A few things relating to HP machines, particularly the 9845 that I'm working on. Firstly, a correction about dismantling the tape drive (this also applies to the 9825 drive that Sellam was working on). I mis-remembered how the motor is fixed -- the screws don't go in from the bottom, they go in from the top (cartridge side), go through the chassis and tap into the plastic disk that holds the motor. Two of them are obvious, where you'd put the cartridge. To get to the other two, you have to remove the head (outside 2 screws on the mounting block, this shouldn't muck up the alignment) and remove the microswitches (tape inserted, write protect), again 2 screws, but the actuating pins fall out, so be careful. When you've done that, it's easy to remove the motor. Secondly, the keys do seem to work as I dsecribed, with one correction. There are 2 lopps through the torroid, and there is a magnet that affects the magnetic properites of that torroid, and thus the coupling between the wires. But the magnet is normally near the torroid and moves away when you press the key. Makes sense, I would have guesed that the magnet saturated the torroid and reduces the coupling between the loops, so this way, uou get a signal when the key is pressed. Thirdly, if you're working on the keyboard encoder board, note that the keyboard controller chip (with a custom number on my board) seems to have the same pinout as the chip in the HP250 (schematics on bitsavers). The LM311 comparator (8 lead TO99 case) is nothing to do with detecting the return pulse from the key matrix, it's used as a relaxation oscillator for the contoller clock. The return signal is detected by a couple of transistors in the CA3046 array (long tail pair circuit). Finally, I've typed out the DIO bus pinout from that HP9826/9836 Pascal system manual that's on bitsavers (this being the piece of info that's going to be the most useful to me at the momnent). If anybody wants this (a plain text file) to avoid downloading/reading the whole pdf, I can post it here. -tony