Hello folks, my first post here.

Recently, I purchased my very first mechanical watch - a Speedmaster moon watch - and have really been enjoying it. I've been using the WatchTracker app to measure its accuracy for the past two weeks and it's been running a reliable +5 SPD, which I'm well pleased with. I wind it every morning upon waking.

My question is this - on at least a couple or three occasions, I've checked the time immediately before and after a full wind, and have been surprised to see the watch gained exactly one second from the wind. This doesn't seem to affect its +5 SPD average. I'm just wondering if a time gain from winding is normal - seems odd to me.

Thanks for any insight. As I say, I'm really enjoying the watch in all respects.

Regards,
DeWayne

 

You really don’t give enough information for me to understand what your concern is. When exactly does the one second gain occur? During winding? After completion of the winding? Is it still within the time indication you expect upon completion of the winding? If so, how long does it take for the watch to pick up one second after completion of the winding? You will need to be checking the Speedmaster against a reference that is much more accurate than the Speedmaster. What reference do you use? With an Omega Speedmaster to perform to within 5 seconds per day, you really have nothing to concern yourself about. Even taking into account the one second gain. Your very first mechanical watch, you say? Get used to the fact that it not going to keep quartz watch time!

 

Apologies if my message was less than clear - I thought that is was.

On the occasions in question, using the WatchTracker app as the time reference, the watch will read - let's say 29 seconds fast before winding. Immediately after winding, the watch will read 30 seconds fast. It gained a second in the process of winding.

And yes, I fully understand the mechanical/quartz difference - hence why I made clear I'm very happy with the watch. I'm simply asking if it's normal to gain time during the process of winding.

Thanks,
DeWayne

 

I think you might be putting too much trust in an app. If you’re concerned have a decent watchmaker take a look at it

 

Stop checking it

And don’t forget if your going to check it we want 5 different positions and in a sauna and a cool room also for it to be a true test.

 

Don't obsess over it, if I'm doing a real world time test I'll measure it after one week .Apps can be wrong, I use time.is

 

Guys, I'm not obsessing over accuracy. I'm simply curious about these things as machines, and wondering - from anyone who knows how they work - if there's an actual reason that winding the watch would make it jump a second forward. That's all.

 

It takes maybe 30 seconds to wind a Speedmaster. I only know of one possible cause for the watch to (apparently) gain one second during winding. The sweep seconds wheel is an idler wheel, driven by the chronograph gear train. That idler wheel has a tension spring beneath it which is there to eliminate lash in the gear train which can cause the seconds hand to occasionally pause for a fraction of a second, then to again proceed as normal. Some folks complain of a “jerky” motion of the center seconds hand when the chronograph is operating. Is there a chance that such an irregularity in the operation of the center seconds hand is giving you the impression that the watch is gaining during winding?

 

No, this isn't a perception of the fluidity or jerkiness of the seconds hand, I'm simply recording the watch's time via the WatchTracker app immediately before and immediately after winding, and the result is a one second increase against atomic time.

 

Well, maybe someone will come along with a more plausible explanation. Best of luck!

 

Well it depends what the author of WatchTracker means by atomic time . I assume that the app is using Network Time Protocol (NTP) so which server is the app querying and which tier is it? Does the app account for variations in network latency the way that for example chrony for computers does by multiple queries to multiple servers or is it a single query? How is 20:14:30.998 displayed and how is 20:14:31.002? Looking at the preview page for WatchTraker (I have no iOS devices) the granularity of the app is integer seconds. The two times in my example may well be displayed as a 1 second difference where in reality they are 0.004 seconds apart.

On any consumer device 1 second means next to nothing, it costs decent money or programming effort to do better. Within sight I have three radio-controlled clocks supposedly getting their time from atomic references, they vary by 2 seconds, my Android phone getting its time from the phone network is another second adrift, and this computer which uses NTP is in the middle of the others. I could get the computer to within a few milliseconds if I needed to, but currently do not bother.


BTW, two of the radio clocks are actually weather stations. They are sitting 50cm from each other either side of this laptop, one shows internal temperature & humidity as 20.5 and 54%, the other 20.2 and 47%. Consumer devices.

 

I find if I wind Speed master and immediately place in freezer , after a week it gains + 1 SPD.
is this normal?

 

You in the freezer with it to make certain it doesn’t run down?

 

Could the added torque during winding increase the amplitude so much that the balance knocks or overbanks (impulse jewel hitting escape lever from behind) momentarily? You could try to listen for it while winding (sudden 'galopping' sound).

I'm assuming you observe this consistently with an app that uses network time and corrects for network delay; I'm not familiar with watchtracker. Watchcheck seems to be ok, when I check it against the time.gov web page/applet.

 

You guys really know how to make a new forum member and new Omega owner feel welcome.

For those who tried to offer a helpful response to my curiosity, thanks.

To the rest, Canuck, chiswick, I'm gonna go ahead and let you have your little clubhouse to yourself. Doesn't appear to be a place I care to hang out.

DeWayne

 

Anybody who is concerned with an apparent one second per variation as a result of winding the watch, is obsessing. Full stop. Keep an open mind and stick around. You might learn a lot.

 

It's not the best place to ask technical questions.

 

I highly disagree this is the best place to ask those questions.

@sonvolt those apps are inherently imprecise over short term less then 8 hour measurements.

Winding does not affect the gear train it just charges the mainspring. Granted wound status of the mainspring affects torque applied to the gear train which can affect time keeping...

But in this case nothing really to worry about.

 

My comment above was tongue in cheek but you have to realise checking a watch in general is at least a 24hr check.
So many things add or subtract to precision. ( position of watch ie; dial up, dial down, crown up, crown down and temperature )

We have had a multitude of this sort of questions over the years and without yourself reading about mechanical watches it is only going to confuse you more.

Here is a start

 

I looked up the WatchTracker App to see how it works. Correct me if I’m wrong, but you have to tap a button when your watch is at a certain time to synchronize and determine the bias. Considering the seconds hand on the Speedmaster is so small, I could see you could easily be a second off just glancing from the subdial to hit the button on your phone. Also, the app notes that if it can’t contact the atomic clock, it defaults to the phone time. Lots of potential for a random 1 sec bias there.

I doubt this is a real effect. Don’t worry about it!