HP262x keyboard to HP120 interface

-----Oorspronkelijk bericht----- Van: cctalk-bounces at classiccmp.org [mailto:cctalk-bounces at classiccmp.org] Namens Tony Duell Verzonden: zaterdag 11 april 2009 20:17 Aan: cctalk at classiccmp.org Onderwerp: HP262x keyboard to HP120 interface For those who are still interested in this (Rik?), I've now built the interface from an HP262x keybaord (actually one 'borrowed' from my HP2623 terminal) to my HP120. The differenve between the 2 keyboard interfaces is quite simple. The HP120 puts the scan counter in the keyboard, the interface being the clock and reset lines for this counter, along with an active-high 'current key pressed' siganl. The HP262x puts the scan counter in the terminal, the inteface is the 7 key select lines (basically the outputs of the counter) and an active-low 'key pressed' signal (open-collector). Although the Hp262x runs the keybaord at 5V, it's all 4000 series logic interally, and can be run at the 12V of the HP120 interface. So the interface between this keyoard and the HP120 is little more than a 7 it counter.

I've now built it, it's a little box with a DA15 socket on one end to take the plug from the unmodified HP2623 keyboard (I obviously wanted to leave the keyoard unchanged so I can still use it with the 2623 terminal) and a short cable coming out of the other end ending in an RJ11 plug to go into the HP120. Inside are 3 chips, all common 4000 series CMOS parts. : A 4024 (7 bit counter). This is linked to the clock and reset lines from the HP120 via the resisotr/diode protection networks as used in an HP150 keyboard. The bottom 6 outputs of this counter are buffred by a 4050 chip and then fed via a 16 pin header plug nad socket to the DA15 socket for the keyboard. This chip may not be necessary, but some counters don't like driving long cables (glitches on the outputs can change the state of the flip-flops). I dout the 4024 sufferes from this, but adding one chip to be sure seems worth it.

The last output from that counter is buffered bu 2 sections of a 4049 in cascade. The reason I used that chip is that I needed a NOT gate to invert the open-collector key-pressed/ signal from the HP2623 keyboard. This I did,, after pulling said signal high with a 3k3 resistor. The output of that NOT gate is again given a diode/resistor protection circuit and fed to the HP120 And it works. It worked first time actually (well, I did test as I went along, but after fitting the last connection , plugging everything in and powering up, it worked). I can type on the keyboard, the correct [1] characters come up. I can use the function keys, configuration screen, etc.

[1] Modulo the fact that I have Danish/Norwegian ROMs in my HP120 (!)

I am now looking at modifying an HP150 keyboard to work with the HP120. More news if I get that working! -tony