[time-nuts] The pendulum problem...

[Brian, WA1ZMS]

Thanks for all the ideas and replies.  Let me see if I can address all
points in just
one e-mail.

1) The clock(s) in question are very costly and to modify them in any way
would
instantly kill the value. These are part of history collection in 100%
original condition.

2) These clocks wind with a crank handle and winding rolls the cable back
onto the
main-wheel drum.  (Chain drive clocks were a ~100 year later design in
America)

3) The pendulum is 1 meter long and takes a full second to travel from one
end to the
other. So 1PPS or 0.5PPS synching is easy to do with a magnet, etc...

4) The escapement is of the anchor type, and as such when you wind the
running weight
you are driving the main wheel backwards. Such an escapement will run
backwards during
the winding and so I lose about 20 seconds or so during the winding.  The
speed of the wind
also can allow for a typical forward second to happen between the clicks on
the drum.
Sometimes I get a loss of 15 seconds, sometimes 20, etc...

5) The pendulum is still swinging during the wind. It's a 1kg weight on a 1m
rod. Takes
lots of energy to stop it.

6) The escapement shaft comes through the front dial to a small second hand
and so you
can see the second hand either pause, run forward, run backwards during a
wind.

I am concluding that without a fancy way to wind such a clock, it will only
be locked to
an external source during a typical 7-day run.  I'm asking for a solution to
a problem that
exists only as a want, not a need. Nevertheless, it is still very satisfying
to hear the tick
of such an old clock as the trigger LEDs on a 5370B blink at the same rate.
It was TVB that
pointed out to me the idea of just how many of our Rb's, Cs's, and OCXOs
will still be
running 200 years from now.  That thought still gives me pause.


-Brian, WA1ZMS