Making a large electromagnet

Discussion in 'The Projects Forum' started by Laserever, Feb 3, 2011.

  1. Laserever

    Thread Starter New Member

    Jan 21, 2010
    5
    0
    I'm making an electromagnet- trying to maximize its lifting force.

    After working with some units with my physics prof, we determined that the length of the wire eventually gets cancelled out, meaning that the less resistance, the less complications. We determined it would then be best to use large wire, large currents, and as many turns as I can muster without putting too much concern into it.

    The construction will be on the order of using 60A of direct current through the windings.

    Now, as for the core, I can't seem to find any soft iron rods that get any wider than about 1 cm. I would like to wrap my large wire around something wider than 1 cm, so I figure that I'll just couple ten of these rods with each other, and wrap the coil around the assembly.

    Here's the question: Would it be better to use 7mm wide rods- allowing for less air space between rods, and making the core more "dense"?

    Or would it be better to go with 10mm rods for an overall wider core, but with overall larger air gaps between rods?

    This is for a levitation assembly, so I think that a smaller- diameter core would keep the levitating object more "centered", but will I lose lifting force?

    Thanks for the advice!
     
  2. JDT

    Well-Known Member

    Feb 12, 2009
    658
    85
    Obviously, the magnetic field is directly proportional to Ampere x Turns. So more current and more turns = more magnetism.

    So by using thinner wire you can get more turns in the space. But wire has resistance. Thinner wire = more resistance. Also more heat generation. So it is all a compromise. At 60A it might be possible to wind the coil with thin copper tubing and water cool. Another possibility is to wind with copper strip which is as wide as the coil so that the heat gets conducted out more easily.

    As far as the iron core is concerned, the core needs to be U or W shaped. Even better will be a cup shape with a central core. The iron core needs to have as big a cross-section area as possible. Also, the length of the magnetic circuit (the length through the iron) needs to be as short as possible. Both these reduce the magnetic resistance. Also needs to be the right type of iron (soft iron). As you are using DC the core does not need to be laminated.

    More turns take up more space so the magnetic circuit gets longer. All these conflicting requirements need to be taken into consideration in the design.

    This is what makes design interesting!
     
  3. wayneh

    Expert

    Sep 9, 2010
    12,123
    3,047
    +1

    The design approach is to start with the copper wire table, which gives all dimensions and electrical properties, and build a mathematical model of how many turns and how much resistance you'll get into the space you have with any gauge choice. Simple? Hardly. Things like insulation thickness matter.

    The choice you make will ultimately depend also on your power source, coil geometry, and cooling strategy. Too small a wire is what we call a "fuse", and too thick a wire will draw all the current you can feed it but may not have enough turns to give the strongest field.

    Magnetic field strength will be proportional to core density, actually permeability. And don't forget that lines of magnetism need a return circuit just like electricity. You'll get more field with a horseshoe pointing both poles at the object to lift.
     
  4. Laserever

    Thread Starter New Member

    Jan 21, 2010
    5
    0
    Horseshoes are for horses. Neat idea though. Would it help if I told that I'm simply trying to scale this project up?

    http://uzzors2k.4hv.org/index.php?page=magneticlevitation

    Thanks for the design guidelines. These are all important things that I've had to consider, and that I will remember during construction.

    I will be using a rewound micro-wave oven transformer for the current supply, so no need to worry about the troubles of having ATXs in parallel- always shutting down from a low load and whatnot. Transformers don't care about shorted secondary leads so much. :)

    I'll post what I'm asking about, and maybe then you'll have a better idea of what I mean.

    [​IMG]
    Right, now you see the cross sections of my two core options.

    Core "A" uses soft iron rods that are larger in diameter, Core "B' uses smaller rods.
    So according to wayneh, the field strength or permeability will be proportional to the density of the core.

    I say core "B" is more dense, but it's also less in overall size consequently.

    Core "A" should have less density but has more soft iron in it overall...

    This is where I am torn!
     
  5. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,692
    901
    Are you sure there is any difference in "density?" If they are packed as shown, the ratio of empty space to filled space appears to be constant or 2:7.

    John
     
  6. wayneh

    Expert

    Sep 9, 2010
    12,123
    3,047
    Yup, that's called a hexagonal close-pack. Same density. Mixing sizes can increase overall density, but it adds complexity. You might consider using transformer steel plates, if you can find it in the right configuration. There are also ferrite cores available - you can even make your own - but I'm not sure the overall permeability would be higher.
     
Loading...