32xx movement problem poll and data thread

[saxo3]

First continuous amplitude, rate, and precision measurements.

Three different 3235 and 3285 calibers in DU position.

Data sampling rate: every 30 s
Duration: 24 hours (2880 data points, each curve)



Amplitudes (after full winding):
3235 excellent
3285 not good
3285 not good

Rates, Timekeeping:
3235 very good
3285 acceptable
3285 not acceptable



Precision:
3235 excellent
3285 excellent
3285 very good

All three 32xx movements are within Rolex precision specs of -2/+2 sec/day
From full winding until 24 hours (test conditions)

[Dennisoul]

After today’s readings at the 60 hour stage not the 48 hour as I’ve written on the paper…my 126334 has been sent back to RSC Kent, GB….as you’ll see by the readings it’s not running well at all, doesn’t even register on some positions…it’s still under warranty..bought brand new from an AD in August 2017







Sent from my iPhone using Tapatalk

[saxo3]

Andy, many thanks for the update.

That's what I meant yesterday, saying "it may not reach 70 hours”. I saw that coming w.r.t. the low amplitudes in vertical positions (9,6,3 up).

You took the right decision to sent it to RSC in Kent.

You may wish to investigate your 31xx watches, to obtain some 32xx vs. 31xx comparisons. It would be nice to see more data from you soon.

[Dennisoul]

Originally Posted by saxo3

Andy, many thanks for the update.

That's what I meant yesterday, saying "it may not reach 70 hours”. I saw that coming w.r.t. the low amplitudes in vertical positions (9,6,3 up).

You took the right decision to sent it to RSC in Kent.

You may wish to investigate your 31xx watches, to obtain some 32xx vs. 31xx comparisons. It would be nice to see more data from you soon.

Cheers for the info Saxo,, that was the final straw this morning when it never registered on the Weishi…
….yes no probs I’ll do some other testing with the 31xx ….I’ll do my Explorer 2 16570 first, then EXp1 214270, 14060m & 116610LN…
When the DJ41 returns I’ll post data for that to see how it runs..


Sent from my iPhone using Tapatalk

[CharlesN]

Originally Posted by Dennisoul

….yes no probs I’ll do some other testing with the 31xx

Hi Andy,
Good decision by you.
The RSC is the best place for the watch.
Itll take around 4 weeks probably but its good to know you have simething else to wear.

Do i take it that you now have the "measuring bug" ?

I look forward to seeing more 31xx data soon and the new improved 32xx data when your watch comes back.

[Dennisoul]

Originally Posted by CharlesN

Hi Andy,
Good decision by you.
The RSC is the best place for the watch.
Itll take around 4 weeks probably but its good to know you have simething else to wear.

Do i take it that you now have the "measuring bug" ?

I look forward to seeing more 31xx data soon and the new improved 32xx data when your watch comes back.

Cheers Charles…..I’ll keep the measuring to every few months, far too easy to get worse ocd than I have lol
Definitely happy to post up results, but will try and keep to every 3 months or so that I don’t get too obsessed….keeps it interesting that way for me….I’ll defo keep up to date with the thread though…. very interesting stuff


Sent from my iPad using Tapatalk

[CharlesN]

I thought I would look at Tic Tac Toc or Toe)

You can also look




Sent from my iPhone using Tapatalk

[saxo3]

You mean Tick, Trick, Track (Huey, Dewey, Louie)



They just discovered the Allan Variance ...

More to come for the friends!

[CharlesN]

Who is who ?




Sent from my iPhone using Tapatalk

[CharlesN]

I have used the data gleaned about my Submariner to make sure that it keeps on time

This shows how I did it ………



Fromm that data I could decide how to leave my watch overnight.

I managed to get it back to "Atomic" time.


And this shows how I got there …….




Sent from my iPhone using Tapatalk

[CharlesN]

To get the Allan Variance we need some more data I think before we use the equation …




Sent from my iPhone using Tapatalk

[saxo3]

Very good Charles, agree 100% to use these modern and professional timegraphers, to produce reliable data with relevance to many watches daily worn on wrists

Of course, not for everybody to understand

[CharlesN]

When doing the measurements showld we also think about the effect of the Sun on the back of the head to see if it gets warmer and changes at all.

Also,

Will sound from beating in a confined area have any effect.

[CharlesN]

I have just done a one hour Amp and Rate test ….. CH position



Looks O.K. I think.


Sent from my iPhone using Tapatalk

[CharlesN]

I think a little explanation may be required here.

This was just a 1 hour test of Amplitude and rate of my Submariner-date 126610 with a 3235 movement.

Fully wound - Dial Up - 2 seconds between data points - 1,800 data points - 1 Hour recording

The top section is The rate and the bottom the Amplitude.

The rate stayed pretty good and constant with an average of +1.7 spd.

The Amplitude also was pretty constant averaging 273 Again, thats pretty good.

Sorry the picture is pretty poor ... I had to photograph it instead of downloading it.

I will do a longer test and eventually work upto a 24 hour one.

[CharlesN]

Last night I did an overnight monitoring of my Sub Date 126610. 3235 movement

I monitored using 2 methods

The first was a simple...

A Method of using an app (WatchTracker) on my iPhone. The app showed a +0.6 spd gain during those 8 hours. (Please see the top picture below).
That is exactly what I would have expected from all the previous averaging tests done on my Witschi. These results can be seen in posts made earlier than today.

My watch gained aproximately 0.6 spd overnight.

It was fully would to start with, and left for 8 hours in the CH position.








The second methodwas equally simple...

I measured the Amplitude, Rate and B.E. over 8 hours. (Please see the bottom 3 pictures).
I started with the watch fully woumd and left for 15 mins to stabilise fully in a CH orientation.
I took measurments every 10 seconds. This gave me a total of 2,880 data points over the 8 hours.

The results are shown below.
To precis the results the X results were ...Gain was +1.6 spd. The B.E. was 0.5 and the Amplitude was 277 degrees.

There was a clear "Dip" in the amoplitude after about 30 mins of running, this is unexplained but i will do another test (But for 24 hours) soon and i will leave the watch for 45 mins before starting to see if this "Blip" vanishes.

Throughout the B.E. was remarkably constant and certainly has nothing to worry about in the slightest.

However .... There was a clear and very noticable "Dip" in amplitude aafter 6.75 hours of monitoring. Again Unexplained.
This obviously slightly reduced the timing of the watch but only very marginally.











A Good conclusion from these tests shows that watctracker showed a 0.6 spd gas in 8 hours which would theoritically equate to a gain of +1.8 spd if left for 24 hours.
The Witschi measures, far more accurately, that the gain is +1.6spd.
Thats darn close to an agreement.


Sent from my iPhone using Tapatalk

[saxo3]

Charles, if I may add my 2-cents ...

Looking at your 8 hour (scaled to 24 hrs) results:

Witschi: 1.6 +/- 0.5 s/d
WatchTracker: 1.8 +/ - x.x s/d
x.x (error) not given
The results are 100 % identical.

For data taking and analysis it is MANDATORY to think about syst. and stat. measurement errors.

Btw, most complicated to measure is 0 (Zero).

The 'dips' in your 3235 amplitude and rate curves I have also seen before for my 3235 SD43:



They explain (in my opinion) why your 72 hour long PR measurements where not smooth:



Amplitude values seem to depend (a bit) on the exact time you measure during the PR.
I observed a similar effect for my 126600 SD43:




PR measurements for 3130 and 3187 seem to be smoother (no significant amplitude dips) than for your 3235:



But there is very little systematic work done on this 32xx/31xx comparison.

An interesting question might be how reproducible these dips are. It seems that they have no impact on excellent timekeeping.

Hence it's more an academic exercise.

[CharlesN]

Academia is good
Excercise is NOT so good.
Oh .... Hang on I was thinking you meant sport ... That idea scared me !

Well As an academic exercise I am now measuring with the identical test as just before but for a full 24Hours.

The "Dips" should be seen there and if they are at the same times then even more research will be needed

[Dennisoul]

Here’s a test I did today on my Explorer 1 MK2 214270 for comparison…calibre 3132
Bought brand new from an AD in June 2019, hasn’t been serviced or regulated…
I’ll do a 24 hour test tomorrow…


Sent from my iPhone using Tapatalk

[saxo3]

Andy, these are your watches & calibers in the queue to be measured?

16570: 3185 or 3186
126334: 3235
116610LN: 3135
214270: 3132
14060M: 3130

[watchtabs]

In all my years around watches, I've never once thought of this.

How do you all keep track of 3 secs in a 24hr period?

[Dennisoul]

Originally Posted by saxo3

Andy, these are your watches & calibers in the queue to be measured?

16570: 3185 or 3186
126334: 3235
116610LN: 3135
214270: 3132
14060M: 3130

Yes Saxo, these are what I’ve got…I’ll get through them all eventually…the 16570 is 3185


Sent from my iPhone using Tapatalk

[CharlesN]

Originally Posted by watchtabs

How do you all keep track of 3 secs in a 24hr period?

Hi,

Its very simple.

There are 3 basic methods,

Ill start from the most basic ...

1). Compare ypur watch to an atomic time signal controlled clock.
The clock will be accurate to somthing like a millionth of a second.

2). If you have an iPhone get an app called Watchtracker. Its very simple to use and it will alsogive you a graph of how your watch is performing.

3). Use a timing device such as a Witschi Chronoscope or a far cheaper device available on Amazon called a Weishi Timegrapher - model 1900 is good.

[saxo3]

Originally Posted by Dennisoul

Yes Saxo, these are what I’ve got…I’ll get through them all eventually…the 16570 is 3185

Andy, please try to measure more often than 0, 24, 48 hours. That will be easier to compare with other results.

The more data along the power reserve the better. I will try to deliver some comparing graphs later.

[CharlesN]

Finally I have managed to take a full 24 hour observation with 2880 data points. (That’s one every 30 seconds)


This shows the full 24 hours with the 2880 data points.
The watch was fully wound before the test began and was left for 15 mins before recording started.
Orientation was CH for the full 24 hours.

The rate is very steady realistically.
The Amplitude seems to have had 4 very quick "Blips" that should be ignored. There were also 2 drops (9h and 16h) which as yet are unexplained.




This is the same as above except in double scale so you see the first 12 hours.




This is the same as above except in double scale so you see the second 12 hours.




This shows the Rate and Amplitude differences. (Beware .. the "Blips" have given false rate readings)




This shows the Rate and Beat Error differences. (Beware .. the "Blips" have given false rate readings)




This shows the 5 positions straight after the 24 hour test period.



All in all this all goes to show that my 126610 is in great order.


Sent from my iPhone using Tapatalk

[Dennisoul]

24 hour results for my 214270 Explorer 1 …3132 calibre…not touched since full wind and left dial up overnight…all protocol followed.



Sent from my iPhone using Tapatalk

[CharlesN]

Originally Posted by Dennisoul

24 hour results for my 214270 Explorer 1

Hi Andy,

When you do a 24 or longer test can you log more data points ?

By that I mean take the readings every few hours - The more the better.

The more data points you give the more we can plot your watch movement and see what it is doing.

I was doing it on my Weishi1900 every 2 hours during the day when i could. Obviously when asleep i took one just before bed time and one as soon as i was awake.

[saxo3]

Originally Posted by CharlesN

Finally I have managed to take a full 24 hour observation with 2880 data points. (That’s one every 30 seconds)


This shows the full 24 hours with the 2880 data points.
The watch was fully wound before the test began and was left for 15 mins before recording started.
Orientation was CH for the full 24 hours.

The rate is very steady realistically.
The Amplitude seems to have had 4 very quick "Blips" that should be ignored. There were also 2 drops (9h and 16h) which as yet are unexplained.




This is the same as above except in double scale so you see the first 12 hours.




This is the same as above except in double scale so you see the second 12 hours.




This shows the Rate and Amplitude differences. (Beware .. the "Blips" have given false rate readings)




This shows the Rate and Beat Error differences. (Beware .. the "Blips" have given false rate readings)




This shows the 5 positions straight after the 24 hour test period.



All in all this all goes to show that my 126610 is in great order.

Hi Charles,

Nice study, it goes to ->>>>>



There is too little progress and interest; hence a good moment for me to take a (short) break until more new 32xx data are submitted, at present nothing new to learn.

[CharlesN]

Hi saxo3

I fully understand you
I am also a little surprised about how little interest there is in looking into the life of an32xx movement.

Everybody here has heard of the problems that are proved to have existed and still do in many watches.

It looks like some people just don’t care, all they want is to have The “Crown” it would appear.

Hopefully you will not be absent from school for too long as we all need to study under the “Professor”.

I will try and see if I can keep things going for the few who do take an interest and calculate the X and D etc.

Hopefully see you soon

Charles

[CharlesN]

I have for the last 24 hours been monitoring my Omega Aqua-Terra.

I started with a full wind and waited 15 mins before I started measuring.
The lift angle was 38 Degrees
Dial up for the entire test




It has a Co-Axial movement ref 8508.

It has left a very interesting and totally different Trace for us all to see …….





The trace VERY clearly shows that as the Amplitude decreases over time the Rate increases and as the Amplitude decreases the rate increases.

This is very different to Rolex watches. It is the opposite almost.



The timekeeping was pretty good also with an x of +1.6 spd
The beat error was 0.0 ... Outstanding.



In real world testing on my wrist, My watch gained about 1.5 seconds a day.

Sent from my iPhone using Tapatalk