Connect with us

Magnets killing laptops.

Discussion in 'Electronic Design' started by Phil Hobbs, Nov 3, 2003.

Scroll to continue with content
  1. Phil Hobbs

    Phil Hobbs Guest

    Your magnet caused core saturation in the inductor or transformer in the
    power supply, reducing the inductance by probably 100 times or more.
    The laptop battery has very low resistance, so there was suddenly
    nothing to limit the current flowing through the power supply inductor,
    and something (probably the switch) blew up.

    Cheers,

    Phil Hobbs
     
  2. Phil Hobbs

    Phil Hobbs Guest

    The problem is that even a small saturated area will have a catastrophic
    effect on the total inductance--it functions like an air gap. The
    effective permeability of the core goes as 1/( integral 1/mu * ds)
    around the magnetic circuit. Since mu is normally somewhere between
    about 100 and 8000, depending on the material, a small saturated region
    will dominate the magnetic resistance, and therefore the inductance will
    drop to a small fraction of its normal value.

    All these magnetic circuit terms are imprecise, but physically
    appealing--magnetic flux isn't confined as strongly as electric current,
    say.

    This is how geomagnetic storms cause blackouts--a small dB/dt over a
    *huge* loop can cause quite a bit of current to flow, all at frequencies
    well below 60 Hz. Transformer cores saturate, and breakers blow to
    prevent equipment damage. Newer transmission technologies are supposed
    to make this better.


    Cheers,

    Phil Hobbs
     
  3. Ian Stirling

    Ian Stirling Guest

    I've just lost the backlight on my toshiba 3110CT, due to placing it
    next to a shelf on which was a magnet. (while running)
    The backlight died, and has not worked since.
    It was a rather strong magnet (2*2*0.5cm), and ended up sticking to the
    back of the display.

    I believe that it's due to saturation of the inductor/transformer in the
    backlight circuitry blowing out the driver chip/transistors, though I will
    have to work out how to open it up and fix it.
    Otherwise the laptop worked fine, and I was able to finish off what I was
    doing, as the display can be sort-of read if the sun is at just the right
    angle to the display.
    Annoying.
     
  4. Magnet? Hard Drive? Get it???

    Larry
     
  5. Dave VanHorn

    Dave VanHorn Guest

    Your hard drive contains two high power magnets.
    They are located very close (obviously not too close) to the platters.

    It takes a lot of field strength to change the data on the disk, and
    magnetic fields fall off in strength VERY rapidly, with distance.

    Sounds like coincidence to me.
     
  6. It was his backlit that failed, not his harddrive...
     
  7. Ian Stirling

    Ian Stirling Guest

    No.
    The hard drive is completely fine.
    Looking at an exploded diagram, the magnet touched exactly where the
    backlight inverter is.
     
  8. Is there a way to shield inductors against this effect of an external
    magnetic field? Like, would a toroidal inductor have helped?

    (I would think not - presumably each part of the toroid would have been
    magnetized in alignment with the external field, and even though on the
    whole that would result in a cancellation as averaged across the whole
    toroid, what matters is saturation grain by grain. But I'm just guessing.)
     
  9. Whelan

    Whelan Guest

    Would this situation be covered by a warranty?
    (Say, you are just an ordinary person who doesn't know much about magnets or
    science.)
    Nan
     
  10. R.Legg

    R.Legg Guest

    This is hardly the type of thing that somebody else is going to repeat
    on working hardware, in order to confirm or disprove. There are too
    many other things that can hobble a laptop as it is.

    Large magnets should be on a specially market shelf.

    If your hard drive is in fact still running, and the backlight hasn't
    been mechanically damaged by the physical force, there's another
    magnet on your machine that needs to be considered.

    This is the one signalling open or shut lid. Perhaps the microswitch
    detector is permanently magnetized shut, signaling a permanently
    closed lid.

    Should be testable (and is likely cheaper to replace than the entire
    display).

    RL
     
  11. Ian Stirling

    Ian Stirling Guest

    I may test it as I try to repair it, swapping the driver circuitry.
    It's not a large magnet, it's a small (2*2*0.5cm) one that I was using
    to stick papers onto a metal shelf.
    There is no magnet there, it's only a microswitch.
    I'd only be replacing (at most) the inverter board (the bit that generates
    the high voltage to drive the backlight), more probably just the driver
    transistors.
     
  12. Dave VanHorn

    Dave VanHorn Guest

    That I'll believe.

    I've done it.

    I've also seen ferrite cores biased with permanent magnets, to decrease
    saturation.
     
  13. Dave VanHorn

    Dave VanHorn Guest

    Yes, but it would be expensive.
    You'd need a mu-metal can that would conduct the flux, and thick enough not
    to saturate.
    Dubious payback. The can would likely cost $.25-$.50 in production
    quantities.
     
  14. Dave VanHorn

    Dave VanHorn Guest

    The field is almost un-noticable, from a few inches away.
    A small NIB magnet, with a surface peak of 2000 gauss is equal to the
    earth's field in 4-6 inches.
    IIRC, it takes something like 700 gauss at the surface of the platter, to
    alter the HD data.
    There's also the magnets in the HD motor, but they are better shielded (part
    of being a better motor) than the voice coil magnets.

    Those curvey magnets you see on the surplus market, for $1 or so, are HD
    magnets.

    Sounds like he accidentally blew the switching converter by saturating the
    core.
    Those things usually run not too far from saturation anyway, since larger
    cores and better materials cost money.
     
  15. Ian Stirling

    Ian Stirling Guest

    Or a small polyfuse/resistor to limit the current, maybe even just a
    design change. Likely $.10 or so, maybe $0.

    My post was mainly to raise awareness that common magnets used for
    sticking paper on metal surfaces are quite strong enough to actually
    kill some laptop circuitry, even though they may have no effect on
    hard drives.
     
  16. Dave VanHorn

    Dave VanHorn Guest

    Assuming that this supply collapsing it's output dosen't cause other
    failures.

    Given
     
  17. Jim Yanik

    Jim Yanik Guest

    I hope you'll keep us posted on what you find and what you have to replace
    to repair the display.

    I wonder;if you bring the display near a RF source,will it cause the
    fluorescent tube to light?
     
  18. The size of the magnet has nothing to do with it's strength. The
    strongest magnets I've commonly seen are the tiny magnets used to hold
    "rave lights". they are smaller than a dime and will stick to each
    other clear through you finger (or, as they are intended to be used,
    they will hold the rave light through an ear lobe).

    I'm not sure that I buy your cause-effect conclusion, but it's not
    impossible. I think it's very unlikely, however.

    Fixing the inverter is probably not feasible (not impossible, mind you,
    but probably more difficult/expensive than replacing it).
     
  19. Dave VanHorn

    Dave VanHorn Guest

    Absolutely it does. The same material, in a 2 inch block will crush your
    finger, and you can tow a car with it. The field intensity is the same, but
    flux is field times area.

    http://hyperphysics.phy-astr.gsu.edu/hbase/magnetic/fluxmg.html
    It's just luck to hit the target, but I've killed flyback converters in
    similar situations.
    Plastic case, <1/2" between the core and the outside of the case.
     
  20. You misunderstood what I was saying. Agreed, a larger magnet of the
    same material is stronger than a smaller magnet of the same material.
    But "same material" wasn't specified. Some smaller magnets can be
    stronger than some larger magnets.
     
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

-