h a l f b a k e r y
"Not baked goods, Professor; baked bads!" -- The Tick

meta:

account: browse anonymously, or get an account and write.

 user: pass:
register,

# Mexican-Waving Newton’s Balls

on a Mobius strip
 (+2) [vote for, against]

An extension of Newton’s Cradle, this office toy will entertain and pacify with its perpetually propagating wave.

A metal, Mobius strip supports rods that penetrate it at regular intervals along its centerline. Each rod can tip up to 45 degrees towards its two nearest neighbors, and its ends are topped with metal balls.

Start the process by swinging a rod, and one of its ends will hit the next rod, which swings to hit the next, and this energy exchange continues one revolution of the strip. Then, the wave is advancing in the same direction, but is on the other “side”. It hits the other end of the first rod, and makes another revolution before the continuing past the starting point.

For unending movement, an electromagnet in the base would exert a pull on the nearest balls, to overcome energy losses. The system could also be made as a large, kinetic sculpture or even as a clock!
Tick, tick, tick, tick, tick, tick…

 — FarmerJohn, Mar 27 2003

what I mean http://www.geocitie...ohnnie/mexican.html

Sounds suspiciously similar to a perpetual motion machine (or are you planning to move the electromagnet?)
 — pottedstu, Mar 27 2003

Pulse it, I would think.
 — FarmerJohn, Mar 27 2003

 Contrary to your description, the wave will progress on both sides of the strip at the same time. Also, as with water in a pond, moving one rod will set off waves in *both* directions (forward and back, so to speak).

 Since the rods are symmetric through the strip, making it a Moebius strip doesn't really add anything to the dynamics: you could do just as well with a flat circular strip.

 And, yeah, something would need to be putting energy back into the system to allow for the losses to friction.

I think you need to build some prototypes as proof of concept.
 — DrCurry, Mar 27 2003

Not only would you get waves propagating in both directions from the electromagnetically-driven ball, you would also get another pair of waves on the “other side” since deflecting a ball would cause the ball on the opposite end of the rod to deflect in the other direction. You might want to drive it at a resonant frequency so as to setup a standing wave.
 — AO, Mar 27 2003

 I tend to disagree with the above annotations, I believe that this will only cause one wave, not two in opposite directions. If you study the diagram, and have spent any time playing with those bally-bounce things, you will see that this moebius chain gang is moving clockwise(at the moment). The leftmost of the two touching balls will be pushed into the next ball on the left and transfer all of its energy, and acheive a state of rest without any other balls/pins being affected.

 This would however only work in Zero G or in perpetual freefall. Because the pins have to maintain a specific position/state in order for the mexican moebius chain gang to work efficiently.

Dr curry, this idea will not work with just a regular circle, as the pins will now be in the wrong position, to continue their journey, this will only work in a moebius.
 — e-alias, Mar 27 2003

 DrCurry://the wave will progress on both sides of the strip at the same time// Not necessarily. It depends which rod is swung first:

 In the drawing, if the left most of the two balls is swung to the left, only one wave starts.If instead the highest ball is swung to the right, two waves will start, but the wave going counter clockwise (on the inside) will stop at the bottom where the gap is.If the two balls farthest apart are hit together, the resulting waves will meet at the top and die out.

 //you could do just as well with a flat circular strip.// The mobius strip with double acting rods is what makes it work. A circular strip would stop after max one circuit.

 //something would need to be putting energy back into the system// See my //For unending movement, an electromagnet in the base would exert a pull on the nearest balls, to overcome energy losses.//!

 AO: //both directions// See first paragraphs in this anno.Time trials or breaking a light beam could drive current switch to the electromagnet.

e-alias: Exactly, though I feel if the rod distances and angles are chosen carefully, gravity should not be a hinder. After the left ball swings left it should stop (delivering its energy to the next) in the correct position for the later wave to the right.
 — FarmerJohn, Mar 27 2003

How is it suspended? (ie. what does it sit on?)
 — PeterSilly, Mar 27 2003

Sorry, still sceptical: build it, please.
 — DrCurry, Mar 27 2003

 Peter: Two or four legs on the bottom edges would woik. I didn't want to complicate the schematic.

Doctor: If this isn't already baked, someone should, since it's one of my few ideas that could relatively easily be put into effect.
 — FarmerJohn, Mar 28 2003

 I agree that the real problem is gravity. Newton's cradle works by allowing the balls to swing freely as pendulums. Gravity would act differently on each of the balls in your design, meaning that you would need to return them to position by some spring device.

Even then, some will lean on the down-side spring more and ruin the continuity of the wave. Your electromagnet would have to be so powerful that it would actually drive the whole wave.
 — ConsultingDetective, Jan 29 2004

 ’Fraid I disagree. Unlike Newton’s cradle, this device does not also use gravity for motion but only the flow of momentum. Well-balanced rods on low-friction axles can assume any angle on any position on the strip. When one end goes upward the other goes equally downward.

The main energy loss would probably be caused by the difference between two neighboring rods’ lines of rotation, being mounted on the twisted, curved surface. Since their ends travel in non-parallel straight lines, they would not meet head-on resulting in losses.
 — FarmerJohn, Jan 29 2004

 [annotate]

back: main index