Followup: Reproducing Impossible Circuits

Last month we caught wind of an impressive display of troll physics. [Fredzislaw100] out of Poland posted a video of a circuit that should not exist. As expected, the comments in our coverage blew up with 200 posts. About half the commenters called a little Adobe After Effects trickery, while the other half offered up an electrical explanation. We’re happy to report that [Alan] successfully reproduced the impossible circuit and earned an incredible amount of electronic wizard points in the process.

[Alan]’s solution uses low-frequency AC with the first two LEDs. The first LED points forwards and the second diode is reversed. Easy enough. For the third LED, [Alan] used high frequency AC with an inductor wired in parallel with the LED and the third switch. For those of you keeping track, that means [Fredzislaw100] put an SMD diode in two LEDs and two switches and an inductor in one LED and one switch.

[Alan]’s build is just a proof of concept – It’s still on a breadboard and doesn’t have the incredible level of polish that [Fredzislaw100]’s has. That being said, [Alan]’s build is most likely very similar given the small glitch at 2:05 in the original video.

Check out [Alan]’s build video below, and for an added treat check out his 2011 Advent Calendar of Circuits.


43 thoughts on “Followup: Reproducing Impossible Circuits

    1. Every last person who claimed “shopped” on that last thread was a moron, and lazy, and doesn’t belong here.

      The circuit was obviously legit. I’ve seen enough shoddy special effects and done enough “impossible” things myself to know that building a circuit that does something impossible is still easier than shopping it as cleanly as that. Do all of us a favor and don’t just up and start shouting “Special effects” whenever you see something a little out of place. Otherwise, you’re no better than George Lucas.

      Mad props to [Alan]. This is similar to what I thought was the best solution in the other thread(DC bias on the 3rd LED) but now that I think about it, this is probably a lot easier to implement.

      1. Or, those people aren’t morons. As the solution presented, albeit quite awesome and creative. I’m still going to call BS on the first one until someone ACTUALLY recreates it. With built in diodes and inductors abound. And don’t get me started on building the AC circuit into a 9v battery.

        TL;DR Recreating the circuitry is the EASY part.

      2. I’m a moron and I’m lazy!

        How else would I learn about this stuff if I didn’t read Hack a Day? Do you think Popular Science is going to cover these sort of things? CNN? My friend Tim?

  1. If you take a look at Fredzislaw100’s video you see Alan’s comment asking if his was similar, Fredzislaw replied with Well, very close…. :)
    The main schematic is the same. I just added a few pF to LED3 to have better LC resonance. In my design HF=~8MHz. VLF=~100…200Hz. Both generators built on one 74132 (are hidden in the battery connector…). Congratulations again!

  2. Glad to see that the few who didn’t just take the easy route and blame it on after effects were correct. I’m not going to check this again so there’s no point in attacking this post, but sometimes you can learn a lot by not just dismissing something as fake right away. Again, kudos for the brave.

  3. How is this in any way equivalent to the original video ? The original video had a SINGLE WIRE completing the loop, not a mesh of connections.

    Even if the inductors were hidden inside the switch, this guy’s scheme still needs more than one wire!

    Hackaday please watch the video before posting it!

    (And take down this post.)

      1. Great that you figured it out :D original could still be edited.

        However, may I rant at 1 thing. STUPID VIDEO’S. Please, if you do something like this, put up pictures. At least of the schematics. Because, honestly, I don’t care to listen to you for 8 minutes. I can read extremely fast, and the only thing I really wanted to see was the schematic you came up with. Videos are the worst documentation format around for just about anything.

      2. @Daid,

        I agree. Video should only be used when there is something that requires it. A few good pictures and a schematic get the point across quicker and also load faster even at high resolution.

    1. If y’all bothered to read the comments before you post you might spot that someone quoted the original videos maker explaining how he did it…

      “The main schematic is the same. I just added a few pF to LED3 to have better LC resonance. In my design HF=~8MHz. VLF=~100…200Hz. Both generators built on one 74132 (are hidden in the battery connector…). Congratulations again!”

  4. This is equivalent to the original video, this circuit could be laid out exactly like the original, but Alan couldn’t be bothered.

    Look at his diagrams and listen to the explanation.

    (Or did I just feed a troll?)

    1. …or the diode-cap, and inductor module in parallel with the LED is so damn small. I don’t know if I’m blind or have bad eye sight, but I don’t see a single component connected to the LED’s other than the LED it is in series with.

    2. I can’t move on with this vid.. lol..

      if I may add:

      Compare how the LED’s light looks like on [Fredzislaw100]’s and [Alan]’s video.

      [Fredzislaw100]’s has that pink-ish glow with an unusual horizontal glare on the camera’s lens. and almost, if not exact, all three LED’s has the same glow.

      [Alan]’s has that irregular dim-red glow that doesn’t glare the camera. his LED’s are diffused at the tip, but i don’t think that’s an issue.

  5. nice to see heckle-a-day in true form with most of the criticism here focused on the presentation and not the actual hack, which is very well done.

    “boo hoo *sob* *sob* i had to watch an 8-minute video.” — are you kidding me? gtfu.

  6. In the original video, at around 1:37, the LED’s flip up while he is soldering the battery leads onto the circuit. You can make out the bottom of at least two of them clearly, and I did not see any components bridging the leads of the LED’s. 0402 or not, there didn’t seem to be anything else attached to them.
    Did anyone else look at that portion of the video?

    1. Note that you see them only from the side, the same side that they display when unsoldered in the start of the video.

      Also note that these LEDs have very large flat metal parts sticking out the bottom of the epoxy package, large enough to hide 0402 or maybe even 0603 components.

  7. Good work Alan. The problem with the internet, well those who use the internet anyway is that there will be a number who will only view the video by Fredzislaw100, believing it’s possible. Not bothering to read the video uploader comments. Even if Fredzislaw100 places the admission of hidden components in the video information box, they don’t read that either.

  8. I appreciate the effort put into his video explaining his take on reproducing the controversial circuitry. Hand drawn schematic on a cluttered bench SHARED! Go dude!
    I’ve thought about ways to pull it off myself and believe his effort is legit. As far as a one to one scaled reproduction, don’t blame him for not being bothered to take it that far. Someone will though :)

  9. Shopped.

    He shows the diagram. This eliminates the possibility of any OTHER diagram, LF, HF or other.

    The diagram he shows has very well known properties from Kirchoff’s law. All three switches must be closed and then all three leds will light.


    “This video is Shopped”.

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.