32xx movement problem poll and data thread

[saxo3]

Originally Posted by dannyp

Yes, I must have missed something, since these directions are perfectly clear !

I'll probably wait another ten days or so, since I'll have to reset the time at the end of the month and don't want to mess with the "experiment" by hacking the watch.

Will also be interesting to try an additional experiment whereby I repeat, but wear the watch every other day.

Start today!
Perform the "experiment" until the end of October, 12 days are sufficient for a first data set.
Can you use the WatchTracker app?

[dannyp]

Originally Posted by saxo3

Start today!
Perform the "experiment" until the end of October, 12 days are sufficient for a first data set.
Can you use the WatchTracker app?

Thanks. Should be easy enough to use! Out of curiosity, is it worth doing an every-other-day test at some point, since that's the most likely scenario for long-term use?

[saxo3]

Originally Posted by dannyp

Thanks. Should be easy enough to use! Out of curiosity, is it worth doing an every-other-day test at some point, since that's the most likely scenario for long-term use?

Do it as I described in 2038, every day.

[Dirt]

Originally Posted by dannyp

Well, now I'm getting a little concerned...

After the post-winding time loss, I wore the watch every day for a few days, instead of every other. Gradually, the time loss dissipated and got closer to zero (gaining back about a second a day). Yesterday I didn't wear the watch. Today I put it back on, and it's roughly +2.5 (but was -1.5 when I took it off on Sunday night).

When I store it, granted, it's dial-up but this is how I've always kept it at night.

Reading through this thread, it seems that the issues have been more pervasive with the earliest 3235 movements, but I'm reaching the six month mark with mine, which I also seem to recall being when some of these popped up.

Does anyone know whether there were specific component changes made to the movement at some point, and whether it was all 32xx or just some?

The movements are mass produced.
Someone in this thread said it's not a matter of if but when, given the fact they're mass produced. Only time will tell
It stands to reason this is the likely scenario given that around 1/4 of the watches are reportedly having problems in the poll.

Hopefully we will hear of a fix, but given all aspects around the nature of the issue, I would imagine Rolex are scratching their heads trying to pin the problem down.
The whole thing is compounded by the declaration by Rolex around the 10 year service intervals. But history gives context there
It looks like Rolex have simply got too far ahead of themselves this time.

[dannyp]

Originally Posted by Dirt

The movements are mass produced.
Someone in this thread said it's not a matter of if but when, given the fact they're mass produced. Only time will tell
It stands to reason this is the likely scenario given that around 1/4 of the watches are reportedly having problems in the poll.

Hopefully we will hear of a fix, but given all aspects around the nature of the issue, I would imagine Rolex are scratching their heads trying to pin the problem down.
The whole thing is compounded by the declaration by Rolex around the 10 year service intervals. But history gives context there
It looks like Rolex have simply got too far ahead of themselves this time.

Would also be interesting to have seen, based on the poll, whether one 32xx specifically was experiencing more problems than others. At least then, maybe, one could deduce whether it had to do with the date function, the second timezone function, day function, or something common to all 32xx series.

[amanbra]

Originally Posted by dannyp

Would also be interesting to have seen, based on the poll, whether one 32xx specifically was experiencing more problems than others. At least then, maybe, one could deduce whether it had to do with the date function, the second timezone function, day function, or something common to all 32xx series.

That one is proving hard to gauge for many reasons. The 3235 would be easily the most plentiful of the 32XX movements in circulation right now and are the bulk of the reported issues.

Baz indicated he saw it on the 3285s and 3255s. The thing is there are far fewer of these two movements in circulation .3255 owners I'd dare say a lot of said owners are unlikely to be sitting at a timegrapher.

On the other hand 3230s are pretty new.

The issue appears to be around are part related to the seconds hand so I believe it's a common component between all the 32XX but i'm not an expert on that stuff.

Rolex did change the lift angle from 55 to 53. I think something has changed. The way the maths works this is not the sort of thing an engineer miscalculates and publishes for 4 years and then goes "oh whoops it's actually 53". This is also not an error you'd expect from Rolex... let alone any proper watch brand.

[dannyp]

Originally Posted by amanbra

That one is proving hard to gauge for many reasons. The 3235 would be easily the most plentiful of the 32XX movements in circulation right now and are the bulk of the reported issues.

Baz indicated he saw it on the 3285s and 3255s. The thing is there are far fewer of these two movements in circulation .3255 owners I'd dare say a lot of said owners are unlikely to be sitting at a timegrapher.

On the other hand 3230s are pretty new.

The issue appears to be around are part related to the seconds hand so I believe it's a common component between all the 32XX but i'm not an expert on that stuff.

Rolex did change the lift angle from 55 to 53. I think something has changed. The way the maths works this is not the sort of thing an engineer miscalculates and publishes for 4 years and then goes "oh whoops it's actually 53". This is also not an error you'd expect from Rolex... let alone any proper watch brand.

By now there should be plenty of 30s and 85s in circulation here, from multiple years, but I see your point. Hopefully all I'll discover is that I haven't been wearing my watch enough (or moving enough while I am).

[HiBoost]

Originally Posted by dannyp

Would also be interesting to have seen, based on the poll, whether one 32xx specifically was experiencing more problems than others. At least then, maybe, one could deduce whether it had to do with the date function, the second timezone function, day function, or something common to all 32xx series.

The RSC watchmaker indicated it was specifically due to friction (wear) on the seconds wheel pivot. As all 32xx movements have a seconds hand, I don't believe it is limited to any subset of the family.

[amanbra]

Originally Posted by dannyp

By now there should be plenty of 30s and 85s in circulation here, from multiple years, but I see your point. Hopefully all I'll discover is that I haven't been wearing my watch enough (or moving enough while I am).

I check my watches once a day on watchtracker when I wear them, I tend to wear them a week at a time as I rotate through them.

If you notice a change that's when you start taking more samples or chuck them on a grapher etc... If your watch stabilises out don't worry about it too much. If it continues to slow down then you prob want it looked at.

[Dirt]

Originally Posted by HiBoost

The RSC watchmaker indicated it was specifically due to friction (wear) on the seconds wheel pivot. As all 32xx movements have a seconds hand, I don't believe it is limited to any subset of the family.

This

[Dirt]

Originally Posted by amanbra

I check my watches once a day on watchtracker when I wear them, I tend to wear them a week at a time as I rotate through them.

If you notice a change that's when you start taking more samples or chuck them on a grapher etc... If your watch stabilises out don't worry about it too much. If it continues to slow down then you prob want it looked at.

Agreed

[dannyp]

Originally Posted by HiBoost

The RSC watchmaker indicated it was specifically due to friction (wear) on the seconds wheel pivot. As all 32xx movements have a seconds hand, I don't believe it is limited to any subset of the family.

Thanks for the link. Seems like an earlier issue so I'd be curious whether there was some fix enacted at some point, and if so, what/when. Doesn't sound like anyone ever did a visual comparison of early 2018/19 32xx and 2020/2021 32xx to see if anything had changed.

[Dirt]

Originally Posted by dannyp

Thanks for the link. Seems like an earlier issue so I'd be curious whether there was some fix enacted at some point, and if so, what/when. Doesn't sound like anyone ever did a visual comparison of early 2018/19 32xx and 2020/2021 32xx to see if anything had changed.

We are waiting for confirmation of a fix which has been implemented at Rolex. To date there is nothing.
All updated parts and service procedures are automatically rolled out throughout the worldwide service network.
I imagine we will know not long afterward all going well

[saxo3]

Controllable and Predictable Timekeeping of a 3235 watch

Since there are many so discussions about 32xx timekeeping and which watch position to choose at rest (for example during nights), I thought the following information and data might be illustrative.

Below is a nice example how to measure and keep excellent timekeeping for my healty 3235 movement, using the simple WatchTrackker app. The explanation of the white line is given inside the graph.



This specific 3235 watch loses time while being worn at top amplitude (about 290 degrees).

It gains time while losing amplitude (about 20 degrees) overnight.

The shown behaviour is movement specific and can strongly depend on the caliber regulation.
Therefore, no two movements are alike.

Choosing the right position at rest, a 32xx movement behaviour and its timekeeping becomes CONTROLLABLE and PREDICTABLE. The same holds for 31xx and other movements.

[dannyp]

Originally Posted by Dirt

We are waiting for confirmation of a fix which has been implemented at Rolex. To date there is nothing.
All updated parts and service procedures are automatically rolled out throughout the worldwide service network.
I imagine we will know not long afterward all going well

From Bas' thread linked above, it sounded as though the issue was at least in part rectified by lubricating the seconds wheel from the factory, and that lack thereof on early models had been an oversight.

[saxo3]

Originally Posted by dannyp

From Bas' thread linked above, it sounded as though the issue was at least in part rectified by lubricating the seconds wheel from the factory, and that lack thereof on early models had been an oversight.

Imagine, you knew directly from Rolex, in Geneva, all the technical 32xx issues they have discovered and solved during the past few years, could you do anything with that information?

Would it not be more interesting to know if your 32xx watch has any issues, and if so, what would be the best (and easiest) way to find out yourself?

[dannyp]

Originally Posted by saxo3

Imagine, you knew directly from Rolex, in Geneva, all the technical 32xx issues they have discovered and solved during the past few years, could you do anything with that information?

Would it not be more interesting to know if your 32xx watch has any issues, and if so, what would be the best (and easiest) way to find out yourself?

Your point is well taken. I suppose that information would be most helpful in understanding:

1. The likelihood that something is actually wrong (was mine built pre or post-fix)
2. What issues overlap with my watch based on approximate dates (to gauge potential severity and the fix)
3. The solutions available, again, partly a guess...

Basically, an all-round game of probabilities that the watch has "issue X" wrong with it to decide whether to service and if so whether to do it right away or wait and see.

[Dirt]

Originally Posted by dannyp

From Bas' thread linked above, it sounded as though the issue was at least in part rectified by lubricating the seconds wheel from the factory, and that lack thereof on early models had been an oversight.

My understanding is that part of the current remedy is to lube the Second pivot where it's not customarily a lube point.
I wonder it it's a measure to kick the can down the road along with mitigating metal dust being distributed around the insides of the movement by keeping it more or less localised
The other part of the problem is that even after lubing the pivot under the conditions set out by Rolex during a warranty service, the problem eventually comes back regardless
There's been instances reported on this forum of some watches going back three times under warranty with the exact same problem.

[amanbra]

Originally Posted by Dirt

My understanding is that part of the current remedy is to lube the Second pivot where it's not customarily a lube point.
I wonder it it's a measure to kick the can down the road along with mitigating metal dust being distributed around the insides of the movement by keeping it more or less localised
The other part of the problem is that even after lubing the pivot under the conditions set out by Rolex during a warranty service, the problem eventually comes back regardless
There's been instances reported on this forum of some watches going back three times under warranty with the exact same problem.

Yep this, my DJ36 same issue again after about 6 months from being back from warranty.

[dannyp]

Well, from the time I started the timekeeping app and now (about 47hrs) it's 1.1sec faster. So it would seem that there aren't any obvious issues with the watch itself.

Now, the interesting question is whether there re problems with my wearing patterns: When I started this, I wound it fully, so even if unworn it should have about 70hrs PR, meaning I don't know whether it's getting sufficient winding each day to maintain it indefinitely.

What I do notice is that it gains 1-1.5sec overnight, then evens out during the day, so that suggests it's doing what would be expected of this series of movements.

I'll post the first graph shot when I have at least a week.

[dannyp]

Here are some graphs for the first two days. I'm not 100% sure how to interpret, or the different things they measure...

[saxo3]

Originally Posted by dannyp

Here are some graphs for the first two days. I'm not 100% sure how to interpret, or the different things they measure...

Good start.
How do you place the watch at rest overnight?

The app graphs are the following:
- Offsets
- Rates
- Variance

The app also has a page where you see the date/time and the measured data: "View data points" Can you show that too?

[saxo3]

This is the relevant graph (Offsets). My interpretation:

- You did not set the watch to the correct atomic time because the graph starts with an offset of about +2 s.

- The lines going down are what happens during daytime.
- The lines going up are what happens during watch at rest.
- If that is right, then you gain more time at rest overnight compared to what you are losing during the day.

- The green line is a coarse fit to your data.
- At present, your watch is running with +0.7 s/d.

- That looks VERY good to me.

For the coming days or weeks:
- do NOT wind the movement,
- do NOT change the watch position at rest (overnight),
- wear this watch every day,
- try to keep the hours at rest similar each day.

[dannyp]

Originally Posted by saxo3

This is the relevant graph (Offsets). My interpretation:

- You did not set the watch to the correct atomic time because the graph starts with an offset of about +2 s.

- The lines going down are what happens during daytime.
- The lines going up are what happens during watch at rest.
- If that is right, then you gain more time at rest overnight compared to what you are losing during the day.

- The green line is a coarse fit to your data.
- At present, your watch is running with +0.7 s/d.

- That looks VERY good to me.

For the coming days or weeks:
- do NOT wind the movement,
- do NOT change the watch position at rest (overnight),
- wear this watch every day,
- try to keep the hours at rest similar each day.

I rest it face-up overnight. As for the other datapoints:

[saxo3]

Originally Posted by dannyp

I rest it face-up overnight. As for the other datapoints:

Thanks. These data points confirm my interpretation from post 2063.

Originally Posted by dannyp

Now, the interesting question is whether there re problems with my wearing patterns: When I started this, I wound it fully, so even if unworn it should have about 70hrs PR, meaning I don't know whether it's getting sufficient winding each day to maintain it indefinitely.

Your watch performs different compared to what you think it should do!

During the day, the movement is at high amplitude due to your wrist movements. The overall rate is slightly negative, therefore the watches loses a bit of time.

During nights, in fixed dial up position, the movement loses amplitude but gains in rate. That means it is running faster. The day-night net effect results in a positive overall rate (+0.7 s/d).

I have measured that in detail for my Sea-Dweller 126600 (3235), as a function of watch rest position overnight.

[dannyp]

So I understand that part, but the question is what happens once it gets past 70hrs, and whether I move enough to keep put going. In theory, that initial wind would keep it going another day without any use. So once that’s past I’ll be curious if patterns change at all.

[saxo3]

Originally Posted by dannyp

… but the question is what happens once it gets past 70hrs….

This question is easy to answer: during the course of the entire PR (about 70-72 hours), the movement amplitude will become continuously smaller, as for every mechanical watch, until your watch finally stops running. Many examples of such PR measurements are shown (graphs) and explained in this thread, mainly by Charles and saxo3 (me).

Originally Posted by dannyp

… and whether I move enough to keep put going.

Nobody can answer that for you. The first systematic experiment, which you are currently conducting, will already give you a first indication.

Originally Posted by dannyp

In theory, that initial wind would keep it going another day without any use.

Yes, but do NOT test it already now and do NOT vary 2 parameters at the same time. That is also the reason whyyou should keep dial up as your chosen rest position overnight. If you change to another rest position, it will most likely also be visible in your timekeeping, which you now measure with the WatchTracker app.

[saxo3]

For dannyp

Here is a nice example of two combined (Timegrapher + WatchTracker) measurements, along the entire power reserve, until the 3285 movement stopped running. For both tests, the watch was all time at rest in dial up position. The experiment was done by Charles.



One can see:

- Excellent time keeping until about 69 hours (see Watch Tracker curves).

- The watch GAINED time during the amplitude decreases in both tests.

- But this behavior is NOT necessarily identical for all 32xx calibers.

[dannyp]

Originally Posted by Dirt

My understanding is that part of the current remedy is to lube the Second pivot where it's not customarily a lube point.
I wonder it it's a measure to kick the can down the road along with mitigating metal dust being distributed around the insides of the movement by keeping it more or less localised
The other part of the problem is that even after lubing the pivot under the conditions set out by Rolex during a warranty service, the problem eventually comes back regardless
There's been instances reported on this forum of some watches going back three times under warranty with the exact same problem.

I can see how lubing from the beginning might solve the problem, while lubing at service would not. With the latter, some damage to the parts themselves may already be done and if they aren't replaced, that damage will not be resolved with lubrication. On the other hand, if lubed from the beginning, perhaps the damage wouldn't occur in the first place?

Of course, if the warranty service identifies and replaces the worn parts, too, well, then I've got nothing...

[dannyp]

Also, the inverse amplitude/time gain/loss relationship makes perfect sense to me as a parent: As my kids get more and more tired (losing PR, amplitude decreasing, they get more and more wild and crazy, right up until the moment they crash)