I've got a non-working HP98035 real time clock module (for HP9825, etc) on the bench. It accepts commands, it sends something back, so I think the bus interface and microcontroller are all fine. The problems are : The real time reads back as 88:88:88:88:88

There is no activity on the crystal connected to the AC5954N clock chip (or at least I think it's a clock chip). Enabling the oscillator testpoint doesn't do a darn thing (well, the control/setting latch U7 changes state so again the microcontroller is doing something). There is no activity on any of the pins on the chip.

It is getting power (it's not a problem with the NiCd battery).

I suspect the chip. Has anyone come across it? It may be a common digital clock/calendar chip, or at least related to one (if it is HP-custom). It doesn't seem to be particularly designed for a microprocessor bus -- the output may be for direct driving 7-segment displays or maybe a multiplexed BCD output to link to a display decoder/driver. The inputs to it are essentially the 'set' buttons, the outputs are the 7 segment lines and digit strobes.

I have the exact same problem. HP98035 real time clock module plugged into a HP9825T, accepts commands, reads back all 8's. Battery is new, is charging and at the correct voltage.

Are you able to set the date and time on your clock? I had disconnected one end of the diode in the charge circuit and powered the clock from a lithium primary cell. I got all 8s when the battery went dead but when it was in that state when I tried to set the date and time any commands after that to the clock hung, I suspect because the nano processor got hung up trying to set the time. After changing the battery it functioned normally again.

It will accept the set time command, but this has no effect. An error read shows code 16, "real time lost". The clock chip seems dead, nothing happening when looking at the test output after sending the test mode request.

You can easily see how they get zapped: they are 2.5V chips, the NiCd battery *is* the voltage regulator. Charging circuit is a simple diode connected to 5V via a resistor. Battery dies, goes high impedance, somebody plugs it in to try it out and poof! Clock chip gets zapped by 5V.

HP were fond of using NiCds as shunt regulators at that time. The did it in many of their handheld calculators (HP20 series 'Woodstock', HP30 series 'Spice', etc). In those it wasn't normally a problem (the calculator electronics drew enough current to pull the voltage down) except in machines with 'continuous memory' (battery backed RAM). There, if the machine it turned _off_ the RAM is the only thing drawing current and it doesn't draw enough to pull the voltage down below the zapping level.

HP should have paralleled the battery with zener with a voltage somewhat above the normal battery charging voltage but below the abs max rating of the power supply pins of the chips. They saved maybe ten cents by omitting that.

The outputs Sa-Sg are standard names for 7-segments, and are read on the processor bus. They don't bother to read segment d, I suppose they can differentiate all the numbers without reading that one. DI must be the Decimal Indicate segment. D1-D3 would be the digit counter. C25, READ and SET are the inputs - we'd have to find what they exactly do.

I wondered if it's actually a digtal watch chip (2.5V could have been a couple of mercury cells in series, LED watches were not uncommon back then). In which case it would not normally have come in a 0.6" wide DIP. Perhaps normally it was a bare chip directly mounted on the watch circuit board or something. The DIP version would be unusual, which is perhaps why we can't find data on it.