[gaijin]

I sent my Omega Speedy Pro in for regulation, and I was called yesterday that it's ready to be returned-- before, it was +11-13 sec/day, now it's been regulated to something like 4.7 sec/day. Since the number is so specific (ie, not 4-1/2 sec), I assume it's done by computer. How accurate are these computer measurements typically? Also, how long is the movement actually hooked up to the computer for these measurements to take place? Seconds? Minutes? Hours?

I initially took this Omega to Swatch in LA for regulation, and when they returned it, they said it was +7.6 sec. In the 3 months since, it never ran slower than +9, usually >10 sec/day. How could they be so far off?

As much as the engineers among us (that would be me :-d) would like movement regulation to be an exact science, it is not.

The main culprits are positional variation and temperature changes.

Our watches run at different rates depending on their positions - dial up, dial down, crown up, crown down, crown left, crown right, and any other odd angle normally encountered in the course of a day's wearing.

They also run at different rates depending on the temperature - faster or slower as their temperature goes up or down.

They are delicate mechanical devices and such variations are the bane of our existence :-s

Here is a picture of my BaliHa'i on my MTG-1000 watch timer:

If you look at the bottom of the display, you will see that the watch is running at a rate of "+000 S/d," or exactly "dead on" with absolutely no gain or loss over the course of 24 hours - based, of course, on about 6 seconds of sampled data and assuming nothing (e.g. position, temperature, etc.) changes.

Does this watch maintain such perfect time when I wear it? Of course not. It's a little bit different every day based on how and how much I wear it, environmental conditions, etc. It's very, very good, but not as perfect as the timing machine says it is.

So, to answer your first question, your watch was hooked up to a timing machine for a couple of minutes and its result is what was reported to you. My machine will only resolve full seconds/day, but more sophisticated machines are capable of tenth second resolution (not really necessary, but we all like accurate sounding numbers, even if they don't mean much - but that's another discussion from an old school analog guy who never really bought into digital ... :-s).

To answer your second question, there will always be an "offset" between what a machine says your watch's rate should be and what it actually is on your wrist when you wear it. Your wear patterns simply do not align with the static testing equipment that was used to regulate your watch. Give the same watch to someone else and they will most likely experience a slightly different rate.

Without taking the time to accurately measure a watch's rate over the course of several days/weeks, then regulating it, then measuring again for several days/weeks and analyzing the results, the best a commercial shop can do is take a "best guess" about the perfect regulation for you. Usually this is a little bit fast because most people prefer a watch that runs a little fast to one that is a little slow.

If all this makes sense, and you want the best results possible, learn how to regulate your watch, buy a timing machine, and prepare for the pursuit of perfect timing to consume a good portion of the rest of your life

HTH

 

[gaijin]

Thanks for the info!

When a watch is timed with such a machine, do most watchmakers only do it in one position (ie, dial up) or in multiple positions? The fact that I was only give one measurement suggests it was only in 1 position, no?

Also, if a watch can be timed in only seconds or minutes, why can't better accuracy be attained? For example, if mine is +4.7 sec, would it not be possible to tweak the regulation screw a tiny bit more and re-measure? The only way it wouldn't be feasible would be if the movements had to be put back into the case, which I don't believe is the case.

Depending on the movement, there can be significant positional variation.

Let's take a hypothetical watch which measures as follows on the machine:

Dial down: +15 s/d
Dial up: -5 s/d
Crown up: +5 s/d
Crown down: -5
Crown left: +15

If only one number is reported to you, it could very well be an average of these 5 positional measurements.

In our hypothetical case, the average would be +5 s/d.

The problem is, "tweaking" the regulation screw a little bit does not yield a linear result for all positions. A small adjustment in one position may result in a big difference in another position.

Without spending a great deal of time, it may not be possible to acheive a better average than +4.7 s/d.

And besides, what it says on the machine is NOT going to be what it does on your wrist.

Please remember, we are talking about a sensitive mechanical device that will change its performance based on how you move every day, what temperature it is, how you rest it when you take it off, how active you are (amount the watch is kept wound), etc., etc.

A matter of a few seconds/day is trivial in the overall scheme of things - and what you observe as +9 seconds today will probably be a little different tomorrow.

Understanding is one thing, obsession quite another :-s