[svorkoetter]

Hi Gaijin,

Thanks for the detailed reply above - definitely find it useful and thinking how I can place my watch so can get average of 0 gain a week (thinking 2 - 3 days out of the winder per week. just need to space it around the power reserve).

Note that as the watch runs down, it may speed up or slow down a little bit, although I suspect Rolexes are pretty good this way. If you want to leave it out of the winder, do it for several short periods (e.g. overnight), not one long period.

A few things I want to understand more:

1.Beat Amplitude - you mentioned this measures the swing of the movement and is a sign of wear/tear/friction - can you let me know what is the range of 'good' 'concerning' and 'bad' etc.

For the vast majority of movements, numbers around 270 degrees indicate a movement in perfect health. Much less than 240 indicates the lubrication is starting to lose its effectiveness, and much more than 300 suggests an incorrect (too strong) mainspring.

2. You mention the measurements were at a lift angle of 52 degs. - what is a lift angle?

It is the angle through which the balance wheel turns while it is actively being propelled by the pallet fork and escape wheel. If you think of the balance wheel as a swing with a child on it, and you are the parent are pushing the swing to keep it going, the lift angle is the angle of swing during which you're pushing.

3. (not related to my watch) Are all the measurements above applicable to any mechanical watch? So in theory if I could put a non-rolex mechanical in the machine I should get some interesting comparative measurements?

Yes. The machine that does this is called a timing machine. One puts the watch in the sensor, tells the machine the beat rate (e.g. 28800 bph for your watch) and lift angle, and the machine does the rest. It's all done via a sensitive microphone that listens to the movement and analyzes the sounds that it makes.

EDIT: PS. I'm a beginning amateur watchmaker, and also being an electronics hobbyist, am in the process of building my own timing machine.

 

[svorkoetter]

Hmmm balance wheel / pallet fork / escape wheel. Too much jargon! So I guess this is something that is dictated by the design of the watch right or is this something the timing machine finds out.

It's dictated by the design of the watch. You have to tell the timing machine the lift angle so that it can compute the amplitude. Back to the swing analogy, if I know how long you're pushing the swing (lift angle), and I know the period of oscillation of the swing (beat rate), I can compute how far it swings (amplitude).

 

[svorkoetter]

Looks like a very healthy watch to me. Average of +2.4 seconds per day at 292 degrees amplitude. Minimal variation between positions.

Does your Milgauss gain a few seconds within an hour or two at the same time every day? Could it be that the reference time you're comparing against loses a few seconds in that time period. For example,

• if your PC clock is fast by 2 seconds per day,
• your PC resynchronizes its clock with the Internet at 5pm,
• you checked the Milgauss against the PC at 4pm and at 6pm,

then it will appear as if your Milgauss gained two seconds between 4pm and 6pm.

PC clocks are notoriously inaccurate (much more so than quartz watches) and can easily drift by seconds per day, which no one notices because most PCs these days resynch against Internet time every day or two.

Try checking your Milgauss against a decent quartz watch, which will at worst be +/- 0.5 seconds per day.

 

[svorkoetter]

A timegrapher only measures the speed (called the "rate") of the watch during the period of time that it is doing the measuring. Any mechanical watch will change its rate throughout the day, to varying degrees, depending on:

• How wound the mainspring is (less of a problem with an automatic that you're always wearing, since it's always almost fully wound).
• Ambient temperature (less of a problem if you're wearing the watch, since your wrist will keep it fairly warm).
• Position (e.g. dial up, crown up, crown down, etc.)
• Sudden shocks, which can speed up or slow the watch down momentarily.

Testing on a timegrapher can really only compensate for position, by taking an average of six positions.

For a good quality watch movement, which your Rolex will definitely have, ambient temperature and position won't have a huge effect, especially not over the course of just an hour or two.

That leaves sudden shocks. Do you do something vigorous, such as playing tennis, golf, or football, during the period where the watch suddenly gains a few seconds? Do you have a car with a bad suspension and you drive it home on a bumpy road, with your watch hand on the steering wheel? That sort of thing is the only thing I can think of that could cause an actual short-term variation in rate (as opposed to a measuring error).

A poster in the other thread where you asked about this temporary speed-up said you should only measure a watch over a long period, but if short-term variations like the one you are seeing occurred routinely without cause, then adjusting a watch on a timegrapher would be useless, which is definitely not the case.