Precision pendulum techniques

There seem to be at least 3 current threads running on this topic, each one started around a specific project but a lot of overlap and cross posting. I wondered if it might be a good idea to have a thread to pick up the various comments and discussions about precision "pendulum science" so that the "project" threads can focus more on the projects themselves?

To kick this off, since the subject of Harrison's clocks has come up, I thought that I'd post this link to a collection of 3 articles that I wrote a few years back, two in Horological Science News with my friend Andrew Millington, and one (summary) in Horological Journal. I put these together for SoD yesterday but they may be of more general interest.

https://drive.google.com/drive/folders/1BiBOr6Ep4ZcK9nwe3kslz8OVEx9blqLr?usp=sharing

**LINK**

Well played, Sir

MichaelG.

John

Thanks for the link to your articles which I've just glanced at now but will read in depth later.

I'm going to the Frodsham workshop tomorrow and will see what info I can "extract" from the guys there that worked on finishing the Burgess Clock B. Of course they also made a replica of Harrisons H3 and finished Derek Pratt's H4.

CS

I knew I had seen this … but it’s taken a while to locate it again

Fedchenko’s rather special pendulum suspension spring is described: **LINK**

https://articles.adsabs.harvard.edu/pdf/1957SvA.....1..637F

MichaelG.

This, from the Letters in Horological Journal, May 1941 may be of interest:

.

With the advantage of electronic camera-sensors at our disposal … this should be relatively simple to emulate.

MichaelG.

.

Ref. __ https://clockdoc.org/gs/handler/getmedia.ashx?moid=57393&dt=3&g=1

Note: Page 180 of the book is page 208 of the PDF, but it is really worth reading the whole chapter.

Edited By Michael Gilligan on 10/02/2023 15:49:51

I feel the need to express my understanding of Precision Pendulum mechanics, and this thread is probably the best place to ‘get it out of my system’.

I expect to be ignored and derided in approximately equal measure … but would be delighted if someone could engage in a reasonable discussion:

The Gilligan Manifesto is brief:

There are just two fundamental features of a time-keeping pendulum

1. its length
2. its angular excursion

and if we are to achieve precision then those are what need to be quantified.

There are many variables to consider …

but these ‘jointly and severally’ serve to vary 1. and/or 2.

MichaelG.

and 3 it’s losses

and 4 any motion outside of the plane of swing

and maybe 5 any oscillation other than simple harmonic in the plane of swing

just thought I would add those for consideration

regards Martin

And air density unless it's running in a vacuum.

.

Martin and Duncan

Thanks for your comments, but I intended that my closing remark should cover what you have mentioned.

MichaelG.

Air density changes the apparent value of gravity, which isn't covered by your 2 points

Flexibility of the rod and suspension are not really external variables but part of the pendulum in my book.

regards Martin

I don't know. I always thought the two fundamental features of a time-keeping pendulum are (1) its period, and (2) its period.

On a more serious note: I've wondered why the old-school "electrical timekeeping" pendulums didn't impulse electromagnetically. After all, they are equipped with an electromagnetic reset of the mechanical impulse mechanism. What was the problem? I suppose it was an inability to regulate an electronic impulse adequately?

Edited By S K on 10/02/2023 20:08:44

Fair cop … both Martin and Duncan

Although, clutching at straws, I might argue that these factors all change the actual or apparent length of the pendulum and/or the in-plane arc of the swing.

What I am trying to do is distill the essence of a real pendulum, as a next step from the simple hypothetical model of a dimensionless bob on the end of a weightless string, and isochronous.

There are endless complexities, but I thought it worth a go at this because the general topic of ‘pendulum mechanics’ seems to be attracting some interest on the forum, and it might help ease people into the subject.

MichaelG.

Edited By Michael Gilligan on 10/02/2023 20:15:12

.

I would commend the ‘Electrical Timekeeping’ book [linked above]

and also ‘Electric Clocks’ https://clockdoc.org/?moid=57392

MichaelG.

Edited By Michael Gilligan on 10/02/2023 20:32:48

I've looked through it, but didn't find a discussion about that. I'm just going to go with an inadequate ability to regulate an electronic impulse, e.g. due to battery voltage declining, etc.

.

I hope you saw my edit … The slimmer volume contains a good history of early developments.

Same author, same ‘old fashioned’ style that you disliked before … but worth the effort of reading.

MichaelG.

.

Edit:__ Do bear in mind that Hope-Jones is describing the failings of the electromagnetic impulsing systems at a time when switches were mechanical things that robbed energy from the pendulum … his Synchronome Switch is perhaps more significant than the change from electromagnetic to mechanical impulsing.

Edited By Michael Gilligan on 10/02/2023 21:02:06

I would consider the action of a roller falling under gravity for a fixed distance down a slope to be a more precise and constant impulse than an electromagnetic pulse even with fairly sophisticated electronics. As has been said the electromagnetic reset of the gravity arm on the synchronome is to ensure that the pendulum remains detached, that no energy is taken from it for the reset.

A further refinement would be to replace the count wheel with electronic sensing and trip the gravity arm from that.

Something which is effectively done in the Shortt Synchronome pairing.

regards Martin

Edited By Martin Kyte on 10/02/2023 22:15:28

John Reynolds described a 'detached falling ball' clock in BHI journal. This had gravity impulse but Hall effect sensors to drop the ball. I do have a copy of the article, but it is probably copyright. I'll dig it out tomorrow and do a precis. My involvement was to provide some sensors and a control board to lift/drop the ball, and divide by the right number to drive a slave clock. It will come as no surprise that this was based on the ubiquitous Arduino.

But having said all that, voltage regulators give a very stable output, and if you actually control amplitude I can't see small variations in impulse mattering over a sensible time frame. Waits to be shot down in flames.

Very true Duncan. I think there is an element of preference that starts to creep in, at least there is with me, as to what you consider to be proper to the pendulum and what can be considered external ‘helps’. I think if I went down the road of electromagnetic impulsing I would end up building an electronic clock. It’s a question of taste really. I’m an electronics engineer by profession and a mechanical’ for fun which I’m sure has a lot to do with it.

Personally I like to think of the pendulum as a mechanical system and anything that directly influences it should be mechanical too. Otherwise I would see it as a hybrid system. As I say that’s just personal taste and nothing wrong with hybrids. Quartz oscillators are part mechanical when all said and done.

regards Martin