HP Forums

Full Version: Chips other than ACT and RAMs going bad?
You're currently viewing a stripped down version of our content. View the full version with proper formatting.

How common is it for HP calculator chips other than the ACT and RAMs to go bad?

In Woodstocks, the most common cause of bad chips is the use of the AC adapter without a battery pack, with a bad battery pack, or with the pack not making good contact. Due to uncharacteristically poor design, this allows the voltages to exceed the limits of the chips.

Similarly, the card reader sense amplifier of the HP-65 (and maybe the HP-67) may be damaged by use without a battery pack.

But have those of you that repair older HPs noticed many cases of other chips going bad?

I'm considering offering tiny modules that can replace the ACT of the Woodstock, Topcat, 67, and 19C calculators. The module would effectively replace the ROM and RAM chip as well, though the anode driver portion of the ROM/anode driver chip would still be required for the display to work.

But I'm wondering whether it's common for other chips in these calculators, or in the Classic or Spice calculators, to go bad. In particular, whether the CRC and PIK chips, or the series-30 CPU or ROM/RAM chips fail often.

Eric

Eric,

It would GREAT if you could produce replacement chips/modules for some of dead circuits in these calculators.

In my experience by far the most problematic of all components in HP LED calculators are the chips in the HP-27. They are, of course, the NMOS chips and are only used in that model. I'll bet that there are more dead NMOS chips in the 27's than there are dead PMOS chips in all other Woodstocks calculators combined.

IIRC the anode driver chip in the 27 is NMOS too, so that might be an issue -- as I think that's one of the frequently failing chips.

-Katie

I'll start with the PMOS ACT/ROM/RAM replacement, since I have more PMOS calculator models to test it in. I could probably do an NMOS-compatible version later.

I don't yet know what the cost of the replacement will be, but it may be high enough that it's not worthwhile for use in the more common models (e.g., 21, 25, 25C).

Replacing an anode driver chip is harder due to the smaller physical size constraint. The electronics isn't that hard to design.

If only the ROM part of a ROM/Anode driver chip fails, my replacement ACT will be sufficient, since it completely ignores the external ROMs and RAMs.

Eric

What chip will you use to replace those old chips? I plan to replace all electronic of my HP-21 with MSP430 plus some drivers for the LEDs and a switched power supply (all from TI, ironic...)

Best regards!

Nelson

A PIC18F microcontroller and a bunch of other stuff to interface to the PMOS signal levels. Packaging it in a module not too much larger than a 22 pin DIP will be somewhat challenging.

I've considered developing a board to replace the entire logic board of the Woodstock series. But that doesn't help any with the 67, 19C, and Topcat series, and I haven't found a source for the style of contact pins HP uses for the connections to the keyboard and display.

In most of the dead Woodstocks I had on my lab table, the 16 pin ROM
chip not connected to the LED display was dead. As it is the same chip type as the ROM/anode driver, it might be that the LEDs act as a overvoltage clamp. This could also explain the burnt out LEDs on some Woodstocks. I never had one where the ROM/anode driver was really dead, except for one (HP29C, sigh). The next common loser is the ACT. So if you build a replacement module that sits in the ACT socket and does not use any ROM, I think you have a solution to fix most dead Woodstocks.