One of the LEDs in a plug equipped with over-voltage protection does not light up when the device is plugged into the 230 V mains. It has worked well for a while since the product was purchased. An inductive load was used with the plug.

After I took it out of its board, I checked it with a multimeter on the "diode testing" mode and found out that it conducts in reverse, with a voltage drop of 1 V. Despite the fact that it does not turn on, my multimeter shows there is a forward voltage drop of around 1.8 V, which seems to be a normal value for a healthy LED.

I was expecting open-circuit and short-circuit failures, but none of these seems to be my case. Since it conducts in reverse, it must have experienced reverse breakdown at some point, too.

Here's a schematic which can give you and idea about the circuit in which my component was:


simulate this circuit – Schematic created using CircuitLab

The LED and its resistor are in parallel with two varistors, 230 V supply and the load which connects to the plug.

Here's a photo of the LED:

enter image description here

What could have caused something like this?

Has anyone seen something like the aforementioned LED has failed? Maybe something very similar?

EDIT: The green LED did turn on with the help of a few batteries, but all the other things still apply; still showing reverse voltage drop of 1 V without turning on (both for multimeter and batteries methods). Still doesn't turn on when connected to the multimeter.

NOTE: The green LED was replaced by a new one in the device it was part of and that new LED lights up without problems.

  • \$\begingroup\$ Have you tried lighting it up properly with a DC source of high enough voltage and series resistance (or constant current source if available)? I don't trust results of diode testing on LEDs much, because it heavily depends on how it is implemented in the multimeter. \$\endgroup\$
    – Arsenal
    Jul 15, 2020 at 11:49
  • \$\begingroup\$ Check also that it's not a back-to-back pair of LEDs designed for AC use. \$\endgroup\$
    – Transistor
    Jul 15, 2020 at 11:54
  • \$\begingroup\$ @Arsenal Tried with a bunch of batteries and a series resistor and yes, this time it did light up. However, I tried the same things with another LED and it turned on in both cases: when doing the voltage "diode test" with the DMM and with the batteries + resistor. What is going on? \$\endgroup\$ Jul 15, 2020 at 14:43
  • \$\begingroup\$ @Transistor No, it seems that's not the case. I managed to make the LED turn on, but only in one direction. Besides, as the images I posted suggests (even though it could have sowed more details), it's just a single green, cuboid-shaped LED. Can't see how there's more than one in such a package \$\endgroup\$ Jul 15, 2020 at 14:48

1 Answer 1


It seems that circuit is originally just meant for a neon bulb, and the manufacturer has simply replaced the neon bulb with a LED.

In that circuit, there is no protection for the LED from reverse voltage being applied over it. LEDs typically have a reverse voltage limit of 5V.

So it will work with AC circuit like that for some time, but the LED degrades faster than in a proper circuit where the reverse voltage is prevented with another LED or diode.

  • \$\begingroup\$ It's true that it's experiencing reverse voltage, but I don't think it goes past the 5V limit at any instantaneous moment. Are you saying that any amount of reverse voltage on the LED will wear it out? \$\endgroup\$ Jul 16, 2020 at 6:21
  • \$\begingroup\$ Based on one green LED datasheet, the limit is 5V reverse voltage at reverse current of max 10uA. With 230V mains, the Live goes up to 325V positive and 325V negative relative to neutral. There is nothing in the schematic that would limit it to 5V over the LED. Why do you think it does not go negative past the 5V limit? \$\endgroup\$
    – Justme
    Jul 16, 2020 at 7:27
  • \$\begingroup\$ Yeah, you're right, those 230V and 325 V appear across the LED when it's in reverse bias. Error on my part. Now, I wonder if, after reverse breakdown successfully ocurrs, will the LED light up? Or could it be that I measured wrong with the multimeter? There's also the thing that when testing with the multimeter in diode mode, it doesn't light up, while other LEDs normally light up \$\endgroup\$ Jul 16, 2020 at 7:43
  • \$\begingroup\$ I don't know the exact details but I would not expect a LED to light up at reverse breakdown. If it does not light up in the original circuit, it's broken, and it won't light up with your multimeter either. There can be damage that has degraded the LED, it may not be a complete short circuit, but it may have developed a resistive path between the LED terminals, so a multimeter is too weak to light it up. \$\endgroup\$
    – Justme
    Jul 16, 2020 at 7:54
  • \$\begingroup\$ This is the stuff I was especially looking for, thank you! I'm going to add your last comment in your answer + something I spotted, unless you have something against these things :) \$\endgroup\$ Jul 16, 2020 at 11:04

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.