Looking for: HP 9000/216 (9816) keyboard

It's a lovelly little machine... I assume you;ve taken it apart. You can remove everyting apart from the PSU by releasign 2 thumbscrews (the expsinon slot cover or expansion board), 2 quater-turn fastners (top cover) and one screw. I don;t know how much RAM you have, most had 2 rows of 4164 chips for a total kof 256K. There';s quite an easy mod to get 1 Mbyte on the mainboard, basically it's fit 41256s in place of the 4164s, add a 74S158/7rF158 in the spare location on the mainboard, I think add a 33R (or is it 22R) resistor amd cut some jumpers. I did it to mine. Argh AFAIK that keyboard is uniqueto the 98716. Mo other HP uses the same keyboard. Actually, is the older version of the 9920 keybaord the same??? The HP120 and HP150 keyboard look to have a simialr interface, but they don't. If froget which way roudn it is, but in one case the keyboard outputs a signal to say it's starting a scan (that is, the next key sent will be logical key 0) and the other takes a signal from the main machien to reset the scan counter. So beacially, the 'start of scan'signal goes in oposite directions in the 2 cases. Anyeway, the keyboard itself contains nothing custom (it's all 4000 CMOS) and scheamtics exist. WHich means you can dedcue the interface details. It would be too hard to make soemthign that will work if you can't find the originak Basically, the keyboard sends the state of each key one after the other. Note that some positions are used for the count (7 or 8 bits, I forget) and direction (one bit) of the tiddleknob. Yes, the coutner for that is in the keyboard. The original keyboard has one odd design feature. It's not an electrical matrix of switches. One eide of each switch is gorunded, the other goes to an input o na 4051 multipleser. It's the only keyboard I've ever seen like that. Where it was done to reduce RFI I do not know. -tony