Connect with us

why doesn't my permanent magnet alternator work?

Discussion in 'Electronic Design' started by matthew, Jul 23, 2003.

Scroll to continue with content
  1. matthew

    matthew Guest

    i just build a small permanent magnet alternator just to see if i
    could do it and make some electricity, however i can't get the
    alternator to work. i've included some pictures so you can see how i
    constructed, wired etc the alternator.

    http://www4.ncsu.edu/~mjdunay/pictures/pma5.jpg
    a wiring drawing

    http://www4.ncsu.edu/~mjdunay/pictures/pma6.jpg
    the actual alternator

    i can spin the alternator up pretty quickly but the little LED i put
    in doesn't even flicker but i'm hoping it's something simple that i
    over looked. to the best of my knowledge i think i've wired the thing
    correctly and i've checked all the connections many times. the
    magnetic poles are switched every other magnet and the coils are
    wrapped in opposite directions. could my LED not work with an AC
    current? do i not have a thick enough gauge magnet wire? is the
    space between the magnets and the iron filings too far? have i wired
    it correctly? the magnet part of the alternator spins quite evenly
    (maybe only swaying a 1/16") so i don't think thats it. i definitly
    would think that the magnets are strong enough to create some kind of
    flicker but i just don't know. any help would be great.

    also if anyone knows of a cheap place to get magnet wire on the net
    let me know.

    thanks
    matthew
     
  2. Ian Stirling

    Ian Stirling Guest

    You can kill the LED by supplying voltage the wrong way round.
    Do you own a multimeter?

    --
    http://inquisitor.i.am/ | mailto: | Ian Stirling.
    ---------------------------+-------------------------+--------------------------
    Lord, grant me the serenity to accept that I cannot change, the
    courage to change what I can, and the wisdom to hide the bodies
    of those I had to kill because they pissed me off. - Random
     
  3. You should get a flicker every time the voltage peaks above 2 volts in
    one of the directions.
    Until you get current moving, you don't need thickness.
    Closer would be a lot better. Are the coils backed up with a
    magnetically permeable short from one magnet to the ones on each
    side? The idea is to get the magnetic path from one side of each
    magnet to its other side to contain as little air as possible. This
    means that all the magnets could be mounted on a disk of iron (since
    this field does not reverse, the conductivity of the iron will not
    circulate eddy current). And there should be a second iron path
    through the coils, that bridges from one to each of its neighbors
    (another disk) behind them, but this iron sees alternating field, so
    its conductivity will act like a shorted turn that circulates
    current. This iron arrangement creates complete flux loops through
    each magnet and both of its neighbors. It also creates quite a bit of
    axial force between the rotating disk and the stationary one. This is
    why this structure is usually made as one cylinder rotating inside
    another, so that the forces from opposites cancel, rather than being
    transmitted through the shaft. Another way to keep the magnetic
    forces out of the shaft is to put magnets that attract each other on
    both sides of the coils, and either rotate the magnets or the coils.
    In either case the attractive force of the magnets is transmitted only
    through the structure that holds them spaced away from the coils, not
    through the generator bearings.
    I would try connecting the coils in series for more voltage, instead
    of in parallel (for more current). Start with the LED across a single
    coil, and make sure when you series the next one, it gets brighter,
    not dimmer. This will prove the series sequence needed. Any series
    parallel combination can be used to trade current for voltage, as long
    as all parallel units are identical (or opposite polarity and reverse
    wired. This all assumes that the number of magnets and the number of
    coils are equal.
    It goes for about $4 per pound on Ebay.
     
  4. John Larkin

    John Larkin Guest

    Are all your coils in phase?

    22 isn't a lot of turns, and your magnetic paths have a huge amount of
    air gap.

    I've lit an LED by dropping a magnet through a coil, maybe 300 turns a
    couple of inches in diameter.

    John
     
  5. Ben Bradley

    Ben Bradley Guest

    In sci.electronics.design, John Larkin
    Looking at the outputs of two coils at a time with a dual-trace
    scope would verify this.
    Now I see, the 22 turns is from the top drawing in pma5.jpg it says
    "22 rappings for each coil." Also, this diagram shows the eight coils
    wired in parallel. That would be good for higher current, but it
    probably provides much less than a volt (unless the rotor is spun
    faster than you would reasonably want to spin it). Putting the coils
    in series might give enough voltage to light the LED, but might also
    give enough reverse voltage to damage it. A parallel diode going the
    other direction should protect it.
    Use an AC voltmeter or DMM on the lowest AC scale to measure the
    output voltage. Compare it with the output voltage calculated from the
    number of turns, magnet strength, gap size and rotor speed, and
    appropriate equations. This would be at least as good an exercise in
    electromagnetics as was building the alternator to begin with, and
    would let you calculate things like how many turns you should put on
    each coil to get a certain voltage at a certain RPM. I have no doubt
    the "appropriate equations" are just a few google searches away.
     
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

-