autowinder woes

[Douglas Davies]

I normally work only on my own watches, and I prefer manual winding vintage watches, so I have almost no experience with autowinders. A friend asked me to "just clean up" his Rolex Oyster Perpetual Cal A 296/645 from 1953. After cleaning (and a new mainspring), it runs fairly strongly (amplitude about 285-290 dial up, fully wound), all positions within about a 25 sec per day window. I decided to wear it around a few days before making a final tweak to the regulator. After 24 hours, amplitude was less by maybe 5 degrees, and after 48 hours, it was off by about 15 degrees. Power reserve is 41 hours, so it was obviously getting some winding, but not staying fully wound. After 72 hours, amplitude was 235 or so, and after 96 hours amplitude was around 190. At 110 hours, at 5:40 a.m., it stopped. I gave it a few twists (the case, not the crown) to get it started and took the dog for a 15 minute walk, which caused the watch to run for about 3 hours. When I hold the watch up to my ear and rotate it back and forth, I can hear the rapid clicks of the ratchet on the center wheel (it is unidirectional) and then the slow clicks of the ratchet on the first intermediate wheel. Based on this, and some calculations I did, I think about 2 hours a day of brisk walking would keep the watch wound. Unfortunately, I am mostly at the computer or workbench all day.

Holding the watch vertically with the back off, and the watch fully wound, the weight of the rotor will not move the winding mechanism. If I push the rotor slightly with a piece of pegwood and then let go, it springs back. Of course if the rotor has some kinetic energy when it engages, it might be possible for it to move the arbor. I considered whether the bridal was sticking too much. I lubricated it with 8217, and when manually winding the watch it does not seem to stick. The watch just keeps winding with no noticeable change in the feel. So I am thinking even if the autowinder delivered enough energy to keep the watch going indefinitely, would it be constantly at the low end of the power reserve? That seems less than ideal.

Any thoughts? Maybe this just not the right watch for a couch potato. I have some questions about lubrication of the autowinding mechanism, but this is already too long.

 

[John Runciman]

Out of curiosity is your watch the same one found at the link below?

Then your testing methodology is leading to confusion. It's hard to tell whether the auto winder is working or not because as you have commented you're not active enough. This is a problem with automatic watches where you shop for repair the watch the customer will take it away come back complaining. Then when the customers complaining the watch goes on the timing machine to verify that the amplitude is what it's supposed to be. Then typically it's explained to the customer that they're not active enough.

But that isn't always necessarily correct. So typically for timing specifications Rolex and other companies are concerned about what the watches doing 24 hours after it's been fully wound up. So in other words if it meets the specifications at 24 hours. So somehow we need to separate whether you're having a inactivity and not actually getting the watch wound up or whether you're actually having a running issue.

They what might be the easiest to do is manually wind the watch up until you feel it's fully wound up. It be nice if you could see the ratchet wheel in other words how many turns it's turned. Because a lot of times people manually winding the watch think that 10 turns the crown is enough To wind the watch up and it's not. Maybe try about 40 or 50 turned to the crown you should be wound up then. Set the watch some place and let it run and see if it runs for 41 hours. Then I'm not sure if there is any kind of test to tell if the automatic is actually working because as you pointed out you're not active enough. Which is a problem with automatic watches.


bidfun-db Archive: Watch Movements: Rolex 645

 

[Douglas Davies]

Thanks for the reply. I am not sure I understand all of your comments. I have tested the watch essentially as a manual winder. I did 10 twenty-four hour tests, starting each one with a full wind. Tested in 5 positions, similar to COSC test. Results: average daily rate gain 3.35 sec, mean variation between two tests in same position, 1.1 sec, highest variation in one position 3 sec. Difference between horizontal and vertical positions (as defined by COSC) 6.25 sec. Highest variation from mean rate was -15 sec for the first crown up test. All other positions within 10 sec. Other than the slow crown up rate, the only other problem was that the dial up/dial down difference was about 7 seconds, which seems high. I am guessing this watch spent many years dial up on the bedside table and the upper pivot is a little bit flattened. So I think the watch is fundamentally sound for a watch of its age and the problem is with keeping it wound.

After some reading up, I am finding that it is not that uncommon for people to need to wind their automatic every few days, either to keep it running or to keep it from settling into the lower range of the power reserve and losing accuracy.

I am wondering if the issue is just lack of activity or if I have caused the autowinder to lose efficiency by improper lubrication. I used Moebius 9101 on the axle bearing (plain metal bearing), the lighter Moebius oil that I use on the train wheels for the intermediate wheel jewels. Maybe the 9101 is too heavy and is slowing down the oscillating weight?

The autowinder (and the ratchet wheel) are connected to the barrel arbor by a spring clutch with two leaves pointing up (to engage the autowinder) and two down (to engage the ratchet wheel). I just put a few drops of 9101 on each leaf of the spring before assembling. Possibly not enough, since it gets spread over the teeth on the respective wheels (which are shaped like the teeth on a clutch wheel).

 

[Douglas Davies]

I put a little more lube on the gear pictured, as well as the corresponding teeth on the ratchet wheel, and then tried to be a bit more active. Watch seems like it will keep running indefinitely with those modest changes. Still doesn't fully wind. Afternoon checks on the amplitude over a period of 8 days ranged from 255 to 290. When I finally stopped wearing it, it ran for 28 hours, versus advertised power reserve of 42. Not going to mess with it anymore, since it is running well. Stayed within + or - 6 seconds over the 8 day period with maximum change on any day 3 sec. Lost a bit at first (after full winding) and gained slightly after that.

 

[roughbarked]

This is quite an old auto winding movement. How many times has it been serviced in it's life? Have you tested the transmission wheels for wear? Checked the barrel bridge for wear? Simply putting a new mainspring into an old watch and expecting that to fix it is probably the wrong way to go about it. Did you properly lubricate the slipper on the mainspring with the recommended lubricant?

 

[Douglas Davies]

Yes, used 8217 on the barrel. Probably could have used a new barrel, but they are not available except used ones on eBay that won't necessarily be any better than what I have.

 

[roughbarked]

True that other barrels could well be worse.