HP Forums

Full Version: HP-29C problem
You're currently viewing a stripped down version of our content. View the full version with proper formatting.

I have an HP29C with a weird problem. When I received it it was apparently working but had a low battery pack (in good condition). I charged the pack overnight and decided to try a battery life test by running a "1 +" loop. About twenty minutes into the test the machine locked up with all 0's in the display (signs and exponents included). It has remained in this state ever since. Discharging the internal caps, etc did not revive it. Any ideas out there?

All zeroes and a locked up machine is not good news. Chances are one of your ROM chips got fried.

I never charge a battery pack inside a "C" series Woodstock because of the danger of overvoltage. Charging in an HP-21 or HP-25 that is powered down is the safe thing to do.


Viktor

I'm rather sure it was not a overvolt problem. I've measured the internal voltages and they are normal. Also the machine was disconnected from the line and had run for about half an hour before it died.

Did you measure the voltages while the charger was plugged in?

Parts of the calculator are powered even when the power switch is off. The idle voltage of the charger is much higher (in excess of 10 Volts) than the nominal battery voltage. If there's the slightest problem (e.g., minor corrosion) with the battery contacts, that is sufficient to raise the voltage that the calculator's internal components receive while the charger is plugged in to dangerous levels.

When MOS components degrade, it's not necessarily an on/off process; it is possible that the excess voltage causes cumulative damage, and the calculator eventually loses its mind during normal (battery) operation.

I don't know for sure of course that this is what happened in your case; in fact, I sincerely hope that it is something else, something fixable! It's just that the symptoms you describe are typical of a fried ROM chip; I've had several HP-29Cs with this problem (and curiously, it's almost always the SAME ROM chip, giving grounds to the speculation that it may be a bad batch or something.)


Viktor

Yes, I checked them with a charger attached and compared them to a known good unit. It does not appear that any excessive voltages make it past the DC-DC converter. I don't think these things are quite as fragile as some people think.

I had one where the battery popped out while running a program while I was out of town. It was exposed to full unfiltered charger voltage for about three weeks with no ill effects. I also had a 25C with an open pack (spring broken between the cells) that sat on charge for a month or so, also with no ill effects.

Do you remember which ROM chip is usually the one to die?

Problem is, in the -C models (25C, 29C, 33C, 34C, 38C) of the Woodstock and Spice
series, not everything runs off the DC-DC converter. In particular, the RAM circuitry runs off the
DC-DC converter when the machine is turned on, but when the machine is turned off, the
RAM contents are maintained by a connection from the battery +ve terminal via a resistor (about 10k IIRC).
The charger output is about 14V off-load, but the charger has a high internal resistance (at least 10 Ohms). That, combined with the charger circuit resistor in the calculator means that when the machine is on, the voltages look 'normal'. But when the machine is off, the RAM alone doesn't draw much current
and the voltage rises. Often to a high enough voltage to cook a ROM/RAM chip.
That's why you should _never_ connect a charger to a 25C, 29C, etc unless there's a known-good battery pack in place.
Incidentally, the DC-DC converter in most HP calculators is designed so that the main +ve output about 6.2V) can never be lower than the input voltage. Normally this is OK (the battery is about 2.7V). But if you connect an overvoltage PSU to the battery contacts you can, again, damage chips.

As Tony Duell explained, the issue is not the voltages AFTER the DC-DC converter, but the fact that parts of the calculator receive the rectified charger voltage directly.

A machine is also less likely to suffer ill effects from the 'raw' charger voltage if it is actually turned ON. That's because in that case, the load represented by the machine significantly reduces the voltage. The most dangerous case is when the machine is OFF, the battery pack pops or has a bad/corroded connection, and some components that draw very low current are exposed to the full charger voltage without providing a load that'd reduce that voltage.

As to which of the ROMs tend to die, I think it's the 1818-0376. I've also had other ROM chips fail on me though.


Viktor