Early Quartz Testing

[Garfield]

I have a mid 80's Seiko 7A38A movement currently in pieces waiting on spare parts from Japan (was dropped from height - a pinion is broken off a rotor, crystal shattered, pusher damaged and there is also corrosion damage). I have sourced all the parts so they are not the problem.

Reading Seiko's data / manual they say to check accuracy using the '10 second gate' presumably on a Seiko tester, there is an adjuster present. I don't have a Seiko tester but I do have a very high spec Tektronix oscilloscope. At some point I am going to need to set the time / rate on this watch so I need to know what the '10 second gate' is / was / does so that I can replicate - would this be some sort of pulse count through an induction loop with the 10 second gate providing the scale ?.

Or is the Seiko tool going to be the only option - these are harder to find than spares for the 80's watches.

 

[Chris Radek]

The "gate" just means to count the rate over that time period. A longer gate gives you a better reading. Many quartz testers will reliably show the rate to 0.01 seconds/day (once calibrated) because they have crystal ovens etc. I would be surprised if you could calibrate to even 1 sec/day with a scope.

If you don't have a special machine, no problem, just let it run 24 hours and note the offset. If you can't tell for sure if it's fast or slow after a day, it's fine!

 

[Garfield]

I shall do that, may need to do more than 24 hours though there are only 6 poositions for the adjuster, each position is plus or minus .26 sec per day.

The trimmer presumably acts on the divider that generates the 1 sec pulses - they do say this shouldn't be done under a light but it is pretty hard not to since the case has to be open to adjust it - your periodic method will work for that too.

 

[John Runciman]

Unless you really obsessed with super accuracy it basically just doesn't matter. As because even a quartz watch when it's in a accurate is so much more accurate than most things we have. For instance if you look at all timing machine like witschi the use it have a setting four seconds per month.

The trimmer

If you read the wording of the Seiko tech sheet it has a switch it doesn't have a trimmer. It's probably using some form of logic regulation and that's why it's specifying a much longer gate times and usual.

 

[Garfield]

As I mentioned only 6 positions - all the 7A38 I've seen they're in the middle, the one I have no different, ignore the condition of the plates - there is quite a bit of water damage in the battery area, I am not sure of the circuits integrity but only when I get it back together will I know. I'm waiting on some parts from Japan.

 

[praezis]

Hello,
according above data table, adjusting is useful only if the error is more than 0.13 s/d.
So you will have to wait for 1 week at least: if your watch deviates more than 1 sec then, you may adjust (1s/7 days = 0.14 s/d).

Frank