Stellaris Transistorized watch troubleshooting and repair.

[LittleWatchShop 141]

I posted about trying to figure out how to get the movement out of the case.  I will share that information here as well.

It is a front-loader.  Remove the crystal and push the lever that is visible through the edge of the dial and the stem comes out.  Then the movement can be removed.

After removing the movement, I connected 1.5 volts at the battery terminals--nothing.

It turns out that this is an unbranded Seiko 3302 movement and I found the manual for it.

 

 

[LittleWatchShop 141]

The first thing I did was measure the coil.  It was fine.  What about the circuit.  I found the circuit in the manual--here it is. I show the schematic and the physical layout.

[LittleWatchShop 141]

Next, I checked the components using a simple two-terminal curve tracer.  All components are working fine with my limited test.

Next, I wanted to check the contacts.  Frankly, it appeared to me that the balance was installed incorrectly and the pallet fork was not interfacing properly with the roller jewel.  I removed it.  Ugh...tiny wires (switch) actually cross above the roller disk.  I am thinking, how do I reinstall this so that the roller disk slips back under the switch wire.  I'm a cowboy...get on with it.  After removing the balance, I saw nothing unusual, so did the reverse operation and managed to navigate under the wires and place the pivots without any torquing of the hairspring.  All back together, but still did not work.

[LittleWatchShop 141]

At this point, I walked away for an hour or so.

I came back with the theory that there was a break in the circuit somewhere and my speculation was that it was along the contact point spring and associated interconnections.

So, here is where I focused.

I did not take a picture of the insulating washer on the bottom.  Interesting feature.  Insulation on top keeps the screw out of the electrical path.  The copper ring underneath connects point A to point B.

I removed it.  It appeared to have some slight discoloration (oxidation on the PCB trace).  I just reinstalled and cranked it down nice and tight (not over tight!).

Then I connected a battery and VOILA...the balance wheel takes off.

In retrospect, I think a different order of tests were in order.

1) testing the coil first was right on.

2) running a curve tracer on the components should not have been the second step...too involved...required building a testing platform...yadayada

3) checking continuity along all paths should have been #2

[LittleWatchShop 141]

OK, I am gonna call this one "done" for now at least.  I cleaned up the case, polished the crystal, installed a battery.

It ran all night and kept perfect (to my resolution.  This watch does not command any value on Ebay, but it is a very nice movement and kinda pretty.  I set it to the right day, but did not want to have to go around to the correct date.

[JohnD 344]

Thanks a lot for that, I have a couple of non-working Seiko 3303's that I can now look at in a new light.

[LittleWatchShop 141]

This watch continues to run fine in a static position--have not found a band for it yet, so no dynamic tests.

I have another identical Stellaris.  I have opened it up and found the same problem with the connector ring.  Fixed that.  However, there is more to this puzzle.  The contact spring was not engaging.  It is adjustable and I did that.  Still does not run.  I know that the electronics are working to some degree because of the rest position of the balance wheel when power is applied (different than when power is disengaged).

I have the technical manual for this movement.  I am liking this movement less and less.  One of the procedures is to clean certain elements using freon.  Really?  Who does that.

I will follow up as I make progress.

[PastorChris 63]

What a fantastic movement! A really interesting blend of the mechanical and electronic. It might be completely common but I especially like the way it is not all covered up in plastic. Relly interesting walk-through as well. Thanks for an interesting thread.

[LittleWatchShop 141]

This thing was running a little fast, so I had to take it apart to adjust.  Very poor design.  You have to fully remove the movement out of the front in order to adjust.

I discovered that the "lever" stem release is also a really poor design.  Once during the process, the lever pushed into the interior of the watch which lifted one of the lever setting levers up and as a result the stem would not engage after installation.  I had to take it out, remove the dial, and trouble to see what was happening.  All I had to do was to pull on the stem-release lever and all was well.  I could have done this without removing the dial had I understood the mechanics.

I am posting a picture of the stem-release lever that I am talking about...for future adventurers.

My Vibragraph just could not pick this one up (maybe need to do some refurbish) so I resorted to old timey techniques.  I did this outside of the case for obvious reasons...with small alligator clips attached to AA battery.  What a pain.

Anyway, it is back in the case with less than a minute a day error.  Good enough for me.  The balance us ultra-uber-super sensitive.  Ugh.

 

[LittleWatchShop 141]

I got this timed to +20 sec/day.  I am gonna stop there!!!!!

Don't hit me.

[GregC 14]

After reading this I think a quick test would be a quick short with a piece of wire between GND and resistor end connected to the switch. This would bypass the switch and energize the circuit. 

[LittleWatchShop 141]

11 minutes ago, GregC said:

After reading this I think a quick test would be a quick short with a piece of wire between GND and resistor end connected to the switch. This would bypass the switch and energize the circuit. 

It would energize the circuit if you choose the pad right at the resistor.  Then what? 

 

[GregC 14]

Sorry, just thinking what could be the first step in the troubleshooting process assuming you didnt already know what the problem was. If the circuit energizes, something should move then the problem is in that part of the circuit. Remove battery and check continuity from gnd to switch. Nothing happens, check coil.

[LittleWatchShop 141]

1 hour ago, GregC said:

Sorry, just thinking what could be the first step in the troubleshooting process assuming you didnt already know what the problem was. If the circuit energizes, something should move then the problem is in that part of the circuit. Remove battery and check continuity from gnd to switch. Nothing happens, check coil.

It has been a few weeks, so the order of steps is cloudy.  I tested the transistor and diode with a curve tracer.  Verified the resistor with a DVM.  Activating the switch by rotating the balance seemed to have no effect. As I recall thinking that the switch never actually passes current.  I don't think I measured the resistance across the copper contact ring.  I just unscrewed it and noticed some oxidation and cleaned it.  I should have measured the resistance in that path.  Maybe I did...dunno.

[GregC 14]

38 minutes ago, LittleWatchShop said:

It has been a few weeks, so the order of steps is cloudy.  I tested the transistor and diode with a curve tracer.  Verified the resistor with a DVM.  Activating the switch by rotating the balance seemed to have no effect. As I recall thinking that the switch never actually passes current.  I don't think I measured the resistance across the copper contact ring.  I just unscrewed it and noticed some oxidation and cleaned it.  I should have measured the resistance in that path.  Maybe I did...dunno.

As the screw had corrosion it broke the circuit so rotating the balance would do nothing. I'm thinking the balance is there to switch the circuit on and off. 

[LittleWatchShop 141]

8 hours ago, GregC said:

I'm thinking the balance is there to switch the circuit on and off. 

Yes, you are thinking correctly!

[LittleWatchShop 141]

I bought another one on Ebay for 10 bucks.  Will see how it goes!