32xx movement problem poll and data thread

[EEpro]

Originally Posted by peterskinner

Worth noting that this thread involves a relatively narrow group, who tend to assume they are representative. But are they? How many 32xx movements have been produced? And how many have been faulty?
I don’t think we have a clue. Self-selecting anecdote on a forum can’t be relied upon, regardless of how sincere.....

Yes, but the same group of people on the same forum don't have a thread about the 3135 or the omega 8800. Just another anecdote, of course.

[saxo3]

Originally Posted by peterskinner

Worth noting that this thread involves a relatively narrow group, who tend to assume they are representative. But are they? How many 32xx movements have been produced? And how many have been faulty?
I don’t think we have a clue. Self-selecting anecdote on a forum can’t be relied upon, regardless of how sincere.....

Since 24 January already 12 posts in this thread w/o any constructive feedback. Go away, voluntarily...

[HiBoost]

Originally Posted by Gab27

Generally speaking, as a snapshot to assess movement health, at what single point in time is measuring amplitude going to be most important? Fully wound? 24 hours out? Or is trend data so important that a single point-in-time measurement (or a series of measurements at the same point-in-time) could potentially miss 'the big picture'?

In short, more data = better. But of course too much data can be overwhelming and takes some level of understanding to interpret. If, hypothetically, I could only get "one" snapshot of data, at a bare minimum I would want dial up 24 hours after full wind and crown down 24 hours after full wind (seconds/day and amplitude for each of those). If those numbers are looking good, you are probably in good shape across the board (at least regarding this specific 32xx issue, obviously this doesn't cover every possible malady a watch movement could face). Fully winding a watch can provide enough power to cover-up an issue and therefore I don't think full wind numbers alone are ideal (though they would still likely have lower amplitude the timekeeping could be correct). At the same time, 2 or 3 days after full wind is less interesting since that likely isn't representative to any actively-being-worn scenarios. Your idea of throwing it on the machine to test an "as worn" condition is a good one as well. If you had taken amplitude data, say, every 8 hours over a 48 hour period from full wind, then you could compare your "as worn" numbers to get an idea of "how wound" it actually is. Essentially you could estimate or infer a power reserve measurement on a watch lacking that complication. I think it is a bad idea to assume any automatic watch will stay at full wind as long as it is worn every day. We all have very different movement patterns. So that is where the merit of this test would come from IMO.

Originally Posted by saxo3

Message to Aeroglyph, Glauco, RomiRolex et al. from the Italian Forum, quoted by Vittorio in post #758.

You are very welcome to share your experiences and the recently (RomiRolex today) announced new data here.

I volunteer to integrate your data into my small data collection and produce (hopefully) some meaningful and comparable graphs for the community.

Your choice!

You have already been invited to the shared google spreadsheet and we of course welcome your creating of graphs from the data stored there. It seems unnecessary for users to cross post collected data on both forums, but the shared spreadsheet allows both sites to produce whatever visualizations/summaries they see fit.

[saxo3]

Originally Posted by HiBoost

You have already been invited to the shared google spreadsheet and we of course welcome your creating of graphs from the data stored there. It seems unnecessary for users to cross post collected data on both forums, but the shared spreadsheet allows both sites to produce whatever visualizations/summaries they see fit.

Indeed, I had a quick look today; à propos "cross post collected data on both forums", why one finds data from HiBoost, Andad, WastedYears, which have already been presented, discussed, and plotted here? Apparently, they start to take some very first data there, which is fine.
I suggest you follow up there. In parallel, maybe you can produce more data from your watch(es).

[HiBoost]

Originally Posted by saxo3

Indeed, I had a quick look today, à propos "cross post collected data on both forums", why one finds data from HiBoost, Andad, WastedYears, which have already been presented, discussed, and plotted here? Apparently, they start to take some very first data there, which is fine.

Yes, we are starting by adding the existing data and will continue to grow it over time. I didn't want to be presumptuous and add your personal data since you were invited to collaborate as well.

Sent from my SM-G960U1 using Tapatalk

[saxo3]

Originally Posted by HiBoost

Yes, we are starting by adding the existing data and will continue to grow it over time. I didn't want to be presumptuous and add your personal data since you were invited to collaborate as well.

[Toivonen]

Originally Posted by HiBoost

In short, more data = better. But of course too much data can be overwhelming and takes some level of understanding to interpret. If, hypothetically, I could only get "one" snapshot of data, at a bare minimum I would want dial up 24 hours after full wind and crown down 24 hours after full wind (seconds/day and amplitude for each of those). If those numbers are looking good, you are probably in good shape across the board (at least regarding this specific 32xx issue, obviously this doesn't cover every possible malady a watch movement could face). Fully winding a watch can provide enough power to cover-up an issue and therefore I don't think full wind numbers alone are ideal (though they would still likely have lower amplitude the timekeeping could be correct). At the same time, 2 or 3 days after full wind is less interesting since that likely isn't representative to any actively-being-worn scenarios. Your idea of throwing it on the machine to test an "as worn" condition is a good one as well. If you had taken amplitude data, say, every 8 hours over a 48 hour period from full wind, then you could compare your "as worn" numbers to get an idea of "how wound" it actually is. Essentially you could estimate or infer a power reserve measurement on a watch lacking that complication. I think it is a bad idea to assume any automatic watch will stay at full wind as long as it is worn every day. We all have very different movement patterns. So that is where the merit of this test would come from IMO.





You have already been invited to the shared google spreadsheet and we of course welcome your creating of graphs from the data stored there. It seems unnecessary for users to cross post collected data on both forums, but the shared spreadsheet allows both sites to produce whatever visualizations/summaries they see fit.

[Gab27]

Originally Posted by saxo3

I always start with full movement winding. You already have a very first idea about the health of your caliber. There are several graphs in this thread, which illustrate the spread of results for different watches with the same 3235 or 3285 movements after full winding.

A characteristic footprint for a good movement is a rather slow decrease of amplitudes over time, nice examples for a 3130 (Andad) and 3187 (saxo3) are in post #414.

Such "amplitude vs. time" measurements are often done in one position only, dial up (DU), which is the most 'comfortable' caliber position (also dial down, DD) with lowest friction.

A coarse picture you obtain measuring only at t=0 and t=24 h. After 24 h the amplitudes must be above 200 degrees, which is the Rolex specification. I would even say above 230 degrees.

Yes, for sure. The most natural way is wearing your watch and wind it with your wrist. You first fully wind it and set the time, then do a reference measurement and simply wear it afterwards. From time to time you measure 5-position-averaged rates and amplitudes. I have done that often in the past, also for watches with other 31xx and 15xx calibers.

The 3130 is a very good movement and the -1.9 sec/day would not worry me too much. Important is that the time loss is consistent, i.e. you always loose more or less -2 sec/day. This can be regulated by a watchmaker. I prefer that my watches run in the positive range.

Yes, probably.

Yes, certainly. It also includes the time where your watch remains at rest, e.g. overnight, that reflects the reality for most.

Understandable.

No, you are very welcome.

To answer all your valid questions with real data, you can have a look at the graph below, which displays several things for you.

This SD43 keeps a high amplitude (in DU position) over a long time, after 60 hours it’s still above 200 degrees. During the same test, I also measured the accuracy (or timekeeping) with a timegrapher-independent app (WatchTracker). The blue curve (no fit, only a guide to the eye) shows that the 3235 caliber is running consistently with about +5 sec/day, which gets smaller towards the end of the power reserve. This +5 sec/day accuracy one can regulate to become less. At present I don't care and prefer to observe how it changes over time.

Thank you so much for answering all my questions!

I have some measurements to take and to observe. It's always exciting to learn more about watches and the technology used with them!

[Gab27]

Originally Posted by HiBoost

In short, more data = better. But of course too much data can be overwhelming and takes some level of understanding to interpret. If, hypothetically, I could only get "one" snapshot of data, at a bare minimum I would want dial up 24 hours after full wind and crown down 24 hours after full wind (seconds/day and amplitude for each of those). If those numbers are looking good, you are probably in good shape across the board (at least regarding this specific 32xx issue, obviously this doesn't cover every possible malady a watch movement could face). Fully winding a watch can provide enough power to cover-up an issue and therefore I don't think full wind numbers alone are ideal (though they would still likely have lower amplitude the timekeeping could be correct). At the same time, 2 or 3 days after full wind is less interesting since that likely isn't representative to any actively-being-worn scenarios. Your idea of throwing it on the machine to test an "as worn" condition is a good one as well. If you had taken amplitude data, say, every 8 hours over a 48 hour period from full wind, then you could compare your "as worn" numbers to get an idea of "how wound" it actually is. Essentially you could estimate or infer a power reserve measurement on a watch lacking that complication. I think it is a bad idea to assume any automatic watch will stay at full wind as long as it is worn every day. We all have very different movement patterns. So that is where the merit of this test would come from IMO.

Thank you very much!

I think I am going to try starting at the full wind and then do every 8 hours or so until the numbers stabilize.

[saxo3]

Originally Posted by Gab27

Thank you very much!

I think I am going to try starting at the full wind and then do every 8 hours or so until the numbers stabilize.

The amplitudes will decrease with time while the rates will (hopefully) remain constant and become negative when approaching the end of the power reserve.
The 3130 caliber has a lift angle of 52 degrees.

[Annan]

I came across this recently and it seemed an appropriate observation for this thread.

[alphadweller]

Originally Posted by Annan

I came across this recently and it seemed an appropriate observation for this thread.

Brilliant

[aayates]

Originally Posted by Annan

I came across this recently and it seemed an appropriate observation for this thread.

of the aisle.

[Driver8]

Just thought I'd bump this up again as it's been one of the most interesting threads in terms of information, and it'll be good to see if there are any new numbers from anyone - good or bad.

[Liga]

Peter Skinner Worth noting that this thread involves a relatively narrow group, who tend to assume they are representative. But are they? How many 32xx movements have been produced? And how many have been faulty?
I don’t think we have a clue. Self-selecting anecdote on a forum can’t be relied upon, regardless of how sincere.....

Originally Posted by EEpro

Yes, but the same group of people on the same forum don't have a thread about the 3135 or the omega 8800. Just another anecdote, of course.

The amount of speculation on this thread is astounding.
All these stats are leading somewhere. Surely.

[JRell]

Yes! The raining on our parade thread is back! Lol

[jssans]

Originally Posted by JRell

Yes! The raining on our parade thread is back! Lol

It’s gonna take a lot more then faulty watches to bring down Rolex. They have transcended usefulness. Rolex is now a status symbol. All hail the Daytona which nobody knows how to use.

[HiBoost]

Originally Posted by Liga

The amount of speculation on this thread is astounding.

All these stats are leading somewhere. Surely.

Can you be more specific as to what qualifies as astounding levels of speculation? We have pictures and testimony from an RSC watchmaker identifying a problem. We have users reporting sudden, significant time loss. We have users reporting problems recurring after repair. We have users recording spd and amplitude for both good and bad specimens. I'm not saying this thread is going to be in the running for the coveted Pulitzer prize in technical horological reporting, but surely the internet is overflowing with nonsense threads more deserving of your contempt.


Sent from my SM-G960U1 using Tapatalk

[EEpro]

Originally Posted by Liga

Peter Skinner Worth noting that this thread involves a relatively narrow group, who tend to assume they are representative. But are they? How many 32xx movements have been produced? And how many have been faulty?
I don’t think we have a clue. Self-selecting anecdote on a forum can’t be relied upon, regardless of how sincere.....






The amount of speculation on this thread is astounding.
All these stats are leading somewhere. Surely.

It will feel like more than speculation when it hits your $13,000 Datejust. I'm not speculating. It happened. Fractions of a second timekeeping for 6+ months then bam, -30 spd. It's crap and defending it is crap. Trying to keep people from talking about it is crap.

[tgoose1]

Sample size in a forum, where folks actually speak up, compared to all Rolex 32xx movements in use is extremely tiny. However, there is a high percentage of reported 32xx similar problems which should be seriously considered as fact. I've worn Rolex Subs almost 50years, my latest for 10yrs without issues as noted from a significant number of 32xx owners. I've considered purchasing a new gold sub recently, but the last thing I desire to do is send a $40k watch back and forth to an RSC multiple times for regulation problems like noted in this thread. I actually wear my Rolex and expect it to perform well all the time as designed. Unfortunately, it will be years before the 32xx has a lasting/trustworthy track-record of precision, don't know if I'll ever purchase a 32xx gold sub. We'll see because I'm running out of time.

[francis]

Originally Posted by EEpro

It will feel like more than speculation when it hits your $13,000 Datejust. I'm not speculating. It happened. Fractions of a second timekeeping for 6+ months then bam, -30 spd. It's crap and defending it is crap. Trying to keep people from talking about it is crap.

Here here

Happened to my datejust as well. ;(


Sent from my iPhone using Tapatalk

[Dirt]

Originally Posted by tgoose1

Sample size in a forum, where folks actually speak up, compared to all Rolex 32xx movements in use is extremely tiny. However, there is a high percentage of reported 32xx similar problems which should be seriously considered as fact. I've worn Rolex Subs almost 50years, my latest for 10yrs without issues as noted from a significant number of 32xx owners. I've considered purchasing a new gold sub recently, but the last thing I desire to do is send a $40k watch back and forth to an RSC multiple times for regulation problems like noted in this thread. I actually wear my Rolex and expect it to perform well all the time as designed. Unfortunately, it will be years before the 32xx has a lasting/trustworthy track-record of precision, don't know if I'll ever purchase a 32xx gold sub. We'll see because I'm running out of time.

There's plenty of great vintage stuff around sans a 32xx movement.
It's always good to return to tried and tested gear that's well proven

[Dirt]

Originally Posted by jssans

It’s gonna take a lot more then faulty watches to bring down Rolex. They have transcended usefulness. Rolex is now a status symbol. All hail the Daytona which nobody knows how to use.

Lol

[Dirt]

Originally Posted by JRell

Yes! The raining on our parade thread is back! Lol

I don't know what you mean by "our parade".
Everyone who has provided data and others has one of the flagship movements.
You know the ones that are usually out the front waving the latest fanboy flags more energetically than the others who are making do with the old gear.

From my perspective, each and every one of them is fully entitled to discuss the issues they're having with the stitching fraying at the edges of the flags just after the parade has started and the splinters in their hands from the poles
Whilst the ones down the back of the parade have never had a problem for decades as far as they know

[peterskinner]

Originally Posted by saxo3

Since 24 January already 12 posts in this thread w/o any constructive feedback. Go away, voluntarily...

Doubt about some of the issues is a perfectly reasonable stance, and criticising the methodology is a form of feedback....so why do you feel you have the right to instruct members with a different perspective to ‘go away’? These debates can’t be controlled by one self-interested group.
Incidentally, I’ve owned several models with this movement, one needed retiming, which I mentioned, the others were and are , fine. But then my view , no matter how evidence based, doesn’t count....because it doesn’t suit you.
For this thread to really work, it needs the widest possible range of views, featuring every perspective.

[PekWatchGuy]

True...so true.....

Originally Posted by Annan

I came across this recently and it seemed an appropriate observation for this thread.

[saxo3]

Originally Posted by Andad

It has been stated in other threads that to wind a 48 hour PR model requires about 40 revolutions of the crown and to wind a 72 hour PR movement requires >70 full revolutions. When I checked my DJ41 from zero PR to an obvious slippage of the mainspring in the barrel it took exactly 40 revolutions.

Originally Posted by Lol-x

My 32XX movement watch just took 77 winding turns with 2 fingers to be fully wound from zero reserve. So no longer can you say 30 turns will give you a fully wound watch.

Originally Posted by HiBoost

I had seen other report 40ish winds used to do it, and now it's in the 70s by a few accounts. So they've almost got it 1 wind per hour of power reserve which is kind of neat. On mine I have not been able to determine when it is fully wound. There is no difference in feel and even holding it up to my ear I don't hear a different sound at any point. I've gone past 80 winds and it just stays the same... was yours pretty obvious or was it a really subtle difference beyond 77?

Some new crown winding results for 32xx movements may be interesting.
I did some systematic measurements for 4 watches, see next post.

[saxo3]

I measured movement rates, amplitudes, and beat errors in two different watch positions (dial up, six up) starting from watch at rest (zero winding) until full winding and beyond, in single steps of 5 crown-winding turns (360 degrees each). I made the following observations.

Common features observed for both calibers:

(1) a steep rise in amplitudes (from 0 degrees) is measured only 1-2 turns after the second hand started moving, i.e. exactly after 20 turns.

(2) The highest amplitudes are measured after 35 turns, i.e. a bit before the spring slip starts.

(3) For both watches the spring slip starts at 38-39 turns.

(4) The measured amplitudes slightly decrease after 40 turns and remain constant until the end of the tests at 70 turns.

(4) For my 32xx watches, there is no gain to wind more than 40 full turns.




Comparison with a GMT-Master II CHNR (caliber 3285) and a Submariner 14060M (caliber 3130)
The 14060M is from 2001, no movement service since May 2013.

[WILLIWALKER]

YM1 loses 11sec / day :/

I will leaver her soon

[Dirt]

Originally Posted by saxo3

I measured movement rates, amplitudes, and beat errors in two different watch positions (dial up, six up) starting from watch at rest (zero winding) until full winding and beyond, in single steps of 5 crown-winding turns (360 degrees each). I made the following observations.

Common features observed for both calibers:

(1) a steep rise in amplitudes (from 0 degrees) is measured only 1-2 turns after the second hand started moving, i.e. exactly after 20 turns.

(2) The highest amplitudes are measured after 35 turns, i.e. a bit before the spring slip starts.

(3) For both watches the spring slip starts at 38-39 turns.

(4) The measured amplitudes slightly decrease after 40 turns and remain constant until the end of the tests at 70 turns.

(4) For my 32xx watches, there is no gain to wind more than 40 full turns.




Comparison with a GMT-Master II CHNR (caliber 3285) and a Submariner 14060M (caliber 3130)
The 14060M is from 2001, no movement service since May 2013.

Solid Gold
That conclusively puts to bed another myth and bolsters information that's been shared by another person who is in the know