The first part I'm attacking is the keyboard. The keyboard is the reason I want to build a "modernized" Psion, I really like this form factor! The plan is to convert it to a USB keyboard since that will work with any modern computer, but before I can do that I need to figure out how it works.

While i probably could have reverse engineered the keyboard by measuring on the connector and pressing keys, I thought it would be easier to disassemble it completely and study the PCB inside - especially since the keyboard connector is pretty small (0.5mm pitch). So I started with popping off the keys and removing the rubber pad. All the keys went into the ultrasonic cleaner for a few minutes (they weren't that dirty, but since it already was in pieces...). The rubber pad and PCB weren't cleaned in liquid, since I might have damaged the conductive carbon surfaces.

Working my way through the PCB with the help of some macro shots and my multimeter, I eventually ended up with the matrix configuration. Some of the function keys get special treatment, I'm guessing to avoid ghosting issues. Otherwise they could have gotten away with having a 6x8 matrix for the 53 keys, this way they could have saved 6 GPIO pins on the CPU. I also designed a quick breakout board to allow me to start writing the firmware for my keyboard.

Speaking of number of pins - the eagle-eyed amongst you will have noticed that pins 1 and 22 are unconnected in the schematic below. The reason is that the keyboard uses 20 pins, but the FPC cable is wide enough for 22 pins! If I would have designed my breakout board with a 20 pin connector, the cable would have been too wide to fit... I think an earlier design might have used some kind of ground loop on the two outermost pins to prevent EMC issues, and it was removed for some reason later in the design process.

The PCB was ordered from OSHpark, but there is plenty of other stuff to do while waiting for it to arrive... :)