32xx movement problem poll and data thread

[FlyinHawaiian]

Originally Posted by saxo3

Correct.

According to your signature you also own a GMT-Master II 126719BLRO (3285) and an Air-King 126900 (3230)?

Yes I do. I just checked the BLRO at t=0, all within specs. Will check t=12 and t=24 later.

The SD43 was very noticeably losing seconds just after a few hours of wearing.

[saxo3]

Originally Posted by FlyinHawaiian

Yes I do. I just checked the BLRO at t=0, all within specs. Will check t=12 and t=24 later.

Thanks. How old are your BLRO and Air-King?
Do you plan to post your timegrapher results for both watches?

[FlyinHawaiian]

Originally Posted by saxo3

Thanks. How old are your BLRO and Air-King?
Do you plan to post your timegrapher results for both watches?

BLRO - 10/23
Air King - 05/23

[SwissSteph]

Hello,

One question: how can you tell if the watch is fully wound during this manual operation? Is there a particular “noise”?

I'm asking this question so that I'm really ready when I receive my “Weishi 1900” to test my two watches. My Excel file is ready, all I have to do is carry out my “60 hours” of testing

Another question, for the moment my Sea-Dweller has been on a “Swiss Kubik” winder since it arrived at my house. Is this enough to keep it completely wound or will I have to do a manual winding just before my tests?

[saxo3]

It is difficult to hear a tiny sound, but you can feel a kind of resistance, some describe it as a "grindy" feeling. Not every 32xx movement has exactly the same behavior. The winder keeps it wound, but it is unclear how much. Make the 40+ full turns of the crown. You can't overwind the movement.

[SwissSteph]

[capice]

Originally Posted by saxo3

It is difficult to hear a tiny sound, but you can feel a kind of resistance, some describe it as a "grindy" feeling. Not every 32xx movement has exactly the same behavior. The winder keeps it wound, but it is unclear how much. Make the 40+ full turns of the crown. You can't overwind the movement.

Doesn't it depend on when the watch was serviced?

[Easy E]

Originally Posted by capice

Doesn't it depend on when the watch was serviced?

Not necessarily. Another factor to consider, it is harder to hear/feel the slip in say a SD43 vs a DJ, due to case and crystal shape and thickness.

[saxo3]

Something interesting.
 
I have investigated all 7 watches that were bought in 2024, measured, and presented with data in this thread. 
 
The different owners (Bigmcmuffins, EasyE, Digiwatch, Maratka, Penelope2017) all used a Weishi timegrapher.
 
I have looked at their measured 5-position averages for the amplitudes (X-Amplitude) and the rates (X-Rate).
 
For each watch I have taken the posted data after full winding (t = 0 h) and 24 hours (t =24 h) later.
 
The different watches are numbered 1,2,3,….7.
 
Look at the two graphs below.
 
What do you see and what is your explanation? I am curious about your suggestions.

[FlyinHawaiian]

Data for my 126719.

[searas]

Originally Posted by saxo3

Something interesting.
 
I have investigated all 7 watches that were bought in 2024, measured, and presented with data in this thread. 
 
The different owners (Bigmcmuffins, EasyE, Digiwatch, Maratka, Penelope2017) all used a Weishi timegrapher.
 
I have looked at their measured 5-position averages for the amplitudes (X-Amplitude) and the rates (X-Rate).
 
For each watch I have taken the posted data after full winding (t = 0 h) and 24 hours (t =24 h) later.
 
The different watches are numbered 1,2,3,….7.
 
Look at the two graphs below.
 
What do you see and what is your explanation? I am curious about your suggestions.

I don't have any 32xx watch, but as prospective buyer follow this interesting thread closely. It seems that there is a lot of variation in the amplitude numbers (t=0 or t=24) even in watches purchased this 2024. What would be the cause?? Is Rolex using both corrected and uncorrected movements in new watches ? Maybe some of these 7 watches are really older than others even if they were sold this year? Or it is just that some of them have seen more use and are more into the degradation path of the 32xx.

Maybe some experts could chime in.

Regards,
Daniel

[Poodlopogus]

Originally Posted by saxo3

Something interesting.
 
I have investigated all 7 watches that were bought in 2024, measured, and presented with data in this thread. 
 
The different owners (Bigmcmuffins, EasyE, Digiwatch, Maratka, Penelope2017) all used a Weishi timegrapher.
 
I have looked at their measured 5-position averages for the amplitudes (X-Amplitude) and the rates (X-Rate).
 
For each watch I have taken the posted data after full winding (t = 0 h) and 24 hours (t =24 h) later.
 
The different watches are numbered 1,2,3,….7.
 
Look at the two graphs below.
 
What do you see and what is your explanation? I am curious about your suggestions.

Hard to say for sure, since none ever dropped below 200, but what this says to me is that the relationship between low amplitude and large time losses is correlated, not causal. In other words, some third "thing" independently causes the watch to lose time and prevent it from getting above 200 degrees.

I recall reading that Rolex designed this movement to keep good time even at low amplitudes (whether that's what happens in practice or not). So it's possible that the two have never been tied together as closely as was suspected.

[Easy E]

Originally Posted by Poodlopogus

Hard to say for sure, since none ever dropped below 200, but what this says to me is that the relationship between low amplitude and large time losses is correlated, not causal. In other words, some third "thing" independently causes the watch to lose time and prevent it from getting above 200 degrees.

I recall reading that Rolex designed this movement to keep good time even at low amplitudes (whether that's what happens in practice or not). So it's possible that the two have never been tied together as closely as was suspected.

The low amplitude to low spd correlation is basically the summary point of this entire thread.

I can’t discern much else from the chart, seems inconsistent to me. Maybe that is the point being made, idk. Without having a similar data set to say all 201x 31s, or a set of Omegas all from the same production cycle to compare to, I’m going with the movements are just not consistent unit to unit.

[Poodlopogus]

Originally Posted by Easy E

The low amplitude to low spd correlation is basically the summary point of this entire thread.

I can’t discern much else from the chart, seems inconsistent to me. Maybe that is the point being made, idk. Without having a similar data set to say all 201x 31s, or a set of Omegas all from the same production cycle to compare to, I’m going with the movements are just not consistent unit to unit.

Yes, but glancing back through the thread, it seems that over time it morphed into an assumption of a causal relationship. Simply a correlation means that watches could run at low amplitude indefinitely and still remain accurate, unless some third factor influences both outcomes.

[PO.Victory]

I’ll take #3.

[tho68]

My Weishi 1900 arrived today (delayed by one day). Here are the initial measurements at t=0.

DU +1,0 270 0,3
DD +0,1 273 0,0
3U -1,7 233 0,3
6U +0,2 246 0,0
9U 0,0 242 0,0

Aver -0,08 253 0,12

Watch is 126610LV (so 3235) bought in April 24 and in use since last week Friday.

More measurements to follow.

BTW, how do you attach those figures? Via the Attachment function?

Regards
tho68

[saxo3]

Thanks for all your replies.

Originally Posted by searas

… It seems that there is a lot of variation in the amplitude numbers (t=0 or t=24) even in watches purchased this 2024. Is Rolex using both corrected and uncorrected movements in new watches ? Maybe some of these 7 watches are really older than others even if they were sold this year? …

You bring up a few very good points.

Originally Posted by Poodlopogus

Hard to say for sure, since none ever dropped below 200….

Yes, it’s not easy to derive.

Originally Posted by Easy E

…I can’t discern much else from the chart, seems inconsistent to me. Maybe that is the point being made, idk. …. I’m going with the movements are just not consistent unit to unit.

It is not inconsistent as the 7 watches are not the same type…

Originally Posted by PO.Victory

I’ll take #3.

Short answer, good choice.

Here are the graphs (again) that we are discussing

[saxo3]

Here is my explanation.

Let's look at the graph of the averaged rates:

We can see that all 7 watches are within the advertised range of -2/+2 s/d after full winding (t = 0 h), i.e., they are correctly regulated.

Watches #1, #2, #3, #4, #5 show a very small change in rate between t = 0 h and t = 24 h.

Watches #6 and #7 both lose a bit more time in the first 24 hours.

Let's look at the graph of the averaged amplitudes:

For the same 7 watches, we see a big difference between the initial amplitudes (after full winding) at t = 0. They all lose significantly in amplitudes after the first 24 hours. Despite this effect, all the watches remain very accurate within the first 24 hours.

In addition, all the watches have an average 5-position amplitude above 200° C. This keeps the watches accurate. What has been described so far everybody can see from my two graphs.

What you cannot see is that I have sorted the seven watches by 32xx movement type.

Watch #1, #2, #3 have 3230 calibers
Watch #4, #5 have 3235 calibers
Watch #6, #7 have 3285 calibers

These three calibers have different "functionalities":

3230: three hands
3235: three hands + date wheel mechanism
3285: three hands + date wheel mechanism + GMT hand

I think this can explain the amplitudes graph: as the number of mechanical movement parts increases, friction increases and with it the loss of amplitude after full winding and along the power reserve.

We see this for NEW 32xx watches measured by different people with different (but the same type of) timegraphers.

Although I am discussing only a very small number of 7 new watches, which is statistically irrelevant, I see a kind of ranking: 3230 are better than 3235, which are better than 3285 watches.

This observation confirms what I have observed since a long time for my own 3235 and 3285 watches. My 3285s are worse than my 3235. I do not think this is a coincidence.

Below the same graphs, but with additional information on the 32xx movement for each watch.

[S.Explorer]

That's very interesting. Anecdotally, I did feel there were much less reported 3230 issues.

[Poodlopogus]

Originally Posted by S.Explorer

That's very interesting. Anecdotally, I did feel there were much less reported 3230 issues.

Yes, but then the question becomes: Is that because there are fewer problems with the 3230 than 3235, and fewer with the 3235 than the 3285 because fewer parts mean fewer problems, or simply the number of watches out there?

Remember, the 85 came out before the 35, came out before the 30. And the 3230 is in relatively fewer models than the 35 (since other than the TT Explorer 36, the 3230 comes in no PM/TT watches), and also has been around longer.

Finally, the watches most sought after, and therefore most popular with collectors, tend to have the 3285 over the others.

However, the overall theory that the more moving parts, the more chances of problems, makes a ton of sense. The above is more just devil's advocate.