Crystal Replacement Woes

[abem]

Greetings,

I've been trying to repair a ladies oval Pulsar for my mom and have been having difficulties. The crystal would work for a few hours or days and then stop. So, today I tried replacing the crystal. That seemed to go fine, but now I find that the third digit is weaker than the others (it wasn't before). Any ideas?

Enlarge (1800x1200): http://img190.imageshack.us/img190/1125 ... 0x1200.jpg


There seems to be some sort of connection between the third digit and the leads to the crystal. If I ground the leftmost lead while the time is running, then only the third digit of the display glows dimly. I thought it could be a problem with a weak solder joint, but I don't see anything wrong. I also thought it might be a weak connection between the board and the display but applying a bit pressure to the display results in no change. I'm just about out of ideas.

Any clues or suggestions would be most appreciated. This seems more complicated than I had hoped!

-abe.

[retroleds]

Abe:
Part of the problem is probably heat the module was subjected to - you have to put a clip(heat sink) on the lead coming out of the module to protect it from heat. And(not trying to be real critical here) but that is a whopping amount of solder, for the job at hand. And the crystal itself needs to have a heat sink on it's lead to protect it - it has a plastic plug in the bottom of that little metal can, and inside it has a grain of a lab grown crystal that is small than a grain of salt. The heat kills.

I wrote this article for a magazine project for a magazine project I ran out of time to continue, it is a good starting point.
http://www.thedigitalwatch.com/Archives/QuickTip_002.htm

[abem]

Ed,

Those tips all sounds like really good advice. I'm not sure what's going on in this case.

The problem with the dim third digit actually appeared as soon as I clipped the old crystal away and before I soldered the new one in place, which casts some doubt upon the heat theory.

Before I soldered the new crystal in place, I placed the leads to the crystal in contact with the module and was able to test out the module with the crystal being held in place through the elasticity of the leads. Even then it exhibited the problem with the dim third digit.

I was hoping that once I soldered the crystal into place, the connection would improve and the problem might go away, but it remained exactly the same. I think I may have either bent the module leads or subjected them to some strain when clipping the old crystal away. I wouldn't expect that to cause this problem, but apparently I've done something to the module to cause the problem.

Fussy little beasts, aren't they? Thus far this weekend, I'm batting 0 for 2 in terms of watch repair stats.

-abe.

[retroleds]

They are fussy little beasts. If you get curious, remove the circuit board, flip over, and pop off the brown-black cover on the back. Holy Toledo! , a lot of fine little wire bond connections. Generally on any electronic watch it is the better idea(may not seem courageous but what the heck) and leave as much of original connection alone, and avoid soldering to a circuit board whenever possible. On those P4 style modules I personally have stopped trying to remove the solder from where the crystal is soldered in. I just snip out the crystal, bend my new one's leads so they can lay across those "figure 8" shaped ends(which you cut off) and lightly and quickly solder to those. Leaves me room for heatsink clips all around, avoid stress the board with additional heat.

Powering up ANY electrical circuit without all components in place is a bad idea. You wouldn't run your car without the fan(serpentine) belt in place, would you? Circuits need components to both control the current and sometimes to throttle it.

I think this statement from my personal website sums up the whole game reasonably,"buying a "mystery piece" is certainly not the worst thing you could be doing with your time or money if you can afford it. If your goal is to become a repairperson the broken ones will be good practice - and you (like me) will certainly break a few along the road of knowledge."

[abem]

Ed,

I may just do that since I'm about ready to give up on this module. Frustrating. I can only imagine how frustrating it must have been for the original developers.

I just replaced the quartz crystal yesterday and left it running overnight. The next morning I found that once again it was back to lighting up a single digit at a time! Now it's flaky AND missing a digit - negative progress. It will light up all of the digits for a few moments at a time, but then eventually (within minutes or hours) it always goes back to single digit mode. I used to be under the impression that this behavior was due to the module not multiplexing the display and that this was a sure sign of the crystal going bad. Is that not the case? Are there other types of faults that exhibit the same behavior? It seemed to me that during testing, the display gets a tiny bit warm - is this normal?

It seems like there are two possibilities:
1) The module is somehow nuking its crystal (overheating maybe?)
2) This isn't a problem with the crystal at all

What's even worse is that I think I may have developed some sort of generalized powerful anti-LED field recently. In the past two days, two other LEDs have stopped working - a Hamilton QED with a pristine module and a P4 (a sizeable chunk of my modest little collection). Am I jinxed?

Both of these exhibit a similar problem with just a single digit lighting up. I was hopeful that I could learn how to replace crystals and then would be able to repair these sorts of problems, but now I don't know. Maybe I should have stuck to software.

-abe.