Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

3-phase question

  1. Apr 20, 2009 #1
    Hello all

    I would like to create a toroid with 3 phase windings connected to a three phase switch connected to 220V or 110V.The windings are wound in such a way that one type of coil is wound around the toroid at a specific location and around the toroid diagonally from the first position of that specific coil. This same process would be done for the last two types of coils. I am, however, very unsure as to how to connect each separate coil with its adjacent coil. I am also unsure as to how to connect the three different types of coils, 6 separate coils total, to a three phase switch. How would you connect these parts together?

    Would you connect one end of one coil of windings to its adjacent coil and connect the other end of that coil to one side of the 3 phase switch? Or would you connect both ends of each coil to the three phase switch? I do not understand how I would connect these three different phases of coils to one another, if needed, and connect them to the three phase switch. Any help would be greatly appreciated.

    check out this youtube video to see the kind of toroid I am trying to accomplish: http://www.youtube.com/watch?v=STnsB5DE9pk&feature=related

    Thanks a lot everyone.

  2. jcsd
  3. Apr 20, 2009 #2


    User Avatar
    Science Advisor


    I think you would have a couple of problems.

    If you are in the US, I think your mains supply is either in phase or 180 degrees out of phase.
    We have 3 phases that are 120 degrees out of phase and that seemed to be what was used in that demo.

    The other problem is that toroid. A toroid is very good at keeping magnetic field contained and yet that demo depends on leakage magnetic field to work. So, it would be very inefficient.
    You might be better off using iron bars pointing inwards towards the center to maximize the magnetic field. They are trying to make a synchronous motor and you can do this without 3 phase. Just have each bar wound in the opposite direction to the one next to it.

    It would be quite a project to rectify your mains supply to give DC, filter it, and then to artificially create 3 phases with electronics. It would be a fun thing to do, but not for nervous people.
    I have seen circuits for doing this and you could probably find them on Internet. Considerable experience with mains voltages would be necessary before you considered such a project though.
  4. Apr 20, 2009 #3


    User Avatar
    Science Advisor

    If you have a three phase line somewhere near you, you can probably get three phase run out to you. You just need to call up the electricity company (also good to see whether or not three phase is, in fact, available near you), call an electrician to install the outlet and internal wiring, and voila, three phases, and probably only a few hundred dollars for the pleasure.

    EDIT: Depending on how much the electrician / carpenter charges to run the wiring where you want it to go, and how much the power company charges to run the line out to you in the first place.
  5. Apr 20, 2009 #4
    This looks like a demonstration of ac induction motors. vk6kro is correct, it is much more efficient to have the poles pointing perpendicular to the axis of rotation. I don't even know how this works with a toroid. Anyway, another option for the 3-phase power source is to use a variable frequency drive:
    You can get drives that convert single phase ac to variable frequency 3-phase.

    As for the connections, you can see an animation here:

    You may also be able to produce the same effect with a single phase power source by using the permanent-split capacitor method.
  6. Apr 20, 2009 #5


    User Avatar
    Science Advisor

    Yes. The only real need for such a demo would be for a school and you could achieve similar results by borrowing a magnetic stirrer from the Chemistry Dept.

    Apart from the obvious rotation of magnets (stirrer bars ), I got one rotating an aluminium disk suspended above the rotating magnet. An electrostatics electrophorus (disk on a stick) can be "persuaded" to rotate with a little lubrication. I made a perspex box for the magnetic stirrer so that the magnet became visible.
  7. Apr 20, 2009 #6


    User Avatar
    Science Advisor

    You're absolutely right. The OP should be careful if trying to reproduce this design because fundamentally it shouldn't even work! If that's really is an iron core torroid with good coupling between coils then I'm afraid that all you'd get is a big 3 phase short circuit. Why do I say this? Well for balanced 3 phase [itex]\sum{i(t)} = 0[/itex] so [itex]\Phi(t)=0[/itex] (for all t). There is no flux therefore no flux derivative and therefore no opposing voltage. That's for an ideal iron circuit of course, in practice there will leakage inductance, but there is no magnetizing inductance with this arrangement.

    I think that there has to be something in the construction of that "iron circuit" that we don't see. It may actually be air cored or perhaps there is some other detail of an iron core that we don't see in the video (eg something that we're not seeing under the glass bowl).

    I’d be interested to hear if anyone has any more details (or hunches) about exactly how this is really constructed.
  8. Apr 21, 2009 #7
  9. Apr 21, 2009 #8
    Sorry, I'm not a motor design engineer, so I cannot answer your question. But if I were, I'm sure I would require a lot more information. Are you trying to do this as a demonstration, like the one in the video? If so, I think I would want the poles pointing upwards from the field core (stator ring) instead of inward. Then I could mount it under a table and spin the ring on the table. There's also a very cheap toy on the market that accomplishes this task, although it operates on a slightly different principle.
  10. Apr 22, 2009 #9


    User Avatar
    Science Advisor

    Maybe it would be a good idea to move to a lower voltage, though, if you are going to be experimenting.
    If you can locate a good heavy transformer that can deliver 18 volts or so that would be good. Most battery chargers would have such a transformer in them. Even if you took some flying leads from the charger for experimenting with. This will make any coils smaller too.

    You would need laminated cores. That is, the cores would be built up of iron plates isolated from each other to prevent eddy currents. The best (and only) source of these would be old transformers.

    When designing transformers, a calculation of 5 turns per volt is a rough guide. Maybe we can apply it here.
    So, if you had a large laminated core, with 18 volts you might start with 18 *5 or 90 turns.
    You want it to have enough inductance so that the current through it is less than an amp or so. Measure the (AC) current if you can, and check the strength of the magnet by attracting a piece of iron to it.
    Normally, a transformer has a closed loop of iron laminations for the magnetic field, but this one would have open loops, so it will tend to hum and rattle a bit.

    This is still a big project and we can only guide you, so mostly it is up to you to make it work.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook