How much time variance is normal on a new sub?

[Chester01]

Think the better question is what degree of worry about a watch's accuracy is normal.

[brandrea]

Originally Posted by Driver8

Well as you know there is a hefty thread dedicated to instances of appalling timekeeping, but yes I will concede that I certainly didn't need a timegrapher to see that my SD43 had something seriously wrong.

My comment was more to the point that I'm equally as incredulous at those who aren't remotely bothered by timekeeping and will happily accept +/-10 secs a day as being perfectly fine from a 4 or 5 figure watch, as I am at those who are bothered by a variance of 0.5 secs a day. As I say, a happy medium is fine IMO.

Well I dropped a fair chunk on my 321 Speedmaster and as I understand it, Omega’s average precision tolerances for non Chronometer/Master Chronometer watches are between -1/+11 seconds per day. This applies to the new caliber 321 movement. (Paraphrased from Fratello).

I’ve never checked the accuracy of any of my watches as I rotate them every couple for days. If they’re off by 30 seconds or more, I’d never know it. nothing incredulous about it IMO, I really just don’t care

That’s not to disparage those who are interested in using time-o-graphs or other devices to measure their timekeeping. If that’s your thing, great and it would appear that for many it is

In my very humble opinion, there a number of factors that are important to collectors, timekeeping for some is important. To others like me, not so much.

I buy my watches more for the intangibles that cannot be measured on a machine. YMMV and that’s ok.

[Kenny G]

I didn’t read all the comments so this may have already been said…

I have an extremely accurate DJ that I’ve now been wearing everyday since resetting the time on July 1st. Today, roughly 36 days later, it’s only +5 seconds which is nothing short of amazing for a mechanical watch (5ish years old and unserviced)

That said, that doesn’t necessarily mean my watch only runs 0.14 seconds fast per day. Sure, on some days it might happen, but when daily wearing a watch long term it will gain and lose time each day. This is what cannot be measured well on a time graph.

In other words, just because a time grapher says +2 s/day, it doesn’t mean the watch will be +60 seconds after 30 days.

Do a real world test over daily wearing it for a month and see where it’s at.

[watchmaker]

Originally Posted by CarlOver

My current Sub for sure deviates differently based on position when stored.

It will do, that is normal.

What I mean is the printed card that claims it will gain in this position, gain a little more in that position and lose in this position; no longer holds true as a rule.

[Andad]

Originally Posted by saxo3

Out of about 250'000 TRF members I have not seen 250 who know the difference between accuracy and precision. This is max. 0.1 % and far lower than your 10 % (1 out of 10)

But the other 249750 members know so many more important things that you don’t know.

But I agree with you that 1 in 10 is 10% and that is the sum of my intelligence.

[saxo3]

Yes for sure no doubt E.
Me more serious to come soon …

[saxo3]

Originally Posted by Kenny G

I didn’t read all the comments so this may have already been said…

I have an extremely accurate DJ that I’ve now been wearing everyday since resetting the time on July 1st. Today, roughly 36 days later, it’s only +5 seconds which is nothing short of amazing for a mechanical watch (5ish years old and unserviced)

That said, that doesn’t necessarily mean my watch only runs 0.14 seconds fast per day. Sure, on some days it might happen, but when daily wearing a watch long term it will gain and lose time each day. This is what cannot be measured well on a time graph.

In other words, just because a time grapher says +2 s/day, it doesn’t mean the watch will be +60 seconds after 30 days.

Do a real world test over daily wearing it for a month and see where it’s at.

I agree. The following Watch Tracker screen shot illustrates what you say:



One can see the daily fluctuations and the overnight compensation by chosing a specific rest position. The fastes and the slowest rates during the tested 2 months were +4.4 s/d and -3.3 s/d.

Before this test, I determined the position dependent rates using my timegrapher:



These data provided an excellent position dependent prediction how much this specific watch is going to lose or gain at rest overnight. The difference between wrist and ambient temperature plays a role.

After 64 days the 3235 caliber was +0.1 s off. Of course that does NOT mean the movement was running fast by only 0.1/64 = 0.0016 s/d = 1.6 ms/d.

Timegrapher are a very useful tool also to predict how much a movement is going to lose or gain overnight. Once you know, you chose the necessary position and one can keep it close to perfect timekeeping.

It is also a lot of fun and mainly works so well because Rolex calibers are very precise, even when they are not accurate in one or two specific positions.

[CharlesN]

Precision Vs Accuracy again it seems but who understands the difference ?

Not enough people here unfortunately.

[saxo3]

Originally Posted by CharlesN

Precision Vs Accuracy again it seems but who understands the difference ?

Not enough people here unfortunately.

ok ok Charles, simply FORGET my last paragraph.

[Andad]

Originally Posted by CharlesN:

Precision Vs Accuracy again it seems but who understands the difference ?

Not enough people here unfortunately.


C’mon Charles,

Is there a Precision/Accuracy forum that you two can find?

[Andad]

My apologies to C and E.

I am just ‘avin a laff.



Please continue with your stats and stuff.

[CharlesN]

Originally Posted by Andad

[B]
C’mon Charles,

Is there a Precision/Accuracy forum that you two can find?

Eddie ...

I thought this was the very forum for that discussion.

Am I wrong ?

Ok, I will create a new thread that deals with the subject.

[Andad]

Originally Posted by CharlesN

Eddie ...

I thought this was the very forum for that discussion.

Am I wrong ?

By your previous post it seems you don’t think you have enough members on TRF who know enough to discuss this subject with you.

I thought a forum with more members who have heard about accuracy and precision would be more rewarding for you?

E.

[CharlesN]

I will look for that Mythical forum.

In the mreantime i will revert to discussing plain and simple "Timekeeping" and how simple it is to attain that.

I even created a new thread on this very forum, all about the simplicity of self-regulation but somehow it just vanished.

I dont think it broke any rules etc.

I really wonder what happened there was a lot of good and simple information contained in the post.

[DJ2020]

Originally Posted by padi56

Have to agree loupes timegraphers phone apps and the rest of todays Rolex toys, can and often will be a Rolex owners worst enemy.

Your right Peter,
They just don't understand how a watch is regulated and all that is involved to keep a watch accurate. All the components that come into play along with all the variables. To tell someone that they have a precision movement and then they ask "well why is it 4 seconds fast every day then?" They just don't get it.

Learning how a watch works and having a full understanding of each components functions would explain a lot. Understanding of the work it takes to get a movement with very little deviation between positions, temperature and isochronism would go a long way in easing there mind and then marvel at the movement on there wrist.

[saxo3]

No Wayne. He is wrong, again, and in every identical post before, imho.
Read post #38.

[Dirt]

Originally Posted by CharlesN

I will look for that Mythical forum.

In the mreantime i will revert to discussing plain and simple "Timekeeping" and how simple it is to attain that.

I even created a new thread on this very forum, all about the simplicity of self-regulation but somehow it just vanished.

I dont think it broke any rules etc.

I really wonder what happened there was a lot of good and simple information contained in the post.

Agreed

[DJ2020]

Originally Posted by saxo3

No Wayne. He is wrong, again, and in every identical post before, imho.
Read post #38.

I have a timegrapher myself, yes it is very useful in many ways. But one should also have realistic expectations of time keeping of their watch and understanding when a watch is running fine, just needs regulation or requires a service. If that knowledge was present, a owner would understand that a watch that's worn that is constantly -5 or +7 seconds a day, everyday is just fine. It just needs to be regulated . And should just get it done.

Imo, Peter is just pointing out a fact that watch enthusiast should know. And not to get to concerned if there watch all of sudden gains or losses a second or two randomly. Rolex does not have a wind indicator on there watches. So one never knows what end of the spectrum there power reserve is. Just a example here. Environmental and physical conditions will absolutely affect time keeping on a mechanical watch. It's un avoidable.

If a timegrapher displays within +/-2, that by no means one
should expect that it will run that way on the wrist. For example I have a Longines HC that has the old L633.5 (2824-2) movement in it. It will run within a second or two after 2 weeks on the wrist 24/7. Very consistent. But on a timegrapher DU it's running +9 spd.
It's not a precision movement at all, but very accurate.But it is regulated to my wearing habits.

[Filipão]

On any Rolex, +2 or +3 spd is not an issue.

Movement is lubed and all is ok.

I'm always concerned when they stop gaining or worse, start losing time.

It means it's time for service, lubrication is not as good as it was.

[premium2]

You are an amazing time tracker for you to have measured 0.38 sec time gain.
For a mechanical watch, 2.5 secs gain is pretty fair.

[saxo3]

Originally Posted by premium2

You are an amazing time tracker for you to have measured 0.38 sec time gain.
For a mechanical watch, 2.5 secs gain is pretty fair.

Agreed.

The OP has not explained how he measured ("…it was gaining about 0.38-.5 seconds a day.") such small numbers or if it is an average value over a certain period.

Maybe, he has measured a deviation between 1.9 and 2.5 s/d over 5 days and then simply divided by 5? Only the OP knows.

Anything up to a few seconds per day is very good for any mechanical watch, imho.

[Dirt]

Come on people.
Obviously it's averaged out over some period and perfectly legitimate

What's the issue?
Jealous that you can't match it?
Enjoy

[saxo3]

Originally Posted by Dirt

Come on people.
Obviously it's averaged out over some period and perfectly legitimate

What's the issue?
Jealous that you can't match it?
Enjoy

Cmon Dirt.

As you know, there is no issue apart that it would have little meaning.

Imagine somebody saying "My new Sub is dead-on since purchase 9 months ago". Would any reasonable person claim that the movement is running consistently with 0 s/d. No.

One can obtain excellent timekeeping over several months and stay even close to an atomic clock reference. But that is a consequence of self-regulation (if possible) and not a mega-accurate watch, see:
https://www.rolexforums.com/showthread.php?t=866906

Anyhow I think we are in the same boat of understanding and there is no envy imho.