...And now it fits. When we demonstrated to the company president he asked if I could do three more. I said "No way!"

First: trying a new antenna. Cut off the end of the board and added something non-metallic, then mounted the coil. Second, trying a new sensor, a piezo accelerometer with an op-amp. We kept the accelerometer, did soemthing else for the antenna.

First: A base unit made into a protocol sniffer using custom firmware and a kludged-on RS-232 port (bottom left). Worked out VERY nicely.

Second: Attempt at kludging in a separate FM radio and using this as receiver so I could capture accelerometer data without wires getting in the way. (Never got it finished.) The dipswitch was used to select the channel, the FM transceivers had three. The wire sticking out to the left and the ground trace made up the dipole antenna.

Here's the transmitter end of the accelerometer diagnostic tool I mentioned in the previous photo. Note the AM transmitter on the right that we used for normal operations.

Underside of the FM transmitter kludge. The AM transmitter didn't have enough bandwidth to do its normal I/O and add extra data, so I was trying to use an out-of-band method to capture the extra. Note the carefully tuned bare-wire antenna.

Sorry about the focus. I needed an indicator on the outside of the case, so I glued this LED onto the board. You can see the inside top of the case in the background with pencil marks and the hole already drilled.

The wand from the previous photo, enclosed in the case. I used a counter-sink bit to get the chamfer.