Searching for IC from Thermohygrograph - Unknown Timer / Pulse generator?

[Ysjoelfir]

Hey everyone, I recently got a used Thermohygrograph which was sold as not functional for just 15€. It turned out to be the batery operated kind with an electric motor driving the drum, instead of the classic mechanical drive. Which is nice, since that thing can actually be switched between 1/7/14/31 days for one drum revolution.
The circuit is incredibly simple. Just the battery, the drum motor and for each time variant one crystal, 1 or 2 15pF caps and a mystery chip WHERE THOSE DICKHEADS SCRAPED OFF THE MARKING! 
I measured the signal pattern reaching the scope and it turned out to be just a simple square pulse with changing polarity (amplitude: +/- 1V (1.2, dropping to 0.8 over the pulse time) and changing delays (~4.2 secs for 24 hour mode) between pulses. The Pulse length is always identical, 48.8ms.

Since the basic circuit is always the same for each time setting the IC must be different. Does somebody know of an IC that does the described thing? I would need the 31 days version (non functional - to bad, that is THE one I would like to use) and ideally a 14 days version (which is not populated).

I attached some pictures for your reference.

[Ysjoelfir]

Also, does someone know of a source for those pens? I did only find a very expensive source for around 16€ for one. Which seems a bit unreasonable if you take into acount the price I paid for that thing.

(sorry for the double post, somehow the software thinks the attachment didn't fit into the first post...)

[fzabkar]

Could it be a programmable microcontroller of some kind, ie same chip with different programs?

Could you program a PICAXE to do the stepping, albeit with a transistor driver?

[amyk]

The low voltage supply and odd pinout makes it unlikely to be a microcontroller. That PCB looks like a very old design, so my bet is on it being a timer ASIC. Late 80s/early 90s IC catalogs might be useful...

[fzabkar]

I'm guessing that the stepper motor's resolution is 0.018 degree per step, which is 20,000 steps per 360 degrees. That works out at 4.32 seconds per step on the 24 hour setting. Does that make sense?

[TurboTom]

What about EM1344 quartz clock chip? 

[amyk]

What about EM1344 quartz clock chip? 

Pinout looks like an exact match (how did you find that?) but I don't think the oscillator frequency is right...

[Ysjoelfir]

I think that it could well be a micro, sure. Sadly, I have no experience with PICAXE, neither do I have any programmer other than an Atmel ICE Basic. Right now I think about designing an replacement using the attiny 202 and two AAs. Could actually be pretty easy to do. Combine that with a small DIP switch to choose the apropriate time setting and done.
But never the less I would love to know what chip they used...

The EM1344 looks interesting, but as amyk already wrote, potentially not suitable for that thing.

regarding the motor: It's not a stepper, it has just two poles. It looks more like a relay with an added gearbox. one tick actually moves the first wheel quite a bit. interesting design. If you want to see it I can take additional photos from the motor section.

Edit: Just learned about bipolar stepper motors with just one winding. So, yes, it definitely can be a stepper. Interesting, I have never seen one like that.

[fzabkar]

The datasheet for the quartz clock chip states that "both the motor pulse period and motor pulse width are programmable by metal mask over a wide range of values". Obviously the EM1344 uses a different crystal, but perhaps these chips were a custom order.