12 Jan 2006, 09:41
I've only ever had two 3502s, so I can't say I've had a lot of experience with them, but in general, they seem quite reliable and well-made; after all, they were the product of over four years of R&D by Time Computer [though I'm not sure I agree with Mike's acessment of them as "bullet-proof"; still, overall, it's a good design, even if I don't particularly care for the bubble-lensed digits].
Anyway, I've never seen a 3502 behave the way you're describing, but I have seen it in a DateII. I prefer to keep all my watches working, and I periodically go through them and check for function [checked once a month for functioning and otherwise left alone, a set of batteries will last nearly two years]. This one was kept in a storage box with nine other watches, and while I was looking at one of the others, the DateII suddenly flashed me the time. I wasn't even positive what had happened, but I set it aside and kept a closer eye on it, and sure enough, about ten minutes later, it happened again. I tried the usual remedies [cleaning contacts, new batteries], but nothing helped. Eventually, I realized I had purchased this watch the previous April, and the blinking happened in January [couldn't have been happening long when I noticed it, or the batteries would have been gone, but the display was still bright, so I was lucky enough to have caught it almost immediately after it started]. Nov., Dec., and Jan. are the rainiest, and thus the most humid, months in San Diego. Here is the treatment I tried:
Removed module and placed complete watch [with case back off], case back, and o-ring seal in toaster oven at 250 degrees F. [about 120 degrees C.] for thirty minutes. It's low enough to not damage epoxy, solder, or rubber, but high enough to drive off absolutely every molecule of H2O, as it is well above the boiling point of water. Allow to cool to 130F. and re-assemble [new batteries, of course, and 130F. will not harm the module; the electronics are subjected to much higher temperatures during the manufacturing process]. You'll know the case temperature is right when you can just barely touch it without it seeming uncomfortably warm [a surface temperature of 140F. renders an object "too hot to handle"].
Result; it is now two years and counting, and the problem has not recurred. I would still be guessing to tell which component on the board is the one that is humidity-sensitive, but I would have to regard this as an abberation, rather than a design flaw, considering how infrequently this syndrome presents itself. Part of the problem is deteriorating case seals; after thirty years, they are either turning as brittle as glass, or into a tarry goo. Either way, the watertight integrity of the seal is compromised, and they are no longer functioning as effective moisture barriers, a situation that can also promote or accelerate corrosion, a far more serious condition.
If you're inclined, give the "heat soak" a try, and let us know if it cures the problem.