Inverse clipping circuit

Discussion in 'General Electronics Chat' started by cmartinez, Jul 30, 2014.

  1. cmartinez

    Thread Starter AAC Fanatic!

    Jan 17, 2007
    3,555
    2,510
    Hello all, I'm trying to design a circuit that would only allow signals above a certain voltage to pass through it, and signals below that voltage to be be blocked.
    Specifically, I want to allow a signal of up to 120 VAC (of only about 2 milliamps) to pass through the circuit, while blocking any signal below 15 VAC
    That is why I titled this post as "inverse clipping circuit", since its output waveform would be complementary to what a zener's would be.
    I want to use the circuit to protect a solid state switch, so as to bypass (but not shunt to ground) any high voltage signal before it gets through it.

    I'm almost certain that a zener (or tvs) diode would be involved, but I haven't seen any examples online on how this could be.

    Please be patient, my area of expertise is digital electronics, and rarely analog, as in this case.
     
  2. BillB3857

    Senior Member

    Feb 28, 2009
    2,400
    348
    Two 15v zeners in head to head series with the load would do it.
     
    cmartinez likes this.
  3. cmartinez

    Thread Starter AAC Fanatic!

    Jan 17, 2007
    3,555
    2,510
    Thanks for the advice, I'm going to try something like the attached figure..

    Which is, as you said, two zeners head to head in series, AND connected in parallel with my switch.
    My intention is both to allow voltages higher than 15V through, and to protect the switch from those high voltages.

    I'll see what happens using my oscilloscope.
     
    Last edited: Aug 1, 2014
  4. BillB3857

    Senior Member

    Feb 28, 2009
    2,400
    348
    Out of curiosity, what is the purpose of the switch? With the circuit you show, the load will see nearly full power even with the switch open. I tried a simulation and with a 50K load, the current was about 2ma. Here are the traces. I assumed when you said 120VAC that you were using RMS terms. That is why the signal generator is set for +/- 169V.
     
  5. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,386
    1,605
    Out of curiosity, when the input is 14V what is the output voltage?

    Also out of curiosity, when the input is 16V what is the output voltage?
     
  6. cmartinez

    Thread Starter AAC Fanatic!

    Jan 17, 2007
    3,555
    2,510
    You're right, I should've elaborated a bit more.

    I'm trying to interrupt a signal being delivered to a piezo-electric actuator, this signal is a "floating" one and has no reference to ground, from what I've deduced so far. The signal starts with a series of quick pulses peaking at 120V, lasting about 0.6 ms in total, and then quickly comes down to around 12V, and that stage can last up to 16 ms. This actuator is for the quick opening and closing of a very special valve in a machine whose behavior I'm trying to tweak a little.
    So what I'm trying to do is to cut short the 12V signal a few moments AFTER the 120V signal ends, and for this end I'm going to use a CPC7514, which is a high voltage analog switch that can in theory withstand pulses of up to 320V. But I'm still not sure if I'm going to fry the poor little sucker, since the load of a piezo actuator is normally a slightly inductive one with a significant capacitive component to it. That is why I'd like to protect it a little from the 120V pulses, if I can.
    Now, I have tried using lots of different components for this purpose, but since the signal is floating, there's virtually no reference ground. And I have not being able to use a transistor nor a darlington, beause they both need a little current to work and the piezo driver is delivering only the current that the piezo component needs.
    You see, it's imperative that the piezo valve receives those 120V pulses in the beginning for it to be able to open up, and the rest of the 12V pulses are there to keep it open. And since I'm trying to cut short the amount of time that the valve stays open, all I'd need to do is interrupt the 12V signal at some point, without ever interfering with the 120V one at the beginning.

    The only thing that's worked so far is when I used an opto-isolated SCR. But once an SCR is triggered, it won't switch off until the signal being delivered to the piezo actuator ends... I've tried using the SCR forced commutation technique, but I can't make it work either due to limits in the capacitive load that the piezo driver will accept.

    And that pretty much sums up what my dilemma is...

    See attached waveform image (ignore the yellow trace), and datasheet. The circuit shown is only an approximate representation of what I think the piezo driver might look like, that is because the driver is encapsulated in resin, and I can't peek inside.
     
Loading...