32xx movement problem poll and data thread

[saxo3]

Originally Posted by rbndylan

Hello,

I'm not going to send it back. The good thing is that in a few months I will be able to compare it with the measurements I have taken and depending on that, I will decide whether or not to send it to the RSC

[MikeyV]

Welp, I sent my watch back in (second time) for the sickness. It was losing 15 sec per day.

I waited as long as I could before the warranty running out.

I REALLY hope there's a permanent fix for this by now - next service will be on me...wish me luck.

I took in my broken DJ as well as my Pelagos which has stopped winding itself. I think the rotor came lose on that one. So I'm watchless for a few months. :(

[saxo3]

Mike, your watch is from 05/2018?

When was the first repair? After this date, how long did it run well until you noticed timekeeping issues for the first time?

[MikeyV]

Originally Posted by saxo3

Mike, your watch is from 05/2018?

When was the first repair? After this date, how long did it run well until you noticed timekeeping issues for the first time?

The watch was bought in June 2018. It went back to Rolex Dallas in May 2019 (was losing over 6 sec per day at the time).

Looking at my record, it ran well from July 2019 (when I got it back) until December 2019 when it was back to losing 3-4 seconds per day.

Last week it lost a minute in 5 days. :(

[csaltphoto]

Originally Posted by MikeyV

The watch was bought in June 2018. It went back to Rolex Dallas in May 2019 (was losing over 6 sec per day at the time).

Looking at my record, it ran well from July 2019 (when I got it back) until December 2019 when it was back to losing 3-4 seconds per day.

Last week it lost a minute in 5 days. :(

I'm sorry but that is just piss poor performance for an expensive watch. I don't understand the apologists. I've said it before but it would not be the same if Rolex didn't make a F big deal about timekeeping, 10 year service intervals and so on. But they do make a big deal about it. There are plenty of expensive watches that do not make any sort of performance guarantee and that's fine. You buy those watches for a different reason (although high horology at the expense of excellence in the one thing a watch is supposed to do well kinda escapes me but different strokes and all that.)

[Scholar]

Watchmaker Al of Archer Watches posted this overview of the issues with the 32 series on Watchuseek and saxo suggested I share it here:

Originally Posted by Archer, post: 56642518, member: 48283

The torque available from a given strength of mainspring is a finite thing. That's somewhat different from the length of the power reserve. A mainspring is a different kind of spring, so I'll explain...

The torque of the mainspring is a function of its width and it's thickness. The power reserve is a function purely of the length, all else being equal.

A typical mainspring designation goes something like this...

1.60 X .10 X 267 - all numbers are in mm.

First number is the width or height of the spring, and is related to how tall the mainspring barrel is.

The second number is the thickness, or commonly known as the strength.

The third number is the length.

The first two numbers can affect the torque delivered from the mainspring, but do in in different proportions. A change in width has a small impact, where a change in strength has a very large impact - it is cubed, so a small change in the strength measurement can have a very significant impact on the torque.

Neither of these are directly related to power reserve. That is determined by the length of the spring - simple relationship between the length of the spring determining the number of turns the mainspring barrel will make, and the gear ratio between the teeth on the barrel, and the subsequent train wheels. Again, all else being equal.

The only relationship they have to power reserve is that when you make a mainspring thinner, you can fit a longer spring in the same sized barrel. That is what Rolex has done. In an effort to catch up to other brands who have extended their power reserves, they have tried to fit more spring in the same space, rather than taking another approach such as two mainspring barrels.

So to fit more spring in the same barrel they made 2 changes:

1 - Thinner barrel wall.
2 - Thinner mainspring.

These changes have consequences. The thinner barrel wall now means that you cannot use the barrel over again, because it's too fragile to open up and close again in service. So the entire mainspring barrel must be replaced at every service. That's a servicing implication, but not necessarily a performance implication.

The thinner mainspring now reduces the torque delivered, which reduces balance amplitude. To now make up for that reduced torque, they did some very un-Rolex like things to the escapement. They made things thinner and more fragile. Smaller surface contact reduces friction, but concentrates the forces and wear that may occur. There's always a give and take in any design, and watch movements are no different in that regard.

So why is balance amplitude important? It's simple - isochronism.

A balance inside a watch is a very imperfect oscillator, so watch companies go to a lot of trouble to help it maintain its period (rate) when there are changes in the distance travelled (amplitude), but there's only so much you can do.

So this means that when amplitude drops, timing changes. How much depends on how big the drop is, but also where in the amplitude range that drop is occurring. Rolex watches don't tend to have really high amplitudes to begin with, compared to many other makers. So on say a standard ETA movement, it's not unusual for amplitudes to be in excess of 300 degrees, where on a Rolex you are more typically looking at 270's or 280's. In fact the point at which there is too much amplitude, and you experience rebanking (also known as knocking), is much lower on a Rolex than say on an ETA 2892, due to the geometry of the Rolex escapement.

When you get to lower amplitudes, such as the low 200's, then some odd things start to happen. There's an amplitude where poise errors on the balance tend to go away (these are only in vertical positions) and it's widely considered that around 220 degrees is where this happens, at least with a traditional lever escapement.

As you drop below that, there's a point where poise errors reverse, so if at full wind amplitude crown left runs faster that crown right, those reverse, and now crown right will run faster than crown left. As you drop lower, the magnitude of those poise errors gets magnified significantly. This is well known with watchmakers and is useful in the context of dynamic poising procedures, so for that I drop the balance amplitude to 160 degrees to magnify the errors, locate the heavy spot on the balance wheel, and make the appropriate adjustments.

So for example I just dynamically poised a watch last week where at full wind the positional variation was 21 seconds across 6 positions (amplitudes were from the mid 280's to around 310). When I dropped the balance amplitude to 160 degrees, those errors ballooned to 57 seconds, so they nearly tripled in magnitude.

So on these 32XX watches, Rolex wants to keep the balance amplitude after 24 hours above 200 degrees to avoid some of the more extreme effects that are seen at lower amplitudes. It is somewhat of a detriment that the amplitude starts lower on these watches to begin with, because you have less of a buffer in the amount that it can drop before you start to see those unwanted effects.

So do they know what the problem is? Yes, I'm sure they do by now. The problem that presents itself is repairing this issue without sacrificing any of the promised performance metrics that this movement is known for. So that means not changing the accuracy and power reserve goals. It's not that they don't know what to do, because they have built plenty of robust movements in the past - it's doing so while maintaining the current performance demands that is the issue.

Hope this helps those who are interested, understand the technical side of this a bit better...

Cheers, Al

[saxo3]

Originally Posted by Scholar

Watchmaker Al of Archer Watches posted this overview of the issues with the 32 series on Watchuseek and saxo suggested I share it here:

Thank you very much for re-posting this information and explanations.

It's by far the best technical part I have read about the 32xx movement issues.

Watchmaker Al (of Archer Watches) should join this thread if he has not done yet …

[CharlesN]

Originally Posted by Scholar

Watchmaker Al of Archer Watches posted this overview of the issues with the 32 series on Watchuseek and saxo suggested I share it here:

This is one of the best and most comprehensive posts I have ever seen on this, or any other forum.

Thank you.

The original post by Al is so clear and pretty easy to understand without any formal technical knowledge.

A True "Gem:" of a post.

[brandrea]

Originally Posted by CharlesN

This is one of the best and most comprehensive posts I have ever seen on this, or any other forum.

Thank you.

The original post by Al is so clear and pretty easy to understand without any formal technical knowledge.

A True "Gem:" of a post.

Al is a true gentleman and was helpful in answering questions about my 321 and misinformation posted here and on other forums.

Sorry to digress from this thread, I just thought I’d give Al some credit where credit is due

[Mountain]

Originally Posted by Scholar

Watchmaker Al of Archer Watches posted this overview of the issues with the 32 series on Watchuseek and saxo suggested I share it here:

Very interesting, thanks for sharing.

However, I’m still not entirely clear why/ how the problem arises in the first place. If the watch has been keeping good time for several months, what happens to precipitate the sudden reduction in precision? Is there a known trigger?

Equally, why would the watch, after repair and keeping good time, suddenly start losing time once more (in many cases)?

Once again, thank you!

[JMGoodnight369]

Originally Posted by Mountain

Very interesting, thanks for sharing.

However, I’m still not entirely clear why/ how the problem arises in the first place. If the watch has been keeping good time for several months, what happens to precipitate the sudden reduction in precision? Is there a known trigger?

Equally, why would the watch, after repair and keeping good time, suddenly start losing time once more (in many cases)?

Once again, thank you!

Based on what Al wrote, it seems like the issue the parts get prematurely worn due to everything being shrunk to mitigate the smaller hairspring

[Easy E]

Originally Posted by Scholar

Watchmaker Al of Archer Watches posted this overview of the issues with the 32 series on Watchuseek and saxo suggested I share it here:

Great post, thanks.

[goodolejr]

Originally Posted by Scholar

Watchmaker Al of Archer Watches posted this overview of the issues with the 32 series on Watchuseek and saxo suggested I share it here:

Thank you, Al.

[alphadweller]

Originally Posted by Scholar

Watchmaker Al of Archer Watches posted this overview of the issues with the 32 series on Watchuseek and saxo suggested I share it here:

Thanks for reposting Archer's comments from WUS. I completely missed that thread. A crystal clear explanation of what's going on with the 3235. The best I've read so far.

[LAS45]

Very informative post. Thank you for sharing


Sent from my iPhone using Tapatalk

[fmc000]

So this basically means that Rolex shoot themselves in the foot and now they're fighting to get the bullet out of it. I'm curious of what the naysayers may have to add now.

[Scholar]

Originally Posted by fmc000

So this basically means that Rolex shoot themselves in the foot and now they're fighting to get the bullet out of it. I'm curious of what the naysayers may have to add now.

The naysayers switched to accusing Al (one of the best watchmakers in the business) of being a nobody who's trying to piggyback on the glory of Rolex. If anything he's saying that the problems with this movement are caused by Rolex moving away from its own traditional approach to movement design…

Originally Posted by brandrea

Al is a true gentleman and was helpful in answering questions about my 321 and misinformation posted here and on other forums.

Sorry to digress from this thread, I just thought I’d give Al some credit where credit is due

I have learned so much from him over the years about everything to do with watches. He writes clearly and takes great macro photos so we can see how these things are put together.

[Dirt]

Originally Posted by Scholar

The naysayers switched to accusing Al (one of the best watchmakers in the business) of being a nobody who's trying to piggyback on the glory of Rolex. If anything he's saying that the problems with this movement are caused by Rolex moving away from its own traditional approach to movement design…



I have learned so much from him over the years about everything to do with watches. He writes clearly and takes great macro photos so we can see how these things are put together.

Thanks for sharing the post by Al.
I imagine there are literally scores of watchmakers either scratching their heads or tearing their hair out over the 32xx movements.
When all is said and done the common denominator is the escapement as the single biggest factor that stands out.
Any other changes apart from the spring barrel seem to be secondary but I suspect there is a deeper concern that may involve a number of converging aspects of the design that are amplifying or causing a core issue that's not obvious.

[michael.brando86]

Hello,

I’ve been tracking the precision of my Rolex Submariner (movement 3230) since I bought it in October 2021, by synchronizing precisely the hour with an atomic clock everytime I set it.

I wear my watch 24/7, shower with it, etc.

Here are my results :

1. +22s in 140d > +0.2 spd
2. -09s in 127d > -0.1 spd
3. -15s in 14d > -1.0 spd
4. -48s in 37d > -1.3 spd
5. -48s in 31d > -1.6 spd
6. -52s in 56d > -0.9 spd
7. -61s in 48d > -1.3 spd
8. -62s in 52d > -1.2 spd
9. -40s in 18d > -2.2 spd
10. -62s in 27d > -2.3 spd
11. -80s in 26d > -3.1 spd

As you can see, the precision for the first 200+ days was incredible, and then it slowly started to slow down until some days ago where it got under the -3 spd mark breaking what I understand is the superlative standard of Rolex (-2/+2).

The watch is 1.5 year old so still under warranty. I’m wondering if it would be useful to bring it to the RSC now.

I would appreciate your opinion on this.

Thanks a lot,
Michael

[Mountain]

And a follow-up comment and question from the final paragraph:

“So do they know what the problem is? Yes, I'm sure they do by now. The problem that presents itself is repairing this issue without sacrificing any of the promised performance metrics that this movement is known for. So that means not changing the accuracy and power reserve goals. It's not that they don't know what to do, because they have built plenty of robust movements in the past - it's doing so while maintaining the current performance demands that is the issue.”

Given the apparent difficulty in achieving the movements objectives, one also has to wonder whether a permanent solution can be found at all, or whether it simply comes down to the skill/ luck of the watchmaker in reassembling one’s watch after a service?

[MikeyV]

Originally Posted by Mountain

And a follow-up comment and question from the final paragraph:

“So do they know what the problem is? Yes, I'm sure they do by now. The problem that presents itself is repairing this issue without sacrificing any of the promised performance metrics that this movement is known for. So that means not changing the accuracy and power reserve goals. It's not that they don't know what to do, because they have built plenty of robust movements in the past - it's doing so while maintaining the current performance demands that is the issue.”

Given the apparent difficulty in achieving the movements objectives, one also has to wonder whether a permanent solution can be found at all, or whether it simply comes down to the skill/ luck of the watchmaker in reassembling one’s watch after a service?

I would GLADLY take a 48 hr power reserve if it meant my watch would run like it's supposed to. Gladly. I'd even pay extra for it.

[Mountain]

Originally Posted by MikeyV

I would GLADLY take a 48 hr power reserve if it meant my watch would run like it's supposed to. Gladly. I'd even pay extra for it.

As would I. Slightly galling if all these problems arise from attempting to achieve a 70hr power reserve, which to me is pointless anyway! Had I known all of this before buying my Explorer, I’d simply have bought one of the earlier 36mm models.

[Easy E]

Originally Posted by MikeyV

I would GLADLY take a 48 hr power reserve if it meant my watch would run like it's supposed to. Gladly. I'd even pay extra for it.

+1, for sure.

[MikeyV]

What does the mainspring talk matter (it is interesting and informative)? These watches run so spot on when fresh (the spring and PR doesn't effect the timekeeping then) we're left to wonder what makes them go south after 6-9 months of continuous running.

Wear or oil or both.

[csaltphoto]

Originally Posted by MikeyV

What does the mainspring talk matter (it is interesting and informative)? These watches run so spot on when fresh (the spring and PR doesn't effect the timekeeping then) we're left to wonder what makes them go south after 6-9 months of continuous running.

Wear or oil or both.

My understanding of it was that the thinner mainspring results in less torque resulting in the lower amplitudes that the movement was designed to run at, in order to create more PR. But lower amplitudes exacerbate poise errors. Also the smaller palette stones, designed to reduce friction in the movement, might have higher wear rates from less surface area to distribute the force. Previously someone posted how to lubricate them correctly to mitigate oil migration. Might not be enough though?

[BruRol]

Originally Posted by csaltphoto

My understanding of it was that the thinner mainspring results in less torque resulting in the lower amplitudes that the movement was designed to run at, in order to create more PR. But lower amplitudes exacerbate poise errors. Also the smaller palette stones, designed to reduce friction in the movement, might have higher wear rates from less surface area to distribute the force. Previously someone posted how to lubricate them correctly to mitigate oil migration. Might not be enough though?

The observations on the main spring dimensions and their impact on torque may be correct but is assumes there has been no change to spring material properties. It's quite possible that the alloy mix used to make the springs has been adjusted to give a thinner spring equal, if not more, bending strength. Thus maintaining the torque figure.

Does anyone have a mass spectrometer kicking about?

[Mountain]

Triangulating all information from the different sources, it seems as if there’s no single issue to point at, rather a possible combination of several. No doubt this is why there seems no obvious solution even after eight years.

Therefore, the question I guess we need to ask ourselves is whether Rolex, or indeed anyone else, can produce a final fix?

Or, whether we should just cut our losses and forget the 32xx series watches?

I’m not sure exactly where I stand now. But, I can be certain the trust I had towards my explorer and Rolex overall has been severely damaged.

[saxo3]

Originally Posted by Mountain

Triangulating all information from the different sources, it seems as if there’s no single issue to point at, rather a possible combination of several. No doubt this is why there seems no obvious solution even after eight years.

Robert, I fully agree.

The following I posted 6 months ago in 3161:

The root cause of the 32xx issues could be a combination of new movement design, new materials, new lubricants, high-performance epilames (thin film coatings), which should prevent oil from spreading across the surface, thereby preventing oil from flowing out of the friction zone.

[Devildog]

Originally Posted by saxo3

Robert, I fully agree.

The following I posted 6 months ago in 3161:

The root cause of the 32xx issues could be a combination of new movement design, new materials, new lubricants, high-performance epilames (thin film coatings), which should prevent oil from spreading across the surface, thereby preventing oil from flowing out of the friction zone.

What does appear to be coming out of all of this is that the movements have a fundamental design "weakness" which has arisen from the marketing pressure to produce 70 hour reserve. A weakness that is exacerbated by a lack of lubrication creating friction and or wear sufficient to reduce the amplitude to a level that timekeeping is significantly affected.

And that the chronology escapement was created by necessity, not advancement in design.

The question I have is that the Daytona has had a 72 hour power reserve for years. Presumably it has a larger barrel?

[steubi1]

Originally Posted by Devildog

What does appear to be coming out of all of this is that the movements have a fundamental design "weakness" which has arisen from the marketing pressure to produce 70 hour reserve. A weakness that is exacerbated by a lack of lubrication creating friction and or wear sufficient to reduce the amplitude to a level that timekeeping is significantly affected.

And that the chronology escapement was created by necessity, not advancement in design.

The question I have is that the Daytona has had a 72 hour power reserve for years. Presumably it has a larger barrel?

Yes. The Daytona movement has a larger diameter because it must not fit into a 36 mm case...
So the barrel has much more volume, I guess...

Tom