I'm intending to control two motors, one in on/off, and another with phase control to adjust the speed. I've already implemented the zero cross detection circuit, which works fine.

But even when applying 5V to the TRIAC driver, the TRIAC doesn't conduct.

When I short the A1 and A2 legs of the TRIAC, the lamp turns on, which suggests that the TRIAC is in fact the problem.

The voltage drop over the MOC3023 photodiode is ~1.15V. With a 220ohm resistor, that should give 17.5mA, which should be enough to turn on.

I've tried with two circuits.


simulate this circuit – Schematic created using CircuitLab

The oscilloscope is showing:

Channel 1: Mains voltage (via 10k resistor)

Channel 2: TRIAC gate voltage

Channel 3: Load voltage

enter image description here

I'm assuming that the gate voltage is following the mains voltage, because the TRIAC driver is conducting.

Is the resistor values too high and there's not enough current through the TRIAC (BTA12)?

Why does channel 3 look so weird? Is it because the TRIAC is conducting partially? It seems to be ranging from +50V to -125V.

I'm aware that an inductive load may cause the circuit not to turn off properly, and that I might have to have a snubber circuit (like in circuit 2) to ensure correct turn off. But I was thinking to introduce that later.

Reference circuits

I've drawn inspiration from these circuits I found online. They seem to use resistors ranging from 100 ohm to 3k ohm. Since I wanted a low standby power consumption, I chose 1k (which is further limited by the 100nF capacitor).

enter image description here

enter image description here

  • 1
    \$\begingroup\$ You are doing something dangerous trying to read mains with a non-isolated scope. It can also give confusing readings depending on where the earth is relative to the mains. Or it can blow the ground lead off your scope. Anyway, you say it is not conducting. Try shorting the MT2 to gate with a 100R resistor (the resistor may burn up if the triac does not trigger so stay away from it). If it turns on, then remove the resistor and try shorting 4 to 6 on the MOC and see if that turns your lamp on. Make sure your load (the lamp) is connected. \$\endgroup\$ – Spehro Pefhany Mar 12 '17 at 3:41
  • \$\begingroup\$ Disconnect the scope entirely for now (use a suitably rated handheld DVM if you want to see what is going on). \$\endgroup\$ – Spehro Pefhany Mar 12 '17 at 3:42
  • \$\begingroup\$ It worries me when I look at that first reference circuit... REALLY Load goes between hot and GROUND ! \$\endgroup\$ – Trevor_G Mar 12 '17 at 14:36

Um..it looks to me like the MOC is not turning on... since there is no R load circuit path. According to the MOC spec it should look like this....

enter image description here

The load is on the other side of the gate on that too...

Also be aware, driving an inductive load with an SCR is a bit different from a resistive load.

  • \$\begingroup\$ Okay, I can try to solder one on. Any suggestions about correct values for the three resistors? It's a bit of a pain to change them. \$\endgroup\$ – user95482301 Mar 12 '17 at 3:30
  • \$\begingroup\$ UNfoortunately it did not say. The gate load being on the wrong side may also be an issue though. Vgt will change if u can get it to switch \$\endgroup\$ – Trevor_G Mar 12 '17 at 3:34
  • \$\begingroup\$ So it makes a difference whether the load is attached to A1 or A2? I thought the TRIAC was bi-directional. \$\endgroup\$ – user95482301 Mar 12 '17 at 3:37
  • 1
    \$\begingroup\$ @trevor. It is incorrect that you need the resistor you point out. It is there only to provide a leakage path just in case you are driving a Sensitive Gate Triac. \$\endgroup\$ – Jack Creasey Mar 12 '17 at 4:49
  • 1
    \$\begingroup\$ @Trevor. It is conventionally as shown above, but it technical makes no difference. An RC low pass filter does not know which side is signal and neutral on a floating AC pair. \$\endgroup\$ – Jack Creasey Mar 12 '17 at 16:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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