32xx movement problem poll and data thread

[saxo3]

Originally Posted by FlyinHawaiian

Data for my GMT II 3285 movement.

Thanks for the data. You have a very nice watch with a unique dial.

As I said above in #5249, already after 48 hours (2/3 of the PR), your 3285 does not have enough power to maintain all 3 vertical amplitudes at a sufficient level to ensure reasonable timekeeping.

The good news is that your watch does not lose time in dial up position at rest. Your watch is not even 1 year old and still has 4 years of Rolex guarantee.

Personally, I would prefer the same watch with a 3185 movement.

[Poodlopogus]

Originally Posted by saxo3

2015: 3235, 3255
2016: -
2017: -
2018: 3285
2019: -
2020: 3230

Oops. For some reason I thought the GMT came before the time-and-date version.

Originally Posted by saxo3

Never heard of this as a Rolex design criterion for the 32xx. They extended the power reserve (PR) from 44 hours (31xx) to 70 hours (32xx) by a completely new movement design. After about 3/4 of the PR, there is not enough power to keep the amplitudes high enough to ensure good timekeeping for all 32xx watches. One design decision was to stay with only 1 mainspring barrel, more wouldn't fit in these tractors?

I thought I'd read that they basically had to do this because amplitude was going to be sacrificed to meet the other criteria, and they thought they'd achieved it (indeed they did, but not long-term).

[FlyinHawaiian]

Originally Posted by saxo3

Thanks for the data. You have a very nice watch with a unique dial.

As I said above in #5249, already after 48 hours (2/3 of the PR), your 3285 does not have enough power to maintain all 3 vertical amplitudes at a sufficient level to ensure reasonable timekeeping.

The good news is that your watch does not lose time in dial up position at rest. Your watch is not even 1 year old and still has 4 years of Rolex guarantee.

Personally, I would prefer the same watch with a 3185 movement.

My t=0 is spot on! Since when we are wearing our watch, the movement should be between t=0 and t=24 anyway right? So, well within COSC parameters.

[Poodlopogus]

Giving this a little more thought...

Here's a great question for the watchmakers that might help us better understand the relationship between more moving parts and worse performance: Are the parts replaced the same on all four 32xx calibres? Or are they frequently different, and do they all require replacements, or are some more just in need of regulation?

This also demonstrates the great conflict between R&D and marketing: Marketing dictates that the "flagship" be released first when a new something-or-other comes out. Hence, the 3255 in the DD first. But it would have been far more prudent, IMO, to release the least complicated version first to get more "field data" on.

The 3230 doesn't require any of the "rapid change" motions that come with a day or date wheel, nor does it need to exert the extra force to keep a fourth hand in motion.

[nikola0406]

https://watchesbysjx.com/2021/05/rol...-analysis.html


Nice read about 32xx movement

[Dirt]

Originally Posted by Poodlopogus

Oops. For some reason I thought the GMT came before the time-and-date version.



I thought I'd read that they basically had to do this because amplitude was going to be sacrificed to meet the other criteria, and they thought they'd achieved it (indeed they did, but not long-term).

Interesting that you think you had read something about the development of this watch calibre. It would be great to know more about it all

We need to be mindful of the fact that watch movements are literally a grab bag of compromises so at all times some criterior has got to be proritised over another and another.
I think this thread has well demonstrated that Amplitude is probably more important in the grand scheme of these things than the mothership had ever appreciated as we know that the 32xx movements were never big on Amplitude.
Perhaps the Chronergy escapement is another mistake in the history of horology?
Also as has been mentioned in theses pages, a dual Spring barrel would be better utilised especially when pursuing much longer power reserves and may be a distinct advantage for a Chronergy escapement.

[searas]

Originally Posted by Dirt

Also as has been mentioned in theses pages, a dual Spring barrel would be better utilised especially when pursuing much longer power reserves and may be a distinct advantage for a Chronergy escapement.

The barrel in the 32xx is supposedly to add 10 hours of power reserve (according to the link in post #5255), but this same barrel has been mentioned as one of supposedly sources of the problems. I will be happy with just 60 hours of power reserve, keeping a barrel similar to those in 31xx, if that would mean a more reliable movement. It seems that Rolex went too far in trying to improve an already reliable and precise movement such as the 31xx.

Regards,
Daniel

[Poodlopogus]

Originally Posted by Dirt

Interesting that you think you had read something about the development of this watch calibre. It would be great to know more about it all

We need to be mindful of the fact that watch movements are literally a grab bag of compromises so at all times some criterior has got to be proritised over another and another.
I think this thread has well demonstrated that Amplitude is probably more important in the grand scheme of these things than the mothership had ever appreciated as we know that the 32xx movements were never big on Amplitude.
Perhaps the Chronergy escapement is another mistake in the history of horology?
Also as has been mentioned in theses pages, a dual Spring barrel would be better utilised especially when pursuing much longer power reserves and may be a distinct advantage for a Chronergy escapement.

Agree that Chronergy is basically stupid. Maybe clever in theory, but in my mind, there were two way, way better options:

1. Forget about gimmicks and just do a better finished version of the Tudor MT movement. The specs are awesome, and any reported systemic issues seem to have been ironed out long ago. They could have changed enough to avoid the perception they were the same.

2. Keep the 31xx for the 36mm pieces and do a proper long-PR movement for the 40mm+ pieces. The vast majority of manufactures do exactly that; look at VC, Chopard, Zenith, Omega, Blancpain, etc. The midsize pieces all have shorter PR than the largest.

[saxo3]

Originally Posted by FlyinHawaiian

Data for my GMT II 3285 movement.

The isochronism of this GMT is not good and worse than the 2 GMT's (03/2024; 05/2024) of EasyE. Your 3285 watch (10/2023) is also a bit older, so it fits the picture. Isochronism numbers and graphs I'll spare you.

[SwissSteph]

Hello everyone,

I received my “Weishi 1900” a little early, which is great. I'm going to start testing my two watches, Explorer II (2024) and Sea-Dweller (2024) on Thursday September 26 (5 days from now).

After reading here and watching a video in French (I'll put the link in my next post), here are the parameters I'll be using. What do you think?

Another question: is it possible to save these parameters in the “Weishi” ... at each “Power OFF” everything is lost for the next “Power ON”?

[FlyinHawaiian]

.

[tho68]

Here are my measurements.

Interesting that the 3U position is significantly lower than the others. Furtunately, it is the position that is least "in use".
Accuracy is still very good after 24h.

I will repeat the test over 60h period next week when more time available.

There 2 questions I have:

1. When I use my watch every day how often does it happen that I get to the situation with less than 24h PR? Quite rarely I would say. BTW, when you walk with the watch how long does it take to fully wind up the caliber?
2. When does it happen that my watch ends up where the amplitue is less than 200 24h after full wind up?

[FlyinHawaiian]

Originally Posted by tho68

There 2 questions I have:

1. When I use my watch every day how often does it happen that I get to the situation with less than 24h PR? Quite rarely I would say. BTW, when you walk with the watch how long does it take to fully wind up the caliber?
2. When does it happen that my watch ends up where the amplitue is less than 200 24h after full wind up?

The same question I had above. While wearing the watch, one should never fall below the t=24 power reserve. So amplitude and rate should stay within +2/-2 sec parameters!


Sent from my iPhone using Tapatalk

[Poodlopogus]

Originally Posted by FlyinHawaiian

Data for my GMT II 3285 movement.

What's amazing, to me, is how little timekeeping deviation there is for DU positions across time intervals. I knew it would typically be one of the two fastest, but didn't realize it could maintain consistency so much better than others.

I don't have a timegrapher, so I tried a different experiment w/my 3230 (that averages about -3.8/d): Fully wound it and rested it 3U. Will take some measurements and see what happens.

[Poodlopogus]

Originally Posted by Poodlopogus

What's amazing, to me, is how little timekeeping deviation there is for DU positions across time intervals. I knew it would typically be one of the two fastest, but didn't realize it could maintain consistency so much better than others.

I don't have a timegrapher, so I tried a different experiment w/my 3230 (that averages about -3.8/d): Fully wound it and rested it 3U. Will take some measurements and see what happens.

Ok, so after about 23hrs it had lost 9.3s when rested the entire time 3U. Maybe tonight I'll re-wind and try 9U, just to see what happens.

Then I'll go to a watchmaker and see if s/he can transplant the MT5400 from my BB54 into my Explorer .

[GradeV]

Originally Posted by searas

The barrel in the 32xx is supposedly to add 10 hours of power reserve (according to the link in post #5255), but this same barrel has been mentioned as one of supposedly sources of the problems. I will be happy with just 60 hours of power reserve, keeping a barrel similar to those in 31xx, if that would mean a more reliable movement. It seems that Rolex went too far in trying to improve an already reliable and precise movement such as the 31xx.

Regards,
Daniel

I think the other compromise was / is thickness. Rolex movements are not especially thin (which is not a bad thing; I think of them as robust) but it is possible that a thicker movement would have affected the case thickness.

As for me personally I’ve not acquired a 32XX and don’t intend to. No hard data from me.

[tho68]

My "practical" expirience confirms what I measured (see below) that my watch is crazy accurate. After around 4 days (including the test period) the deviation is 1s max.

[Dirt]

Originally Posted by tho68

My "practical" expirience confirms what I measured (see below) that my watch is crazy accurate. After around 4 days (including the test period) the deviation is 1s max.

I'm hearing you.
I get the same practical results out of my 13 year old 31xx daily driver that's on my wrist as i tap.
And it's not the best 31xx i've ever owned throughout the last 20 + years with respectable Amplitudes to go with it

[saxo3]

Originally Posted by tho68

Here are my measurements.

Interesting that the 3U position is significantly lower than the others. Furtunately, it is the position that is least "in use".
Accuracy is still very good after 24h.

I will repeat the test over 60h period next week when more time available.

All measured amplitudes are in the optimal range after full winding and are above 200° after 24 hours.

It notice that the DU, 3U, and 9U rates all show a rather large difference between t = 0 and t = 24. Do you have any explanation for this?

The rate changes in 3U and 9U compensate each other so that the X value remains almost unchanged after 24 hours.

[Poodlopogus]

Originally Posted by saxo3

All measured amplitudes are in the optimal range after full winding and are above 200° after 24 hours.

It notice that the DU, 3U, and 9U rates all show a rather large difference between t = 0 and t = 24. Do you have any explanation for this?

The rate changes in 3U and 9U compensate each other so that the X value remains almost unchanged after 24 hours.

Would it matter where the watch was in the date change cycle when each measurement was taken?

[Poodlopogus]

Originally Posted by searas

The barrel in the 32xx is supposedly to add 10 hours of power reserve (according to the link in post #5255), but this same barrel has been mentioned as one of supposedly sources of the problems. I will be happy with just 60 hours of power reserve, keeping a barrel similar to those in 31xx, if that would mean a more reliable movement. It seems that Rolex went too far in trying to improve an already reliable and precise movement such as the 31xx.

Regards,
Daniel

Originally Posted by GradeV

I think the other compromise was / is thickness. Rolex movements are not especially thin (which is not a bad thing; I think of them as robust) but it is possible that a thicker movement would have affected the case thickness.

As for me personally I’ve not acquired a 32XX and don’t intend to. No hard data from me.

If I've understood other things I've read by watchmakers elsewhere, the primary compromise for the sake of PR was mainspring thickness. However, I don't think it would be as simple as "use a shorter, thicker mainspring, lose 10hr PR, maintain better amplitude" because of the escapement design. I think that was meant to be used in conjunction with a thinner mainspring.

I think I also recall watchmakers here saying that the barrel design itself just makes it a disposable, rather than serviceable, component.

The real problem, from my perspective, was in insisting that 36mm and 40+mm watches use identical movements. NOBODY else does this, other manufactures have a movement for large watches distinct from their midsize counterparts.

So, had Rolex kept the 31xx for 36mm watches and designed a larger movement for the larger watches, fewer compromises would have been required.

[saxo3]

Originally Posted by Poodlopogus

Would it matter where the watch was in the date change cycle when each measurement was taken?

Yes!
@tho68: at what exact time of day did you take your measurements?

[tho68]

when exactly did you measure?
->
I did the measurements at around 6.40pm. 24h apart.

It notice that the DU, 3U, and 9U rates all show a rather large difference between t = 0 and t = 24. Do you have any explanation for this?
->
The delat is in the -(20-25) range. No idea, why it is like this.
I just started a full 60h test period. I started the test at 8:15pm. Is this ok?

[Easy E]

Originally Posted by tho68

when exactly did you measure?
->
I did the measurements at around 6.40pm. 24h apart.

It notice that the DU, 3U, and 9U rates all show a rather large difference between t = 0 and t = 24. Do you have any explanation for this?
->
The delat is in the -(20-25) range. No idea, why it is like this.
I just started a full 60h test period. I started the test at 8:15pm. Is this ok?

I believe you want to be between the 8s (8a-8p). I think the date wheel adds friction otherwise.

[saxo3]

Originally Posted by tho68

I did the measurements at around 6.40pm. 24h apart.

The delat is in the -(20-25) range. No idea, why it is like this.
I just started a full 60h test period. I started the test at 8:15pm. Is this ok?

Both are ok if the watch was set to this time.

The reason for this is as follows:

The engagement of the 32xx date change mechanism reduces the amplitudes and also impacts on the rates.

To avoid this measurement "artefact" one should not measure during the periods 23:00 - 01:00 and 05:00 - 08:00. I have reported this several times in this thread.

So far no one has been able to tell me why 32xx amplitude drops also occur at 05:00 - 08:00. During a complete PR scan these amplitude reductions appear 6 times, always at the same time intervals.

The graphs below are continuous measurements (2 watches) over the entire power reserve with measured data points every 60 s, see post #1663 (27 June 2021).

[Poodlopogus]

So I did a little test over the past few days, trying my 3230 resting 3U or 9U for about 36hrs each (with a full wind before initiating each). 3U was about -10s/d while 9U was about -5s/d. Definitely out of spec but was expecting the opposite in terms of which position ran slower. Maybe because this is a non-date watch? Now trying DD...

[saxo3]

Originally Posted by Poodlopogus

So I did a little test over the past few days, trying my 3230 resting 3U or 9U for about 36hrs each (with a full wind before initiating each). 3U was about -10s/d while 9U was about -5s/d. Definitely out of spec but was expecting the opposite in terms of which position ran slower. Maybe because this is a non-date watch? Now trying DD...

Your test method (w/o a timegrapher) is time consuming but in principle ok.

-10 s/d in 3U is a strong indication for a too low amplitude.

I don't see how you can predict the rate in any horizontal or vertical position.

The rate(s) in one or two position(s) cannot be out of specification but I see what you mean.

[Poodlopogus]

Originally Posted by saxo3

Your test method (w/o a timegrapher) is time consuming but in principle ok.

-10 s/d in 3U is a strong indication for a too low amplitude.

I don't see how you can predict the rate in any horizontal or vertical position.

The rate(s) in one or two position(s) cannot be out of specification but I see what you mean.

I say "definitely out of spec" because I've typically always rested my watches DU, and DU it's about -1.5 s/d. Therefore, unless DD proved to be gaining 10+ s/d, it would be very hard to average out to +/- 2 s/d (especially since 6U is often a time-losing position as well).

I'm of course talking about Rolex's timekeeping specs (advertised everywhere), not its amplitude specs (not shared by Rolex with the end user).

And I agree, it's time consuming, but I'm not so obsessed to the point of getting a timegrapher, so I'm instead merely testing the end results. Fortunately, I do have a couple other watches to wear during ;).

[CharlesN]

After almost 4 years of collecting data, producing graphs, collating information and many useful discussions about the 32xx movements I would like to suggest that this thread becomes a “Sticky” in this .. "The Rolex WatchTech" forum.

Any positive feedback from the TRF management would be most welcome.

Here's Hoping ...

[CaptT]

Originally Posted by CharlesN

After almost 4 years of collecting data, producing graphs, collating information and many useful discussions about the 32xx movements I would like to suggest that this thread becomes a “Sticky” in this .. "The Rolex WatchTech" forum.

Any positive feedback from the TRF management would be most welcome.

Here's Hoping ...

Makes sense to me - this is by a large margin the most engaging thread in this WatchTech section (and easily be top 10 w/in the General section). For the time being - while 32xx are being produced and scrutinized - it would seem reasonable to make this a sticky.

I particularly appreciate all the constructive work folks are putting in to compile and analyze data around these movements' performance, and the resulting dialogue about possible solutions/changes/enhancements to the 32xx movements in light of their widespread performance issues.