Magnet Material for Free Pendulum Clock

[Anonymous]

May I comment as a total know-nothing greenhorn?

It was my impression that the timing of a pendulum clock is solely set by the length of the pendulum itself. If the pendulum is set swinging it will go back and forth at the same rate … tick-tock. However because of mechanical (friction) losses it's amplitude will gradually reduce down to zero in the absence of some energy input either by weights in a mechanical clock or by the magnetic impulse in a solenoid clock.

I don't understand the reference to clock accuracy wrt the solenoid material. Surely it's the synchronicity (is that a word?) of what's driving the solenoid that matters.

Have I made a total twerp of myself?

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Peter Greene 🇨🇦 on 06/11/2022 00:52:25:

May I comment as a total know-nothing greenhorn?

[…]

I don't understand the reference to clock accuracy wrt the solenoid material. Surely it's the synchronicity (is that a word?) of what's driving the solenoid that matters.

Have I made a total twerp of myself?

.

Happy to answer those points:

1. Yes, of course !!
2. Yes, synchronicity is a word [see Carl Jung]
3. It’s more complicated than first appears … please have a read about what constitutes a ‘Free Pendulum’ and start discarding some of your perfectly reasonable assumptions and approximations. The devil is in the detail [which is why I remarked about accuracy]
4. Not at all !!

MichaelG.

[John HaineParticipant @johnhaine32865]

Just to expand on Michael's comments. The "free" pendulum should be subject to no other forces than gravity, a very short impulse provided to supply the energy lost over the previous swing applied exactly at "bottom dead centre", and air drag (though ideally it will be in a vacuum to eliminate this). The impulse can be from an electromagnet and this might use a ferromagnetic core. If there is any remanent magnetism in the core there could be a resulting additional force on the pendulum that could affect its period. If the remanence could change, for example through repeated magnetisation by the coil, temperature effects, or aging, it will affect the pendulum period so should be avoided.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Michael Gilligan on 06/11/2022 07:55:43:

[…]

please have a read about what constitutes a ‘Free Pendulum’ […]

.

I suggest this is probably a good place to start: **LINK**

https://adsabs.harvard.edu/full/1931PA…..39…66B

[ the website is old and clunky, but it’s an amazing resource ]

MichaelG.

.

P.S. __ @ John Haine : Thanks for the additional comments, John 

 

Edited By Michael Gilligan on 06/11/2022 10:22:19

[peak4Participant @peak4]

I had this page open when I retired last night, and refers to the Shortt – Synchronome clock

http://www.royalobservatorygreenwich.org/articles.php?article=1327

There's a brief into on Wiki HERE

https://en.wikipedia.org/wiki/Shortt–Synchronome_clock

Bill

[SillyOldDufferModerator @sillyoldduffer]

Posted by Peter Greene 🇨🇦 on 06/11/2022 00:52:25:

May I comment as a total know-nothing greenhorn?

…

Have I made a total twerp of myself?

Not at all. Asking the question makes you a new recruit into the fascinating world of pendulums. Welcome, but be aware it's addictive!

An ideal pendulum would do as you say, but the engineering problem is there's no such thing as an ideal pendulum because the timing of a real pendulum is effected by many factors:

• The size of the swing, smaller the better
• friction losses in the suspension
• suspension alignment errors causing the pendulum to fly in a curve
• The length of the pendulum varying with temperature
• Changes in air pressure varying the buoyancy of the bob and the amount of work it does moving through sticky air
• Humidity, also alters the density of air, and it can cause dimensional changes too
• Gravitational changes, particularly tidal. Also, because gravity isn't identical around the world, a pendulum adjusted to run accurately in London has to be readjusted to run accurately in New York.
• Vibration – man-made and earth tremors causing the pendulum to swing off course
• Materials changing over time. For example Invar has an excellent temperature coefficient, but the alloy is rather unstable and the resulting tiny dimensional changes can cause trouble to a
• Disturbances due to applying the impulse needed to keep the pendulum going as it gradually loses energy due to friction. Designing and making an effective mechanical device to detect where the pendulum is and to apply the minimum necessary impulse force to keep it moving without shocking the system is very difficult.
• Disturbances because 'tick' and 'tock' are unbalanced

It's fairly easy to make a pendulum clock that keeps time within about 30 minutes per day. However, the very best mechanical clocks manage about 1 second per year and are extraordinarily ingenious. For amateurs, pendulums open a rich field of experimentation and interest. With modern materials and components, how close can a hobbyist get to the performance of a 1921 Short-synchronome? Is it possible to make a pendulum sensitive enough to detect tidal forces? It's endless, including the possibility raised in this thread that residual magnetism could effect time keeping…

Dave

[Clive SteerParticipant @clivesteer55943]

Peter

Your impression is correct but the practical implementation of an accurate pendulum based oscillator is not trivial.

A major difficulty is that the bob weight on most simple pendulums has a circular trajectory which introduces a timing error( circular error) if the swing amplitude changes and the error is a square law function of amplitude change. So the timing is far more sensitive to amplitude variations than it is to say variations in pendulum length.

There have been numerous attempts at solving the "amplitude" issue with different types of escapements, methods of powering or reducing energy loss issues etc. However many of these solutions interact and when personal opinion is added to the mix there is a degree of divergence as to the "best" solution.

Things can be "simplified" by negating circular error in the first place and one solution designed by Fedchenko largely achieved this and produced one of the most accurate pendulum clocks.

Unfortunately by that time quartz and atomic clocks could out perform his clock.

Such is life.

CS

[Michael GilliganParticipant @michaelgilligan61133]

Magnetic tinkerers may find the magnetometer in this ‘Physics Toolbox’ useful:

**LINK** https://www.vieyrasoftware.net

There are Android and iOS versions available … and, thus far, I am very impressed.

MichaelG.

[Martin KyteParticipant @martinkyte99762]

Posted by Peter Greene 🇨🇦 on 06/11/2022 00:52:25:

May I comment as a total know-nothing greenhorn?

It was my impression that the timing of a pendulum clock is solely set by the length of the pendulum itself. If the pendulum is set swinging it will go back and forth at the same rate … tick-tock. However because of mechanical (friction) losses it's amplitude will gradually reduce down to zero in the absence of some energy input either by weights in a mechanical clock or by the magnetic impulse in a solenoid clock.

I don't understand the reference to clock accuracy wrt the solenoid material. Surely it's the synchronicity (is that a word?) of what's driving the solenoid that matters.

Have I made a total twerp of myself?

I'll give you another significant word, Isochronous. In a clock oscillator (pendulum and the like) it means that the period does not vary with amplitude. Unfortunately a pendulum does so vary, but gets more isochronous at smaller amplitudes. A balance wheel however can be made to be isochronous by adjusting the spring rate to match the balance wheel. This is why people go on about circular error and constant impulsing in pendulum clocks.

regards Martin

[John HaineParticipant @johnhaine32865]

Posted by Michael Gilligan on 06/11/2022 11:47:40:

Magnetic tinkerers may find the magnetometer in this ‘Physics Toolbox’ useful:

**LINK** https://www.vieyrasoftware.net

There are Android and iOS versions available … and, thus far, I am very impressed.

MichaelG.

Michael, that's a VERY cool app, thanks for pointing it out.

[Author]

Posts