32xx movement problem poll and data thread

[whatsthetime?]

Originally Posted by saxo3

The introduction dates of all 32xx movements:

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

Thank you.

[the dark knight]

Probably false alarm/hope, but I noticed something "strange" with my very new (took delivery 2 weeks ago) 126610. It just stopped 2 hours ago, last wore it this morning. But I didn't notice the huge loss in time I've noticed with every other 32xx watch in the roughly last 24 hours of the PR. This morning it was still -6s TOTAL for the roughly 2 weeks I've owned it.

Being such a new watch I haven't thought of putting it on the timegrapher but I may do so just to check it out.

[saxo3]

Do a full winding (40 full crown turns), measure in 5 positions (amplitudes, rates, beat errors), and post the results.

The data will be very interesting for a new watch (08/2023).

[the dark knight]

DU : +1s/d, 270, 0.0ms
DD : 0s/d, 267, 0.2ms
3U : 0s/d, 237, 0.0ms
6U : +1s/d, 232, 0.4ms
9U : +1s/d, 238, 0.4ms

[saxo3]

Looks very good to me.
The 0.4 ms beat errors are a bit high but within specs (max 0.5 ms).

[rbndylan]

Hello !

A little comparison with the measurements in May, rather strange, what do you think ?

12 Mai : FULL
DU : +0.7: 269 : 0ms
6H : -3.7: 211: 0ms
9H : -5.1: 215: 0.1ms
3H : -2.2: 219: 0.2ms
DD : +0.9 : 258 : 0ms
Moyenne : -1.88 / 234.4 / 0.06

28 Août : FULL
DU : -0.4 : 281 : 0ms
6H : -0.2 : 238 : 0ms
9H : -4.8 : 230 : 0.1ms
3H : -3.1 : 224 : 0.2ms
DD : +0.9 : 253 : 0ms
Moyenne : -1.52 / 245.2 / 0.06

[saxo3]

Originally Posted by rbndylan

…what do you think ?

I would take more data along the power reserve, as you did in May 2023 (test 2, 4099)

[rbndylan]

Hi Saxo3, are you okay? Yes, I would do it in a few weeks but there is a nice difference between the values. I wonder if I was completely wrapped or not. Moreover, when I go up via the crown, I usually stop when I hear a small noise (gravel), it seems to me it's when the spring slides above, right? Or do you have to continue anyway? Or else it doesn't correspond to that because usually I don't need 40 towers of crowns

[rbndylan]

Another question, how can a watch change its measurement like that? I heard say for example over 5/6/7 years, a 31xx that was constant level measurement. I agree that it will change on the wrist but that the timegrapher? Do you have an explanation?

[saxo3]

Originally Posted by rbndylan

Hi Saxo3, are you okay? Yes, I would do it in a few weeks but there is a nice difference between the values. I wonder if I was completely wrapped or not.

That can have several reasons: (a) the caliber has changed a bit during the past 3 months (b) you have not used the same measurement procedure (c) you do not give errors for the individual numbers (d) the watch was not fully wound etc.

Therefore, I do not know the origin of the reported changes.

Originally Posted by rbndylan

Moreover, when I go up via the crown, I usually stop when I hear a small noise (gravel), it seems to me it's when the spring slides above, right? Or do you have to continue anyway? Or else it doesn't correspond to that because usually I don't need 40 towers of crowns

The "small noise" has been reported many times for 32xx movements, it does not harm, it's the friction of the mainspring inside its barrel; some calibers show it more pronounced than others; you can continue winding. For a full 32xx wind you need to perform 35-40 complete crown turns, 360 degrees each turn, I have measured that.

Originally Posted by rbndylan

Another question, how can a watch change its measurement like that?

I can not answer this for your watch and your setup.

Originally Posted by rbndylan

I heard say for example over 5/6/7 years, a 31xx that was constant level measurement.

My 31xx are much more reliable than my 32xx watches, also the long-term timegrapher reproducibility.

Originally Posted by rbndylan

I agree that it will change on the wrist but that the timegrapher? Do you have an explanation?

I don't know.

Below is my comparison and how I see your numbers. If your timegrapher data are ok (I do not doubt them), then your watch seems to be better now in its vertical positions, after full winding at least.

[rbndylan]

Thank you Saxo. Regarding the noise, when we start to hear it, it is mandatory that the main spring slides and therefore it is completely reassembled?

[saxo3]

Originally Posted by rbndylan

Thank you Saxo. Regarding the noise, when we start to hear it, it is mandatory that the main spring slides and therefore it is completely reassembled?

Probably yes, to be sure try to count exactly the turns to achieve full winding.

I have a few questions about your timegrapher measurements:

- After full winding, how long do you wait (stabilization time) before you start with the first measurement?
- What is your measurement duration in each position?
- How long do you wait after each position change before you start with the next measurement?
- How much variation do you see in the rates and amplitudes during each individual measurement?

[rbndylan]

I do this : It’s the same that the measures of May

Wind your watch fully
Switch on timegrapher
Set 53, 12s, 99.9
Leave watch paused 15min in DU
Position DU -> 6h -> 9h -> 3h -> DD
Measure 2min DU
Take values
Change to 6h, leaving measurement running for 2min
Restart measurement for 2min and take values
same thing then for 9h, 3h and DD

[saxo3]

That looks all good. For my better understanding:

- Your Weishi 1900 measurement periods are selectable: 2 s, 4 s, 8 s, 12 s, 20 s, 30 s, 60 s.

- You took 12 s and measure 2 min, that means you have 10 measurement points, correct?

- If you would note all these 10 points, which are each averaged over a 12 s period, what is the variation (or scatter between) these 10 data points? This gives you an idea about the precision.

[rbndylan]

We talked about this back in May. I don't really understand what this corresponds to, so I left the default at 12s.

[rbndylan]

I've just rested my watch overnight for two days in DU and the other night in DD. Here's the result:
Test: DD = 0.7 in 10.8h
Test: DU = 0.2 in 10.6h

Whereas before, DU saved more time at night than DD. This corresponds to what I saw on the timegrapher. One question, how could my watch change its value in a few weeks/months?

[saxo3]

I would repeat the timegrapher measurements:

(a) Full winding, wait 15 min.

(b) Measure rate and amplitude in 5 positions as before: 12 s, 2 min in each position, 2 min wait after each position change.

(c) Repeat (b) for 2-3 times.

(d) Post all data.

This gives you an idea how reproducible your timegrapher data are.

[rbndylan]

The timegrapher gives the same thing, and it matches what I've observed with the DU and DD over the last two days. I don't understand why the movement has "changed" its behavior and hasn't stayed like it was a few months ago.

[saxo3]

ok, then I can't help you any further

[Easy E]

Originally Posted by rbndylan

The timegrapher gives the same thing, and it matches what I've observed with the DU and DD over the last two days. I don't understand why the movement has "changed" its behavior and hasn't stayed like it was a few months ago.

Your numbers are not that far off from each other. Given mechanical nature of the movement there will be slight variations in the data across a time span - whatever that may be. If you watch your meter during the readings you will see slight fluctuations in values. Your final reported numbers are only a snapshot of the particular moment your recorded the values off the meter.

Again, overall you don't really seem to have that dramatic of a variance from May to current - not enough to worry about anyway. If you want to get that granular and that specific your data collection has to also be super granular.

[saxo3]

Originally Posted by Easy E

Your numbers are not that far off from each other. Given mechanical nature of the movement there will be slight variations in the data across a time span - whatever that may be. If you watch your meter during the readings you will see slight fluctuations in values. Your final reported numbers are only a snapshot of the particular moment your recorded the values off the meter.

Again, overall you don't really seem to have that dramatic of a variance from May to current - not enough to worry about anyway. If you want to get that granular and that specific your data collection has to also be super granular.

I had the same discussion with CharlesN in August 2021.

When he repeated 10 times his timegrapher measuremts (rates, amplitudes) in the same position, he got different results and we were able to determine measurement errors.

Look at post 1939 and the achieved rate results in DU, his numbers varied from 0.7 s/d to 1.8 s/d.

[rbndylan]

can a watch take a long time to stabilize?

[77T]

Originally Posted by rbndylan

can a watch take a long time to stabilize?

Stabilize the state of accuracy?
Well, that depends upon many moving parts.

These data from timegraphers show variances in different positions at room temperature without any external factors like arm movement etc.

Since a watch is always in some state of decay from our notion of time (artificial vs astronomical) - I’d say they never “stabilize”

But the break-in period for a movement is usually complete within a year.

Sent from my iPhone using Tapatalk Pro

[padi56]

Originally Posted by rbndylan

can a watch take a long time to stabilize?

Take my advice get rid of your timegrapher machine,then perhaps you will then start to enjoy wearing your Rolex watch for what it is a mechanical wrist watch marvel.

[GradeV]

Originally Posted by padi56

Take my advice get rid of your timegrapher machine,then perhaps you will then start to enjoy wearing your Rolex watch for what it is a mechanical wrist watch marvel.

For me, part of the enjoyment of wearing a mechanical wrist watch marvel is the knowledge and assurance that the movement will chug along reliably and accurately for years and years.

Those words right there on the dial - superlative chronometer - play a big role in that.

That Rolex can provide all of the above in a robust and stylish package is the icing on the cake.

[padi56]

The words Superlative Chronometer been on Rolex dials for the past 50 odd years.All it means in the real world to bare the word chronometer on the dial it must be tested at the Swiss COSC and past a test to a AVERAGE of -4+6 seconds a day..

[Dirt]

Originally Posted by padi56

The words Superlative Chronometer been on Rolex dials for the past 50 odd years.All it means in the real world to bare the word chronometer on the dial it must be tested at the Swiss COSC and past a test to a AVERAGE of -4+6 seconds a day..

As we all know, Rolex have been the only manufacturer to put that on the dial and it has only ever been a marketing strategy on their part.
Now since the overnight change of the Chronometer hangtag at their directive to the Superlative Chronometer hangtag. Rolex do a preliminary COSC test, which is limited to the movement just as it always has been.
However the new hangtag signifies the watch is now classified by Rolex that it is indeed a Superlative Chronometer as defined by Rolex, which undergoes additional testing of the assembled watch itself and critically has the movement fitted into the case. It surpasses the COSC testing of the movement.
The exclusive Rolex Superlative Chronometer certification standard is held to the much touted + or -2 seconds per day.
That is the new standard which Rolex has set for themselves in their finished product.
It is documented and that is the standard they work to at the RSC even after the watch has left the factory. Without exception.

[Easy E]

Originally Posted by Dirt

As we all know, Rolex have been the only manufacturer to put that on the dial and it has only ever been a marketing strategy on their part.
Now since the overnight change of the Chronometer hangtag at their directive to the Superlative Chronometer hangtag. Rolex do a preliminary COSC test, which is limited to the movement just as it always has been.
However the new hangtag signifies the watch is now classified by Rolex that it is indeed a Superlative Chronometer as defined by Rolex, which undergoes additional testing of the assembled watch itself and critically has the movement fitted into the case. It surpasses the COSC testing of the movement.
The exclusive Rolex Superlative Chronometer certification standard is held to the much touted + or -2 seconds per day.
That is the new standard which Rolex has set for themselves in their finished product.
It is documented and that is the standard they work to at the RSC even after the watch has left the factory. Without exception.

So then why is it so wrong to expect the -/+2 advertised performance from a $10k or greater piece?

[Dirt]

Originally Posted by Easy E

So then why is it so wrong to expect the -/+2 advertised performance from a $10k or greater piece?

To be completely honest, i don't know
As you state given that price point and the fact that it's promoted, i think it's fair and reasonable to have high expectations.
After all Rolex won't back away from it will they?

On reflection I'm not necessarily sure that anyone is stating categorically that it's wrong. Just that it's perhaps being unrealistic or pernickety or something or other

I can say that I have had Rolex watches for a good many years that have run to the Superlative standard day in day out.
Until this new movement came out there was no possibility that one should expect anything different going forward.
Sadly it's on record that Rolex basically needs to go back and do more work

I'm set, as every COSC Rolex I've had with the 31xx movement has met or surpassed the new "Superlative" accuracy standard for the last 25+ years and they continue to do so without fail with a bit of routine and timely servicing
It's the sort of stuff that builds brand loyalty

[Dirt]

Originally Posted by 77T

Stabilize the state of accuracy?
Well, that depends upon many moving parts.

These data from timegraphers show variances in different positions at room temperature without any external factors like arm movement etc.

Since a watch is always in some state of decay from our notion of time (artificial vs astronomical) - I’d say they never “stabilize”

But the break-in period for a movement is usually complete within a year.

Sent from my iPhone using Tapatalk Pro

Some say that the notion of a break-in period is a figment of imagination.
I don't know?
The watchmakers of old would refer to it
If it's the stuff of fantasy, why would they speak of it?

Personally I've had 3 new Rolex watches that slowed a little(maybe 1.5 - 2 seconds) within the first 2 weeks, which was consistent with what was reported back in those days on the forums.
In recent times(11 years) it hasn't necessarily been quite so pronounced. Maybe a little more nuanced and much quicker if at all

I think it's hard to say these days.
After all how much break in occurs that would have a noticeable effect to the end user after a watch movement has been tested for about 2 weeks by COSC, then a further test period by Rolex when it's cased up in a completed watch to verify it runs to the Superlative standard.

Like everything mechanical.
After an intial run in period, the bits are actually wearing out to varying degrees