Analog Meter Cycle Life?

[The Doktor]

 I'm considering building a clock using three analog meters.  Each meter would have a custom scale, 1 going from 1 to either 12 or 24 for hours, and two going from 0 to 60 for minutes and seconds.   I am curious as to whether or not analog meters suffer from metal fatigue failure after a very large number of full-scale deflections. The meter displaying seconds will be experiencing 1440 full-scale deflections everyday. That is about 525,600 deflections per year. I will be using cheap meters from aliexpress that look like the old round Weston meters. Are these likely to fail rather quickly, or do they have the potential to last for a few years in the service?

[strawberry]

higher sensitivity meter should come with thinner spring.

clock spring go 1 cycle per second 24/7.

[Kleinstein]

The springs / tension bands are rather soft and not made to store lots of energy. So they should run with rather low stress and pretty much last forever. So no fatigue expected.
To keep the calibration the meter movements also don't want to have fatigue to the springs.
It at all there could be a bit wear when going back from 60 to 0 seconds very fast, hitting the lower mechanical limit.

[807]

Sounds like a cool project. Is it your own design?

[robert.rozee]

similar project here:
https://www.thebackshed.com/forum/ViewTopic.php?TID=6621


cheers,
rob   :-)

[CatalinaWOW]

While I don't think there will be any wear issues in the meters, I do agree that the return to zero will be a problem.  Possibly damage as Kleinstein mentions, but also real issues in properly representing the first few seconds of the minute.  I would guess that the mechanical time constant of the meter is a large fraction of a second, and either the damping is small so there will be substantial ringing, or it is large and slew rate limitations will make the keep the meter from returning to zero for more than a second.

Perhaps setting the seconds meter to read 30 seconds full scale, and having a reverse scale from 30 to 60 could be a solution.  Running the seonds meter with a triangle wave instead of a sawtooth would eliminate the need for a near instantaneous transition.  Taken to a high enough degree you could have the scales be backlit and only the appropriate one illuminated during each half minute.

[Conrad Hoffman]

I'd be worried about pivot wear with pivoted meters. Taut-band is probably better for this. Any of these things are designed with a huge factor of safety with regards to material properties, so metal fatigue shouldn't be an issue.

[robert.rozee]

hmmm, i do wonder if any of those cheap meter movements are taut band? or just hairspring and bearings. i've certainly seen plenty of hairspring meter movements that have worn out.

the count up/down idea is a neat one. for hours you could have a blue scale counting up from 12 and 1 to 11 above the scale line, and down from 12 and 1 to 111 below in red. then use a couple of LEDs, one red, one blue, to indicate which scale (AM or PM) to use. the same for minutes, except 5 to 30 in steps of 5 counting up (labelled "past"), and 25 down to 5 in steps of five labelled "to".

ie:
12    1   2   3   4   5   6   7   8   9  10  11         AM
-+----+---+---+---+--+---+---+--+---+---+---+---+-
      11  10   9   8   7   6   5   4   3   2   1  12    PM

and for minutes:
  0    5    10   1/4   20    25   1/2        past
-+-----+-----+-----+-----+------+-----+-
        5    10   1/4   20    25                  to
 


cheers,
rob   :-)

[CatalinaWOW]

Pivot bearings and hairspring are quite robust.  Every mechanical watch had one, usually with a one second period.  And even the cheap ones lasted more than a couple of years.  The good ones with jewel pivots last for decades.  Your meter representing second is accumulating life (wear) sixty times more slowly.