Timing tolerances - a guide to understand how they work

Posts
2,399
Likes
6,935
Is there a single inexpensive timegrapher that can be used for manual calibers 1861 & 861 and auto calibers 1045 and 565?
This one is free. Works well even with my laptop's' microphone.

 
Posts
818
Likes
3,126
Omega’s tolerances for this watch are that the Delta number must be no greater than 15 seconds, so on that measure this watch passes. The tolerances for the average daily rate are from –1 to +11 seconds per day, and since the average here is 0, this also meets the requirements Omega set out.
Given the number of measured positions, the limits of tolerance, and the maximum delta, it should be possible to calculate the maximum loss or gain in any one position, beyond which it is mathematically impossible for the watch to be in spec.

A lazy day project for someone with decent Excel skills, perhaps?
 
Posts
23,140
Likes
51,646
Given the number of measured positions, the limits of tolerance, and the maximum delta, it should be possible to calculate the maximum loss or gain in any one position, beyond which it is mathematically impossible for the watch to be in spec.

A lazy day project for someone with decent Excel skills, perhaps?

Excel not required: +21/+6/+6
 
Posts
602
Likes
548
I don't expect the timekeeping to change when I case the watch...
Thanks for the response.

You prompted me to read a bit more about the certification process and thought among others this was an interesting read https://www.govbergwatches.com/blog/chronometry-explained-cosc-certificate/ . I guess one of the main differences that this highlights is that a COSC certificate for a specific calibre is not the same as the timing of an individual watch.
I certainly have non chronometer watches that run well within COSC and in my view that is testament to the skill of the watchmakers that have serviced them in the past.
 
Posts
252
Likes
234
Great article Al!!! I always love when you do your write ups or even the advice you have given me before.

I have been out of the omega game for a couple years was a one watch Rolex guy, but always loved the brand. I just got a 2502.33 so gonna hang out on this forum for a little

I can use this as a guide for some amateur testing with my watch app. I am going to put it in a different position tonight then run a test. Day one was 9.5 sec will be interesting to see tomorrow.
 
Posts
1,076
Likes
1,787
Al, where can I send you a six pack of beer to? Thanks for sharing your knowledge.
 
Posts
87
Likes
28
Thanks for putting this together. While it’s useful it still leaves a question to be asked - in regards to METAS movements, why does Omega explicitly state that this movement should never lose time and should always be in the 0-5 sec per day range? I have a new PO8900 that’s a week old that is about -1 sec after 24 hours. The first 15 hours on my wrist it’ll lose 1.5 seconds and then when I rest it crown up at night it’ll gain back 0.5-0.75 seconds but still be a hair slow by the end of that 24 hour period. So the fast position for this watch isn’t quite fast enough.

If Omega knows that daily wearer habits will be the most impactful in the accuracy why make a blanket statement that METAS watches should always be in the 0-5 second range? If it wasn’t for that comment by Omega I wouldn’t be bothered with -1 per day but the fact they say this makes me think something is wrong with the watch. Any insight?
 
Posts
2,219
Likes
4,946
I have a new PO8900 that’s a week old that is about -1 sec after 24 hours. The first 15 hours on my wrist it’ll lose 1.5 seconds and then when I rest it crown up at night it’ll gain back 0.5-0.75 seconds but still be a hair slow by the end of that 24 hour period. So the fast position for this watch isn’t quite fast enough.

Out of interest, what are you checking this against?

Chris
 
Posts
87
Likes
28
Out of interest, what are you checking this against?

Chris

The Official US Time app on IOS
 
Posts
2,219
Likes
4,946
The Official US Time app on IOS
I don't know that one and don't use IOS but, I use https://time.is

When you're dealing with one second changes as you say, it would be worth checking with another source to be sure.

Chris
 
Posts
87
Likes
28
I don't know that one and don't use IOS but, I use https://time.is

When you're dealing with one second changes as you say, it would be worth checking with another source to be sure.

Chris
You have a point but I also time my other PO8900 with this app and that one shows accurate readings so I think the source is good.
 
Posts
29,117
Likes
75,248
Thanks for putting this together. While it’s useful it still leaves a question to be asked - in regards to METAS movements, why does Omega explicitly state that this movement should never lose time and should always be in the 0-5 sec per day range? I have a new PO8900 that’s a week old that is about -1 sec after 24 hours. The first 15 hours on my wrist it’ll lose 1.5 seconds and then when I rest it crown up at night it’ll gain back 0.5-0.75 seconds but still be a hair slow by the end of that 24 hour period. So the fast position for this watch isn’t quite fast enough.

If Omega knows that daily wearer habits will be the most impactful in the accuracy why make a blanket statement that METAS watches should always be in the 0-5 second range? If it wasn’t for that comment by Omega I wouldn’t be bothered with -1 per day but the fact they say this makes me think something is wrong with the watch. Any insight?

I’m not aware that Omega ever explicitly states that the watch will never run slow. In fact the whole point of my post is to explain why that isn’t so.
 
Posts
87
Likes
28
I’m not aware that Omega ever explicitly states that the watch will never run slow. In fact the whole point of my post is to explain why that isn’t so.

This is what I was referring to
 
Posts
29,117
Likes
75,248
This is what I was referring to

No idea what the source of this is, as it looks like a post made on a forum or something. But note the first bullet point above they refer to the average rate, and again as I’ve explained in detail in my first post in this thread, if the average can be 0 then the numbers that make up that average can be less than 0.

Read the first post again, as that explains it all.

Cheers, Al
 
Posts
87
Likes
28
No idea what the source of this is, as it looks like a post made on a forum or something. But note the first bullet point above they refer to the average rate, and again as I’ve explained in detail in my first post in this thread, if the average can be 0 then the numbers that make up that average can be less than 0.

Read the first post again, as that explains it all.

Cheers, Al

Looks like this is the full article

https://www.watchprosite.com/page-w...ed-master-chronometer-part-1-the-live-report/

But yes, to your point it doesn’t look like it’s coming from Omega. Your first post explained it all. I think this particular watch is spending a lot of time in a slower position which is skewing the average. I wear it on my left hand and crown up happens to be a fast position so I’m going to switch to wearing it on my right hand to see how that switches things up, if any.
 
Posts
272
Likes
445
Al, thank you for taking the time to write this up. As someone just getting into automatic and mechanical watches, this is great info to see and learn. I also noted a lot of questions on this forum about deviations from actual time, and was curious. You have now clearly explained to me how its tested.
 
Posts
87
Likes
28
Edited: Question was already answered in a different thread
Edited:
 
Posts
2,399
Likes
6,935
But regardless of that, I don't see how you get sub-second precision (in a single reading) comparing either to your watch.
The app Timestamp Camera permits to take a video with a phone with a precision of the timestamp.displayed. You can take a video of the watch as it is running and obtain subsecond accuracy of the timing of the movement of the seconds hand.
Edited:
 
Posts
12
Likes
20
If you have any questions on this, please let me know.

Cheers, Al

Hi Archer, I have POC 9900 which I bought new from Omega Boutique in December 24, 2018.

Let's say hypothetically the watch was tested by METAS in December, 2016.

So by the time I bought the watch, it was hypothetically 2 years old. The Result of METAS testing of my watch was + 0.1 spd.

My questions:

Is it possible that if METAS would test my watch now, the result would be slower, since the watch had been sitting in the store for 2 years?
From my limited experience, an automatic watch will run slower as it get older. For example: when I bought my Breitling Chronomat in 2005, the watch gained on average 5 spd. One year later the watch gained only 4 spd.

My POC now on average lose about 1 spd. (worn 5 days a week, 13 hours a day). My concern is the watch will lose about 2 spd one year from now, and will lose about 3 spd two years from now.

Thanks in advanced. (Btw, sorry for my bad English).