Connect with us

Triac question

Discussion in 'Electronic Design' started by Tom, Nov 21, 2003.

Scroll to continue with content
  1. Tom

    Tom Guest

    What extra precautions must I take if I wish to switch on and off a
    transformer (say a radio or TV connected to the mains) as opposed to
    just a non-inductive load such as an incandescent lamp? I realise there
    will be a lagging current - will it switch off ok?

    Tom
     
  2. The inductance will produce a spike of voltage when it switches off,
    even though this will occur at very nearly the current zero crossing.
    This is not the ignition coil type of spike that you get when you
    interrupt a big current passing through an inductance. It is just an
    artifact of the phase shift between voltage and current. That phase
    difference means that there is not zero volts across the supply line
    at the same moment there is zero current passing through the
    inductance. The TRIAC stays on till the current is almost zero, then
    switches off. At that moment, the line voltage appears across the
    TRIAC, and this sudden change in voltage might retrigger the triac
    into conduction.

    The normal precaution is to place a series RC network across either
    the TRIAC or across the inductive load. It is slightly more effective
    across the TRIAC, but causes the transformer to see some leakage
    (capacitive) current when the TRIAC is off. The capacitor is what
    provides the reduced change in voltage rate, with the resistor
    limiting the inrush current at TRIAC turn on. Common values are 47 to
    220 ohms and .1 to ,47 uf. The cap has to be rated for across the
    line operation.

    The other problem with using a TRIAC to feed a transformer is that is
    the turn on delays from zero crossing are not exactly equal on the
    positive and negative half cycles, the small DC component of the
    resulting voltage will saturate the transformer core, causing big
    current pulses that heat the transformer and TRIAC. Not so serious
    for very tiny transformers, but big trouble for large ones.
     
  3. Gordon

    Gordon Guest

    John, does that mean that you would use a 47ohm and a .1uf cap for say a
    1/3HP motor and go for the 220ohm .47uf for a 2HP motor and none for a hot
    water heater ?

    If you could point to some reading material I'd appreciate it.

    Thanks

    Gordon
    my address needs gj in front
     
  4. CFoley1064

    CFoley1064 Guest

    What extra precautions must I take if I wish to switch on and off a
    Hi, Tom. You basically have two types of solutions, relays or solid state.
    For relays, you want a snubber across the load, which prevents arcing across
    the relay contacts, extending their life. For an RC snubber, here's how to do
    it. Pick a resistor such that the load current plus the resistor current will
    equal more than half but less than the total current the relay is rated for.
    Put that R across the load in series with a cap (start with, say, an 0.1uF
    line-rated cap) and expose the relay contacts. Increase or decrease the value
    of the cap until the visible arc across the relay contacts when they open just
    disappears. That's a pretty good cap value for your snubber. You can also buy
    these RC snubbers as Quencharcs (R and line voltage-rated C in one package)
    from Newark and other distributors.

    When you're switching inductive loads with thyristors, you first think of using
    inverse-parallel SCRs. Since each SCR blocks reverse, you'll be able to turn
    your load off (if current is above maximum switch-off current when the voltage
    crosses zero, a triac won't turn off).

    One easier solution is the Teccor Semiconductor alternistor, which acts like
    inverse parallel SCRs without the hassle of two gates. You might want to try
    looking at these -- I've had some success switching inductive AC loads with
    them...

    http://www.teccor.com/web/menuitems/products/alttriac.htm

    Also, if you're doing a one-off, many solid-state relays have inverse-parallel
    SCRs instead of triacs -- read the manufacturer's specs. More expensive, but
    it's a "nuke it" solution, and easy to interface without having to worry about
    being at line potential.

    Possibly a better question for sci.electronics.basics?

    Good luck, and take care with the mains -- they're not tame.
    Chris
     
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

-