WatchUSeek Watch Forums banner

1 - 20 of 50 Posts

·
Registered
Joined
·
9 Posts
Discussion Starter #1
Hello,

I'm new to the watch world. I recently purchased a Bell & Ross 03-92 Steel. I hear on podcasts people saying, "my watch keeps great time." How do I properly test a watche's precision? Is this done generally over a 24 hour period or longer?
 

Attachments

·
Registered
Joined
·
384 Posts
I'd say the most readily accessible way would be to download the WatchCheck app for your phone and use it to build a daily log of how your watch keeps time against NTP servers.
 

·
Registered
Joined
·
12,121 Posts
I just go to time.is and set my watch according to this. Then something the next day (around 24 hours later), I compare it again. That's as much as I worry about, but then I change watches every day and by the time I get back around to the first watch, it's long since run down/needs resetting. Hence, I don't worry about accuracy or precision.

BTW, welcome to the madness. You and your wallet are in for a treat. ;)
 

·
Registered
Joined
·
3,169 Posts
Find an accurate time source, I like the “Emerald and Sequoia” time app. Wait until the second hand on your watch reaches 12oclock and pull out the crown. Set the correct time and wait for the app’s seconds to reach “0”, now push in the crown. Now your watch is perfectly synchronized with an atomic time source.

Check your watch against the same app in 24 hours and see how much time it’s gained or lost.

Enjoy!


Sent from my iPad using Tapatalk
 

·
Registered
Joined
·
164 Posts
I recently went through the same process. I checked the watch against my atomic clock app once daily for seven days. I found that the difference was little on the weekdays when I wore it more than on the weekend . I averaged out over the 7 days.


Sent from my iPhone using Tapatalk
 

·
Registered
Joined
·
7,570 Posts
I hear on podcasts people saying, "my watch keeps great time." How do I properly test a watche's precision?
I often wonder exactly the same thing. What does that statement actually mean?

Depending on how they measured, there is quite a range of possibilities what it actually means. Most often, poeple either compare their watch to an atomic clock or time.is after 24h, or they wear the same watch for some weeks and compare the total offset after a week or month.

However, all mechanical watches are subjected to differences in temperature, their position in earth's gravity field (-> positional variance), differing main spring torque (-> isochronism), etc. So the accuracy can vary depending on use time, summer/winter, how the watch is stored over night, etc. Once we take all that into account, the answer gets more complicated. Since the industry had the same problem, the Swiss standardized the testing proceduce as the COSC chronometer certification process. But other standards are also used for non-Swiss watches or by some manufacturers to go beyond COSC requirements.
 

·
Registered
Joined
·
184 Posts
I enjoy using Watchtracker ($5) on iphone. I typically check "worn" rate, vs "overnight" rate to come up with daily accuracy. It's easy to use and helps me keep my current watches regulated. For example, the current watch I've been wearing the past two weeks loses time while I wear it at a rate of about -6 seconds per day. Dial up, over night, it gains at approximately +4 seconds per day. Just checking those two conditions, which is very easy to do, it provides appropriate graphs and stats. Depending on how long I wear the watch affects it's 24 hour accuracy.
 

·
Registered
Joined
·
7,570 Posts
In fact, I'm always surprised how few people here seem to own a timegrapher. These devices don't cost a fortune ($150) and measure the precision in all positions within a few minutes. In addition, they inform about beat error and balance amplitude. Since I do buy grey market / used watches, it's a huge advantage being able to check all this straight away whenever I get a "new" (= possibly old and unserviced) watch. It saves a lot of money to own one.
 

·
Registered
Joined
·
184 Posts
differing main spring torque (-> isochronism)
Being new to automatics, I hadn't read up much on this. On my current new watch that I've been wearing, I noticed that if I manually wind it to what I think is max wound, it loses more time. I haven't manually wound it for a week, and after the first day or so, it has gone from losing time to on some days actually gaining time. As of this morning for the last four days (96 hours), it's lost -1.2 seconds total, or average of about -.3 seconds. I'm interested to see if I wear this watch daily 9 to 16 hours if in fact I ever actually do have to manually wind it. And if so, it won't be a full 40 turns.

Anyway, tldr, thanks for pointing me toward this term: isochronism. I'll be interested to read more about it. It seems this Miyota 9015 prefers to be mid-wound, but I would think it would be faster fully wound. Either way, I went from mildly disappointed during the first few days of ownership losing 10 seconds a day, to just wearing and enjoying the watch that has settled into very acceptable accuracy.
 

·
Registered
Joined
·
7,570 Posts
Being new to automatics, I hadn't read up much on this. On my current new watch that I've been wearing, I noticed that if I manually wind it to what I think is max wound, it loses more time. I haven't manually wound it for a week, and after the first day or so, it has gone from losing time to on some days actually gaining time. As of this morning for the last four days (96 hours), it's lost -1.2 seconds total, or average of about -.3 seconds. I'm interested to see if I wear this watch daily 9 to 16 hours if in fact I ever actually do have to manually wind it. And if so, it won't be a full 40 turns.

Anyway, tldr, thanks for pointing me toward this term: isochronism. I'll be interested to read more about it. It seems this Miyota 9015 prefers to be mid-wound, but I would think it would be faster fully wound. Either way, I went from mildly disappointed during the first few days of ownership losing 10 seconds a day, to just wearing and enjoying the watch that has settled into very acceptable accuracy.
Don't get obsessed with this and bear in mind that a day has 86,400 seconds. Even 10 s/d off means it's 99.99% accurate. No other mechanical measuring device in the world reaches remotely that level of precision. When you start logging split seconds per day, that is fully within the normal scatter of any effect you can imagine, including the measuring error of the app you're using.

The movement specification for the 9015 is -10/+30 s/d and up to 40 s/d positional variance when fully wound. So if your watch is actually precise to a few seconds per day, that's exceptional performance.
 

·
Registered
Joined
·
2,582 Posts
I really don't think you can ever really know what your watch is doing. On a timegrapher it may do one thing and on the wrist something else all together. Temperature, movement, orientation, state of wind etc can all play a factor and cause wide swings in accuracy day to day.

There are so many variables and factors the best you can really hope for is an average over an extended period of time.
 

·
Registered
Joined
·
5,290 Posts
I use a quartz watch. Just set them both to the exact time and see how much the mechanical deviates over time.
 

·
Registered
Joined
·
9,498 Posts
. . . but I would think it would be faster fully wound. . . . . .
The greater the tension in the mainspring the greater the impulse imparted by the pallet fork.

The greater the impulse is to the roller jewel the higher the balance amplitude is.

The higher the amplitude the longer the period of the oscillation.

The longer the period of oscillation the slower the movement advances.
 

·
Registered
Joined
·
2,582 Posts
The fascination with accuracy is a funny thing. I think the vast majority of people measure dial up, off the wrist and have never heard the word isochronism. They see a number and that number becomes how accurate their watch is. It's simply not even close to true.

Even a COSC watch with an excellent isochronism profile is going to vary day to day and the only way to really establish the real number in is on wrist averaging. Don't get me wrong there has to be a set standard that gives an objective value (timegrapher in multiple positions) but it does not give what you can expect in the real world.
 

·
Registered
Joined
·
3,553 Posts
I set mine aginst an atomic G-Shock. I wear it for a day or three before switching watches. If I'm not signifcantly late for anything and don't miss the first pitch at 7:10 I figure I'm OK.
 

·
Registered
Joined
·
2,582 Posts
The greater the tension in the mainspring the greater the impulse imparted by the pallet fork.

The greater the impulse is to the roller jewel the higher the balance amplitude is.

The higher the amplitude the longer the period of the oscillation.

The longer the period of oscillation the slower the movement advances.

But how much does the amplitude increase? Enough to actually affect the timekeeping? Does the speed of the oscillatation increase as well, therefor negating or reducing the effect
 

·
Premium Member
Joined
·
63,255 Posts
Set your watch using Emerald or Time.is then you can check it again after 24 hours to get a rough idea of how it performs. For a more accurate reading run it for several days before checking it.
 
1 - 20 of 50 Posts
Top