Triac-switched mains receptacle with failure sensing

Thread Starter


Joined Jul 25, 2012
I'm building a 'remote-control' for three receptacles that are barely accessible. They'll be powered nearly 100% of the time, but sometimes the items plugged into them need to be power-cycled.

I'm a low-voltage DC/TTL kinda guy. I know my way around simple AC circuits (I'd have no qualms about wiring a room, for instance), but I've never tried solid-state switching of AC before.

Forgetting about the ~100% duty cycle, below is a simplification of what I'm got so far. SW1 turns triac D1 on (and off). So far so good, but my actual problem comes with detecting faults in the VO1-D1(gate) line and D1 itself -- specifically, VO1's diac's conductivity not tracking SW1 (i.e., conducting when SW1 is off or failing to conduct when SW1 is on), or D1 similarly conducting (or not) contrary to intention.

Complicated by detecting the failures even if there's nothing plugged into the receptacle completing the circuit.

My theory of operation for this is as follows:
  • SW1/R1/VO1 control D1's gate; when SW1 is closed, VO1's photodiac conducts and the gate follows the voltage at D1's T1 terminal. When SW1 is open, D1's gate voltage should be zero and D1 should not conduct.
  • R2/T1 form a follower for D1's conductivity, even when the receptacle isn't in use.
  • D2/D3 clamp T1's primary voltage to 6V. R2 drops the voltage across T1's primary to that approximate range, so D2/D3 are intended to provide nominal protection against surges and small spikes.
  • T1 is a 1:1 transformer, so when D1 is conducting so should VO2.
  • When everything is operating correctly, both inputs of NAND gate IC1 should be low when SW1 is open, and high when SW1 is closed. Any problem with VO1 or D1 should cause the inputs to be different, driving FLT high.

In looking at this, I think I need to make the following changes:
  • Move D2/D3 to T1's secondary side, and insert a resistor in the secondary circuit. (Duh!)
  • Rewire IC1's input 2, because I think right now it'll cause VO1 to conduct regardless of SW1's position.

I have the feeling I'm over-engineering this, but as I mentioned I've never had to work with sensing AC flow on a is-it-there-or-not basis before.

I've asked about this on ScienceForums.Net and PhysicForums.Com, and the current schematic incorporates some comments made there.

I'd appreciate any comments, suggestions, or constructive criticism..



Joined Mar 24, 2008
Welcome to AAC!

Now and then the forum software sees a link in a post and moderates the post. This happened in your case. I have approved the thread, and once you have 10 or more posts it won't happen again.

Have fun!