32xx movement problem poll and data thread

[mato123]

Here are numbers for my watch:

Reference: 126600
Caliber: 3235
Warranty from: 12/2021
Tested: 2/2025

[saxo3]

Originally Posted by mato123

.

Thanks, the numbers look good. What timegrapher and lift angle setting did you use? Do you know if this watch has been serviced before you bought it?

[CaptT]

Originally Posted by phranxinatra

Actually, there is leaked information. In this video, David Bredan (abtw) talks about how a friend of his, who worked at the Rolex factory said that for a very long time they couldn't simulate this problem. They couldn't figure out what was causing the problem, so they stained the lubricant to see where it was migrating within the movement.

https://www.youtube.com/watch?v=WepR0X17kwY

Interesting. And not surprising they couldn’t simulate the issue.

Based on the number of guests that view this thread (especially relative to other threads) it would seem *someone* finds the data collection and dialogue here interesting.

From yesterday:

[mato123]

Originally Posted by saxo3

Thanks, the numbers look good. What timegrapher and lift angle setting did you use? Do you know if this watch has been serviced before you bought it?

I bought generic "Multifunction Time Grapher 1000" of Ebay. To be honest, the cheapest one I found. Lift angle = 53 degrees. I asked about the service history and I was told it wasn't serviced.

I asked if they heard about 32xx issue while trying couple of watches in their office. They had Explorer II recently that ran fine when they purchased it but was slow few week later when they wanted to sell it, so it went to service.

[mato123]

Forgot to mention that the watch was in mint condition, clearly previous owner didn't wear it daily. I wonder if that is related to "good" numbers from the time grapher. I'll see how it holds up after few months of daily wear.

[saxo3]

Originally Posted by CaptT

Interesting…

Not for me as I don't understand Hungarian.
You screenshot from yesterday is also unclear.

[saxo3]

Originally Posted by mato123

…I wonder if that is related to "good" numbers from the time grapher...

No.

[CaptT]

The “someone” I was referring to is the 23 guests:



I find that interesting.

And of course there are a number of members that follow and contribute here that also take an interest in this thread and the information exchange.

[Amy_Moses]

After nearly three years on the wait list for a 124270 Explorer I gave up and bought one from Watchfinder London in Feb 24 with Aug 23 on the card. It looked unworn. I wore it 24/7 and it ran a ~0.7 secs per day slow.

After about 6 months it started losing time almost in step changes - I tried to post a chart but can't. I haven't got a timegrapher. Timings were done by setting it against a GPS clock and then tracking the indicated vs actual time in a spreadsheet and calculating the secs/day loss.

I didn't knock or drop it, I had no lifestyle changes, I didn't stop or start any sports or other activities, I don't do manual work and I didn't change the measurement method. After the first step change I occasionally gave it a full wind to see if there might be a power reserve problem, but it made no difference.

It was losing 15 secs/day when I took it to RSC London, St James' Sq. in Feb 25 - almost a year to the day after I bought it. Their email said they would carry out a movement service under the guarantee. It was with them for 14 days and I've had it back a few days and it's gaining about 0.5 sec per day.

This was meant to be my retirement and last-watch-purchase-ever-honest-no-I-really-mean-it-this-time watch but now there'll always be that nag in the back of my mind that I've got a lemon and it will start losing time again - probably just after the warranty runs out. I'm contemplating flipping it for a 114270 or maybe the black lacquer Aqua Terra.

Also, when I was doing my pre-buying research the warranty card seemed to be an absolute necessary pre-requisite yet no one asked for the card at any stage in this repair. Neither was I given a service card or any paperwork when my watch was returned. If it hadn't been the London RSC I'd have been worried by this - is this normal?

[phranxinatra]

Originally Posted by saxo3

Not for me as I don't understand Hungarian.
You screenshot from yesterday is also unclear.

I have written the gist of it. Basically the oil migrating to wrong places caused the low angle issue. And Rolex have fixed this issue.

[Easy E]

Originally Posted by phranxinatra

I have written the gist of it. Basically the oil migrating to wrong places caused the low angle issue. And Rolex have fixed this issue.

Except they haven't

[saxo3]

Originally Posted by phranxinatra

I have written the gist of it … And Rolex have fixed this issue.

Nonsense.

Originally Posted by phranxinatra

2017 SD43 started to lose time after a year. Until that it kept perfect time. I took it back to the boutique but they only regulated not serviced. Took it back again 2 weeks later, regulating again. Sold it and bought a new MK2 last August, it's pretty accurate but only 6 month old.

Do you still own this 126600 Sea-Dweller (3235) form 08/2020?
If so, how does it perform after 4.5 years?
What are the amplitudes after full winding and 24 hours later?

Answer with facts (data) and not with a video that the vast majority here cannot understand (the language).

[saxo3]

Originally Posted by Amy_Moses

... Neither was I given a service card or any paperwork when my watch was returned. If it hadn't been the London RSC I'd have been worried by this - is this normal?

No.
Buy a timegrapher and measure the amplitudes after this RSC service, it will serve you later on as a reference … and welcome!

[mato123]

Originally Posted by Amy_Moses

After nearly three years on the wait list for a 124270 Explorer I gave up and bought one from Watchfinder London in Feb 24 with Aug 23 on the card. It looked unworn. I wore it 24/7 and it ran a ~0.7 secs per day slow.

After about 6 months it started losing time almost in step changes - I tried to post a chart but can't. I haven't got a timegrapher. Timings were done by setting it against a GPS clock and then tracking the indicated vs actual time in a spreadsheet and calculating the secs/day loss.

I didn't knock or drop it, I had no lifestyle changes, I didn't stop or start any sports or other activities, I don't do manual work and I didn't change the measurement method. After the first step change I occasionally gave it a full wind to see if there might be a power reserve problem, but it made no difference.

It was losing 15 secs/day when I took it to RSC London, St James' Sq. in Feb 25 - almost a year to the day after I bought it. Their email said they would carry out a movement service under the guarantee. It was with them for 14 days and I've had it back a few days and it's gaining about 0.5 sec per day.

This was meant to be my retirement and last-watch-purchase-ever-honest-no-I-really-mean-it-this-time watch but now there'll always be that nag in the back of my mind that I've got a lemon and it will start losing time again - probably just after the warranty runs out. I'm contemplating flipping it for a 114270 or maybe the black lacquer Aqua Terra.

Also, when I was doing my pre-buying research the warranty card seemed to be an absolute necessary pre-requisite yet no one asked for the card at any stage in this repair. Neither was I given a service card or any paperwork when my watch was returned. If it hadn't been the London RSC I'd have been worried by this - is this normal?

Similar situation here. I wanted a new watch with a 5-year warranty from an AD because of these issues but got tired of waiting. Then, I found a pre-owned one in my city from a reputable grey market dealer for €2,500 under retail. My thinking was, I could set aside the savings for a service if needed. I’m happy to give Rolex a chance to get it right, but if this becomes a recurring problem, I'll sell it right after the second service and buy something with 31xx movement or a different brand. It's sad that one has to think like this about a new watch. It really affects your enjoyment.

[Amy_Moses]

Originally Posted by saxo3

No.
Buy a timegrapher and measure the amplitudes after this RSC service, it will serve you later on as a reference … and welcome!

Thanks, Saxo3.

I recall reading that that timgraphers have a hard time with the 3230 movement because of the Chronergy escapment. Is that correct?

I'm in the UK and can get a Weishi 1000 for £175. Any good?

[Easy E]

Originally Posted by Amy_Moses

Thanks, Saxo3.

I recall reading that that timgraphers have a hard time with the 3230 movement because of the Chronergy escapment. Is that correct?

I'm in the UK and can get a Weishi 1000 for £175. Any good?

Reading the 3230 is no problem. The 1000 will work effectively, but I think the Weishi 1900 is a bit better and should be just a click over $200.

[saxo3]

Originally Posted by Amy_Moses

Thanks, Saxo3.
I recall reading that that timgraphers have a hard time with the 3230 movement because of the Chronergy escapment. Is that correct?

You are welcome. It is not correct.
The amplitudes of co-axial movements, such as those of some Omega watches, cannot be correctly measured by the Weishi and most other timegraphers.

Originally Posted by Amy_Moses

I'm in the UK and can get a Weishi 1000 for £175. Any good?

Not so good either, buy a Weishi 1900 for £225, its rate resolution is higher: 0.1 s/d compared to 1 s/d of a Weishi 1000.

The 1900 manual you find here:
https://www.cousinsuk.com/PDF/products/8445_W47107.pdf

PS: I am not a fan of Weishi timegraphers and don't intend to advertise them!

[phranxinatra]

Originally Posted by saxo3

Nonsense.

Do you still own this 126600 Sea-Dweller (3235) form 08/2020?
If so, how does it perform after 4.5 years?
What are the amplitudes after full winding and 24 hours later?

Answer with facts (data) and not with a video that the vast majority here cannot understand (the language).

Nope, I sold the SD. I have a Pepsi (2021) and a Sub date (2022), both bought brand new from boutique, and both had to be serviced last year. The video I posted was an answer to SwissSteph, who wrote there was no leaked information from Rolex. I gave you the translation, 'cause obviously no one understands Hungarian here. Not that I agree with what was said in the video.

[saxo3]

Originally Posted by phranxinatra

Nope, I sold the SD. I have a Pepsi (2021) and a Sub date (2022), both bought brand new from boutique, and both had to be serviced last year.

Thanks for the clarification.
In total, you had three 32xx watches (or more?) that developed the issue and needed "service" within the 5 year warranty period.

[saxo3]

This thread often discusses the results of timegrapher measurements. It does not seem obvious to everyone how to use a timegrapher, i.e., the orientation of a watch in relation to the timegrapher sensor (microphone).

Below is an illustration of how to correctly install a watch to take measurements. This applies to all timegrapher.

Never place any thick 'protection' between the crown and the microphone as this will attenuate the movement signals. Best is you put nothing between the crown and the microphone.

A worst case example (wrong watch orientation plus a thick microphone cover) you find here.

[fsprow]

Well put. I also point out, as I have here before, that the Weishi Chinese Timegrapher is OK for relative measurements but not very precise, especially for amplitude. Comparing several example with readings on my calibrated Witschi machine, I have often found amplitude reading errors of 10-20 degrees.

[saxo3]

Originally Posted by fsprow

Well put. I also point out, as I have here before, that the Weishi Chinese Timegrapher is OK for relative measurements but not very precise, especially for amplitude. Comparing several example with readings on my calibrated Witschi machine, I have often found amplitude reading errors of 10-20 degrees.

Interesting, I have a few questions for you.

You say that a Weishi is "not very precise". How do you measure the precision for amplitudes or rates with a Weishi timegrapher?

Can you give us a concrete example that demonstartes the amplitude errors of 10-20 degrees? I mean: the same watch, measured under identical conditions: both after full winding, both in 5 positions, measured on a Weishi and your Witschi which gives a direct a direct comparison of amplitudes and rates.

Which Weishi and Witschi timegraphers (models) do you use for such a comparison?

[fsprow]

Checked my notes -- for example measurements of a Patek 5164A with full wind at basically the same time, all I saved were dial up. Five measurements each machine. Witschi Watch Expert IV, Weishi 1900.

SPD: Witschi +1.8; Standard deviation 0.3 spd
Weishi +1.1; " 0.9 spd

Amplitude: Witschi 298 degrees Standard deviation 3 degrees
Weishi 311 degrees " 5 degrees

[Easy E]

I’m not trying to be argumentative, and full disclosure, I only have ever used a 1900. I’m just asking to check it off, are you sure you had the LA the same for both sets of tests? That would easily explain the variance otherwise.

[saxo3]

Originally Posted by fsprow

Checked my notes -- for example measurements of a Patek 5164A with full wind at basically the same time, all I saved were dial up. Five measurements each machine. Witschi Watch Expert IV, Weishi 1900.

SPD: Witschi +1.8; Standard deviation 0.3 spd
Weishi +1.1; " 0.9 spd

Amplitude: Witschi 298 degrees Standard deviation 3 degrees
Weishi 311 degrees " 5 degrees

Thank you for the numbers, measured in one position (dial up).
I hope you appreciate my long answer below!

My question about the precision data of your Weishi 1900 was a tricky one, because your Weishi cannot measure precision, but your Witschi does.

I understand that you took 5 measurements to get some statistics for your Weishi 1900, which resulted in a std value (standard deviation) of +/- 0.9 s/d. This value is highly dependent on many parameters, such as what settings you used on your Weishi 1900, how long each measurement was and how often you repeated it (you did 5 times).

You are probably aware that your Weishi 1900 data are based on a very low statistics compared to your Witschi data. Just an example: it is advertised in this thread that the data taking in each single position should be 2 minutes. The Witschi can take data every 2 seconds (or even faster). This results in (at least) 60 Witschi data points during 2 minutes, which is 12 (60/5) times more compared to your Weishi.

Let's take a look at your numbers:

Witschi Watch Expert IV Rate: 1.8 +/- 0.3 s/d
Weishi 1900 Rate: 1.1 +/- 0.9 s/d

These two rates are "the same" (= overlapping ranges) within your determined standard deviations.

Witschi Witschi Watch Expert IV Amplitude: 298 +/- 3°
Weishi 1900 Amplitude: 311 +/- 5°

These two amplitudes are outside your standard deviations and therefore different. It seems that you started measuring with your Weishi 1900 immediately after the full winding, measured very high amplitudes, and then continued with your Witschi. This could explain the difference in amplitudes because you did not wait long enough (15 minutes) for the caliber to settle after full winding. The comparison under identical conditions is different (see my suggestion below)

Conclusion: First, I am not trying to dispute your measurements, but to understand the differences in your numbers and how you got them. Secondly, the Witschi Watch Expert IV is certainly the much better instrument. The main disadvantage of the Weishi 1900 is the lack of precision information; the quality of this cheap Chinese product is not comparable to any Swiss Witschi timegrapher.

Suggestion: It would be nice to repeat this comparison under identical measurement conditions for the same or another 4 Hz (28800 A/h) watch. The watch should be at room temperature, i.e., off the wrist for a few hours (overnight) before starting the measurements. I suggest the following detailed procedure:

(a) Full movement winding (first)
(b) Place watch on your Witschi Watch Expert IV timegrapher
(c) Wait 15 min for caliber stabilisation
(d) Measure 2 min in dial up position, repeat 4 x, note all results
(e) Change watch to a vertical position (3U or 6U or 9U)
(f) Wait 2 min for caliber stabilisation
(d) Measure 2 min in this vertical position, repeat 4 x, note all results

Continue with:

(e) Full movement winding (second)
(f) Place watch on your Weishi 1900 timegrapher
(g) Wait 15 min for caliber stabilisation
(h) Measure 2 min in dial up position, repeat 4 x, note all results
(i) Change watch to a vertical position (3U or 6U or 9U)
(j) Wait 2 min for caliber stabilisation
(k) Measure 2 min in this vertical position, repeat 4 x, note all results

In total, this comparison takes 2x (15 min + 5x2 min + 2 min + 5x2 min) = 74 min

All selectable timegrapher parameters must be the same on both instruments!

[fsprow]

You have provided a very thoughtful and useful analysis. The data I presented were after getting together with a group of other watch enthusiasts who had wondered how accurate their Chinese timegraphers were. My measurements were as described, but I might clarify that in each case, the watch was fully wound. I waited before recording measurements, not 15 minutes, but perhaps five minutes. And all of the parameters, lift angle, etc. were set the same on both timegraphers. I have never owned a Weishi. My measurements were of machines provided by my friends, and so I don’t have the opportunity to repeat. However, for the three Weishi machines tested all deviated significantly in amplitude.
Thank you again.

[fsprow]

Originally Posted by Easy E

I’m not trying to be argumentative, and full disclosure, I only have ever used a 1900. I’m just asking to check it off, are you sure you had the LA the same for both sets of tests? That would easily explain the variance otherwise.

Yes they were set the same. Thanks.

[mato123]

I watched a video the other day in which they mentioned how Omega prepares their watches for METAS certification. They don’t use a timegrapher to measure accuracy in the final stage; instead, they monitor the watch hands for 24 hours and compare it to a reference time.

I guess it makes sense. When you put a watch on a timegrapher for a few minutes, you only measure how accurate it is in those few minutes and then pro-rate it for 24 hours. However, when you observe the hands over 24 hours, you get the true 24-hour deviation.

[saxo3]

Originally Posted by fsprow

You have provided a very thoughtful and useful analysis. The data I presented were after getting together with a group of other watch enthusiasts who had wondered how accurate their Chinese timegraphers were. My measurements were as described, but I might clarify that in each case, the watch was fully wound. I waited before recording measurements, not 15 minutes, but perhaps five minutes. And all of the parameters, lift angle, etc. were set the same on both timegraphers. I have never owned a Weishi. My measurements were of machines provided by my friends, and so I don’t have the opportunity to repeat. However, for the three Weishi machines tested all deviated significantly in amplitude.
Thank you again.

Thank you for your kind reply and for providing more details.

I now see why your statement in post #5571 "…the Weishi Chinese Timegrapher is OK for relative measurements but not very precise, especially for amplitude. Comparing several example with readings on my calibrated Witschi machine, I have often found amplitude reading errors of 10-20 degrees." is not correct because I am convinced that the route cause is not the timegrapher!

In my opinion the key point is that you did not wait long enough (only perhaps 5 minutes) for the movement to stabilize after full winding.

If you start measuring too early, the amplitude values will scatter at high numbers. The amplitudes will decrease a little over time until they stabilize. This usually takes approximately 15 minutes. I think you have not waited long enough and therefore your Witschi and Weishi 1900 timegrapher data are not comparable, as they measured the same caliber but in different (unstable) conditions after full winding.

I have another question: did you wear the watch until shortly before you did the full winding and then started the timegrapher measurements a few minutes later?

I appreciate the open discussion with you.

[fsprow]

Good points - however my experience, at least with the Witschi, is that the numbers stabilize well within a five minute period. Perhaps not for the Chinese machine. And the watch had been worn until an hour or so before winding and measuring (in my recollection!)

Frank