HP-97 f key

[Pierre]

Hi!
A friend recently give me a nice HP-97. Unfortunately, the x2 and the f keys don't work. For the latter, it is very annoying. I took the machine apart: the keyboard contacts are ok, no dirt, no corrosion (as they are in gold). I cleaned them nevertheless, but the problem remain.
Any idea ?

Pierre

[Tony Duell]

I assume that the x2 key is the 'x squared' key, right?
Electrically, the HP97 keyboard is a rectangular matrix of keys. 4 columns and however many rows are needed (14?). The 2 keys you mention -- f and x^2 -- are not in the same row, and they're not in the same column.
Assuming all the other keys work correctly, there can't be a problem with the keyboard interface circuitry. ROM failure is, I suppose, possible (the machine could lose the routines needed to handle these keys, but _very_ unlikely. No, the only
likely cause is bad contacts on the 2 keyswitches, or broken connections (PCB traces, vias) to them. That's what I'd check first.

[Pierre]

Thank you Tony for your help.
All circuitry and contacts are OK. May be for the f key there is an electronical problem ? I don't know how the calculator work internally, but I suppose that the strike on the f key have to be kept somewhere in memory until a second key is pressed ; so may be there is a problem with this memory ?
It is a bit difficult to test this calculator when taken apart, particularly the keyboard, as when the printer and the card reader are not connected, it doesn't work.
All the best
Pierre

[Tony Duell]

It's possible there's a logic fault, but I think it's fairly unlikely. I still suspect the keyboard itself.
As I mentioned, the keyboard is a 4*14 matrix. The 4 columns go to the ACT chip via the first 4 pins on the 18 pin connector on the keyboard PCB (the pins are numbered from the end furthest from the 9 pin connector, that is to say they are numbered from the printer side of the machine).
The 14 rows are scanned by the display cathode driver chip -- the 20 pin chip on the keyboard PCB, possibly marked 1820-1629.
Remove the keyboard from the machine (or at least remove the printer, logic PCB, and support plate). Now connect a continuity tester between pin 1 on the 18 pin connector and pin 4 of the cathode driver chip. You should get continuity when the 'f' key is pressed.
Then try connecting the tester between pin 4 of the 18 pin connector and pin 12 of the cathode driver chip. This time you should get continuity when x^2 is pressed.
If the keyswitches are OK, then it would appear to be a ROM problem on the logic board. As you have a parts unit (I think), you might try swapping out the 18 pin chip (1818-0225) ROM0 on the logic board, and then the 5 little 8 pin chips on the same board with the ones in your other machine. Those are the ROMs.

[Pierre]

Tony,
You were right. There was a bad contact under the two keys. I had to slightly bend the contacts to make them work. It was a bit difficult to find the good balance between full contact and no contact at all, but I finally succeed.
Many thanks for your wise advices,
Pierre

[Tony Duell]

Happy to help, and I'm pleased that another fine old HP calculator is working again :-)