I read that acceptable amplitude is generally considered between 270 to 310. My local watchmaker had finished a service on one of my watches, and showed me the results where the amplitude was 240 in a vertical position. I said I thought that was low; he said acceptable in that position. Any thoughts? Thanks.

 

Depends on the age of the watch? I received a Constellation from a dealer who said it was serviced but it was running a minute slow. The amp was only 240 and I thought it was low (hence I assumed he didn't service it). But a friend who does watch making, said its actually serviced and fine for a 60 year old watch.

Also, it seems modern pieces like Rolex Ceramic subs that seem to reach the high 290s to even 300s....(had a sub that was serviced and it was around there)

 

What's the watch and calibre?

Assuming these are at 30 minutes after full wind, then some calibres might not achieve more than 240 degrees in vertical positions. Here's a document from Witschi who are the big manufacturer for watch testing machines showing amplitude, rate etc. You could use these in general but, they wouldn't apply to all old movements with wear (and parts not replaced) and even some specific manufacturers and calibres. H means horizontal (Dial up and down) and V is for vertical (crown up/down, for example) and you make these tests at 30-60 minutes after winding fully.

The more important is the 24 hour after full wind amplitude where companies tend to specify a minimum amplitude, in any position, of between about 160 and 210 degrees, depending on the calibre. Witschi are showing a minimum of 212.5 degrees for a men's watch above but, I wouldn't read too much into that if it is different to the manufacturers spec.

Not every watchmaker is going to try and get the amplitude or accuracy to a particular level, especially if it's a cheaper service. These things can eat up a lot of time.

For me, if it's a 56x/75x in the second post and it only has 240 degrees in horizontal at full wind (+30 to 60 minutes) plus it's running a minute slow, then I wouldn't let it out of the door. I'm assuming these are vintage watches so, just made an example based on that but, if it's an 1120 Constellation, then the figures are far too poor.

Cheers, Chris

 

Thanks, Chris. It's a Day-Date ref 1803, cal 1555.

 

It's very important to know the caliber, because what is "acceptable" will vary wildly between manufacturers, and even between movements made by one manufacturer.

First, minimum acceptable amplitude checks are really never done at full wind, but are typically done 24 hours after full wind. I think when people make very general statements about what amplitude should be, this is sort of an extrapolation of how the full wind horizontal amplitude relates typically to that formal 24 hour check.

I do the same thing when I get a watch in - fully wind it and check the horizontal amplitude, just to get a feeling for how it's running, but this is different than a formal tolerance or factory specification.

What is checked at full wind is the maximum amplitude, and that is done in a horizontal position, since that it where the amplitude should be the highest.

Looking at the Rolex specs I have for the Cal. 1555, there are a few things to note:

1 - The minimum amplitude, 24 hours after full wind, should be no less than 200 degrees in any position - typically the lowest positions will be pendant (or vertical) positions, so as long as the amplitude on your Cal. 1555 does drop more than 40 degrees in the 24 hours after it's fully wound, it meets the specifications based on the numbers you have provided.

2 - The maximum full wind amplitude should be no more than 280 - this is common across the 1500 series movements, because higher amplitudes will cause rebanking on these calibers. The more modern movements (3000, 3130) have the max. amplitude at 300, so the geometry of the impulse pin and pallet fork horn allow for more rotational displacement before rebanking occurs. So based on this if someone was taking the high end of what you have read is "acceptable" it would be 30 degrees above what Rolex considers the maximum, and the watch would probably be rebanking.

Hope this helps.

Cheers, Al

 

Thank you very much, Al. Under your number 1 above, just to be clear: did you mean, as long as the amplitude doesn't drop more than 40 degrees?

 

Instead of looking at manufacturers' specs, for knowledge's sake one should focus the attention on the relationship between the amplitude of the spinning balance wheel and the disturbance given by the escapement. In fact, manufacturers choices are always the result of a compromise for various reasons and might not put the principles into practice. Ideally, the larger is the amplitude, the less is the disturbance of the escapement on the isochronism of the torsional oscillator. As an example, coaxial escapements can work at a lower amplitude because in principle their disturbance on the isochronism is reduced.
A few advanced horological treatites, such as the magnificient Théorie Générale de l’Horlogerie by Léopold Defossez, deal with this matter in depth.

 

Thanks for weighing in. Is this relationship quantifiable, or done via visual inspection?

 

It is certainly quantifiable, albeit not via a simple visual inspection. It depends on both geometrical and mechanical parameters of the oscillating system. However, as a general rule, in the past quality movements tended to show large amplitudes and low lift angles.

 

The minimum is stated - I'm sure you can figure it out from there.

 

I'm not a watchmaker it must be said, however I agree with @Archer and @ChrisN that just because a watch is showing good amplitude and lines, doesn't mean it's perfect/acceptable to a watchmaker with high standards.



My recently received Landeron 248 Nicolet chronograph will be due in for a service at some point, owing to slight misalignment of the hands and minute counter, though for someone who just likes watches, these would be almost impossible to spot.



My cheap timegrapher shows 0s/d at full wind with 312 degrees of amplitude! Great right? Well, the 1.1ms beat error isn't particularly ok, and the lift angle is wrong for a L248 - it should be 42 degrees.

That's adjusted and the difference is stark:



Especially when considered to @Mitka and @Christian's work on a L248 on full wind:



Dial down takes some time to adjust and I can see some minor errors - my knowledge is lacking in what the blips are, though I would assume any combination of dirty jewels/palets/poise/etc.



With the chronograph running the amplitude at full wind is low. It does seem though that the watch may have had a service in the past though there is no marking to suggest this.

 

Looks a nice watch

When you're testing these, you should have the crown against the microphone and the dial up (you can place dial down but, this is not the typical way), so the crown should be on the left in your first picture and then the microphone will stand the best chance of picking up the noises. Then you rotate the microphone. If the strap gets in the way, do up the buckle.

In the dial up picture, it is likely that what you're seeing on the screen is one dirty pallet stone and that gives the slightly irregular look to one line. The beat error is a little high but it isn't too bad at 1.1 ms. Some movements have a movable stud carrier so setting the movement in beat is easy as you move the stud carrier - on these, you have to turn the collet at the balance so, it's very unusual to get a beat error exactly zero although I would usually aim for lower than you're seeing.

I think the dial down is due to the microphone not picking up the sound well - try it by mounting dial up and turning the microphone.

For me, and I think this is what you're saying as well, you can use this tool to identify problems but, you can't use it to state that everything is fine. It's only testing the train and escapement and there is a lot more to the watch than that plus, a dry movement will often have good amplitude so, it's not a definite indication of a watch not needing a service.

Cheers, Chris

 

Would the other part for you be a visual inspection of the movement, particularly if the service history is unknown?

 

Timing machine won't always tell you the condition of the movement with certainty, so a visual inspection is always recommended...

 

@Chris - yes, that's more or less what I meant. A watch may look like it runs well, however as @Archer has alluded to before, and on eBay listings that say the watch is Serviced, visual inspection is definitely needed.

Full wind chronograph with crown facing microphone dial up:




Dial down:



This watch will be seeing a service soon, I don't plan on ruining how nice it is (and thank you for the compliment ^^)

 

Yes, the timing machine doesn't tell you anything about the lubrication of the keyless works, for example. In fact, dry jewels in the train can still show good amplitude and a single position reading won't tell you anything about how the balance and hairspring are set up for multi position running. Al has a thread here, somewhere, showing wear in an 1120 (?) keyless works due to lack of lubrication and it can all get expensive quickly.

These are not too bad, with under 30 degrees drop when the chrono is on. I'd agree though, have it looked at as the amplitudes are a bit low and it seems to have an unknown service history.

Am watching Germany-Korea and Mexico-Sweden while enjoying a beer, or two. Far more important than watches.... Soccer (to those not UK based) is a game where "22 men kick a ball about for 90 minutes and then Germany win". Having lived in Brazil and Spain, I'm not convinced about the last part but as a supporter of England, who always have problems with Germany, "come on Korea and Mexico".

Cheers, Chris

 

90 Minutes? Seems like many hours when I've watches a game...all for a "nil-nil" tie...

 

also worth noting that if you timing machine isn't set to the right lift angle for the particular calibre the amplitude reading may not be correct.

 

So, Canada didn't qualify then, Al

Germany out so England's chances just doubled!

It won't be, as you say, but I think @GuiltyBoomerang showed that a few posts before.

Cheers, Chris