1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

How would this NOT work as a perpetual work/motion machine?

  1. May 12, 2012 #1
    So perpetual work/motion machines are impossible, but I've toyed with this idea for a hypothetical, super massive, planetary-scale electrical energy generator for a while.

    Suppose you had 2 very large moon/planet-sized spherical magnets that were far enough so as to not attract magnetically but close enough to attract gravitationally. You then set these magnetic spheres in motion so that they start orbiting a common center of mass.

    You then enclose the shared orbit in a hollow, stationary torus. You then coil the torus with wires. My idea is that as the magnetic spheres orbit around their common center of mass, they move along the inside of the hollow torus coiled with cables and generate an electrical current in it forever.

    And yet it couldn't work forever. I just want to know how or why this system couldn't go on indefinitely.
  2. jcsd
  3. May 12, 2012 #2


    User Avatar
    Science Advisor
    Gold Member

    The energy would be lost due to electromagnetic radiation. Oscillating magnets in any configuration would produce EM waves which carry energy away from the system.

    Also, the magnets would produce eddy currents in the wires (that is the principle way it extracts power), but these currents produce an opposing field to these magnets slowing them down, transferring momentum from the magnets to the torus.

    That being said, it would take a heck of a long time to rob an earth sized magnet of its rotational kinetic energy in this way.
  4. May 12, 2012 #3


    User Avatar
    Science Advisor

    If you want to know what the actual source of the force that will slow down rotation of your moons/planets is, it's the coil. When a magnet moves and induces current in a coil, the induced current generates its own magnetic field. That magnetic field will always be oriented in a way as to slow down the moving magnet. This is why it takes certain amount of torque to turn generator, and why it allows to convert mechanical work into electricity.
  5. May 12, 2012 #4
    How could you ever distance them so that they are affected by gravity and not the magnetic field?
  6. May 12, 2012 #5
    By "making" the intrinsic magnetic field of the planets very weak. Then the results of the magnetic force between the planets wouldn't be seen for long time scales. Of course, after a very long time it would perturb the orbits, and I'm not sure if the OP recognized that or not.
  7. May 12, 2012 #6
    ah. I am only just now starting to study E&M (excluding basic freshman E&M), so I wasn't sure if it would be possible considering how weak gravity is.
  8. May 12, 2012 #7
  9. May 13, 2012 #8


    User Avatar
    Staff Emeritus
    Science Advisor

    If I'm not mistaken, doesn't the strength of a magnetic field fall off with the inverse cube of the distance instead of the inverse square like gravity does?
  10. May 13, 2012 #9


    User Avatar
    2017 Award

    Staff: Mentor

    For a dipole field, this is right, and therefore a larger distance does help to make gravity stronger than the electromagnetic force (from the dipoles). Higher orders would be possible and would fall off even quicker. However, you could just use two magnets with a rope in between and rotate that (or use rails and a single magnet), no need to get some fancy planet-sized magnets.

    Anyway, the method of power extraction itself would slow down the movement.
  11. May 13, 2012 #10


    Staff: Mentor

    Exactly. Each joule of electrical or thermal energy extracted would reduce the orbital KE of the planet by 1 joule.
  12. May 13, 2012 #11
    If I understood you correctly, you use the spheres' magnetic field to sweep the turns at different parts of the torus, and generate varying electromotive force.

    But, if you remember Lenz's Law, the current that flows through the coils (you must draw current to get positive work from the system, an open circuit does not generate power) opposes the cause of its generation, in this case, the relative motion of the spheres with respect to the torus. Thus, the induced current generates a magnetic field that tries to slow down the spheres.

    As they slow down, the gravitational foce forces them to go on an orbit with a smaller radius (remember that the centripetal acceleration is caused by the gravitational attraction of the spheres, so in equilibrium [itex]v^2/r = G M/(2 r)^2 \Rightarrow r = G M/(4 v^2)[/itex], the radius of the orbit is inversely proportional to the square of the orbital speed). Thus, the spheres start spiraling towards eachother. They either hit the walls of the torus, or hit the planet around which they are orbiting.
  13. Jun 20, 2012 #12
    Its not a bad idea, but its not perpetual energy. Basically you're talking about making a massive alternator that uses gravity to give you your mechanical motion and converting it into usable energy.

    It looks like it will last forever, but it won't. The laws of thermodynamics still apply and eventually gravity will implode the system, and for that reason it can not be called perpetual energy. However, due to the massive scale you are talking about, once set up, it could last millions of years, just not forever.
  14. Jun 20, 2012 #13


    User Avatar
    Science Advisor

    The best perptual motion machine I have ever seen is this: Start with a vertical wheel on an axle. At the point on the wheel directly to the right of the center attach a 9 kg mass clearly labeled "9". Rotate the wheel, say, 60 degrees clockwise and add another such mass. Continue this around the wheel.

    As the wheel turns the masses will turn over as they go past the bottom point and be labeled "6" instead of "9"!! Since the masses on the right side are always larger than the masses on the left, the wheel will continue to turn!
  15. Jun 20, 2012 #14


    User Avatar
    Science Advisor

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook