1
\$\begingroup\$

I am using an Omron G3MB-202P SSR to switch an AC outlet on and off as depicted in the below sketch.

enter image description here

This works fine. However, when connecting a phone charger to the "controlled" AC outlet while the relay is off, the charger goes on and off every 1-2 seconds. Why is that so?

I have tried the same out with a different SSR, namely the Fotek 25DA and found that it does not result in such a "reset loop".

\$\endgroup\$
2
  • \$\begingroup\$ What is the minimum load stated in the SSR datasheet (that you didn't link)? \$\endgroup\$
    – Transistor
    Mar 30 '18 at 12:43
  • \$\begingroup\$ The Omron SSR's datasheet (now linked) does not specify a minimum load current, but it says here that most Omron SSRs' mininum load current is specified at 0.1A. \$\endgroup\$
    – gnzg
    Mar 30 '18 at 13:11
2
\$\begingroup\$

There is typically a snubber across the thyristor in an SSR. The snubber consists of a resistor in series with a capacitor, with a total impedance that is the (vector) sum of the two. There are also leakage currents through the thyristor but they are negligible under normal conditions.

This is specified on the Omron datasheet as a "leakage" of 1.5mA at the Japanese standard voltage of 200VAC, so you can assume it to be 10% more at 220VAC. That is enough current to get your charger to attempt to come to life (and it often does similar things to LED lamps).

enter image description here

The Chinese SSR also shows a snubber, with even higher "leakage" current spec. The fact it doesn't do a similar thing may have more to do with Chinese cost-cutting vs. Japanese fastidiousness than what the designers intended. i.e. they may have substituted an alternistor (aka snubberless triac) for the triac and deliberately omitted the snubber. From the spec sheet it should be worse.

You can probably get the circuit to behave without the crude method of throwing away power by putting a somewhat bigger "snubber" in parallel with the load. Perhaps something like 100 ohms fusible resistor in series with a few uF (the capacitor should be rated for X operation).

\$\endgroup\$
4
  • \$\begingroup\$ Thank you for the super detailed answer Spehro! The reference to the snubber is a good point, and I will look into that. For now, I will stick to a MOSFET relay rather than a thyristor, just for the sake of ease of use. \$\endgroup\$
    – gnzg
    Mar 30 '18 at 16:07
  • \$\begingroup\$ The Fotek is not a MOSFET relay, of course. \$\endgroup\$ Mar 30 '18 at 16:20
  • \$\begingroup\$ Seeing as the Fotek one is also a thyristor, can the difference in the resulting current leakage between the two be condensed to how good their snubber circuits are? \$\endgroup\$
    – gnzg
    Mar 30 '18 at 16:32
  • \$\begingroup\$ As I said above, I suspect one is there and one isn't. It has to "leak" current to be effective. But there is no way to know without dissecting the unit. \$\endgroup\$ Mar 30 '18 at 16:34
1
\$\begingroup\$

Leakage currents increase in Triac higher current rating and is lower in MOSFET SSR’s.

The charger builds up a rectified voltage to some threshold then starts only to discharge and restart as a slow relaxation oscillator.

\$\endgroup\$
2
  • \$\begingroup\$ Thank you for the succinct answer Tony. Would perhaps adding a low impedance resistor for the leakage current in parallel to the AC output of the SSR work to nullify the leakage current? \$\endgroup\$
    – gnzg
    Mar 30 '18 at 13:46
  • \$\begingroup\$ Yes, usually 1W tungsten Xmas bulb will do or even two in series \$\endgroup\$ Mar 30 '18 at 14:00

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.