No subject

Sat May 1 12:37:47 UTC 2010

The main issues in designing a FP are the sustaining  system, avoidance or
damping of elliptical motion and safety considerations in case the wire
breaks. Of course keeping fingers and draughts away is also a consideration.

Sustaining systems are mostly electromagnetic, either with a ring shaped
electromagnet at the top near the suspension point controlled by optical
sensors, or one or more coils below the centre point with a magnet on the
bob. This acts as both a sensor and "motor". There is also a reluctance type
driver described using mains frequency solenoids. The most elegant system is
the parametric one where the suspension point is oscillated up and down
sinusoidally at twice the pendulum frequency and there are no horizontal
forces acting on the bob at all. I found a very complex mathematical
analysis if that. It would be an interesting challenge using optical sensors
and a stepper perhaps to move a cam or crank to realise that.

Avoidance of elliptical motion and increasing the Q of the oscillator is one
of the reasons why most FPs are so long with heavy bobs. Despite this I
found some articles on short FPs, including one hanging from the wall and
used as a clock with a pendulum less than a metre long. Elliptical movement
is often controlled by a "Charron ring" which interacts with the wire to
limit ellipsoidal movement. There are also magnetic eddy current damping
systems described and one elegant method which uses a precisely timed pulsed
sustaining system to cancel elliptical motion.

As was pointed out, FPs are not primarily time keeping devices but there is
a relationship between the period of their precession and the rotation of
the planet, which is also dependent on Latitude. Perhaps unsurprisingly,
someone has described a FP clock which required an electronic system to stop
it for a few hours in the middle of the night to sync its movement to the 24
hour cycle. One can easily see GPS control creeping in there :-)

I hope the discussion continues, It's been great so far. I'll keep the group
posted on progress.



More information about the time-nuts mailing list