Automatic Watch Regulation
Discussion
The COSC Certificate should indicate in which position the movement loses least time, and so you can try leaving it in this position overnight.
For example, my 7750 loses least time when positioned "crown down" so that's how I store it.
For only -5seconds, I think you could do more harm than good opening it up.
For example, my 7750 loses least time when positioned "crown down" so that's how I store it.
For only -5seconds, I think you could do more harm than good opening it up.
BOR said:
For only -5seconds, I think you could do more harm than good opening it up.
Not if it's a "confirmed" -5 seconds according to his wear pattern. In that case it's a very easy adjustment to measure the rate in one position and regulate it 6 seconds / day faster in that position. That'll come damn close to a 1 s/d gain as long as he keeps to about the same wear pattern and within a second or two for any normal use.One thing that can throw them out on factory regulation, btw, is wearing on the right arm. That gives it far more time in the crown up position than is usual, and crown up is where much of the positional error is usually "dumped" during positional timing because not many people walk round with their arms above their head.
BOR said:
Thanks.
Just out of interest, do you know why different positions have different effects on accuracy ?
I assume it is somehow related to bearing friction?
The short, non technical, answer is "many, many reasons" but that's probably not useful Just out of interest, do you know why different positions have different effects on accuracy ?
I assume it is somehow related to bearing friction?
Normally (assuming no major faults) the biggest effects are frictional from the balance pivots.
Friction at the pivots will reduce the amplitude (degrees of rotation) of the balance wheel, and a simple balance / spring system will run slightly faster at a low amplitude than at a high one. Which is why basic watches tend to run faster as they wind down rather than slower, as you might intuitively expect.
When the watch is sitting dial up, or dial down, the balance is horizontal and it's running on the very tips of its pivots, which are typically around .003 inch diameter. That's the lowest frictional position so the watch runs with large amplitude and slowly.
When it's "on edge" the balance is resting on the sides of the pivots, friction increases, and the amplitude drops off - typically by around 30 - 40 degrees of rotation. That makes the watch tend to run faster when it's crown down, up, left or right.
But there are also tolerances and clearances in all the moving parts, so their precise positions wrt each other will also change in every position, which also affects power transmission -> amplitude -> rate. So there will be a difference between, say, crown up and crown down.
The next effect is introduced by the hairspring, which should "breathe" in and out perfectly concentric to the balance axis. But there's no such thing as perfection in machines, so the hairspring will create a side-load on the balance pivots as it expands and contracts. That side load will vary depending on amplitude, as the (slightly off-centre) spring expands or contracts more or less and it'll either add to or reduce any side load from the weight of the balance, so affecting amplitude itself.
If the spring is significantly off-centre then, as it breathes, its centre of gravity will also tend to rise and fall (depending on the position of the watch) which takes energy (reducing amplitude) and affects the effective strength of the spring itself by either adding to or reducing its force, again depending on position.
The last major effect is any fault in static balance of the entire balance wheel / staff / roller / hairspring assembly - which in effect gives a "heavy spot" on the balance rim. This will either add to or subtract from the spring's force depending on position of the watch and the amplitude of the balance.
A heavy spot can cause, for example, a loss in the crown up position at full wind but a gain in the same position at low wind, while giving the opposite (gain followed by loss) when the watch is crown down.
There are a lot of other, minor, factors (such as poise of the pallet fork) that will have an effect but the above cover most of it.
Most of the errors are unavoidable to some extent or another so fully timing to positions involves minimising such differences as far as possible then introducing other deliberate "errors" (such as slack between the index pins and hairspring) to approximately compensate for any residual effects. It's the point where watchmaking moves away from science and into art.
The consistency is, frankly, astonishing when you look at it in mathematical terms.
If you consider the limit for COSC of +6 d/day, that represents an error of 0.0069%, consistently, day-in day-out for a spring-powered, mechanical, item based (in general) on principles established well over a century ago.
Still not as good as my 1978 quartz which manages around +0.04s/day, but that's just down to running at around 8k times the frequency - the errors on individual beats cancel out better
If you consider the limit for COSC of +6 d/day, that represents an error of 0.0069%, consistently, day-in day-out for a spring-powered, mechanical, item based (in general) on principles established well over a century ago.
Still not as good as my 1978 quartz which manages around +0.04s/day, but that's just down to running at around 8k times the frequency - the errors on individual beats cancel out better
Variomatic said:
True, but bear in mind that COSC certification is for the bare movement, before casing, and carries no guarantee of what it'll do in the finished watch 
Very true, but Omega will honour COSC as a standard to which their watches are regulated, usually much higher level of accuracy, and will adjust for free under warranty Easy regulation then. Strap it to the machine, take a reading in one position, then adjust to read 5 sec / day faster in that position. Should put it within 2 s/day or so for your wear pattern. Doesn't even matter if the machine agrees or not with what you've been getting from it (they don't always!)
I agree that the accuracy of an automatic can be astonishingly good. While a "cheap" movement, my Spitfire Chrono only need resetting for time changes, otherwise it is on the button. And that after eight years wear with no servicing. My Monster is under a minute a week, which at the price is pretty spot on in my book.
Gassing Station | Watches | Top of Page | What's New | My Stuff