[time-nuts] Measuring the accurcy of a wrist watch

Ulrich Bangert df6jb at ulrich-bangert.de
Thu Apr 17 04:14:39 EDT 2014


many thanks for your help. Will give it a new try over Eastern!

Best regards

> -----Ursprüngliche Nachricht-----
> Von: time-nuts-bounces at febo.com 
> [mailto:time-nuts-bounces at febo.com] Im Auftrag von Tom Van Baak
> Gesendet: Mittwoch, 16. April 2014 19:44
> An: 'Discussion of precise time and frequency measurement'
> Betreff: Re: [time-nuts] Measuring the accurcy of a wrist watch
> > Tom,
> > 
> > can you explain what exactly you understand by "a large 
> coil of wire"?
> Sorry, by large I meant a large number of turns; the coil 
> itself is quite small. Rather the winding one myself I just 
> used the pickup coil from an old cheap plastic self-impulsed 
> pendulum clock. The wire is extremely fine and there must be 
> thousands of turns since the spool diameter is only 15-20mm 
> and the net resistance is 3.5k. Here are some iPhone photos I 
> just took:
> http://leapsecond.com/pages/Junghans/coil.htm
> > Did you make the easurements on the Junghans with a DIY 
> sensor or with 
> > one of the commercially available?
> Both. The commercial ones sold by Bryan Mumford are 
> excellent; his instrument includes signal conditioning, 
> adjustable high gain, and other useful features. It's meant 
> for watchmaker types with no electronics background. It works 
> perfectly out of the box.
> The Junghans wristwatch is extremely well engineered for 
> long-life and the leaked magnetic signal is the weakest of 
> any watch I've measured. Still, it can be measured. The 
> placement of the pickup coil on the watch face needs to be 
> optimized for best "reception", or any reception at all for 
> that matter.
> By contrast, a typical AAA-battery desk/wall quartz clock 
> movement generates a huge magnetic signal. It is so clean 
> that you can clearly see both the start (+) of the impulse 
> and the end (-) of the impulse about 30 ms later. In fact I 
> suspect it's actually 31.25 ms, or 1/32 s, since that's 1024 
> cycles of a 32.768 kHz oscillator. See:
> sensor placement: 
> http://leapsecond.com/pages/Junghans/quartz-clock.jpg
> output to scope: http://leapsecond.com/pages/Junghans/coil-aa.gif
> > I have made some basic tests with a coil coming from a 
> loudspeaker's 
> > cross over network. It has a few hundred windings, R=1.3 
> Ohms, 2.3 mH, 
> > but the only thing i receive with this coil is a strong 10 Mhz 
> > signal...perhaps no real surprise in a time nuts laboratory.
> I suspect your 1.3 ohms means the number of turns is far too 
> low. I don't see any RF here, nor even very much 50/60 Hz.
> /tvb

More information about the time-nuts mailing list