LED glows during power outage

Status
Not open for further replies.

gthomas785

Member
Location
MA
Occupation
Physicist
Hi Everyone, I am not an electrician but this isn't a DIY post I'm just curious about something I observed. Wondering if anyone on this site has any insights.

The other night we had a power outage. A storm came through and presumably knocked a branch onto a wire, my whole street was out as well as a couple of adjoining streets. However, My neighbors around the corner still had power. I live in a single family house with 240/120 single phase service and share a transformer with 4 other houses on my street.

So I get up in the middle of the night to use the bathroom and notice that the LED night light plugged into the GFCI above the sink is glowing. Nowhere near the usual brightness mind you, but bright enough that I can see my way around the bathroom. I unplugged it and it went out, plugged it back in and it came back on. As I watched it longer, I noticed that the brightness was kind of slowly pulsating, it would fade away (like a capacitor discharging) and then suddenly come back on. The time between pulses was completely random. Keep in mind none of the other lights in the house were doing anything like this, it was a total blackout. Kinda freaky actually. I didn't bother to do any further testing because it was the middle of the night and I wanted to go back to bed.

We've had power outages in the night before and this is the first time I've seen this. So I'm trying to figure out why it happened. I have a couple of ideas.

I've seen the "LEDs glow when off" threads on here and it's usually blamed on either a switched neutral or some capacitive coupling across switch terminals. But this is different, a power outage on my whole street, those scenarios would not apply.

Obviously, there was some small amount of voltage on the line that was enough to light up my tiny little LED night light. But how? My first thought was, my neighbors had a generator running and maybe they managed to somehow backfeed to my house. But the voltage was so low, their generator should have been way overloaded for that to be the case.

The second thought, which I think is more likely. My neighbors around the corner still had power, and I assume that our neutral wires are connected together either through the POCO neutral or the metallic gas and water piping. And, since the "hot" wire was not hot but still connected from my house, to the street through the transformer, there would probably be a very large capacitance available on the hot side. So even a tiny amount of voltage on the neutral/ground coming from the neighbors might supply enough current to drive an LED. What do you think of that? Any other ideas? What would cause that "pulsating" effect?
 

Rock86

Senior Member
Location
new york
Occupation
Electrical Engineer / Electrician
This may sound silly... but I had the power off while working on a receptacle one time and I got a small shock when I touched the hot leg. circuit breaker was off, no shared neutral, but... there was a cell phone plugged into a charger on an adjacent receptacle. I'm not saying with 100% certainty, but when I unplugged it, the line was safe to touch. I assumed at the time that the charger was cheap and was able to back feed on to the line. I never tested the theory after that.
 

synchro

Senior Member
Location
Chicago, IL
Occupation
EE
The glowing is likely due to capacitive coupling between adjacent conductors within cables of the distribution system, where at least one conductor is still hot.

I suspect the slow pulsation is caused by the charging and discharging of a capacitor as you said, where this capacitor is within the LED light after rectification to produce DC. Once enough voltage is on the capacitor then the active circuitry may start to function and draw current which then discharges the capacitor, and the slow pulsating cycle begins again. This type of behavior is characteristic of what is called a relaxation oscillator.
 

gthomas785

Member
Location
MA
Occupation
Physicist
Hmm.. the cell phone charger idea is interesting, I did have my phone plugged in.

But if that voltage is coming from a cell phone battery, then what about all the other low-impedance loads that must have been connected at that point (fridge compressor, etc since power had been out for a few hours) they should pretty much suck the voltage down to zero, right? This wasn't a case of an isolated branch circuit and I can't imagine that my phone has enough juice to keep the whole house at 3-5 volts for very long.. But maybe I'm not thinking about this the right way.

Capacitive coupling in the distribution system makes sense. What's surprising is that I've never seen it happen before. We only have a single phase leg on our street, but I believe there are 3 phases down the hill where the fault occurred. I guess it could be just the unique circumstances of this particular power outage that would have allowed some energy to be coupled into our line and it was enough to make the night light glow.

As for the pulsing, yes that makes total sense. There is a rectifier with what I guess is a pretty big filtering capacitor in the night light, since it also takes some time to go fully dark if I unplug it while the power is on. Depending on what type of voltage regulator it has, that capacitor could probably charge up even with very little voltage present and push the LED over its threshold.
 
Status
Not open for further replies.
Top