Hi how did you get on with the G757? My board like yours appears to be in great condition, but the LCD only comes on if I heat the back with a hairdryer for a few seconds. It then lights up perfectly (all segments) for just a few minutes. Then i'm assuming it goes off once it's cooled. Makes me think it's the zebra strips expanding and contracting. Although could even be the solder etc. making better contact once it's hot. Question is could things like the capacitors come to life when hot, but die when cooler?
I doubt that it's related to intermittent contact. There are many articles that describe the aging process of ceramic capacitors that use barium titanate (BaTiO3) - an example being Google for "digikey capacitor aging what to expect" - they all say much the same thing - ceramic capacitors lose 3% of their capacitance per decade and that they can be rejuvenated by heating them above the Curie Point of around 125°C. Resoldering does much the same thing by localized heating - there is no minimum time period, the criteria is that the temperature of the capacitor must exceed the Curie Point. In my case, I carefully monitored the oven temperature using a thermocouple probe and followed the same guidelines provided by murata - prior to distribution, they oven bake their capacitors for 90 minutes. Only the bare PCB goes into the oven and I placed it on a small piece of kitchen towel (the LCD display and zebra stripe connectors would not appreciate the same process). Since my first attempt, I bought a G757-4000 from ebay. This one exhibited the same tell tale symptom of a fading display that responded to warmth. This one had also experienced mild battery leakage. So I stripped it down, gave it a good wash with mild soapy water and cooked the PCB for 90 minutes before reassembling it. It's also been working without problem - the battery leakage damaged a tiny piece of PCB track that connected the tiny filament bulb, so the light doesn't work ... but it looks like new and now has a very clear display. The biggest problem that I had was when the tiny gold spring back panel piezo connector jumped across my kitchen worktop - fortunately it's magnetic but it still took me an hour or so to find it ! I never did get any helpful feedback from Seiko about the precise values of the capacitors but I do strongly believe the fact that localized heating can rejuvenate them and my experience tends to substantiate the fact.
@@craigkellum So much responsibility ... it worked twice for me and I'm pleased to say that both watches have exhibited perfect behaviour except for the light function on the 4000 model. I hope that your experience will prove to be just as good. 😊
Curious ! - How sure are you that the capacitors are 180nF (anyone talked to Seiko ?) ? I've also got a G757, at first I thought that it was a contact problem and mine shows the same symptoms. I've found that if I keep the module warm(ish) by sitting it on my router then it all springs into life. If you measure the voltage on the small test point at the bottom right corner of the black plastic rectangle - with a new battery, it's 1.57v but then jumps to 2.53v when the upconverter's running. So it looks like the switching of the drive transistor(s) might be stalling. Anyway, have you a progress report ? Just another note - you'd think that it'd remember the time setting as the upconverter probably only relates to the LCD ... I've found that mine forgets the time and effectively reboots, so maybe it's to do with whatever drives the whole logic of the watch - and that might be the main 32.678KHz oscillator - maybe it's the culprit if it keeps stopping ? Might put a scope on it if I can find a suitable test point. ... Diarized account and thoughts : ... Put a scope on the test point where I measured 2.53v and there's actually a 1.5v 1.024KHz square wave offset by 1v. ... It does look as though heating the PCB might be a good option - I hadn't appreciated that ceramic capacitors age significantly over the years but heating them above their Curie Point temperature can rejuvenate them (wish that applied to people) - so barium titanate (BaTiO3) has a Curie Point of 120 - 130°C ... so cooking them for a couple of hours should restore their original capacitance (and that's exactly what the ceramic capacitor manufacturer does before their capacitors are sent to distributors); some have said that re-soldering a capacitor has the same effect.. ... Might have a go at this - the LCD will not appreciate being cooked, so only do it to the bare PCB. ... Tried cooking PCB at 150°C, cleaned and re-assembled - all working again ... stopped after a couple of hours. The 4 capacitors didn't seem to be all the same value, looked more like the first 3 are 100nF and the last being 220nF (the battery capacitor is maybe 100nF) - when tested before cooking they were all significantly lower ... cooking them did result in increased capacitance but still a blank display and time resets; think that it might be the master 32.678KHz oscillator, maybe the crystal ? ... A few days have now elapsed and some despondency. Dropped a line to Seiko to see if they'd tell me the values of the capacitors and any help about the notoriety of the watch to behave badly after a battery change - yes, this watch has quite a reputation for dying / display instability when the battery is changed. Seiko were friendly and answered by telling me that they can not offer a repair service as they no longer have the parts ... I wrote back to ask if they'd just tell me the component values and to see if there was a circuit diagram. In the meantime, the now "cooked" PCB had suddenly restarted and seems to be behaving itself. It sat for several days without problem on my router and I then thought let's freeze it, so I put it in my refrigerator for a couple of days and it's still working - time still precise to the second even after surviving near 0°C temperatures ! Later, it became an ornament in my sitting room without problem. So it's still defying logical fault diagnosis - it's looking stable but I bet it fails as soon as I put it back in its case and that back is really hard to get off ! I'm kind of concluding that there is an intrinsic design flaw - maybe it doesn't like a brand new battery or something about its characteristic like its impedance - maybe 1.57 volts is too much juice and it needs to be 1.55 volts ... who knows !? ... Back in its case and stable ! I've tried leaving it in warm places, tops of radiators and then cold treatment in the 'fridge and its been steady as a rock ... the watch had been very temperamental, as others have experienced, I am tending to think that either the battery has now "settled in", "cooking" the board did it or it was down to some ethereal intervention ! For those who might follow my PCB cooking example, I used a domestic Bosch oven (fan assisted) and a voltmeter with a thermocouple probe, free mounted on the middle shelf, next to the PCB that sat on a piece of paper. The PCB was cooked at 150°C for 2 hours (the Bosch dial setting was surprisingly accurate !). ... Oh, yes, Seiko Service UK were very friendly but will not or can not provide any design details like the capacitor values - maybe someone else can have a go ? ... Watch this space ! - It's still going no matter where it is ! I think that I'll now quit while I'm ahead - really inconclusive opinion and diagnosis ... it might've worked if I'd just left it alone with its new battery, maybe cooking did cure it or maybe one of the gold flashed contacts were not good despite having been cleaned and again maybe it's just down to "magic". ... My watch is now fully working and here it is : drive.google.com/file/d/16jeu8kO6ckfpDKHmJ3UV2R-MSZBvF5lf/view?usp=sharing I bought it in Bury, Lancashire in 1981 - it had a flat battery in the shop and they knocked it down to £25.00 on the understanding that the guarantee was void. They wouldn't replace the battery but another jeweller did and it worked for many years. I bought the watch because I liked it and it was sometime later when I became aware of the Roger Moore / James Bond fashion accessory. ... 29 Oct ... watch still working without any problems, even after the "hour change" ... this watch has a "bad reputation" vis-à-vis new battery replacement and I now suspect the stability of the 32KHz oscillator to be the culprit, basically bad design. One thing is clear is that heating it for 2 hours at 150°C has some rationale and that, in my case, it has now continued to function for over 2 weeks. In the absence of design component values, the idea of capacitor replacement is a real gamble and I feel lucky that this wasn't my first practical solution. 🤓👍🦊 01 December ... Watch still working without any problems ... Have you changed the capacitors yet ?
Hi how did you get on with the G757? My board like yours appears to be in great condition, but the LCD only comes on if I heat the back with a hairdryer for a few seconds. It then lights up perfectly (all segments) for just a few minutes. Then i'm assuming it goes off once it's cooled. Makes me think it's the zebra strips expanding and contracting. Although could even be the solder etc. making better contact once it's hot. Question is could things like the capacitors come to life when hot, but die when cooler?
I doubt that it's related to intermittent contact. There are many articles that describe the aging process of ceramic capacitors that use barium titanate (BaTiO3) - an example being Google for "digikey capacitor aging what to expect" - they all say much the same thing - ceramic capacitors lose 3% of their capacitance per decade and that they can be rejuvenated by heating them above the Curie Point of around 125°C. Resoldering does much the same thing by localized heating - there is no minimum time period, the criteria is that the temperature of the capacitor must exceed the Curie Point. In my case, I carefully monitored the oven temperature using a thermocouple probe and followed the same guidelines provided by murata - prior to distribution, they oven bake their capacitors for 90 minutes. Only the bare PCB goes into the oven and I placed it on a small piece of kitchen towel (the LCD display and zebra stripe connectors would not appreciate the same process).
Since my first attempt, I bought a G757-4000 from ebay. This one exhibited the same tell tale symptom of a fading display that responded to warmth. This one had also experienced mild battery leakage. So I stripped it down, gave it a good wash with mild soapy water and cooked the PCB for 90 minutes before reassembling it. It's also been working without problem - the battery leakage damaged a tiny piece of PCB track that connected the tiny filament bulb, so the light doesn't work ... but it looks like new and now has a very clear display.
The biggest problem that I had was when the tiny gold spring back panel piezo connector jumped across my kitchen worktop - fortunately it's magnetic but it still took me an hour or so to find it !
I never did get any helpful feedback from Seiko about the precise values of the capacitors but I do strongly believe the fact that localized heating can rejuvenate them and my experience tends to substantiate the fact.
Thanks so much for the response on this. Really appreciate it. I'll try the same process of carefully heating it. Nothing to lose really.
@@craigkellum So much responsibility ... it worked twice for me and I'm pleased to say that both watches have exhibited perfect behaviour except for the light function on the 4000 model. I hope that your experience will prove to be just as good. 😊
Curious ! - How sure are you that the capacitors are 180nF (anyone talked to Seiko ?) ? I've also got a G757, at first I thought that it was a contact problem and mine shows the same symptoms. I've found that if I keep the module warm(ish) by sitting it on my router then it all springs into life. If you measure the voltage on the small test point at the bottom right corner of the black plastic rectangle - with a new battery, it's 1.57v but then jumps to 2.53v when the upconverter's running. So it looks like the switching of the drive transistor(s) might be stalling. Anyway, have you a progress report ? Just another note - you'd think that it'd remember the time setting as the upconverter probably only relates to the LCD ... I've found that mine forgets the time and effectively reboots, so maybe it's to do with whatever drives the whole logic of the watch - and that might be the main 32.678KHz oscillator - maybe it's the culprit if it keeps stopping ? Might put a scope on it if I can find a suitable test point.
...
Diarized account and thoughts :
...
Put a scope on the test point where I measured 2.53v and there's actually a 1.5v 1.024KHz square wave offset by 1v.
...
It does look as though heating the PCB might be a good option - I hadn't appreciated that ceramic capacitors age significantly over the years but heating them above their Curie Point temperature can rejuvenate them (wish that applied to people) - so barium titanate (BaTiO3) has a Curie Point of 120 - 130°C ... so cooking them for a couple of hours should restore their original capacitance (and that's exactly what the ceramic capacitor manufacturer does before their capacitors are sent to distributors); some have said that re-soldering a capacitor has the same effect..
...
Might have a go at this - the LCD will not appreciate being cooked, so only do it to the bare PCB.
...
Tried cooking PCB at 150°C, cleaned and re-assembled - all working again ... stopped after a couple of hours. The 4 capacitors didn't seem to be all the same value, looked more like the first 3 are 100nF and the last being 220nF (the battery capacitor is maybe 100nF) - when tested before cooking they were all significantly lower ... cooking them did result in increased capacitance but still a blank display and time resets; think that it might be the master 32.678KHz oscillator, maybe the crystal ?
...
A few days have now elapsed and some despondency. Dropped a line to Seiko to see if they'd tell me the values of the capacitors and any help about the notoriety of the watch to behave badly after a battery change - yes, this watch has quite a reputation for dying / display instability when the battery is changed. Seiko were friendly and answered by telling me that they can not offer a repair service as they no longer have the parts ... I wrote back to ask if they'd just tell me the component values and to see if there was a circuit diagram.
In the meantime, the now "cooked" PCB had suddenly restarted and seems to be behaving itself. It sat for several days without problem on my router and I then thought let's freeze it, so I put it in my refrigerator for a couple of days and it's still working - time still precise to the second even after surviving near 0°C temperatures ! Later, it became an ornament in my sitting room without problem. So it's still defying logical fault diagnosis - it's looking stable but I bet it fails as soon as I put it back in its case and that back is really hard to get off ! I'm kind of concluding that there is an intrinsic design flaw - maybe it doesn't like a brand new battery or something about its characteristic like its impedance - maybe 1.57 volts is too much juice and it needs to be 1.55 volts ... who knows !?
...
Back in its case and stable ! I've tried leaving it in warm places, tops of radiators and then cold treatment in the 'fridge and its been steady as a rock ... the watch had been very temperamental, as others have experienced, I am tending to think that either the battery has now "settled in", "cooking" the board did it or it was down to some ethereal intervention ! For those who might follow my PCB cooking example, I used a domestic Bosch oven (fan assisted) and a voltmeter with a thermocouple probe, free mounted on the middle shelf, next to the PCB that sat on a piece of paper. The PCB was cooked at 150°C for 2 hours (the Bosch dial setting was surprisingly accurate !).
...
Oh, yes, Seiko Service UK were very friendly but will not or can not provide any design details like the capacitor values - maybe someone else can have a go ?
...
Watch this space ! - It's still going no matter where it is ! I think that I'll now quit while I'm ahead - really inconclusive opinion and diagnosis ... it might've worked if I'd just left it alone with its new battery, maybe cooking did cure it or maybe one of the gold flashed contacts were not good despite having been cleaned and again maybe it's just down to "magic".
...
My watch is now fully working and here it is :
drive.google.com/file/d/16jeu8kO6ckfpDKHmJ3UV2R-MSZBvF5lf/view?usp=sharing
I bought it in Bury, Lancashire in 1981 - it had a flat battery in the shop and they knocked it down to £25.00 on the understanding that the guarantee was void. They wouldn't replace the battery but another jeweller did and it worked for many years. I bought the watch because I liked it and it was sometime later when I became aware of the Roger Moore / James Bond fashion accessory.
...
29 Oct ... watch still working without any problems, even after the "hour change" ... this watch has a "bad reputation" vis-à-vis new battery replacement and I now suspect the stability of the 32KHz oscillator to be the culprit, basically bad design. One thing is clear is that heating it for 2 hours at 150°C has some rationale and that, in my case, it has now continued to function for over 2 weeks. In the absence of design component values, the idea of capacitor replacement is a real gamble and I feel lucky that this wasn't my first practical solution.
🤓👍🦊
01 December ... Watch still working without any problems ... Have you changed the capacitors yet ?
Thank you for sharing your experience