Quartz Crystal replacement question.

[Sully008]

Hello all,

Can anybody tell me if this sounds right for a QC used in the MkIII module?

Freq. 32.768KHz
Load Cap. 12.5
Oper. temp.: -10ºC to +60ºC
Freq. Tolerance: +/-20ppm

This would be the "tuning fork" type QC as listed on pg.889 @ mouser.com. I'm going to try and get one from a local parts supplier, otherwise I'll have to buy online.

TIA,
Mike

[clockace]

hi mike, the frequency is right but i don't know about the rest. i just put one in a pc i am working with that looks like the round cylinder used in the MKIII/IV's. it looks tiny compared to the originals, but it lit up. i may have other problems with this one tho', since the date came up 16:03. i thought it may be the different qc, so i pulled the cylinder and put in the one that looks like the original. it also lites up all functions but the date is still 16:03. i can change the date but the month won't change. i also have 2xvarta40's in it because i ran out of the 80's. i don't think they have enough poop in them because they drain while i am trying to set it. the extra 80mah obviously is necessary. i will get some more 80's, but i think the date thingy is something else beyond me. counting this one i have gotten 4 out of 6 none runners going. . did you get to zantech's site? they don't have their whole catalog online. how much is mouser wanting for the qc's? i am done for the night. peter

[charger105]

G'day Peter.
Quite often when you power up digital watches, crazy values are present in the registers. All you have to do is start setting them, and it should kick back into the correct running sequence once you reach the end of the register.

So, if this is a 6 reed switch MkI, try setting the date to 29, and the time to 23:59, and the month should kick over to 17 in a minute. Keep doing this, and I reckon that it'll eventually go from 19 back to 01. It should then cycle correctly from 01-12 as expected.

Good luck.
Andrew.

[rewolf]

Hello all,

Can anybody tell me if this sounds right for a QC used in the MkIII module?

Freq. 32.768KHz
Load Cap. 12.5
Oper. temp.: -10ºC to +60ºC
Freq. Tolerance: +/-20ppm

This would be the "tuning fork" type QC as listed on pg.889 @ mouser.com. I'm going to try and get one from a local parts supplier, otherwise I'll have to buy online.

TIA,
Mike

That one should work. Practically all 32kHz watch QCs have the same technical specs, they're pretty interchangeable, so focus on the mechanical properties (size). Worst case it runs some 10ppm off-frequency if the QC's and the circuit's load capacitance do not match (standard value is 12.5pF).
Low frequency crystals (below 100kHz) are always tuning-fork type (at least I have never seen any other).

[Sully008]

Hi all,

I love this site! So many knowledgeable people. There are some forums I belong to that have some pretty useless people. But you guys rock!

Peter, I went to zantech's site, like you said, not everything's on there. Mouser has QCs ranging from $1-up. But, if I can get it locally, it'll save me huge on shipping/taxes. 4/6 Synchronars is impressive! I've got a couple of modules here to test still, but I think I may have fried one or two messing around.

rewolf: Thanks for the confirmation on the specs. I'm off to the store this morning.

[clockace]

Posted: Fri Oct 24, 2008 11:00 pm Post subject: : Quartz Crystal replacement question.

--------------------------------------------------------------------------------

G'day Peter.
Quite often when you power up digital watches, crazy values are present in the registers. All you have to do is start setting them, and it should kick back into the correct running sequence once you reach the end of the register.
andrew down under, awesome suggestion. i tried it and it went from 16:30 to 5:01. from there i think i can handle it. thank you again. i think you are the poster who suggest replacing the sc and qc on my dot matrix and that worked 100% as well. this site is alright. now if i just knew what i was doing............... regards, peter

[Sully008]

Well, I tried replacing the QC tonight and no difference. I still get only ~.23v @ the battery pads. The only thing I haven't changed are the resistors attached to the QC. I guess I'll try that next.

[clockace]

mike, that's a real bummer. it seems like you are going backward. i looked at some of your first posts and on one of them you said you got the digits to lite up. maybe you should go back to that spot and try to figure out what you did from that point that made it worse. the mkIII/IV must be significantly different than the mkI/II's i have been working on. of course if i didn't get any sign of life out of them either i don't have the foggiest idea where to go from there. i think all of them that i have gotten going were only minor problems with things i could deal with by replacing a few parts. i know which end of the plug to put in the wall, but don't know how to turn on the lamp when it comes to electronic circuits. i hope some of the other geniuses on this forum can steer you in the right direction. i am going to try opening my sunwatch sometime soon. i hope i have better success than the mkIV i tried to open last march. that one will never work again. i sent it to howard for repair and he laughed and said it would be better to build another one, which he is doing for me. peter

[charger105]

Hi Mike.
Working on MkIIIs and MkIVs is quite tricky. It is very easy to destroy a working module. The IC's wire bonds are exposed, as are the wire bonds on the display. The circuit diagrams you have been looking at on the other site are for MkIs and MkIIs, and are generally not applicable in your situation. The solar cell connection points are the same however, so no problem there. Make sure you haven't shorted any tracks or pads where you've attached the solar sell.

So, if you've touched the IC by accident, then the module is likely to be damaged, and cannot be repaired (sometimes the ICs are encapsulated, but I suspect yours is not).

If you've touched/bumped the display, then the diplay may be damaged and if so, can't be repaired either. Sometimes the display on these models is encapsulated, in which case it is more robust. If it is only the display that is damaged, then a new display can be transplanted, although this can be tricky for the un-initiated.

I suggest you do replace the resistors (1.6M and 22M), since this is all that is left. After this, if it seems to be working, with segments or digits out, then you might want to consider changing the display........of course these symptoms could point to a defective IC as well.

Good luck.
Rgds,
Andrew.

[Sully008]

Hey guys,

Thanks for the replies. I'm not sure what happened with this module. I the board and display is still pretty much enclosed in the potting gel. When I first attached the 3v source to the two wires, all the digits lit up. I even grabbed the magnet from my Pulsar and activated the reed switches. At that point I was a happy camper and put it away until I took it out again and tried to power it up. That's when I only got one digit. I can't explain it. I didn't poke or prod anything else, and I didn't remove the gel from the board. The chip is still covered.

I'll see what I can do about changing the resistors. I've got a couple of boards here with them still attached. I haven't given up hope yet!

[rewolf]

If only 1 digit lights up it's usually the QC. You replaced it.
A high-ohm resistor in parallel with the QC (Usually 10 or 15M, so I quite probably in your case it's the 22M) helps with oscillator startup. You can decrease it to at least 4.7M - maybe this helps. I "healed" a few (non-Synchronar) quartz oscillators that way.
Solder flux on the QC pins can also hinder oscillation - you could try cleaning them with flux removal (denatured alcohol or isopropyl alcohol will also do).

[Sully008]

If only 1 digit lights up it's usually the QC. You replaced it.
A high-ohm resistor in parallel with the QC (Usually 10 or 15M, so I quite probably in your case it's the 22M) helps with oscillator startup. You can decrease it to at least 4.7M - maybe this helps. I "healed" a few (non-Synchronar) quartz oscillators that way.
Solder flux on the QC pins can also hinder oscillation - you could try cleaning them with flux removal (denatured alcohol or isopropyl alcohol will also do).

Bah. I tried replacing the resistors and no go. I measured the ones I took off the board and they measured ~20M and 4M. The QC is not polarity dependent, is it? It doesn't matter which way it goes on? Thanks for all the suggestions, much appreciated!

[rewolf]

I measured the ones I took off the board and they measured ~20M and 4M. The QC is not polarity dependent, is it? It doesn't matter which way it goes on?

No, QCs don't have a polarity, you can't do wrong as long as they have 2 only connections.
Have you tried lower values for the ~20M resistor, like 10M or 4M7?
(I admit I haven't read all the previous postings, maybe there are hints that point to somewhere else. Besides that, I've never ever touched a Synchronar, only seen a circuit diagram of the MK1, IIRC).

[Snetzka]

One thing about being in a totally different time zone is that i miss out on all the question / answers..

Rewolf is right, Crystals (Xtal's or QC's) with two leads are not polarity conscious, and even the three legged Varity where the 'extra' lead is just a ground connection.

Your best bet is to try lowering the bias resistor which might 'kick' the oscillator in to life. If this fails it looks like the oscillator circuit is at fault and your only remedy would be to 'force feed' it with a external oscillator. (Which takes up more space, and worse still extra power consumption.) or use it for spares. Unfortunately simple oscillator circuits (as I assume) used in this instance can be prone to failure if shorted out for any period of time.