Chaotic pendulms

A few years ago I made a 'chaotic' double pendulum. The only record I have is is a Youtube video of a prototype.

I've been approached by a local shopholder who would like to put one in her window if I could make it run continuously.

I think that if I could by some means deliver a periodic impulse to the main beam to compensate for frictional loss it should continue to show chaotic behaviour. It would have to surely?

It's a bit muddy in my mind though and I'd appreciate any suggestions of how to tackle this.

Robin.

Edited By Robin Graham on 08/09/2023 23:24:35

Edited By Robin Graham on 08/09/2023 23:25:06

..scratches head..

"What are you doing..? ..your coffee's gone cold.."

..50 minutes an hour have slipped by..

more anon.

Those desk toys may give some inspiration

could be addictive...

Edited By Ady1 on 09/09/2023 10:19:42

There's a way to drive pendulums with a magnet on the end that interacts with a coil. The coil is connected in a simple transistor circuit which, when the pendulum swings past the coil in either direction, is triggered to pass a current through the coil which repels the pendulum to keep it swinging. For an example see:

**LINK**

...but there are loads of links on the web. I've used the principle in a clock but using an Arduino instead of a transistor.

**LINK**

I think that 4QD of loco speed control fame used to sell a self-contained unit to do this.

I thought it was Minimag chap, perhaps both. The neat trick will be regulating it so it doesn't just go faster and faster in one direction.

Duncan, you're quite right, it's Minimag. The SPC1 or SPC3 (or is that no longer made?).

**LINK**

The SPC1 has a cored coil which makes it unsuitable for clocks but the SPC3 is un-cored. But for this application cored might be fine. Adjustment would probably be either changing drive voltage (1.5 - 4.5Vdc) or just proximity to magnet.

Robin that's cool - how did you make the bearings please? I'm quite tempted.

Looks like Robin has been volunteered to write it up in ME or MEW, will make a change from the endless series about valve gears (I'm allowed to say that)

Thanks for replies.

DiogenesII: ".50 minutes an hour have slipped by.."

I think you may have anticipated my next question! When I first made one of these things I wondered what would happen if I somehow generated 'ticks' on a clock hand from, say, the secondary pendulum. Backward or forward ticks according to the sense of rotation. I reckon that 'on average' the hand would go in one direction albeit erratically.That's based on a perhaps half-baked notion of 'handedness' . If the pendulum is started in a clockwise direction say, then the the total angular momentum (call it positive) will diminish because of friction but can never become negative.

Ady1: Thanks for the links. Interesting. Those globes look beautiful.

Thanks for the info about electromagnetic pulsing (and to Duncan as well). I shall experiment.

The bearings are bog-standard deep groove double shielded. Probably NSK, nothing special. I did make a prototype using ceramic ball bearings but the small improvement in running time wasn't worth the cost.

Robin.

Edited By Robin Graham on 09/09/2023 22:42:59

Edited By Robin Graham on 09/09/2023 22:58:02

I do not understand how this thread relates to the OP. That aside, there was such a device in (I think), CR Macintosh Hillhouse at Helensburgh. That seemed to run forever powered by the breeze. As long as you could time the impulse an electromagnet would provide the energy to sustain operation.

Fascinating things to watch. Are there published designs around?

Martin