Yes, it picks up the ticking of the balance wheel, and you adjust it by seeing the error, then opening the back and adjusting the spring control, then put it back and measure again, until it shows the error is under 1 second a week or so. An iterative process, but it works well. Others use a light source and a photocell to detect the movement of the balance wheel or the escapement inside the watch to get the information.
I have in the past use a piezo buzzer as a vibration pickup for my frequency counter, have it taped to the side metal while i adjust at the back of automotive mechanical clocks, means you can trim one in in 40 seconds as opposed to 2 days,
where as most of these ones tend to be based on high gain microphones, you can still get away with a piezo if you have a lot of gain,
Isn't it a little ironic to adjust mechanical watches using a machine based on a quartz oscillator?
1 second a week would be extremely good for a mechanical watch, the "high end" COSC certified ones are only -4/+6 seconds per day (measured over 5 positions and 3 different temperatures).
I can regulate my low end Seiko automatic by placing it face up or face down. Face up it runs a little slow, face down it speeds up... This is quite typical behavior of mechanical watches.
Isn't it a little ironic to adjust mechanical watches using a machine based on a quartz oscillator?
If you want a practical and accurate watch, get a 10$ Casio. It will be orders of magnitude better than even the most expensive swiss mechanical movements assembled by nude virgins under moonlight. Mechanical watches are terrible timekeepers.
Time to replace the little 377 coin cell in my cheap $5 watch. It stopped 15 seconds to midnight........
In the 70's I worked for a watchmaker,he used a similar device that picked up the ticking of the watch but as the display was a printed paper roll to save paper he would just turn on the speaker of the device and listen to the ticks, he would then adjust the watch accordingly, within a minute or so he had the watch adjusted to a second or two a week. And he did not have a beard or a tongue sticking out while making the adjustments, a poke with the tweezers and it was done.
Things like this have been around for a while. Here's a nice little gem of a show I like a lot
you can see one in action.
Good watch makers tools are expensive, those so called cheap watchmakers screwdrivers found on e bay and in the pound/Dollar shops don't work out cheap the bend and break or the handles tun on the blades or they slip and damage the watch as in electronics good equipment costs but pays for itself in the end.
1 second a week would be extremely good for a mechanical watch, the "high end" COSC certified ones are only -4/+6 seconds per day (measured over 5 positions and 3 different temperatures).
I can regulate my low end Seiko automatic by placing it face up or face down. Face up it runs a little slow, face down it speeds up... This is quite typical behavior of mechanical watches.
Isn't it a little ironic to adjust mechanical watches using a machine based on a quartz oscillator?
If you want a practical and accurate watch, get a 10$ Casio. It will be orders of magnitude better than even the most expensive swiss mechanical movements assembled by nude virgins under moonlight. Mechanical watches are terrible timekeepers.
5 seconds a day is around 58ppm, while a typical quartz watch is in the single-digit ppm range, so it's around an order of magnitude better at most.
See
http://tf.nist.gov/general/pdf/2276.pdf£184 seems very reasonable for a specialist piece of kit like this.
I wouldn't be surprised if they were $1k+ before the Chinese got into the game
This one actually appears to be a clone of another Chinese one, the "TYMC MTG-1000", which costs roughly twice as much.
...and a PC containing a suitable microphone will work, using this software:
http://tokeiyade.michikusa.jp/download.html
Less people wear watches now than they used to as they've been replaced by phones.
5 seconds a day is around 58ppm, while a typical quartz watch is in the single-digit ppm range, so it's around an order of magnitude better at most.
My watch has has a very common 7s26 movement, it's specified at -20/+40 seconds per day. That's about 300ppm.
There's also an Android app to perform these measurements,
WildSpectra. It depends a bit on the watch and the microphone in your device if it works though, my watch is too quiet.
Less people wear watches now than they used to as they've been replaced by phones.
Pretty much only watch enthusiasts wear mechanical watches, I doubt they will switch to a smartphone. The vast majority wears (wore?) a quartz watch, this machine is useless for those watches anyway.
Thoughts?
I bet you that you can do it with a smart phone app: it has a microphone, with adjustable gain (20-30db typically) and a graphics display + canned user interface.
Otherwise, it is not that hard to make your own.
So you've made one then?
If I make one, what will you do in return?
Here's a look at the innards of a similar one, plus some info on how they work.
Thanks really nice explanation. Yea they misspelled the description. I bet the manual they include is equally as bad.
From a modern mcu perspective, those things are quite simple in term of basic functionality: you use a microphone to listen to the beats. With that, you time a known pulse train. Count until a certain number of beats have passed and you read your timer counter. Done.
So you've made one then?
If I make one, what will you do in return?
So I take it that's a "No" then!
The vast majority wears (wore?) a quartz watch, this machine is useless for those watches anyway.
The one in the Secret Life of Machines video linked earlier works with a quartz digital watch.
[ The vast majority wears (wore?) a quartz watch, this machine is useless for those watches anyway.
Really ?
This Seiko Caliber 7545A says otherwise.
[ The vast majority wears (wore?) a quartz watch, this machine is useless for those watches anyway.
Really ?
This Seiko Caliber 7545A says otherwise.
No point in adjusting a quartz with another quartz, you'll need something significantly better. Your measurement instrument must be at least an order of magnitude better than the device under test. Also the device posted measures amplitude and beat error of the balance wheel, both are useless for quartz.
My misunderstanding : I thought you were saying that quartz watches cannot be adjusted. The "not with this device" got lost in the translation...