Connect with us

12 volt light swith

Discussion in 'Photovoltaics' started by Douglas Gaulin, Feb 8, 2004.

Scroll to continue with content
  1. What are most people using for a 12 volt wall light switch? From what I
    understand, it is unsafe to use a regular 120 volt wall swith in a 12 volt
  2. Nice to get something that's actually DC rated and fits a standard
    box/plate, but these do tend to be rare.

    The main difference is that arcing is more prevalent with DC, since AC
    tends to self-extinguish. If you can't get anything else, the standard
    rule of thumb seems to be: derate the 15A AC wall switch to about 5A
    DC (around 60W max per switch in a 12V system).

    FWIW I bought some a couple of automotive rocker switches rated for
    20A DC, but they were of such poor quality that I decided I didn't
    want them mounted in my walls and assigned them to some low-current
    non-critical tasks outside the house.

    The moral: even buying a DC-rated switch doesn't necessarily guarantee
    that you're further ahead.

  3. Guest

    How about a backwards diode?

  4. Nick Hull

    Nick Hull Guest

    What about simply putting a big cap across the switch to supress the arc?
  5. Works for inductive spikes, but not DC current from (say) a resistive
  6. daestrom

    daestrom Guest

    Along with a 'free-wheeling' diode across any inductive loads. Duane's
    schematic shows just such a diode on the load side of his circuit.

  7. Guest

    Wow, I must say that your post was informitive (as I understood most
    of it).

    I am the guy who put an alternater on his bicicle to get extra energy
    when exersizing. The DC /AC arcing that you explained about; is this
    why when I turn off the curcit (I use an ac light switch on the
    negitive, just before the battery, taped to the handlebar), the volt
    meter on the alternater swings all the way to 17 Volts until I stop
    pedaling (and while it dosen't charge, it becomes easyer to pedal)?

    I am using an alternater from a Chev 350 engine (Yes it is to big), is
    there any danger in what I have described above? I have been
    experimenting with solar and bicicle generated electricity, and I
    might be at that point where a little knowlage is dangerous. so I just
    thought that I should ask if there is any risk.


  8. Why not just use one of those mercury switches? The contact is broken
    quickly enough, and there is plenty of gap.


  9. I read all the replies.

    If I'm going to be adding circuitry, I would be inclined to use the
    switch to operate a relay with the appropriate ratings.


  10. Roger_Nickel

    Roger_Nickel Guest

    Rather than using an auto alternator you may get better efficiency by
    using an old industrial stepping motor as an alternator. These typically
    have two centre tapped windings and a permanent magnet rotor so that you
    do not waste energy energising a feild winding. They also seem to be
    made to closer tolerances and with smaller gaps in the magnetic circuit
    than the the typical auto alternator, which should further improve the
    efficiency. Old mainframe computer printers are full of these motors.
    You could rectify the outputs from the two windings separately and then
    connect the outputs in series and feed the combined output into your
    battery via a solar power cell power regulator. This is not ideal, your
    battery would not charge until the output of the generator exceeds
    battery voltage. It would be better to use a switchmode power supply to
    generate the 13-14 volts you need to charge the batteries from "any DC
    input" but this would be a special design project. I have played around
    with using a small step motor as a generator for driving high efficiency
    white LED's in a hand wound torch and the idea works well ,,, all you
    need to do is scale it up.
  11. For the few milliseconds that the FET is on. Average power
    dissipation is probably immeasurable unless you flick the switch on
    and off continually...
  12. daestrom

    daestrom Guest

    If I'm not mistaken, when the switch is 'on', R1 & D1 will turn on the
    mosfet continuously. The current *thru* the mosfet will be nil since the
    source to drain is shorted by the switch. But when opening the switch, the
    mosfet is already 'on'.
    But your 'given's may not be true (see above).

    But it does seem like a potentially hazardous mode of failure. Perhaps a
    slo-blo fuse after Q1, before where the switch ties back in (on the load
    side). Sized to blow if the load current flows through the mosfet for any
    significant time, yet doesn't carry normal load current (bypassed by switch
    when 'on'). This would also increase the resistance of the mosfet branch
    slightly, allowing for more 'degradation' in the switch contacts.

  13. daestrom

    daestrom Guest

    Low current, slo-blow fuse in series with mosfet. If the failure you
    mention occurs, fuse blows, current through mosfet drops to zero. But now
    you have an 'arcing switch' that isn't able to interrupt the load.
  14. Honestly, I haven't looked.

    If they aren't available, using them for any kind of mass marketed item
    is out. On the other hand, getting them from stores that sell used
    construction materials shouldn't be a problem.

  15. daestrom

    daestrom Guest

    I don't understand why you think the mosfet is 'off' when the switch is
    'on'. When the switch is 'on', the resistor R1 and diode are applying the
    voltage needed to bias the gate of the mosfet to 'on'. The only reason it
    has little current flow is the source and drain are at the same potential,
    but they are not at the potential of the gate (thanks to the zener diode).
    With this negative voltage applied to the gate with respect to the drain,
    the mosfet is on continuously while the switch is on.

    The R/C will take a moment to charge when turning the switch *ON*, but that
    is irrelevent since the switch is carrying the current when it is on. When
    the switch is turned 'off', the capacitor will quickly discharge through the
    second switch contact shutting *off* the mosfet. The RC time constant would
    *only* be an issue if turning the device on, then turning it back off before
    the RC time constant. *THEN* you would be opening the switch contacts
    before the mosfet had turned on. Frankly, it's not clear what purpose the
    capacitor serves unless just to limit the dV/dt on the zener when turning
    the switch 'on'.

    As far as fusing, something on the order of 1 amp slow-blow may be adequate.
    Although load currents would normally be higher than 1 amp, load current
    only flows through the mosfet-fuse combination for fraction of a second when
    turning the device off. Not long enough to blow a fuse with an internal
    time-delay created by thermal mass (i.e. 'slo-blo'). If the switch's second
    contact fails to close and turn off the mosfet as you suggested could
    happen, then load current flows through the mosfet-fuse combination until
    the fuse blows. If the actual load current is low enough that it doesn't
    blow the fuse, then it is also low enough to avoid damage to the mosfet.

  16. And maybe an LED in series with it to show (by a brief flash of light)
    that it's still functioning. Otherwise, if the fuse blows, the arc
    protection is missing, but the operator doesn't know it...
  17. Don Kelly

    Don Kelly Guest

  18. The idea is to make and break the connection quickly, and to provide a
    large enough gap to break the arc if it does happen. Any mercury switch
    I have ever seen has a large gap -- if for no other reason than to keep
    the inevitable splashes from closing the switch.

    Hey, I wonder if the lights flicker in an earthquake when you use those

    Anyhow, I have never had a problem switching DC, and I switch some
    pretty big currents when I'm working with some of the accessories in a
    car. I understand the theory of why it's harder to switch DC, but I
    have never actually seen it cause trouble.

  19. Alpinekid

    Alpinekid Guest


    I just want to jump in here and say how nice it is see this kind
    of exchange going on. I'm really enjoying it and I'm learning a lot.

    My feeling on Ray's last point is that we do not see it very often
    because of over-engineering. As time goes on and we design switches to
    tighter specs and we use cheaper material and demand more performance
    and longer life spans we will see it more.

    In the "olden" days when you might have a kitchen light switch that gets
    uses once or twice a day and has a 1/2" gap with heavily plated contacts
    and it switches maybe 5 amps you may never see it.

    But now days you have a cheap switch designed by the Tie-i-wan
    corporation and made in Cheapistan
    with plating a few millionth thick, almost no wiping action, very small
    gap size and light loading, its a different story. It also might be
    a car switch that is switching heavy loads like to the electric motor.

    I glad to get the theory and I'm thankful for such a clear and well
    reasoned design.
    Thanks for taking the time to explain it. Keep up the good work and the
    good flow of information.

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