Ask Hackaday: Did you catch the Grammys?

posted Feb 19th 2012 8:01am by
filed under: Ask Hackaday

Although award shows aren’t necessarily our thing, [T. D.] sent in something that piqued our interest. His friends recently got back from the Grammy awards where they witnessed thousands of LED bracelets blinking in time to a performance by Coldplay. A little bit of YouTubing pulled up this video that demonstrates the effect (because that video will probably be taken down shortly, just pick something from this link).

[T.D.]‘s friends brought one of these bracelets back with them and like a good Hackaday reader, he cracked it open. This is the precious board pic that [T.D.] sent in. We’re pretty confident that the IC is an ATMega48PA, but beyond that we’re not quite sure how these bracelets can, “light up and flash at precisely the right time” as [T.D.] puts it.

From what we saw on the Grammy broadcast, it’s possible these bracelets merely flashed whenever the user clapped their hands. A circuit that simple doesn’t require a microcontroller, so we’re left wondering what the heck is going on here. If you’ve got an idea of how these choreographed light display bracelets work, drop a note in the comments.

EDIT: Commentors have pointed out these wristbands are called ‘Xylobands.’ There’s a great video of these wristbands in action at the 2011 X Factor finale.



64 Responses to Ask Hackaday: Did you catch the Grammys?

  • Nomad says:
    February 19, 2012 at 8:18 am

    There seems to be something on the backside judging from Circuitboard design. Obviously a two-layer board.

    Are there pics of the backside?

    Reply Report comment
  • Ivan says:
    February 19, 2012 at 8:21 am

    Perhaps it’s a mesh network that is smart enough to tell when everyone is clapping at the same time and activate neighboring bracelets.

    Reply Report comment
  • Jonathan says:
    February 19, 2012 at 8:23 am

    I see an ‘X2′ crystal and a glob of epoxy, probably hiding a raw die chip.
    I suspect it’s some kind of radio link.

    Reply Report comment
  • xchip says:
    February 19, 2012 at 8:24 am

    how about just using the rtc?

    Reply Report comment
    • pockpock says:
      February 19, 2012 at 11:59 am

      I don’t know but it seems to me this thing is meant to be synchronized to the song and it surely would be organizational pain following a schedule for a live show that exactly. Also you would have to synchronize the RTCs in factory and I guess that’s too complex. Just put in a microcontroller and flash it with a very simple algorithm, no need for calibration or anything besides flashing the MC which pretty sure is done automaticaly. It’s a one-night-use braclet don’t need to overengeer that…

      Reply Report comment
  • T.M. says:
    February 19, 2012 at 8:27 am

    There is a wire that goes around the wrist, inside the bracelet. My guess is that this is just a crude antenna to pick up the pulses to blink.

    Reply Report comment
  • Shawn says:
    February 19, 2012 at 8:39 am

    I’d agree with Jonathon. Just someone hiding in a control booth somewhere controlling a transmitter to tell the watches when to start. And then it’s just timing.

    Reply Report comment
  • Chris says:
    February 19, 2012 at 8:42 am

    After a bit of googling, apparently they’re called “xylobands” and Coldplay have used them before. The inventors seem pretty protective of their intellectual property, but in an interview he did say,

    “Basically they’re an RF-driven wristband. They’ve got a receiver in them which receives data and that data tells them how to flash and when to flash, basically.”
    Source: http://www.coldplay.com/newsdetail.php?id=906

    There’s probably a transmitter at the FOH.

    Reply Report comment
  • Jacob says:
    February 19, 2012 at 8:45 am

    X2 is definitely a crystal, and there seems to be a lot of pogo-points for programming them (to a frequency?) So I’d guess it’s some kind of radio-link system between the bracelets and sound-to-light software running on a computer backstage

    Reply Report comment
  • greg says:
    February 19, 2012 at 8:48 am

    i also believe its radio nadef but the comment above i beleive is wrong the “wire” or wires your seeing is most likely driving the leds on the bracelet . if you do look a little closer you can see the metal cross shaped section above the board has a thick trace going to it and on my phone i cant really see all that great but it doesnt seem to be tied or grounded on the other “top” side and thus i beleive thats the antenna .

    Reply Report comment
  • jsmxbox says:
    February 19, 2012 at 9:08 am

    http://glowbands.com/product-details/

    Reply Report comment
  • tim says:
    February 19, 2012 at 9:08 am

    The wires to the leds can power the leds and work as antenna at the same time. With longer antenna you can work at lower frequency, which give more range.

    Reply Report comment
  • Zhen says:
    February 19, 2012 at 9:09 am

    http://www.dailymail.co.uk/sciencetech/article-2073124/X-Factor-2011-Coldplay-light-Wembley-giving-audience-radio-controlled-glow-bands.html

    Reply Report comment
  • Paul Campbell says:
    February 19, 2012 at 9:26 am

    2032 is a coin cell battery – the “2032+” is one side of it – I thing the metal structure is a spring that holds the battery

    on the LHS you see a bunch of inductors (L3 L5) – I think this is the RF matching for the antenna (one or both of the wires in the band)

    it’s receive only so probably no zigbee chip hiding there – might be just a single chip RF demod, or even a half dozen transistors hiding somewhere

    Reply Report comment
  • Alex Rossie says:
    February 19, 2012 at 9:41 am

    This sounded pretty cool, but just being RF controlled LED blinkers is very lame.

    I was hoping there was an elegant way to make these LEDs detect when to blink.

    Reply Report comment
  • Max says:
    February 19, 2012 at 11:07 am

    I also think C2, L3, L5 / C3, L4, L6 are impedance matching for the wires, both used as antenna and led supply. Wavelength at the given frequency of about 860Mhz as statet in the product-page returns in a wavelength of around 35cm. So the wires can easily form lambda/4 antennas, signal being decoded unter the epoxy blob and interpreted in the avr, then flashing the leds in the received pattern. The avr runs on its internal clock, the epoxy-RX-blob needs its own acurate X2 crystal. nice design, but not very cheap, considering the price of the avr.

    Reply Report comment
  • Retroplayer says:
    February 19, 2012 at 11:19 am

    Why not just make it ultrasonic and pipe the “tone cue” into the PA? Seems simple enough. Nobody hears the tone being transmitted over the PA except the bracelets.

    Bada-bing-bada-bang… done. And worlds cheaper than RF, etc…

    In fact, I would be surprised if that isn’t exactly how they work. If not, I want royalties for the idea from the company when they slap their forehead and switch to that system. lol

    An interesting bonus is that anyone recording the audio from the show would likely also record the tones, which would help them identify bootlegged and unauthorized recordings.

    Reply Report comment
    • nes says:
      February 19, 2012 at 1:14 pm

      Difficult to say without knowing exactly what’s under the glop-top, but a microcontroller with integrated RF transceiver (e.g. Chipcon cc11 series) only commands a roughly 25% premium over the equivalent ATMega device.

      I guess that reliable ultrasonic detection would require at least a piezo element and some custom electronics as opposed to an all-in-one IC. Also, it would be expensive to make work if the PA is not capable of ultrasonic frequencies.

      Reply Report comment
    • jonzilla says:
      February 19, 2012 at 5:02 pm

      Lady Gaga already did it, though instead of ultrasound I believe they used a high frequency, but sub-20kHz tone. No need for separate ultrasonic speakers, since most people can’t hear high frequencies, especially when masked by lower frequencies.

      It’s a good idea; far cheaper to implement than RF.

      http://www.wired.com/underwire/2012/02/sonic-notify/

      Reply Report comment
      • Retroplayer says:
        February 19, 2012 at 9:15 pm

        Good point. I am assuming most speakers are capable of reproducing frequencies beyond the 20KHz, and this is typically true, but PA systems may filter that out. I would say that most people would not be able to distinguish a tone near the upper limit anyway. Especially if it were brief. I know my hearing doesn’t go that high (but I am 38) though I can still feel the pressure on my ears of tones I can’t distinguish.

  • RicoElectrico says:
    February 19, 2012 at 11:57 am

    I’ll just leave it here… They list radio frequency as 869.5 Mhz.

    Reply Report comment
  • Justin Case says:
    February 19, 2012 at 12:02 pm

    Shift color, they do not ???
    Weak, this is.

    Cool would they be if on the crowd a bar graph of audio displayed.
    (For profit if implemented, copyright I do)

    Reply Report comment
    • Brian Benchoff says:
      February 19, 2012 at 12:32 pm

      Better yet: assign each wrist band a unique ID based on seat number. Everybody’s a pixel.

      Reply Report comment
      • daniel says:
        February 20, 2012 at 2:13 pm

        And for general seating (like, a pit) where you wouldn’t know where people are seated (standing), have the client (wristband) perform different duties based on RSSI. It wouldn’t be perfect, but over a pit of a few thousand people, it’d probably average out to look okay. Also, without triangulation you could only do spherical shapes with your ‘pixels’. But its a start. Someone get on it!

      • daniel says:
        February 20, 2012 at 2:21 pm

        ^ and i’m claiming that idea! haha =)

  • Tuckie says:
    February 19, 2012 at 12:12 pm

    Why not just make this into a smartphone app for concerts?

    Reply Report comment
  • Mojo says:
    February 19, 2012 at 12:52 pm

    If these were repeater type radios, with very limited range, you could get a cool game-of-life type explosion effect. I’m claiming the rights on that puppy.

    Reply Report comment
  • Alan says:
    February 19, 2012 at 1:01 pm

    You can notice that different wrist bands are sometimes flashing differently than others. Pretty unique idea.

    I wonder if they gave different “sequenced” wrist bands out, or they can actually determine what other wrist bands are around them.

    Reply Report comment
  • kf5jak says:
    February 19, 2012 at 1:06 pm

    http://www.xylobands.com/glowbands-product_info.php

    “Xylobands™ are operated using our proprietary software that can be downloaded onto your laptop.
    The laptop would be connected by cable to a transmitter box and antenna.”

    Reply Report comment
  • Microguy says:
    February 19, 2012 at 5:56 pm

    It’s NOT clapping. I couldn’t get much from the videos. They focused on the bands and not the wrist bands. But I could see, they didn’t blink together, I mean, not ALL of them.

    Parts of it looked kinda random, parts were synchronized, but not ALL of them were.

    At first, I was thinking maybe sound, like the old color organs we used to see. But that didn’t sound or look right either. It looked some some beat to one sound, and some to another.

    So I’m better they simply send out a simple signal telling the (wrist) band to bear a certain color at a certain time, or blink a certain pattern, based on say a pre-programmed song.

    Thinking about how *I* would do it….

    I’d maybe have a couple types of wrist bands. Then send out a single code telling it when to blink. The wrist band, blinked a pattern 9or color) depending on the type. So at any one code, one band may blink the other may not. (or different colors blink).

    That way, you can get various visual effects across the room, but still only sending out one simple code at a time.

    Similar to…
    All blink
    only blue blink
    only red blink
    etc…

    The device sending the code, would be more similar to the old color organs. (assuming they actually sang and played at the time, and didn’t use a tape or lip sync)

    Reply Report comment
    • evilgreenie says:
      February 20, 2012 at 5:41 am

      Much better video of the wrist bands
      HERE
      Looks like a preset set of blink patterns triggered by wireless.
      The different colors look good – very well distributed throughout the arena with none of the clustering you’d expect with pure random patterns..

      Reply Report comment
  • Jason Knight says:
    February 19, 2012 at 7:39 pm

    Thank you so much for rick-rolling me into watching five videos filled with some of the most annoying whiney out of tune rubbish (that I kept waiting for the song to start since it seemed to be a five minute long intro) this side of the tone deaf crack who… I’ll stop there since I ended up killing the audio before I shoved red hot pokers in both ears.

    … and STILL not seeing any “blinking in tune with the music” in any of them… though the cute blonde standing behind them doing a slow clap in sync with nothing was particularly amusing.

    Reply Report comment
  • Lexor says:
    February 19, 2012 at 9:04 pm

    Here’s an interview with the inventor.

    Reply Report comment
  • d hamilton says:
    February 19, 2012 at 10:39 pm

    Where can I purchase 10 or less units ??

    Reply Report comment
  • ScouttMan says:
    February 20, 2012 at 12:15 am

    Hey that chip kind of looks like the TI CC2511F32 microcontroller. It the same one used in the wixler and is a 2.4ghz transceiver and microcontroller combo.
    Just a guess;)

    Reply Report comment
  • xorpunk says:
    February 20, 2012 at 2:35 am

    The silicon holds and image from xylo’ software that handles RF based programming.. The straps are just LED strips..

    Too bad it didn’t get into the hands of ‘hackers’ or we’d see what the code on the silicon does..

    Reply Report comment
  • MattQ says:
    February 20, 2012 at 8:15 am

    One thing that’s interesting is that the crystal is so far away from the micro. Maybe it’s a timing element for something else as well? the x2 above it makes me wonder if it has two differential outputs. Without seeing the other side of the board, it’s entirely conjecture. My guess is a simple one way RF transceiver. That way they can receive simple RF signals that say “Do routine A” when they play some part of a song, and then “Do routine B” when they get to the right moment. I bet it’s really simple like that.

    Reply Report comment
  • xobit says:
    February 20, 2012 at 2:04 pm

    I can’t see the mixer but if there is one the x-tal could be the local osc. (LO) and the output from the mixer could feed the micro – a simple receiver

    Reply Report comment
  • jaik says:
    March 3, 2012 at 8:56 am

    Hi,

    our glowband started to flash for a short period yesterday night.

    We don’t know how to activate the lights, but it seems as if an reverse engineering is possible by a simple method (that we try to find out).

    Greeting

    jaik

    Reply Report comment
  • neuron says:
    March 5, 2012 at 11:54 am

    How did they implement the RF receiver? What circuit did they use?

    Reply Report comment
  • d hamilton says:
    March 8, 2012 at 12:42 pm

    The receiver is under the black blob on the board, look closer at the photo above.

    Center left under X2.

    Reply Report comment
  • d hamilton says:
    March 8, 2012 at 12:45 pm

    I would like to get one of these bands.

    Please let me know if someone is willing to send their band over to Colorado usa.

    Reply Report comment
  • Tobin says:
    April 20, 2012 at 2:19 pm

    Realize this is kinda an old thread, but with Coldplay on their new tour there will be a lot more of these. From what I understand they are frequency synced, probably to a radio controller. I was thinking it might be sweet to figure out how to adjust it so that it blinks when I get a cell message. Would also need to do some adjustments so that it fits my wrist. If anyone goes to the concert. Go back to the gates after entering and ask for a few more. It worked at my show.

    Reply Report comment
  • Tobin says:
    April 20, 2012 at 2:23 pm

    There is also this sight for the people that make them. http://www.xylobands.com/glowbands-product_info.php

    Reply Report comment
  • Tobin says:
    April 20, 2012 at 4:57 pm

    Oh and this might help anyone. http://www.google.co.uk/patents/US20110025912?dq=jason+regler&ei=fOORT4T-J4jmiALUwPCIDg

    Its a patent by the creator of the wristbands. It uses a dtmf to communicate. Not sure how this is done exactly.

    Reply Report comment
  • TheFiend says:
    April 23, 2012 at 12:42 pm

    I put a switch in mine so I can turn it on and off at will. They are indeed radio controlled and I believe each colour is a different frequency, so they can work separately. Having each one on a separate frequency so they chase around the stadium would be great, but likely that would be WAY too hard to make sure it works right as people tend to switch seats and move around and what not… I wish they came stock with a switch on the inside so you can actually use them after the show… would be great to put it on my leg when I am riding my bike at night BUT they use enough freaking solder that I didn’t have to use any extra.. just melted what they had and stuck the switch in! Which frequency they use, I am not sure yet… I might get to talk to it later (I wish I got 2 of them)!

    Reply Report comment
  • xenocide0 says:
    April 25, 2012 at 8:25 pm

    Looks like they are a 60 kHz-wide signal centered on 869.5 MHz

    more info: http://stevencrowley.com/2012/03/27/coldplays-wireless-hardware-startup/

    Reply Report comment
  • gstar42 says:
    May 2, 2012 at 1:31 am

    > I put a switch in mine so I can turn it on and off at will.

    That’s exactly what I want. How did you do it?

    Thanks…

    Reply Report comment
  • BLUskreenOFdeth says:
    May 3, 2012 at 11:23 am

    Have one sitting in front of me. Will attempt to dissect and take pictures. Doesn’t light up when you clap your hands. All I can say is this device made an already great act better. Now I must know how it works!

    Reply Report comment
  • costas says:
    May 3, 2012 at 11:25 pm

    I have a xyloband from the coldplay concert @ san jose last Saturday, I want to activate it at home!! any ideas how to do it??

    Reply Report comment
  • mregala says:
    May 18, 2012 at 5:45 pm

    Coldplay’s European tour has started today, so expect more hacking in the days to come!

    Some friends of mine got it working by just pressing hard the middle of the plastic case – that ruins the RF features though.

    Anybody has more information besides the following?
    http://www.google.co.uk/patents/US20110025912?dq=jason+regler&ei=fOORT4T-J4jmiALUwPCIDg
    http://stevencrowley.com/2012/03/27/coldplays-wireless-hardware-startup/
    http://www.caskater.com/2012/05/02/dissecting-the-xyloband/

    Fisher

    Reply Report comment
  • lpereira says:
    May 19, 2012 at 8:22 am

    I got mine yesterday. During the concert, some xylobands just light with RF order. Some also light pressing it. And some stay permanently blinking! :P
    I open it today. they have 2xCR3216 (6V) batterries and 1xCR3216 (3V). CR3216 look to be for RF functionality, and CR2032 for manual actuation?!?
    CR2032 has a little gap in the contacts. If I press it, the lights turn on. If I do it without the 2xCR2016, it dosnt turn on…
    Probably I will replace it with something costumized in order activate my xyloband (and the one’s from my friends) base in some easy RF signal. Of course I will replace my friends hardware too.

    Any suggestion about what hardware to use? Space is too limited for arduino mini… :S

    Reply Report comment
    • mregala says:
      May 19, 2012 at 10:44 am

      The problem with activating them is knowing what type of signal. As in, in the offical site they say they operate with a propriatary software wich sends __ RF signal. I’m thinking there’s more to that than just sending a simple signal in X frequency – more likely, a protocol.

      If we had access to that software and the antenna the reverse engineering process would be smooth

      Reply Report comment
      • lpereira says:
        May 19, 2012 at 12:26 pm

        Hi mregala! I did not intend to use the original hardware, nor to decode the proprietary protocol. That’s to much above my capabilities :P

        In my previous comment, you should read “2xCR2016 (6V) batterries and 1xCR2032 (3V)” instead of “2xCR3216 (6V) batterries and 1xCR3216 (3V)”…

        I just would like to consider a small microprocessor with RF interface. Easy to program (like Arduino), and capable to operate from low capacity batteries, like CR2016.
        Maybe some PIC (DIL8)?!? I have an EasyPIC5 development board stored from 5 years ago. Maybe it could be useful!?

  • anthormigo says:
    May 21, 2012 at 9:34 am

    Tomorrow I will have one. I will take a look but don’t expect too much of me :)

    Reply Report comment

    • Leave a Reply

    XHTML: You can use these tags: <a href="" title="" rel=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <pre> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

    Hack a Day serves up fresh hacks each day, every day from around the web as well as hacking related news.

    Send us your hacks






         



    Hacks

    Resources