I've been designing a smart home system for a while and one of its main features is remotely controlling all the lighting.

I'm using a pretty simple design where the main components are an ESP8266 (talking to a home server), an opto-triac MOC3052 and triac BT136. In essence, the switching part is similar to this circuit (borrowed from elsewhere):


I did not include the snubber in the circuit (because I'm not connecting any reactive load) and now I'm wondering whether I should have.

Everything works splendidly, except that especially in one room, the TRIAC seems to shortly self-trigger when some other lights on the same floor are turned on or off, which causes an irritating blink of the light.

My guess is that this is due to long wires between my circuit and the actual light, so enough power gets induced when other lights are toggled nearby, which is a sufficient spike in voltage for the TRIAC to trigger.

Is my guess correct and is this something the snubber will resolve?

  • \$\begingroup\$ Common mode choke if it's due to EMI. Can you measure on your gate pulse in circuit? \$\endgroup\$
    – winny
    Commented Sep 20, 2016 at 8:29
  • \$\begingroup\$ 56r and 100r resistor value are which watt? \$\endgroup\$
    – Jaimin
    Commented Jan 9 at 7:09

2 Answers 2


Triacs have a dV/dt specification, above which they can self trigger.

Interrupting loads on the same circuit (so lights in adjacent rooms) can easily cause spikes on the supply.

A snubber will go some way to help, it will increase the threshold, but may not be enough. As these spikes are differential mode, a common mode filter may not be enough.

  • \$\begingroup\$ What would you suggest to do then? \$\endgroup\$
    – LubosD
    Commented Sep 20, 2016 at 9:48
  • 1
    \$\begingroup\$ @LubosD I'd add the snubber, and I'm pretty sure that it will be sufficient. The snubber will limit the dV/dT (rate of change of voltage) below a level where capacitive coupling can trigger the triac. \$\endgroup\$
    – jms
    Commented Sep 20, 2016 at 10:32

[PROBLEM] I've been facing this issue for a week and I think I've finally solved it. To understand why you're facing this issue, there are 2 phenomenon that need to be understood. 1)- switches are physically moving metal contacts and when they are a particular distance away from each other, the voltage between them is sufficient to cause an arc. The ark frequency changes with the distance and effectively a <5us , ~1000V spike is created across the live and (inverse spike) neutral line . 2) - triacs have 2 important specs, the dV/dt limit and the gate current. The spikes mentioned above are usually in the range of 100-400 V/us so the triac you buy needs a to have a specification above that.

[SOLUTION] all my statements are based only on my observation and are not hard and fast rules

Hi-com triacs generally have 1. a high dV/dt limit around 400-1200 V/us 2. a gate current of around 35-50 mA 3. Are stated to be great with inductive loads or not require an external RC snubber (snubberless) 4. Operation in 3 quadrants (3Q)

While other (sensitive gate) triacs have 1. a low dV/dt limit around 50-200 V/us 2. a gate current of around 5-30 mA 3. Are stated to be used for lighting and dimmers 4. Operation in all 4 quadrants (4Q) ( means it works easily with negative gate current )

So pick a triac with these characteristics like the BTA series rather than BT series which are better suited.

  • \$\begingroup\$ I m trying On BTA16 But same problem is getting , I have used 560ohm instead of 56ohm and 330ohm instead of 100ohm and not using snubber \$\endgroup\$
    – Jaimin
    Commented Jan 20 at 4:25

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.