Quartz waveforms and more fault finding
Like Tree3Likes
  • 2 Post By Joe Horner
  • 1 Post By Archer

Thread: Quartz waveforms and more fault finding

Results 1 to 7 of 7
  1. #1
    Member
    Join Date
    Nov 2011
    Location
    Anglesey, UK
    Posts
    1,551

    Quartz waveforms and more fault finding

    Just acquired a new toy, courtesy of Amazon (also available elsewhere) which I thought might interest some of you. It's a £20 digital oscilloscope that came in kit form:

    Took about an hour to assemble, and obviously the first tests were on watch circuits. I knew what to expect, but never had a way to see for myself. Until now :)

    The patients were a couple of ETA 955.112 movements I had lying around at home. One was completely blocked, the other was running but due service based on not being done for at least 5 years. Results in following posts.
    Attached Images Attached Images


  2. #2
    Member
    Join Date
    Nov 2011
    Location
    Anglesey, UK
    Posts
    1,551

    Re: Quartz waveforms and more fault finding

    First the non-runner.

    Connecting the scope across the coil terminals showed a stream of 25 pulses which lasted around 25 milliseconds in total. That repeats every second, with reverse polarity (so the next stream would be "downwards" one second after this one).

    If you look closely, you'll see that the first 10 are wider (ie: power is applied for longer) than the last 15. That's because those first 10 are the "normal" drive pulses and the rest are an attempt by the circuit to free the blocked train.

    Name:  CompletBlockage.jpg
Views: 40
Size:  146.5 KB

    Interestingly, if you remove the rotor completely, then you get exactly the same effect. That's because the circuit senses the rotor moving by the back EMF (reverse voltage) that the rotor generates in the coil by moving.Name:  NoRotor.jpg
Views: 36
Size:  182.1 KB

    You can see that by running it with only the rotor fitted, which amplifies the back EMF because the rotor can move really fast with no other load. Note the sharp spike downward just after the drive pulses:

    Name:  RotorOnly.jpg
Views: 39
Size:  163.3 KB
    Last edited by Joe Horner; 2 Weeks Ago at 02:05.

  3. #3
    Member
    Join Date
    Nov 2011
    Location
    Anglesey, UK
    Posts
    1,551

    Re: Quartz waveforms and more fault finding

    So, what of the healthy(ish) movement? I've reduced the timebase on these next two to show the pulse train more clearly. Note that the whole train is about 8 milliseconds long, instead of the 25mS for the blocked train above.

    In its "due service" state, you get the normal drive pulses but no obvious back EMF. There's still just enough for the circuit to sense but it'll likely miss some and apply extra drive, at least on some seconds. The main effect of that will be a gradually reducing battery life as power is supplied to the coil more often and for longer:

    Name:  SlightRestriction.jpg
Views: 36
Size:  122.9 KB

    After a quick clean & oil, there's now a clear back EMF dip below the baseline at the end of the main pulse train. It's nowhere near as pronounced as the "rotor only" example because the rotor is slowed by having to drive the rest of the train, but it's there and the circuit will see it clearly:

    Name:  GoodTraceStretch.jpg
Views: 36
Size:  157.5 KB


    The same trace but with the time-base compressed to match the ones in the previous post. The back EMF "dip" is a little clearer here, and (comparing to the blocked train above) so is the difference in the amount of power needed each second:

    Name:  GoodTrace.jpg
Views: 36
Size:  138.1 KB


    And that's why a quartz watch that's overdue service will start to eat batteries!
    maillchort and Klaus Tickalot like this.

  4. Remove Advertisements
    WatchUSeek.com
    Advertisements
     

  5. #4
    Member ExpiredWatchdog's Avatar
    Join Date
    Feb 2019
    Location
    Silicone Valley, California
    Posts
    574
    Joe: Thanks for the thorough analysis, I’m gonna have to dust off the Tectronics 7736 & see what I get on my ETA 912.002 once I finish up the service.
    If it's a good meter, it's a Fluke.
    Hang my self when I get enough rope. (It's a Tubes lyric, from the mid 70's; song's called "White Punks on Dope". Enough said?)
    The difference between 12 and 21 is dyslexia.

  6. #5
    Member
    Join Date
    Nov 2011
    Location
    Anglesey, UK
    Posts
    1,551

    Re: Quartz waveforms and more fault finding

    Just keep in mind that different calibres may have different details in the trace (number of pulses, exact timings etc) but the overall features should be the same for any with power compensation (basically anything worth servicing nowadays!).

    The key points are more power delivered to a tight or blocked train and back EMF as the rotor moves (will always be clearer with the rest of the train removed!)

  7. #6
    Member
    Join Date
    Apr 2009
    Location
    Canada
    Posts
    4,806

    Re: Quartz waveforms and more fault finding

    Quote Originally Posted by Joe Horner View Post
    Just keep in mind that different calibres may have different details in the trace (number of pulses, exact timings etc) but the overall features should be the same for any with power compensation (basically anything worth servicing nowadays!).

    The key points are more power delivered to a tight or blocked train and back EMF as the rotor moves (will always be clearer with the rest of the train removed!)
    Interesting comment about what is "worth" servicing. Typically the only quartz movements I service are vintage where a replacement is no longer available, or originality is a factor for a collector. So I'm far more likely to service a watch that does not have asservissement than I am to service one with it. Most modern quartz movements are cheaper to replace than service, so at least in my shop the newer movements are the ones not "worth" servicing. I suspect we have a slightly different client base though, so I get what you are saying.

    Just for the sake of completeness, I'll show graph of the motor pulse from an old school "dumb" quartz movement that has an open loop system, and no feedback from the movement of the seconds hand:



    That is my own personal CWC issued G10 by the way. The motor is on 100% of the duration of the pulse, and it just "assumes" that the hand ticked one second forward when the pulse happened.

    Here's a modern movement pulse graph showing the chopped pulse, the pulse width, and the drive level:



    I can often tell if the movement is in good shape without powering it using the machine, just by placing it on the inductive sensor on the timing machine and looking at the drive level. If it's high, I know that likely the consumption numbers will be high when I do the remainder of the electrical checks without the battery.

    Although many watch collectors consider quartz watches to be lesser watches, with some describing them as a "battery and dividing circuit" there's a lot more going on than most people realize in these.

    Cheers, Al
    Joe Horner likes this.

  8. #7
    Member
    Join Date
    Nov 2011
    Location
    Anglesey, UK
    Posts
    1,551

    Re: Quartz waveforms and more fault finding

    I thin the main difference in "worth" servicing is likely access to the replacements.

    Most of the newer ETA (and, obviously Omega etc) range are getting scarce, if not effectively unavailable without that magic account. Which makes them worth servicing unless you're seeing enough to justify the costs of investing in accountship!

    Must admit the Witchi dedicated display does a nicer job than my new toy with a boggo generic 'scope layout, and interesting to see the difference in wave shape between the sharp square wave voltage measured across the coil and current draw with a leading slope from the coil inductance.

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

    Posting Permissions

    • You may not post new threads
    • You may not post replies
    • You may not post attachments
    • You may not edit your posts
    •