Connect with us

Home made powdered iron cores

Discussion in 'Electronic Design' started by Ethan, Feb 11, 2004.

Scroll to continue with content
  1. Ethan

    Ethan Guest

    Has anybody tried to make thier own powdered iron cores? I am trying
    to mix up epoxy and iron filings from the machine shop at work
    (relativly large chunks of cheap steel). So far I am just trying to
    get a mold to work right. I know just the steel itself has a huge
    magnetic hystereses, is there any way to reduce this? I am hoping
    using a powder will at least reduce eddy current losses. Does anyone
    have experience with this?

  2. I have made ferrite powder and epoxy cores, just to see how high I
    could get the permeability. I used the lids from spray paint cans as
    molds for toroid cores. I think I got the permeability up to about
    10. The mixture was stiff as bread dough.
  3. R.Legg

    R.Legg Guest

    If your application frequency is low enough for you to expect iron
    filings to work without overheating, then you should really be fooling
    around with gapped C or UI core sets, if the available material
    permeability is too high.

    If you reduce the density of the magnetic material mechanically,
    without changing anything else, the power loss density can actually

    Hysterisis is a characteristic of the magnetic domains in the material
    used. When air is the magnetic material, the circuit is blind to it's
    distribution, but all material forming the circuit will still share
    the flux density changes imposed.

  4. Big John

    Big John Guest

    Hi Ethan,

    I think the most important question is - What is your application? Are
    you making a choke or a transformer? What frequency are you looking at
    (60Hz, audio, or RF)?

    For lower frequencies laminated steel plates are normally used. They are
    made from electrical sheet steel (which is actually an iron-silicon alloy).
    They are thin and are punched into various shapes. Rings are used for
    motors; E's, I's and C's for transformers and chokes. Eddy currents losses
    are reduced by stacking these laminations (or winding them in the case of
    cut "C" cores) to make up the magnetic cross section needed. The
    laminations have a non-conductive coating that prevents currents from
    flowing axially through the stack. Inductors also have a fixed air gap to
    prevent magnetic saturation.

    For high frequency applications powdered ferrite or similar material is
    used. Cores used for chokes have a non-magnetic filler mixed in to provide
    a distributed air gap.

    If a powdered iron core is what you want there is a home brew way of doing
    it. You can use magnetite sand collected from a beach or other sandy soil.
    Wrap a large magnet in a plastic bag and tie a string to it. Drag the
    magnet in the sand. Magnetite grains will collect on the outside of the
    plastic bag. After you have enough sift it through a fine screen to get a
    uniform small particle size. Then mix the magnetite sand with epoxy and
    squeeze it into a wax mold. When the epoxy hardens use hot water to melt
    the wax away. This works for small cores. I've never tried it, but I'd
    guess that larger cores made this way will probably crack as the epoxy

    While you can experiment with home brew materials for some low power hobby
    project, anything you make for the real world should be made from
    commercially made magnetic laminations or cores with the proper insulation,
    wire, and coatings. They are far better than anything you can make on your
    own. There are places that sell small quantities of both cores and
    pre-punched plates in standard sizes with good tech data.

    To get started try Arnold Engineering for electrical sheet steel and
    Magnetics for ferrite cores. Google from there.

    Hope this helps,
    Big John
  5. I read in that Big John
    >) about 'Home made powdered iron cores', on Thu, 12 Feb
    There ARE 'electrical steels', available in bar, rod, sheet and plate
    forms just like ordinary steels. They have very low but not zero carbon,
    and may have other trace elements. The laminations should indeed not be
    called 'steel' because they contain virtually no carbon; they are
    'silicon-irons'. Their mechanical properties are very different from
    those of steel.
  6. Ethan

    Ethan Guest

    I don't have a real application for this. I am just playing around
    for a hobby. A while ago I got an old book on magnetic amplifiers,
    and want to play around with various amplifiers, transformers and
    motors. However some of the magnetic circuits are pretty odd shaped,
    such as a magnetic amplifier with the two coils on quadrature axis, so
    they are not magneticly coupled, but can change the permeability of
    the core so current in one coil can change the inductance of the
    other. To get this type of magnetic circuit I will have to make it

    I have had trouble making a laminated steel core, mostly because I
    don't have the right tools in my garage, but a wax mold should be
    within my capabilities.

    The idea of using magtatite beach sand sounds cool, so I will have to
    look into this more to find the right kind of beach. I wouldn't
    expect great performance from this, but for my lack of an application
    the gee-wiz factor is compelling.

    If I get something to run and characterize it, I will post it.

  7. No, you can buy U-shaped laminations and cut them in half, or even E-
    shaped, which are easier to get, and cut away the centre leg. It can be
    difficult to find suppliers of small quantities, but they do exist. Or
    you can salvage them from old transformers, although U-shaped are not
  8. N. Thornton

    N. Thornton Guest

    I can suggest another possible approach: making a core from pvc coated
    garden wire. You can wind your core any shape you want, cut wires to
    length and tape them together, etc.

    Regards, NT
  9. I read in that N. Thornton <>
    If you can still get 'florists' sprigs', they are VERY soft iron wire
    with a reasonably insulating oxide coating, and might well make
    excellent core material.
  10. R.Legg

    R.Legg Guest

    You'll find that this kind of magnetic regulator is refered to as a
    'parametric transformer'. In small signal applications, it becomes a

  11. N. Thornton

    N. Thornton Guest

    I've got a book that recommends making transformers that way, but it
    didnt occur to me it might be 'excellant'. The picture of one they
    made shows that most of the core is taken up by gaps between the rats
    nest of wires - and all the loose ends probably acted like mini
    speakers too. But if thats all youve got, fair enough I guess.

    Regards, NT
  12. I read in that N. Thornton <>
    Well, of course you have to make the thing carefully, as with anything
    else, if you want it to work well. And I was referring to the magnetic
    properties of the material, not any particular construction.
  13. CBarn24050

    CBarn24050 Guest

    TVs use mag amps for scan correction, you might be able to modify 1 for your
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day