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

Can an exterior magnetic field be created using permanent magnet induction?

  1. May 13, 2007 #1
    I'm trying to make an interior rotating mangetic field that will induce electron flow in an encapsulated closed loop coil. What I'm constructing at this time (not finished yet) is a rotating (0 - 1600rpm+) neodymium permanet magnet setup in a Halbach Array assembly. I understand that inorder for the coil of wire to produce a field it must be changing, correct? A constant rotating field will not produce/induce the field desired. How can I produce changing electron flow in the coil? I suspect that once this is figured out, the coil will produce a field on the interior (between the coil and the Hallbac Array rotor) as well as the exterior of the coil assembly. The strongest "exterior" field is what I'm trying to create as well as, once the Halbach rotor has spun down to zero rpm there will be no magnetic field produced.

    If my description is not clear, I'll try to draw what I'm doing but I'm pc graphicly challenged.

    Thanks for all you help.
    Last edited: May 13, 2007
  2. jcsd
  3. May 14, 2007 #2


    User Avatar

    Staff: Mentor

    Yeah, a clear drawing would help a lot.
  4. May 15, 2007 #3
    I'm in the process of posting a drawing, please be patient, I would like it to be as accurate as possible. I'm having some problems drawing it since the the unit is a sphere. Each view is similar to the other which may tend to confuse a mental visual picture.

    Thanks for your patience
  5. May 22, 2007 #4
    Here is a picture of what I'm making see attached

    This is a work in progress and any input would go a long way.

    Thanks again guys

    Attached Files:

  6. Jun 11, 2007 #5


    User Avatar

    Hmm.. I'm still not quite understanding
    what you're trying to accomplish.
    You talk about creating an exterior
    magnetic field. In very PARTICULAR
    circumstances flux is confined
    almost exclusively in the interior of a
    high permeability closed region (e.g
    inside an magnetic 'shield' hunk of
    iron/ferrite or in a ferromagnetic
    toroid or so on; in such cases the magnitude of flux external to the
    permeable shield can be insignificant,
    the closed flux loops pass almost entierly
    through the highly permeable regions
    rather than exiting them.

    If you have dipolar (N<---S) "bar magnets" the flux will exit the north
    pole and eventually form a closed loop
    returning to the south pole. More flux
    will concentrate in the interior of high
    permeability regions than elsewhere
    along the loop. In general the field will
    expand to fill all space to SOME extent
    getting progressively weaker as one
    goes further from the poles. The
    response of magnetic materials to a field
    can be highly NONLINEAR, but
    electromagnetic/static magnetic effects
    are basically LINEAR except where/when
    non-linear materials transport the
    field. Given the assumption of
    linearity, the field from any ONE of your
    dipole "bar magnets" can be sketched
    and estimated independent of the
    existence of any other of the bar
    magnets in the multipolar array.
    The resultant field at any point in space
    in this linear assumption will simply
    be the vector SUM of each of the fields
    due to the individual field sources
    (each individual bar magnet) acting
    independently. So that may simplify
    the first estimate of what the external
    field will be -- just picture the field line
    loops extending outward into all space
    from a single bar magnet going from
    its north to south pole with a rapid
    diminishment of field intensity as the
    distance from the poles increases.
    Then realize that the fields from other
    nearby poles will either diminish or
    reinforce the field at a given point in space depending on whether the fields
    from the other poles cancel or
    reinforce at the point in space in question.

    The winding's topology eludes me; you
    draw it with a sinuous shape along a
    circular band, and to me that looks like
    it's something like a helical solenoid
    wound around the path of a toroid.

    Compact tightly wound toroidal
    solenoids produce flux that's
    concentrated fairly uniformly in the
    area of their interiors and the magnitude of field they produce outside
    of the interior of the torus's cross section
    is very small (nearly nonexistant) if
    the toroid is of a high permeability
    material e.g. a ferromagnetic one.

    If you have a closed loop of wire,
    a changing magnetic field in the interior
    of the loop will induce a changing EMF
    of circuital voltage in the loop and the
    EMF voltage will therefore create a current in the loop opposing the change
    in external magnetic field through the
    loop. One can imagine a single loop
    'ring' and a single bar magnet next to
    the loop; when the bar magnet and
    wire ring loop are coaxial (the loop could
    slide around the bar magnet; the
    pole of the bar magnet pointing directly
    into the interior of the loop), the bar
    magnet's field passing through the loop
    interior will be maximum. When
    the bar magnet is perpendicular to the
    loop effectively NONE of the flux from
    the bar magnet will pass THROUGH
    the hole in the loop, so in that orientation the flux through the loop will
    be essentially zero. Rotate the loop
    face with respect to the bar magnet's
    pole face and you'll get varying degrees
    of bar magnet flux penetrating the loop
    depending on the angle between the
    field and the loop's face as you go from
    parallel to perpendicular.

    Your ?toroidal solenoid? winding will
    generate EMF accordingly to the
    vector sum of the external flux that
    passes through its windings' interior
    from the external fields in its environment. It's not clear to me
    what the point of the winding is, or
    that the net flux enclosed by its path
    will ever be significantly different than
    zero since as you go around the circle
    the vector sum of the bar magnet
    fields look to be zero, assuming the
    bar magnets are of uniform strength,
    and geometry. I presume the couple
    bar magnets you drew with double
    heads "<--->" vs "<---" is just an error
    of drafting and that they're just two
    more that follow the obvious 90 degree
    pattern of rotation. I further assume
    that the bar magnets are of a constant
    distance from the winding, the winding
    is uniform geometrically, and all
    materials in the toroid and between it
    and the bar magnet poles are uniform.

    If you want changing flux in the winding
    you could seemingly just hook it up to
    a power source that generates the right
    current in the winding to create the
    desired field strength in it. Being
    a toroidal winding you'll get much less
    field strength outside the torus than
    within in, at least if it's wound tightly
    so that the solenoidal / toroidal
    approximation holds.

    Finally, regardless of the ~ 2000
    RPMs involved, many aspects of this
    problem will seemingly behave
    essentially identically as if the apparatus
    was not in motion. To be precise,
    magnetic fields and currents that vary
    in space and time as slowly as a few
    thousand cycles per second can be
    accurately modeled by the mathematics
    that assume the fields are just static
    at any given moment and that no
    strong high frequency dependent
    effects will be relevant. Hence you
    can use the magnetostatic and "DC"
    approximations for the coil and the
    bar magnets to estimate the fields that
    exist in and around the apparatus at
    any given moment. That would make
    it relatively easy to use well known approximation equations of magnetic
    field strengths due to coils and dipoles
    to analyze the resultant external field
    at any point in space due to such an
    assembly of permanent magnets and
    less so due to your coil depending on
    the materials and geometry involved.

    I'm sure you could even model it in
    some magnetostatic or other kind of
    mechano-magnetic modeling software,
    though just getting approximate results
    from equations by hand may prove to
    be more illuminating with much less
    investment in effort modeling the system.

    What is this for, anyway?
  7. Jun 11, 2007 #6
    i didn't read any of the above weirdly formatted post but here goes. first of all this is a sphere of magnet wire? i really can't understand what this thing is suppoed to be. draw an orthographic view
  8. Jun 13, 2007 #7
    Yes this is a sphere of magnetic wire. To be more specific it's two half spheres connected together that house an interior PM array configured on a dual axis rotor. Orthographically the image on the left is the top view, the image on the right, since it is a sphere, can be the front and the side view.
    Last edited: Jun 13, 2007
  9. Jun 13, 2007 #8
    i can't imagine how you would wrap the magnet wire around the sphere??? what is the point of this? i think it will be very hard for you to get an external field considering toroids and solenoids have very negligible external fields.
  10. Jun 13, 2007 #9
    I will need to respond to your post in parts, I'm exceedingly busy, my appologies xez. I am under the understanding that a Hallbach Array augments the field on one side of the array essembly to the point that the opposite side is 50x weaker? If this is not the case than my electromagnetic induction exercise ball is in the toilet.

    Thanks again for your comprehensive input to my post.
  11. Jun 13, 2007 #10
    what is the point of this thing. why don't you just describe what effect you want.
  12. Jun 14, 2007 #11
    I a nut shell... It is my understanding that... when a magnetic field is induced into a ferromagnetic close loop circuit, the opposing force that is created is allways equal and opposite in nature. So, the center Haullbach PM rotor inducing a current in the surrounding windings will create an opposing field between the rotor and the windings. Since the center rotor is now in motion the opposing force will enhance the internal spinning, reducing secondary input to keep the rotor spinning. It is my hope to be able to induce this surrounding magnetic field and utilize the exterior field for futher research in mobile and switchable (On and Off) magnetic fields that could be used for NDT or therapeutic porposes like a magnetic bracelet, EDS and so on.
    Last edited: Jun 14, 2007
  13. Jun 14, 2007 #12
    it won't work. there will be no external field and the windings won't enhance the field, they'll oppose acting like brakes on the rotor.
  14. Jun 14, 2007 #13
    The winding topology was drawn as a sinuous shape for reasons of ease of drawing. The intent is to have the closed loops, wire or high permeable ferromagnetic shims machined to accomodate the exterior sphere shape

    The windings are to create an opposing field that was induced by the (what is allmost a monopole) halbauch array. But as you have mentioned the net exterior flux field will amount to near zero configured/designed as I have in the drawing provided.

    An external power source is outside of the scope of the intended end goal, thanks anyway xez. The only external source that can be utilized is man-power, even if that man-power is aided by mechanical means.

    Do you have a recomendation re: software?

    Primarily EDS
  15. Jun 14, 2007 #14
    ElectroDermal Screening?
  16. Jun 14, 2007 #15
    electrodynamic suspension/ levitation
  17. Jun 15, 2007 #16


    User Avatar

    Ok, I apologize for not initially understanding / visualizing
    the bit about the Hallbach array's purpose and function / topology.

    I believe your understanding of that is essentially correct, it
    can generate a field that's oriented and enhanced
    externally to such an properly constucted array.

  18. Jun 15, 2007 #17


    User Avatar

    Here are my thoughts relative to the above.
    a) When the strength of a magnetic field changes so
    that there's more or less flux through the space enclosed
    in a conductive loop, a circuital EMF is induced in the loop
    that causes an 'eddy current' flow around the loop.
    The eddy current flow is such that the magnetic field
    it generates opposes the CHANGE in the magnetic
    flux through the conductive loop. So when the
    exterior field increases, the current through the loop
    wll tend to decrease the flux through the loop by
    generating an opposed magnetic field that tends to
    cancel the increase of the flux through the loop.
    Conversely when the exterior field is decreasing the
    flux through the enclosed loop, the induced eddy
    current gives rise to a magnetic field that increases
    the magnetic field through the loop, opposing the
    diminishment of the enclosed flux. In all cases the
    induced eddy current and its magnetic field will act
    as a 'mechanical brake' on the relative motion of the
    exterior magnet and the coil such that the field in the
    eddy current loop/coil will tend to remain as it is
    and resist mechanical motions that would cause the
    flux through the loop to either increase or decrease.

    b) Magnetic fields passing near permeable
    not previously magnetized ferromagnetic materials will
    cause an attraction between the magnetic field source
    and the ferromagnetic material. The flux will be drawn
    to pass more completely in to / through the
    highly permeable 'soft' ferromagnetic material and
    so there will be attraction between the material and the
    magnet source; a withdrawing magnet will be 'braked'
    by this attraction whereas a magnet moving toward
    highly permeable material will be accelerated toward
    it by the greater attraction the closer it gets.

    The way to enhance the spinning of a rotor with
    permanent magnetic pole(s) is to cause an attraction
    between the pole and space in front of the pole/rotor's
    direction of travel, and/or cause a repulsion between
    the pole and space behind the pole's / rotor's direction of

    In common generality relating to permanent magnet
    rotors it's not possible to use passively
    induced eddy currents or static permeable materials to
    cause net augmenting acceleration of a rotor in a
    perpetual way; that'd lead to a 'perpetual motion'
    or 'free energy' type of circumstance.

    Of course as in a motor you can use dynamically
    generated externally placed/forced magnetic fields to accelerate
    or maintain the motion of a rotor, but that's not by
    drawing power / field energy from the rotor to create
    the accelerating fields augmenting the rotor's
    continued motion.

    As long as the net momentum / energy of a closed
    system isn't changed (i.e. not self-increasing and
    also not decreasing either) you can contrive ways
    to alter the system / motion in arbitrary ways without
    violating energy / momentum conservation laws,
    though of course there may be other physics that
    limit the facility / spontaneity of the system transitioning
    from one equal energy state to another...
  19. Jun 15, 2007 #18


    User Avatar

    There is no reason it's different to create an Hallbach
    array with electromagnets than to create one with
    permanent magnets. So to the extent you want a
    pseudo-uniform switchable (on/off)
    'monopolar' flux emerging from the exterior of a
    cylinder or sphere, you can just directly build the array
    in part or in whole with electromagnets and turn it
    on or off as desired. The mechanical rotations and
    external shell winding isn't (in this limited sense) needed.
  20. Jun 15, 2007 #19


    User Avatar

    Generally there are a few ways this can be done:

    a) magnet pole repelling another similar magnet pole

    b) magnet pole attracting a dissimilar magnet pole
    (which may be a pole induced in a 'soft' ferromagnetic
    substance by the presence of the first magnet)

    c) magnet pole repelling a diamagnetic material; usually
    this needs EXTREMELY strong magnets since most things
    are only VERY weakly diamagnetic; superconductors being
    one notable exception.

    d) repulsion of an eddy current region by an increasing
    magnetic field.

    e) attraction of an eddy current region by an diminishing
    magnetic field.

    Generally these all need some kind of
    spatial or electrical / mechanical stabilization to perform
    any kind of stable levitation / suspension against gravity
    and anything that'll perturb the system from rest.

    As for modeling software for magneto static or
    induction machines / motors.. hmm.. ANSYS emag or
    ANSYS multiphysics, maybe. Be prepared for sticker
    shock as to high prices, as well as to find that the act
    of modeling a system is sometimes more complex and
    time consuming than just building it or analyzing it by
    thought / manual calculations.

    The fact that you're talking about complex coil and machine
    shapes as well as moving parts especially complicates
    FEM type modeling efforts; it's possible, but not trivial.
  21. Jun 15, 2007 #20
    why do you post like the tha xez, it is so annoying
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook