enter image description here

Source: http://www.amb.org/audio/epsilon24/

I have been trying to understand the operation of the above circuit. To simplify, I am ignoring K1, Q1, R4 and D2. I also assume there is a jumper across T1 and T2. Since I assume there is no K1, I assume that R3 is connected to 12V rail and the LEDB junction is connected to the drain of Q3.

As I understand, this is a circuit which is used to power up a device from mains (through a 12V relay) using a momentary switch. When the circuit powers up C3 will charge up to 12V. Q3's drain is also at 12V. Q3's gate, because the capacitor charges up to 12V, should be at almost 0V and therefore it should be off.

What I don't understand is what happens when the switch is pressed: I assumed that Q3's drain is at 12V and the cap is also at 12V. But if that is the case, nothing should happen when the switch is pressed because both points are at 12V.

So I must be wrong with some of my assumptions.


1 Answer 1


When power is first applied to this circuit, voltage is applied simultaneously to the top of R1 and to the top of R5 (via the output relay). However, Q3 has a significant amount of gate capacitance, so Q2 will turn on first, shorting the gate of Q3 (and the current coming from R1) to ground. This will also prevent C3 from charging. The circuit will be in the "off" state.

The first time the button is pressed, C3, which is discharged, will pull the base of Q2 low enough to cut it off. This will allow Q3 to turn on, which pulls its drain low, turning on the relay and cutting off Q2. The circuit will be in the "on" state. Now C3 can charge up via R1 and R2.

The next time the button is pressed, the voltage on C3 will turn on Q2, which will short the gate of Q3 to ground, turning it off. This will also discharge C3, but Q2 will be held on by the voltage on the drain of Q3, via R5. The circuit is again in the "off" state.

Obviously, if you simply hold the button down, it will slowly toggle between the two states.

  • \$\begingroup\$ Out of curiosity, is there a simple modification that could be made to prevent the circuit from indefinitely toggling in the event of a short across the button? \$\endgroup\$
    – JYelton
    Dec 28, 2012 at 17:39
  • 1
    \$\begingroup\$ You could try putting a capacitor in series with the switch (same value as C3), and a very-large value bleeder resistor in parallel with the new capacitor. Make sure that the bleeder current can't operate the circuit. The R-C time constant of the added components will essentially enforce a minimum time between operations. \$\endgroup\$
    – Dave Tweed
    Dec 28, 2012 at 17:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.