Fixing Christmas Lights And Shocking Yourself Silly


As [Medhi] was setting up his Christmas tree, he found a string with a few broken lights. Because he’d bought a cheap string of lights wired in series, of course one bulb was burnt out, rendering the entire string useless. His original game plan was to search through the entire strand for the broken bulb, but that’s the easy way out. His backup plan was to zap the broken bulb out of the string. After a few hours of figuring out what that meant, he came up with a way to fix a broken string of lights.

When a bulb burns out, the filament breaks creating an air argon gap between the two electrodes. By sending a huge voltage down the string, it should fire an arc through that gap, illuminating the burnt-out bulb for a brief time.

Experiments with socks and low humidity commenced, but it wasn’t until [Medhi] stuck his finger in a lighter that he found a better source of high voltage sparks. [Mr. Brows] connected the piezoelectric element to the plugs on his string of lights and… nothing happened. At least until he plugged the lights back in. Then, strangely, they worked. The reddit thread for the video says this behavior is due to an anti-fuse built into the bulb. When enough voltage goes through this anti-fuse, a thin sheet of insulator breaks down and allows dead bulbs to short themselves.

Hackaday head honcho [Mike] just got this method of finding dead Christmas lights to work, replacing 14 bulbs in a string of 100 lights. This leads us to an interesting question: why isn’t this simple method of fixing a string of Christmas lights common knowledge? You would think something this useful wouldn’t be introduced to the world via a YouTube video where someone  repeatedly burns and shocks himself. You can, of course, buy something that does the same thing, but this is far too simple of a solution for a classic problem to pass under our noses for this long.

40 thoughts on “Fixing Christmas Lights And Shocking Yourself Silly

  1. Well, the anti-fuse is probably a way for the burnt bulbs to fix themselves. In series, each bulb would normally have a reduced voltage drop, but when a bulb is burnt out, that one bulb will have the entire 120 VAC at its terminals. If the threshold for the anti-fuse’s breakdown is below 167 V or so, it should automatically short and return the lights to functionality. This is probably just a result of poor tolerances.

    1. Would it not be a bad idea if they “self fixed” for the reason described in the video? Each bulb that goes short circuit means a slightly higher voltage is seen by the rest of the circuit and increases the stress on the remaining bulbs. If they self fixed you could end up with the string of bulbs going into a run away condition where the bulbs are blowing, then going short circuit until eventually the whole string is a short circuit connected to the mains….bad idea….

    1. Exactly – I bought one of these last year, they are awesome. Pull the trigger, and it shoots a high-voltage blast through the string. I don’t think i pulled the trigger more than 10 times to fix many a non-working sections last year. You can do it with the string plugged in, and plug the gun into any bulb socket, so the string just lights up when it’s good to go. Being curious, I had to do a little research after I got it to see how this ‘magic’ worked… Plug in a set of series bulbs, plug this ‘gun’ in the place of any bulb, pull trigger and they magically work? It was like one of those brain teaser led/switch youtube videos at first…..

    2. Typically when a light bulb burns out, the shunt trips and the light strand stays on. Problem is that when the filament doesn’t burn out, but breaks during off season storage, the shunt never trips.

      Back to this device… I have one and it is great for relighting strands. The only problem is that every time you shock the sucker it weakens the other working bulbs just a little. Too much shocking and you are replacing 20 bulbs in a still lighted strand.

      1. Too many shorted bulbs in a string would cause a cascade failure anyhow, with all lights popping out due to overvoltage, because the remaining bulbs will see higher and higher voltage across them every time one goes.

  2. I use a 120V to 6V transformer backwards (120v side to dead light string power plug, 6 volt winding to a 12 volt 1A wall wart via a switch between the wall wart and backwards transformer). This has a lot more joules behind it than a piezo lighter. I find if the string is long enough to be two or three stings – removing one bulb from each of the working ‘substrings’ helps a lot to make the zap effective.

    I’ve had an outdoor LED string go bad due to LED leads rusting/corroding off after 3-4 holiday seasons. Moisture gets into the sockets and eventually rusts/corrodes the LED wires. This is a real problem to LED string life, as the LEDs themselves should last 50K hours. It’s too much trouble to try and weather seal a new string. The problem is this type of failure is difficult to find except to go LED by LED. Granted I purchase big box inexpensive sets, but I’ll guess they are all made the same and by the same few companies unless you go $$$$$ commercial grade.

  3. I’ve seen Christmas light bulbs with an extra loop of wire around the two wires that anchor the filament (and supply power to it). I thought the idea was that when the filament failed, the two wires sprung apart slightly and made contact with the loop. This then shorted out the dead bulb, allowing the rest of the string to function. However, having read this, I wonder if it was voltage-sensitive anti-fuse?

    1. No, you’re right. The idea is that when the filament breaks, the tension between the prongs releases and creates a short with the wire winding at the bottom. It doesn’t always work :( And if enough of them break the voltage on the other bulbs begins ramping up and they last less and less. So change them when you see them, don’t wait.
      If you haven’t any spares, you can get away with twisting the bulb in the socket to force a short in the bulb base (tangling the leads up w/ each other).

  4. This brings back memories of the sixth grade, when I and a classmate discovered you could remove the element from a lighter and use it to shock people. For a while it was even more fun than fire.

    It was great fun.. for us.

  5. Something doesn’t add up here. Aren’t those “anti-fuse” things supposed to keep the rest of the string lit when one of the bulbs burns out? He didn’t stretch out the string to see if there was one bulb out, but they were all out. Why would they all go out if the burned-out one had one of those bridge things? Then he zapped it, and then the string came back on. I wish at that point he’d have stretched the string out so we could see if one was still out, the anti-fuse having acted, or if the HV spark had somehow spot-welded the burned filament back together.

    The last time I messed with Xmas tree bulbs I was just a kid, and when a string went out, we’d lay it on the floor, take a brand-new bulb, and swap out the first bulb in the string. Then if they didn’t come on, we’d leave the new bulb where it was and take the just-removed bulb and swap out the second one. Then use THAT bulb to change out the third one, until the string came on, in which case we’d throw away the bulb that we’d just swapped out.

    I’ve also seen modern strings where the sockets have a shunt so that you can change out a bad bulb while the string is lit.

    1. The anti-fuse is a short segment of coated wire wrapped around the filament feed wires and electrically in parallel with the filament. The insulation coating (probably an oxide coating) prevents it from conducting any electricity. If the filament blows, the idea is the insulation on the wire breaks down and then effectively shorts out the filament (which was open anyway). Sometimes the insulation on the shunt wire does not break down – causing the whole string to go dark. High voltage sparks (like from the gadgets they sell) are plenty enough voltage to break down the shunt wire insulation. Even if there are many bulbs out – enough sparks will break down the stubborn shunts and allow the string to light – just get the dark bulbs replaced because the remaining lit bulbs are sharing the voltage usually dropped by the burned out bulbs – and they will be stressed. For example if it’s a 50 bulb string using 2.5v bulbs, and you have 7 or 8 burned out, the 42-43 lit bulbs each now have roughly 3v on them (20% overvoltage).

  6. We used to use a small no-contact AC detector (the $10 kind you buy at the hardware store to check if a wire is live) and run it down the string till it stopped beeping. Doesn’t work so well on the LED strings (being DC and all).

  7. Apparently they use the same system on street lights. A whole string of them are wired in series, and each bulb has a big antifuse, a disk of some material or other (sorry for the lack of information here). When a bulb blows, the entire however-many-thousand volts ends up across the disk, and pops through it.

    Of course they could just wire them up in parallel, but series saves on wire, depending how you loop them.

  8. This is weird. Here (at 220V) these anti-fuses work quite well.
    Is this problem maybe due to the 120V not being enough to blow the anti-fuses?

    Not sure about the point with the raised voltage either, did anyone measure the resistance of a closed anti-fuse? Maybe it fits and compensates for the bulb just fine…

  9. I’ve never heard the term anti-fuse before, but the “auto-shunts” are made of anodized aluminum wire wrapped around the filament posts 4 or 5 times.. The trick is to get the anodization juuuust thick enough to not short during normal operation, when there is 2-5 volts (RMS AC) potential across the bulb, but juuuust thin enough to arc thru to the filament support posts and spot-weld itself in place when the filament goes open circuit and there is then 120 volts of potential across the bulb. There is a very tight tolerance for the anodization thickness, and being cheaply made in 3rd world countries, it is often out of tolerance (too thick) and fails to arc over at 120v when the filament goes out. The much-higher-than-120v pulse from the piezo element is usually enough to cause the arc thru the anodization to occur (when 120v isn’t enough ‘cuz the anodization is too thick) and weld the shunt into place.

    My father claims to have independently invented the same trick in 1960 to fix a piece of lab equipment where he worked. It used a bunch of #44 bulbs in series, and he wrapped anodized aluminum wire around the contacts at the base of the bulbs so that it kept working (and not ruining the experiment) when a bulb burned out.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.