Bulova Accutrone 214 1.55V silver oxide battery fix - diode part search

[e-pirate]

Hi folks!
After Dave's video about Bulova Accutrone 214 Spaceview I thought: "well... why not?" and after a month of search and waiting I bought 1967 model from eBay. The watch is in nearly perfect cosmetic condition. They were stated as "serviced" and they seem so. When left on the bench they run dead spot on, not a single second off by 24 hours. But when on my wrist... well... 25 second off forward each hour.
As mentioned before, this watch is designed for 1.35V mercury battery and do not feel good with modern 390/ 394/936 1.55V silver oxide. Some of them do, some of them don't. Mine seem to be "over-active" and run too fast when on the wrist.
There are two common solutions: "rephase" which is more "watch repairman" thing and adding a diode in series with the oscillator to lower the voltage by .2V. I'm definitely not going to get into "rephasing" story, but the diode seem to be a good solution that I can do myself. Moreover, there is an commercial semi-official solution called Accucell-1 which is not more then a 395 cell with a diode:

Sure, being an electronic hobbyist, I'm far away from the idea of buying dirt chip diode with same value cell for 15 buck shipped a piece each 7-9 month! The only problem, that I was not able to figure out the diode part number. Not by searching, neither by figuring out what this particular "SW" mean. I believe this is some kind of Schottky diode with 0.2V drop on 10 uA load.
First of all, I'm interested in figuring out the part number of this particular SMD diode used in "official" proved solution. If not.. any suggestions are welcomed.

[Chalcogenide]

I would try with a signal schottky like a 1N5711. The datasheet shows roughly 250mV drop on 10uA load at ambient temperature. However, depending on the peak current, a slightly beefier diode might be needed (like a BAR42).
Definitey no need for an ultra-low drop schottky since the current we are talking about is so low. The PMEG2005AEL barely drops a few millivolts at 10uA of If.

[e-pirate]

I can stay with any package I will be able to solder by hand. But I don't thinks "the lower is better" rule does apply my case. If the drop is to low for that amount of current that turning fork coils consume - there will be no effect or it will be intermittent. The turning fork is running at 360 Hz meaning it is 360 teeth a second to be pass by the jewels. This is 31M a day! That why I will prefer to use a proven solution (i.e. the diode on the picture) - lots of possible things to screw up here.

[salbayeng]

This is usually a pretty good website for SMD codes:
http://www.s-manuals.com/smd/sw
(but I couldn't find it there).

The part looks similar to the very common schottky diode BAT54ws (this has the code "S0")
Get the data sheet for that, it will give you a starting point.

There are several hundred candidates for your application here:
http://au.mouser.com/Semiconductors/Discrete-Semiconductors/Diodes-Rectifiers/Schottky-Diodes-Rectifiers/_/N-ax1mjZscv7?Rl=ax1mjZerbyZ1z0z5meZ1yztawsSGT

Your package looks like SOD-523 (about a 0603) or a SOD-323 (bit smaller than 0805)

[e-pirate]

This not the first time I see BAT54 here, will take a look at.

BTW, to understand the sizes, here is the movement pic (aprox. 30 mm in diameter):


And here is the schematic:


P.S. Oh, the case code could be "MS" as well as "SW" 

[Gyro]

I hang my head in shame at the mere thought of such sacrilege, but there looks to be a inviting space in the component cavity in that photo to accept a diode between the -ve case connection and the emitter of the transistor.

...Slinks away before being lynched 

[e-pirate]

I hang my head in shame at the mere thought of such sacrilege, but there looks to be a inviting space in the component cavity in that photo to accept a diode between the -ve case connection and the emitter of the transistor.

Yap... this is exactly what I'm thinking about! There is a free standing pin I can use to look it visually attractive. The only trouble is to choose that correct diode.
An I found some interesting article from guys fighting same problem: http://rick_oleson.tripod.com/batt_adapt_us.pdf.
The BAT83 or BAS70 seem to be the best candidates.

[Gyro]

An I found some interesting article from guys fighting same problem

Wow, that looks about as comprehensive investigation and writeup as anyone coupld hope to find!

[e-pirate]

Wow, that looks about as comprehensive investigation and writeup as anyone coupld hope to find!

I found it accidentally while digging through hundreds of forum pages.. good luck, actually. But non of the mentioned diodes have "SW" or "MS" code, unfortunately. So, the Accucell-1 uses some other diode. Which could be potentially better because this battery replacement is especially designed for Accutrons. I can purchase one and salvage the diode from it, but I still fill bad about paying over 15 buck for a 1 dollar cell and 10 cent part and wait about 3-4 week.

[Seekonk]

It would be a science project, but I wonder if adding a B-E resistor would prevent the overdrive. 

[Gyro]

I found it accidentally while digging through hundreds of forum pages.. good luck, actually. But non of the mentioned diodes have "SW" or "MS" code, unfortunately. So, the Accucell-1 uses some other diode. Which could be potentially better because this battery replacement is especially designed for Accutrons. I can purchase one and salvage the diode from it, but I still fill bad about paying over 15 buck for a 1 dollar cell and 10 cent part and wait about 3-4 week.

The list of diodes and their test characteristicson page 4 looks very comprehensive. I'm sure there must be as much information as you need there, even if the package code doesn't match - there are parts with both WS and SW markings but none of them is a sensible match.

[e-pirate]

It would be a science project, but I wonder if adding a B-E resistor would prevent the overdrive.

Well, I would prefer not to experiment on this. Basically cause pocking around in a precise mechanism with a hot soldering iron isn't the best idea. Besides, this watch is half a century old. Literally. As less operations I can make without getting movement from the case - is better. I'm not a mechanical type guy. I feel more comfortable with a soldering iron and my scopes and multimeters then with a watch tweezers. I'm afraid of damaging something there. To be honest. If a diode trick will work, I will attempt to solder a small one in the parts compartment at the top corner between emitter and the case, if not... well, I'm lucky to have an responsible seller that offered me a full refund. Destroying this masterpeace is the worst scenario possible.

[ian.rees]

Apologies for the slightly OT post, but I've been working with Rob Berkavicius (the guy responsible for the schematic and diagram above) and would highly recommend his Accutron service.  I wouldn't recommend the diode mod to the watch itself - it's really not required to get the watch working properly, and these things are quite fragile.  Your watch is probably "due" for a service anyways, so might as well combine that with an adjustment that'll have it working fine with new batteries.

We've been working on re-winding the coils for Accutrons, and are pretty close to starting shipping.  They are amazingly engineered devices - lots of subtleties, even in the way the wire is wound in the coils.  -Ian-

[e-pirate]

Apologies for the slightly OT post, but I've been working with Rob Berkavicius (the guy responsible for the schematic and diagram above) and would highly recommend his Accutron service.  I wouldn't recommend the diode mod to the watch itself - it's really not required to get the watch working properly, and these things are quite fragile.  Your watch is probably "due" for a service anyways, so might as well combine that with an adjustment that'll have it working fine with new batteries.

We've been working on re-winding the coils for Accutrons, and are pretty close to starting shipping.  They are amazingly engineered devices - lots of subtleties, even in the way the wire is wound in the coils.  -Ian-

There are two opposite opinions here. Some people suggest re-phasing, while others suggest a diode. Even Rob suggests Accucell-1 on his page here http://members.iinet.net.au/~fotoplot/accbat.htm. Mine seem to be just over-active because they keep perfect time when left unattended. I'm able to get BAS70 here in Moscow in local (well.. if "local" is a correct word for the city 40+ km in diameter) store in SOT-23 case. This will at least let me know if this is en electronics or mechanics problem. I will simply strap the diode between negative output of the battery and the case - exactly the way it made in Accucell-1 and see what happen next.

[hr8799]

It has been awhile since anyone posted to this thread, but I was wondering if you made any progress tracking down the accucell diode part? 

I am also curious if you were successful in making your own modified battery for 1.33v.  I'm in a similar boat with a couple of accutrons that do not work well with standard 1.55v batteries, even after servicing and rephasing.  What was your final solution?   I am tempted to follow something along the lines of the what is described in the PDF you posted above.  Anyway, i'm very curious to read what your final solution was! 

[Cyberdragon]

I am also curious if you were successful in making your own modified battery for 1.33v.

You can buy 1.35V zinc batteries from Weincell. Why people tear up antique watches to use the cheapest batteries is beyond me.

[hr8799]

Thank you for the suggestion.

There is no need to worry about the watch, it is safe with me!  My interest in this thread was motivated not by modifying a watch, but rather a battery.  The PDF linked above describes one example, and the Accucell represents another such modification. In both cases, the intent is to leave the watch in original condition.  I treat all antique/vintage timepieces with great care and respect, and I hope that others will as well.

Regarding your suggestion, I have read several posts (on watch-specific forums) where zinc batteries have been tried and/or suggested.   There is no need to repeat that discussion here, but I encourage you to check it out if you are interested.  What I took away from those discussions was that zinc batteries presented some risk to a watch (potential for leakage) and that was enough to scare me off.  Hearing aide batteries are also a bit short-lived. 

Anyway, it is fun to experiment, think of alternatives, and try to learn something along the way.   

p.s. I realize now I should have just sent a PM to the person who started this thread.