Connect with us

Solid State Relay (SSR) with peak firing SCRs for inductive loads

Discussion in 'Electronic Design' started by P E Schoen, Jun 24, 2013.

Scroll to continue with content
  1. P E Schoen

    P E Schoen Guest

    I have used SSRs with peak firing for inductive loads (transformers), but
    they are generally non-stock items and are rather expensive:
    http://www.wolfautomation.com/products/30434/peak-switching-ssr-single-phasebrcarlo-gavazzi-rm1c
    http://www.galco.com/buy/Crydom/PSD2450

    So we thought we might use some random firing SSRs which can be triggered
    for phase angle modulation or with externally generated peak firing for
    initiation:
    http://www.newark.com/crydom/cwd2490-10/ssr-panel-mount-280vac-32vdc-90a/dp/80K5802?Ntt=cwd2490-10

    We have quite a few of these or similar units in stock which were bought for
    another product that has now been redesigned. I thought it should be easy to
    make a small circuit that would be powered from the 5V to 12V DC control
    signal and sense zero crossing or peak of the 80-265 VAC load signal across
    the output terminals to determine the peak point and fire the SSR at that
    time. However, these random-firing SSRs are also mostly special order with
    long lead times, minimum quantities, and high cost. Crydom seems to be the
    only manufacturer for this type SSR, but there may (and should) be others.

    So, I have thought that it may be better to build the entire SSR from
    discrete devices or dual modules, which are easily available in many
    ratings, and we use them in many products, but with a trigger board that has
    two separate, isolated gate drive circuits, and associated circuitry, which
    results in a rather large, complex, and expensive board.

    I would like to duplicate the drive circuits of the peak firing SSRs, which
    must be fairly simple. I have considered using optoisolators with SCR
    outputs, such as:
    http://www.digikey.com/product-detail/en/TLP748J(TP1,F)/TLP748JTP1FCT-ND/2021328
    and simply using the anode voltage of the main SCR to drive its gate through
    the opto-SCR. But I am worried that the sudden application of the peak
    voltage (up to 375 volts) to the gate might cause damage, although when the
    gate triggers, the anode voltage will drop almost instantly to 1-2 volts.
    And at that voltage, I'm not sure the opto-SCR can maintain gate drive after
    the initial firing. With the opto-SCRs on for the remaining period of time
    determined by the control signal, there will be times during the zero
    crossing where there will be no voltage to trigger, and there may be some
    distortion until enough voltage is present to cause the main SCR to conduct.

    Since it is a highly inductive load, there will be current in the SCR at the
    time of the applied voltage zero crossing, and if the gate is not triggered
    the current will keep it in conduction. If it does stop conducting and
    voltage once again appears on the anode, the opto-SCR will trigger the gate.

    Are these peak-firing (and the ubiquitous zero-crossing firing) SSRs really
    so simple as to use opto-SCRs to drive the gated from the anode voltage? Or
    do they use some sort of trigger pulse transformer for the gates? We have
    found that it is necessary to maintain DC gate drive on both SCRs throughout
    the entire cycle of the sine wave. The other possibility is that the main
    SCRs could be LASCRs, but I have been unable to find any commercially
    available in the size I need (40-150A, 120-265 VAC and possibly up to 500
    VAC).

    Thanks,

    Paul
     

  2. how about an FETs/IGBTs?

    I have been thinking about a short circuit protected master switch for a triac controller board

    A sil dc-dc converter, optocoupler, ir2127 and a pair of FETs or IGBTs back to back


    -Lasse
     
  3. Phil Allison

    Phil Allison Guest

    "P E Schoen"

    I have used SSRs with peak firing for inductive loads (transformers), but
    they are generally non-stock items and are rather expensive:
    http://www.wolfautomation.com/products/30434/peak-switching-ssr-single-phasebrcarlo-gavazzi-rm1c
    http://www.galco.com/buy/Crydom/PSD2450

    ** Very nice, not really that expensive IMO.


    I would like to duplicate the drive circuits of the peak firing SSRs, which
    must be fairly simple. I have considered using optoisolators with SCR
    outputs, such as:
    http://www.digikey.com/product-detail/en/TLP748J(TP1,F)/TLP748JTP1FCT-ND/2021328

    ** Looks a bit wimpy for use with large SCRs.

    This is one of the better ones:

    http://www.vishay.com/docs/83689/83689.pdf

    Despite the term "optotriac" in the title, the output uses SCRs - the
    various dv/dt figures are all 5 or 10kV /uS.


    and simply using the anode voltage of the main SCR to drive its gate through
    the opto-SCR. But I am worried that the sudden application of the peak
    voltage (up to 375 volts) to the gate might cause damage,

    ** You need a series resistor of about 220 ohms to limit peak amps.


    although when the
    gate triggers, the anode voltage will drop almost instantly to 1-2 volts.
    And at that voltage, I'm not sure the opto-SCR can maintain gate drive after
    the initial firing.

    ** It is desirable for the trigger device to turn off ( ie fall below
    holding conditions) as soon as the main SCRs have fired - so they can
    commutate off alone, unimpeded by the trigger having to do like wise.


    With the opto-SCRs on for the remaining period of time
    determined by the control signal, there will be times during the zero
    crossing where there will be no voltage to trigger,

    ** Huh ?

    With an inductive load, zero current will not coincide with zero volts -
    SCRs and Triacs remain on until the current is below holding conditions.


    and there may be some
    distortion until enough voltage is present to cause the main SCR to conduct.

    ** Usually a very small issue.

    Since it is a highly inductive load, there will be current in the SCR at the
    time of the applied voltage zero crossing, and if the gate is not triggered
    the current will keep it in conduction. If it does stop conducting and
    voltage once again appears on the anode, the opto-SCR will trigger the gate.

    ** That is how it works, long as you keep driving the LED.

    Are these peak-firing (and the ubiquitous zero-crossing firing) SSRs really
    so simple as to use opto-SCRs to drive the gated from the anode voltage?

    ** Bet ya they do.

    Or
    do they use some sort of trigger pulse transformer for the gates? We have
    found that it is necessary to maintain DC gate drive on both SCRs throughout
    the entire cycle of the sine wave.

    ** Really ? Why?


    .... Phil
     
  4. P E Schoen

    P E Schoen Guest

    "Lasse Langwadt Christensen" wrote in message
    I have thought about something like this, but SCRs are generally less
    expensive and better suited to high power non-PWM applications. After
    another search I found the following:
    http://www.crydom.com/en/Tech/Tips/SSR the inside story.pdf

    It shows two SCRs connected in anti-parallel and a simple switch closure
    between the gates causes the module to turn on. Then they show the use of a
    rather complex opto-transistor driving another SCR and a FWB for
    bidirectional operation of the switch. But it seems that this would require
    several volts to maintain gate drive after turn-on. It may not be too
    critical, but it seems that it may be enough just to use a resistor to limit
    the current to the gate at turn-on.

    I considered using a small inductor so that the current would not have such
    a high peak at initial turn-on. For some 55A 1kV SCRs I may use:
    http://www.mouser.com/ProductDetail...=/ha2pyFadujd9vKNrwSrCFk3fegDfrcvo9vWU/YdEow=
    gate turn-on time is about 2-3 uSec, and peak gate current is 4 amps, so I
    would need an inductor of about 200 uH. Probably a small one like the
    following:
    http://www.mouser.com/ProductDetail...GAEpiMZZMsg%2by3WlYCkU9E4kXPnClt9kZd7tCtDMeQ=
    220 uH 100 mA 10 ohms and only about $0.35.

    Thanks for the suggestions.

    Paul
     
  5. Ecnerwal

    Ecnerwal Guest

    At the risk of depleting the supply, they are often (but unreliably)
    available by each, at short leadtimes and low cost if you peruse a
    certain popular action site that rhymes with "Be Prey." You have to do
    your own lookup most of the time - the sellers often don't know much
    about them. Some of the buyers may not either. Keep an eye out, and you
    might find something that works for you...
     
  6. P E Schoen

    P E Schoen Guest

    "Ecnerwal" wrote in message
     
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

-