problem--pulsar datecommand

[clockace]

last week my datecommand started flashing the date continuously. tried new cells, no joy. tried stronger magnet on reeds. no joy. tried heating, cooling. no joy. yesterday i gave it a bath. white vinegar, water, alcohol, blow dry and warm up. joy!! has been performing perfectly for about 30 hrs now. any ideas as to what the problem could have been?? peter

[Old Tom]

Surface conduction by leaked battery electrolyte- These old LED watch circuits are voltage switched and it really needs very, very little leakage current through the paint and accumulated gunge of 30+ years to make them misbehave (think Ohm's law and a clean circuit resistance of perhaps 200 MegOhms compared to mucky corcuit where it may drop to 200 KOhms). I must admit I always wash a circuit board first before I reach for the soldering iron- many quartz oscillator "faults" are simply the surface conduction ruinning the bias in the oscillator circuit and stopping the oscillator dead. Most people cure this by changing the crystal in which case the flux in the solder cleans off the contaminants and a miraculous recovery is made- I have tested many, many replaced crystals and practically all were good.

Vinegar is for fish & chips! Dilute formic acid has far fewer foody contaminants.

[richard_uk]

Could it still be contaminants if they work when they're warmed up then fail when they cool down or is that definitely a faulty oscillator?

Edit - I'm talking LED/LCD watches in general here not specifically Pulsar.

[clockace]

thx old tom. i kind of suspected that to be the problem, that's why i washed it. it is still working this am here in az, now almost 48 hrs so i guess i got the problem fixed. not familiar with formic acid but clean white vinegar has pretty much worked for me. interesting comment about washing the board before soldering all over the place. makes good sense. thx again. peter