I've just bought some mains coil relays from eBay and they are fitted with a led indicator. It looks like the LED has a series 100k resistor connected in parallel with the coil. I always thought that leds had relatively low PIV and if reverse biassed would let all the light out in big rush!
I can't imagine these leds to be anything special, the relay including DIN rail socket was only £3
So how come it is working?
It probably has a standard diode in reverse across the LED.
That indicator circuit is costing you just over half a watt.
To say nothing of the not inconsiderable back-EMF of the coil.
Perhaps they are special, with at least a series diode.
As a mass produced item, effectively a solid-state "neon", it needn't be expensive.
Yeah but that 100k resistor limits the current to 2mA.
I've some Omron MY2IN mains relays with LED indicators. The diagram of the coil part printed on them is:
+----------+ 13 -----+ +----- 14 | | +----+ +----+ | +----------+ | | | +-----|>--/\/\/\-----+Mains is applied across 13 and 14.
What do you think they are doing? It's neat and I've taken the cover off and checked.
Some LEDs actually consist of two LED diodes, of either the same or of different colours, in inverse parallel (sometimes ambiguously called 'back-to-back' IIRC). For something made in volume, putting the two diodes on the same die would probably be the cheapest approach if you wanted the result to work on AC.
They should have used an RC snubber, something like a 0.1uF capacitor with a 100 ohm resistor in series, as a current limiter. But that probably costs more to manufacture.
These are very similar Omron MY2N-J - They came from an Ebay discounter with the makers name obscured by some paint but a little solvent revealed the information.
I will take one apart to investigate further when I get an idle moment and also try a standard LED in a similar configuration.
I can see no sign of series or shunt diode.
It could be that the coil is also a transformer for the LED, although I haven't seen that done.
There are dual red/green LED's comprised of a red and a green in parallel back to back, so you can light either one by suitable DC polarity, or both to make yellow on AC. Maybe this is what you've got?
Yes that was an idea that occurred to me too but the transparent cover does not look as it that is the case.
once you have a coil, making it an (auto) transformer is trivial..
once you have an LED. packaging it with a reverse diode, or indeed a full wave bridge, is also trivial.
Too little information, but as an ex electronic designer, that's the sorts of solutions I would go for.
You can work out an equivalent RC circuit like this...
Need to know the peak current LED can handle - most are rated 20mA continuous, and can handle non-repetitive peaks of more, say 40mA. Peak current happens when circuit is energised at peak mains voltage with the capacitor discharged (capacitor looks momentarily like a short circuit). [1]
Peak mains is 340V
340V / 40mA = 8500 ohms.Now we need a capacitor with an impedance at 50Hz to make up the difference bwteen the original 100kohms and this 8500 ohms.
100kohms - 8500 ohms = 91500 ohms.C = 1 / (2 * pi * f * Z), so
1 / (2 * pi * 50 * 91500) = 0.0347µFSo, adjusting to standard component values, you want an 8k2 resistor in series with a 0.033µF capacitor, which gives an LED current of
2.3mA (instead of the original 2.4mA).Power consumption drops to 0.046W, less than 1/10th of the original.
Note that this circuit requires that the LED conducts both directions, either back-to-back LEDs or LED back to back with a standard diode.
[1] Actually, there's a worse case if circuit is energised very shortly after it was de-energised and the capacitor still holds some charge in the opposite sense to the mains voltage, in which case 340*2 is the max voltage, but this is an unlikely scenario you might choose to ignore.
Some experimental results:
A cooking grade red led out of my stock shows a reverse voltage breakdown of about 15v. This diode in series with 100k resistor taken up in ac volts on a variac seems happy (for several minutes at least) at 280v ac (the maximum I can go to. The red Led in the relay shows no such reverse breakdown up to 100v (again the maximum my bench supply will do)
So whilst a standard led would appear to survive AC with a suitable current limiting resistor, at first sight the ones fitted in the relays would appear to have a series diode inside the package. However the forward voltage on the relay led and my standard led measured at identical currents are the same within 10mV which would suggest that the relay diode cannot have an extra series diode.
Until proven otherwise, I can only assume that the relay led is a special device with a very high >100v reverse breakdown voltage.
I'm off to lash up a higher voltage variable supply now.....
diagram
although
A transformer is what it appears to be. On my relay there is no connection from either 13 or 14 to any of the LED and resistor wiring. I'll admit at first casual inspection I did think the metal bits all joined together but real close examination showed that not to be the case and there are separate fine wires disappearing into the coil from the bits that the LED/R connect to.
I was surprised to learn that the trick to getting more light from LEDs was to pulse them with very high peak currents (for, of course, very short periods of time so as to keep the average dissipation within reasonable limits).
I would guess that they might tolerate a pulse of a few amps if it was only for a few microseconds. Only the manufacturer would know details about the thermal inertia of the LED plus its mounting, but next time, instead of being like you very conservative, I'm going to just try a standard RC snubber as provided ready made in one not-too-expensive package, and see what happens.
It would be even less if you could get away with a 100 ohm series resistor.
I worried about that sort of thing (and lightning-induced transients) around 20 years ago, but the LEDs are still functioning in a circuit similar to yours. They have, I think, dimmed a little, but it was a cheap surplus three-wire red+green LED pair, which I intended to appear yellowish when both were on and power was applied to the freezer, and a warning red when the power had been turned off at the switch.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.