The unsolved secret of David Jones' perpetual motion machine [video]

"Unsolved {riddle|puzzle|mystery}" are the usual forms for a headline of this type in the UK. Part of the Beeb may have briefly lost the rule book.

Maybe "unknown solution" could work too, since it was an unsolved mystery, then someone solved it but didn't share the solution, so it's a secret to anyone else.

"Unknown secret" or "hidden secret", maybe.

My first thought was also a long-lasting battery and an electrostatic motor similar to those used in that bell. That along with the large moment of inertia of the wheel and very good (possibly magnetic suspension?) bearings, makes it possible for it to continue rotating for a very long time.

Also, for a second looking at the title, I thought it was going to be this guy: https://en.wikipedia.org/wiki/David_L._Jones_(video_blogger)

What was your guess?

They're available on AbeBooks last time I checked. Buy "The Inventions of Daedalus" and if you like it also "The Further Inventions..." which are both fantastic.

Alternatively you could read through his columns in back issues of New Scientist and Nature at your local library but won't get the additional commentary and goofy drawings.

Finally, there's a documentary on his perpetual motion machines (possibly mentioned, haven't rtfa yet) which is more about the man than the machines. I had to write to its creator to get a copy though!

Nature and New Scientist online archives? (sub. required).

Supposedly this one was made by David Jones in 1999 in a Vienna museum.

https://www.youtube.com/watch?v=2dpmP_H1LBM

He actually made four, it seems. Sibling comment shows one but I haven’t seen the others yet.

> every criticism I've ever seen of people's attempts at perpetual motion always seem to involve the fact that some part will wear out eventually.

Every informed criticism of perpetual motion machines starts and ends with the second law of thermodynamics.

Any experts will usually leave it at that, because people who want to believe in a perpetual motion machine, are going to believe in a perpetual motion machine and any deriving discussion is usually not that interesting.

I can't think of any serious criticism of a perpetual motion machine that is based on that. What are some examples you've run across?

Statements such as "some part will wear out eventually" can be indeed a bit confusing. What they really mean is "that machine is moving because it's consuming energy stored in some of its parts, and it will eventually deplete it completely". Would you argue that a car is a perpetual motion machine, provided that somebody "repairs" it by refilling the fuel tank once in a while? Same thing. A machine that moves by extracting the energy stored in one of its parts is not a perpetual motion machine, it's just a machine that runs on a very unusual fuel. On the other hand, there might well be the case of a real perpetual motion machine that needs repairing once in a while- in that case, however, it should be possible to prove that the energy stored in the part that needs replacing is not the source of the movement, either because it's not enough, or because it is going somewhere else- for example dispersed as heat.

It's difficult to make a blanket statement, but the idea behind alot of these is to take something that required energy to produce, break it down somehow, and claim that the released energy is "free" and that makes it perpetual motion.

If your machine has positive power output you have a perpetual motion machine. If you have a machine the has positive power output but also consumes a gear (and the power required to repair the gear is higher that the power output before it broke) then you have a gear-destroying machine.

The biggest objection is that any work results in dissipation, and any machine which seems to violate this is quietly cheating with some input of energy.

Aka Thermodynamics

Edit: what possible objection could you have to this?

So build one you can maintain without stopping it.

(New account, but I know the rules, so don't auto-downvote. Read. Thanks!)

Strangely enough, magnets do weaken over time, and we can show it from conservation of energy. If a magnet and paperclip are placed nearby on a desk, they will zip towards each other. This momentum has to come from somewhere. It cannot come from the paperclip, because paperclips do not attract each other, so it must come from the magnet. The magnet must therefore somehow store energy, and when its magnetic field causes motion, that motion must somehow consume the magnet's energy, even if the magnet's magnetic field doesn't appear to change.

Commercially, magnets are produced by heating above a certain temperature[0] and then coercing the heated metal's magnetic fields into alignment. Since this process consumes energy, you could imagine that the equivalent amount of energy expendable by the magnet over its lifetime is bounded by that process. This might seem silly, but it has some corollaries which we can directly experimentally observe. For example, a nail gun correctly oriented relative to the Earth's magnetic field will temporarily magnetize ferromagnetic (steel, usually) nails while the nail is in motion, by the same sort of domain-alignment process involved in permanent magnetization. (Please do not do that experiment without taking proper safety precautions. Maybe we can talk Adam Savage into doing it for us.)

In terms of calculations, we can use the "demagnetizing field"[1] but I think that it is not a very good name. Nonetheless, it does show that any time that there is a strong magnet, there is an automatic effect which will slowly slowly slowly weaken it.

More generally, any maintenance done on a machine violates the closed-system assumption. You'd have to include the cost of maintenance if you wanted to show that you're violating thermodynamics. Because magnets are so long-lived, they might seem perpetual, but they are not. They are finite energy stores and they do run out eventually.

[0] https://en.wikipedia.org/wiki/Curie_temperature

[1] https://en.wikipedia.org/wiki/Demagnetizing_field