:
http://sound.westhost.com/project66.htm. The only changes I made were in the choice of transistors (ZTX214 and ZTX300) and the use of switched resistors for setting the gain instead of a potentiometer. When it's set for full gain, it's a little noisy but the mic signal is very strong. Medium gain seems to be a good compromise, at least with my PC's "line in" port. Which, incidentally, is far from a noise-free signal path. Will be interesting to try it out on a better quality sound system.
The actual construction is captured for posterity as a time-lapse video, shot with a Mini-DV camera, an IEEE-1394 interface and 'dvgrab' on my Linux box. I set it to grab every eighth frame, which makes the playback run at eight times normal speed. I had to find a 60W filament bulb for the work light, to avoid awful colour shifting problems with the CCFL. Once I've got the video file onto YouTube or Vimeo, I'll post a link.
So now I'm back on the expanded scale voltmeter for the bubble car. All attempts to fix up the old Tandy multimeter have failed. It's got a flaky connection somewhere deep inside (it's two PCBs full of CD4000 series chips and a DVM chip, sandwiched together). My original plan to use a fine 80mm square Simpson panel meter has been thwarted by the fact that there just isn't room behind the panel on the charger for it to go. I have, however, found a smaller 100uA meter which I have fitted with LEDs for illumination. There is room below the ammeter for that one, and I have found the right zener diode to make the circuit work.