I have a circuit meant to manage a shutter (or a shutter motor to be more precise). The motor has two power lines - depending on which power line one connects power, it moves up or down.

I designed the following circuit to operate the shutter: SSR failing to turn off circuit

RY1 and RY2 are operated via a uC / Mosfets.

I now have the problem that the SSR (RY2) does not turn off even though there is no more current flowing through its control pins (voltage ~20mV, current <1uA). The SSR only turns off when the shutter reaches a final position (fully open or fully closed) - then the motor stops automatically, hence draws no more current on the 230VAC line, and the SSR then stops conducting.

I've already applied a varistor to protect the SSR (R2) because the motor produces some nasty voltage spikes when it reaches an end position. I also added a snubber circuit (R1 / C1) and tried different values (100/100n, 22/100n, 47/47n) to reduce the dV/dt rise across the SSR (as described here), but so far I haven't been successful.

The motor draws roughly 110W.

Is there an "easy fix" for this circuit or should I replace the SSR with a mechanical relay?


  • \$\begingroup\$ You haven't shown the limit switches between RY1 contacts and the motors and these are what ultimately turns off the motors. Question: if RY1 changes state while RY2 is off does the shutter move? i.e., Are we looking at leakage current through the SSR or do the motors stay off until the first time the SSR is 'used'? \$\endgroup\$ – Transistor Dec 11 '15 at 19:19
  • \$\begingroup\$ The motors stay off when RY1 changes (unless RY2 is conducting of course). In fact, the uC makes sure that RY1 only changes during "no load" situations, i.e. when RY2 is non-conducting. When RY2 is turned off via the control pins, it keeps conducting until the motor stops because it reached an end - this can be as much as 40 seconds after the control pins have no more current across them. I've verified the turn-off time using a test-light connected to the output of RY2, and it stops conducting exactly when the motors stops (+-100ms). \$\endgroup\$ – final Dec 12 '15 at 10:41
  • \$\begingroup\$ Good. That proves that it's not permanent leakage through the SCR or varistor that's running the motors. See my answer. \$\endgroup\$ – Transistor Dec 12 '15 at 11:10

You might find that a AQH3223A works better. The AQH3213A is a "zero-crossing" type SSR, while the AQH3223A is a "random" version. Zero crossing types can have issues with motor loads.

From https://www.panasonic-electric-works.com/pew/eu/downloads/ds_x61_en_ssr_technical_information.pdf

4) When controlling loads using zero- cross voltage types in which the voltage and current phases differ, since the triac sometimes does not turn on regardless of the input state, please conduct sufficient tests using actual equipment.

  • 1
    \$\begingroup\$ It says that the triac sometimes does not turn on, my issue is that it does not turn off. But I will give it a shot - I'll pick up some AQH3223 today and will hopefully be able to test them tomorrow. Thanks for the idea! \$\endgroup\$ – final Dec 11 '15 at 16:12
  • \$\begingroup\$ I tried the AQH3223, unfortunately with no luck - the load still doesn't turn off. I assume (after reading some additional material) that the motors have a very low power factor and hence most SSRs don't manage to turn off. \$\endgroup\$ – final Dec 12 '15 at 10:37

I wonder is the snubber in the wrong place? As you have mentioned, dv/dt can be a problem with triacs and it may be causing a false triggering.


simulate this circuit – Schematic created using CircuitLab

Try snubbing the triac. And have a look at Choosing TRIAC snubber resistor for multi-purpose switching.

  • \$\begingroup\$ I will try that shortly and will let you know if it helps! \$\endgroup\$ – final Dec 12 '15 at 18:59
  • \$\begingroup\$ I tried placing the snubber across the triac as you described. I used the snubber configurations 47/47n, 310/47n, 390/100n, but with no real success. However, there was a slight improvement - when power across the SSR's control pins was gone AND the switching relay switched, the SSR sometimes (roughly 50% of the time) managed to turn off. This was not the case without the snubber across the triac (I had at least 20-30 tests before and the SSR never turned off) \$\endgroup\$ – final Dec 12 '15 at 20:47
  • \$\begingroup\$ A quick read of the article linked in my answer suggests that a higher value of resistor may be required with high inductance loads. You can try adding resistance in series with your snubber. e.g., If the snubber has 100Ω inside then add another 100Ω in series, then 200Ω, then 400Ω. I wouldn't go too high with the R value. Sorry I haven't time to do some calculations but I'd be interested in learning what works for you. Be careful with the mains! \$\endgroup\$ – Transistor Dec 12 '15 at 21:08
  • \$\begingroup\$ Yeah, I already tried higher resistor values as written in my comment above... I have two "one-piece" snubber components here, one with 47Ω/47nF, one with 22Ω/100nF. I added a 270Ω resistor in series with the 47Ω/47nF snubber and a 370Ω resistor in series with the 22Ω/100nF type, but as mentioned with limited success :/ \$\endgroup\$ – final Dec 12 '15 at 21:43
  • \$\begingroup\$ Another thing I stumbled across is that according to link, I use a SSR which is too small for the job. Assuming a ~90% efficiency of the motor, I have a 0.1kW motor, which would require at least a 5A SSR according to the guide (possibly even 10A), but I only use a 1.2A SSR. \$\endgroup\$ – final Dec 12 '15 at 21:59

Assuming that you're using the higher-current version, I'd suggest that you look into the motor characteristics. If the motor current and voltage have significant phase shift, and you're using a zero-voltage cutoff unit, it's entirely possible that the SSR never sees zero voltage and current simultaneously, so it never turns off.

  • \$\begingroup\$ Unfortunately I don't have a datasheet for the motors - they're 10 - 15 years old (I have two kinds of them). Assuming that you're right, how would I change the circuit to have it turn off? \$\endgroup\$ – final Dec 11 '15 at 16:17
  • \$\begingroup\$ @WhatRoughBeast, the triac in the SSR can only see current. AFAIK it will always turn off when current is zero. \$\endgroup\$ – Transistor Dec 11 '15 at 19:05

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.