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

Dimension an electromagnet

  1. Oct 16, 2006 #1

    I like to research the possibility to build an electromagnet that is quit big and has a lot of strength. As I really don't know much about magnetism, I don't see how to start with this.

    Anyone that please can help me with some formulas?

    I like to make a sort of excel or matlab program where I can put in some numbers
    • magnetic contact area
    • desired magnetic strength
    So the program gives me how big my electromagnet should be and of course how much electricity it takes per sec.

  2. jcsd
  3. Oct 16, 2006 #2


    User Avatar
    Gold Member

    brief explination

    an electromagnet is a collection of a very large ammount of wire wrapped around a hunk of metal. (cunductable) eg. a nail. you wrap this wire around this nail (using magnet wire because it has a very thin insulation layer) then you need to consitder the size of the wire, and how much current you can pass through it WITHOUT MELTING IT!!!!!!! you can pour current through it but you have to keep it at a reasonable tempearature. even if this includes submerging it in liquid nitrogen, etc....
    i dont have any specks on a electromagent but i do know that with the desired direction, there is ither a positive charge or a negitive charge.

    one method is to make a coil with your fist and stick you thumb out the side.
  4. Oct 17, 2006 #3
    thx for your reply.

    The fact that I was asking this question on this forum is because I can't affort to use the trial and error method as I will build a large electromagnet.
    The dimensions should go up to a cross-section of 40*40 cm^2 (15.75 *15.75 inches^2) or even up to 100*100 cm^2 (=39.4*39.4 inches^2). Cross section will probably be circular, as this is the most easiest, I guess.

    This magnet will therefore require a lot of material.
    Electrical power is not of that importance (at least not yet) as I can use a 125A (cos phi : 3) connection.
    The magnet itself should run for short cycles only, some 10 to 30 seconds. So I guess heat production will be controlleable.

    Therefore I like first to calculate my magnet into the smallest details (dimension of core, number of layers and turns of wires, the used current but most important of course the created magnetic attraction field and how deep it is working)

    I hope someone can give me some formulas (with some explanation how to use) where I can start with.

    Thx a lot!
  5. Oct 17, 2006 #4


    User Avatar
    Gold Member

    i dont know any equations, but i know what you mean by a very large electromagnet.
    by the way, 30 secconds could melt your wire significantly so you must be carefull
    USE MAGNET WIRE! you can pile wire much higher (more revolutions) with magnet wire. because it is dipped in a insulator, not "surrounded" like normal wire.
    i dont have equations, i suggest you look on google or something. im sure its out there somewhere. good idea posting it.
  6. Oct 17, 2006 #5
    Magnetic field = (number of turns of solenoid) * Current * permeability / (pi*diameter of turn)

    MKS dimensions
  7. Oct 18, 2006 #6


    User Avatar
    Gold Member

    Keep in mind that a magnet does not have to be large to be powerful. Back when I was a kid, Edmund Scientific Company (which later became EftonScience) sold a unit that was about 2 inches in diameter by an inch high. It ran on 2 'D' cell batteries, and could lift 500 pounds.
  8. Oct 18, 2006 #7
    thx for your replies!

    I know the dimension aren't related linear with the strengt, but I need really both : big magnetic contact area and a big attraction field.

    I need 2 N/mm^2 (145 pounds force / sq.inch ) at a few mm (+/- 0.1inches) of the magnetic area.

    If I will make this magnetic contact area, what should I do as best :
    • make a solenoid with soft iron core
    • place several U shaped magnets under a plate (in roset form)
    • other possibility...
  9. Oct 23, 2006 #8
    @ moderator :

    this thread can be closed. I examined my problem and I have found that it isn't feasible.

    But anyway : thx to everyone that has replied!
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook