The Museum of HP Calculators

HP Forum Archive 13

[ Return to Index | Top of Index ]

Recalibration of HP-01
Message #1 Posted by Thomas Reins on 17 May 2003, 12:30 p.m.

Hi. I have a 1977 model HP-01 watch that is losing much more than the 30 seconds per year these were originally calibrated for. I'd like to recalibrate it, if possible. Does anyone here have any knowledge, either direct or as a web reference of this procedure? I can't seem to locate any potentiometers on the board at all, there seems to be no adjustments to tweak - though admittedly I've been unwilling to completely disassemble it without further information... I'd really like to get this thing back up to specs - as it is now, it loses about a second per day - 00:06:15 per year. Does anyone here remember anything about the method to correct this?

      
The crystal oscillator engineer speaks
Message #2 Posted by NH on 18 May 2003, 4:28 p.m.,
in response to message #1 by Thomas Reins

Hello,

I have done design work on little crystal oscillator circuits previously. Some of them were for 32.768 kHz, others were for 14.31818 MHz .

Anything I came up with was strictly ho-hum, using a dedicated CMOS cell that was designed on a silicon chip layout software. Little more than a 2-transistor logical inverter really.

You bias it to the halfway point, like, if it is a 3V supply, then you bias it to 1.5V . It acts like a little amplifier, and with the right load capacitors, you get a reliable oscillator. If you mind your P's and N's, maybe low-current consumption also.

BACK TO YOUR CALCULATOR WATCH, I have 3 suggestions, and the 3rd suggestion is the one to focus on:

1. you might conceivably try to change out the crystal. Probably it is in some tiny oddball metal-can, but I have seen some very small packages for 32.768 kHz. NO, I dont even know if your HP-01 runs on 32.768 kHz it might be something else. Also, if you did replace it the result may be worse than if you left it alone. You could solder the original back in, unless you had destroyed it with the extra handling, which is a 50-50 chance.

2. If you can see little load capacitors (typically there are two) each of which goes from either terminal of the crystal, to ground, then you may have something. Increasing the capacitance typically slows it down, less capacitance increases the speed. Since yours is running too slow, you'd need to have less capacitance. BAD NEWS, these capacitors are often integrated onto the custom silicon die (thats how I did the ones I designed).

3. You could leave it alone. If you lose one second per day, that's nuthin'. Twice a year you have to reset it for daylight savings time. And what most of us do with a watch that runs a little slow, is you set it about 3 minutes fast. After a couple months, it will be right on, and after that, it will be a little slow.

Perfectionism is a dangerous game when dealing with old nostalgic items. You have something thats pretty good, and it works. Start messing with it and you break it, then you have nothing (because its been discontinued by those clueless marketeers so there is no factory building you more).

Good luck,

- Norm

      
Re: Recalibration of HP-01
Message #3 Posted by Ellis Easley on 18 May 2003, 5:33 p.m.,
in response to message #1 by Thomas Reins

30 seconds per year is less than 1 part per million! The "pretty good" canned oscillators that go in computers are spec'ed 100 ppm, I think - I don't know how that compares to "broadcast standards". My DigiKey 2000 catalog has oscillators in +/- 100, 50 and 25 ppm. Also 32768 cylindrical can crystals in three sizes: 5.1mm long X 1.5mm dia, 6.2 X 2.1 and 8.2 X 3.1.

I wonder how they got such a high accuracy on the HP-01? It must have some kind of adjustment, but as David Smith says the module is non-repairable, maybe it is a thick film hybrid, laser trimmed.

            
Re: Recalibration of HP-01
Message #4 Posted by Dave Shaffer on 18 May 2003, 7:07 p.m.,
in response to message #3 by Ellis Easley

I've never seen the inside of an HP-01, but many other digital watches, even the cheap Casios, have a little variable capacitor somewhere inside for diddling the oscillator frequency. If you can get inside, look for something with a screwdriver-like slot. Try turning it (carefully, and not too much!) and track the time for a few days to see if things got better or worse. Repeat as necessary.

If you just want accurate time, for $30 you can get a desktop clock that receives the low frequency (60 kHz) signals from WWVB. It stays within a few microseconds.

If that's still not good enough, let me know. I'll take you with me the next time I go (radio astronomy) observing. We know the time to within a few billionths of a second at most of our stations. After we do a Very Long Baseline Interferometry experiment, we know the relative time betwen antennas all over the world to about 0.1 billionth of a second. (Or, at least what it was a week ago, when we did the experiment!)

                  
Re: Recalibration of HP-01
Message #5 Posted by Chan Tran on 18 May 2003, 8:58 p.m.,
in response to message #4 by Dave Shaffer

I didn't know that the HP01 has such a good accuracy. Most of the watches I can find have the spec of +/- 15 seconds/month which is a lot worse. May be the 01 was calibrated using software similar to the AF factor in the 41 clock module. By adjusting the AF factor in the 41 I can achieve similar accuracy. Actually my 2 HP41 gain less than 5 secs a year.

      
Re: Recalibration of HP-01
Message #6 Posted by David Smith on 18 May 2003, 9:19 p.m.,
in response to message #1 by Thomas Reins

The HP01 crystal is not 32768KHz. I think it is 768KHz, but dont know for sure. It has been ages since I opened up my HP01. I seem to remember that the crystal is one of the few components that could possibly be changed. I don't remember if there is a tuning cap on the crystal either.

            
Re: Recalibration of HP-01
Message #7 Posted by Thomas Reins on 19 May 2003, 12:01 p.m.,
in response to message #6 by David Smith

Well, it's cased in plastic at the moment, and I don't even know how it's supposed to be removed, whether it pops out or screws out, and I don't want to force it either way. Most of the electronics I've calibrated/repaired were larger things - oscilloscopes, spectrum analyzers, &c... Bigger parts, and a schematic diagram... How did you go about getting the innards out of the case without damaging the whole board?


[ Return to Index | Top of Index ]

Go back to the main exhibit hall