tracking a bicycle speed


    May 19 2007 | 3:24 pm
    Hi.
    Can someone help me figuring out wich sensor or device would be best
    to track the (rotation) speed of a fixed bicycle? The speed data is
    then suposed to control the triggering and speed of some movie clips
    with Max/MSP/Jitter in osX, using some analogue to digital converter,
    wich, by the way, I would apreciate some advices about.
    I've worked with sensors before, like flex, light, pressure and
    others, but none of these would be of any use for this project.
    thanks
    rd

    • May 19 2007 | 3:46 pm
      I'd look into a Hall effect sensor paired with a magnet or magnets.
      You really need something hi-speed, and these should be quite accurate.
      There was a cool project at the NYU ITP department that used that to
      drive a persistence of vision LED display in the bike wheel spokes.
      Peter McCulloch
    • May 19 2007 | 5:04 pm
    • May 19 2007 | 9:04 pm
    • May 20 2007 | 12:44 pm
    • May 20 2007 | 1:02 pm
      You could try one of those old light sets that has a small generator
      with a wheel that rubs against the tyre - as you cycle faster the
      light gets brighter - so a simple a to d would get you a value you
      could work with. (although you might have to check the voltage
      doesn't get to high and damage your a to d)
      HTH
      Nick
    • May 20 2007 | 7:42 pm
      hello and thanks to everyone for the quick answers.
      Julien, you were right. There is indeed a thread that started in
      April 2006 called "Bicycle as a Max/MSP interactive controller ",
      wich seems to have escaped my first searches.
      The mouse hacking > hi object solution would of course be the
      cheapest and easiest to execute, and I'll certainly try it first, but
      I'm guessing it won't be as robust and effective as, probably, the
      hall effect sensors plus magnets.
      Has someone really tried with the mouse? Did it work?
      rd
    • May 20 2007 | 9:32 pm
      i used it, but on old school roller skates. worked nicely. you will
      have problem with mouse off screen but this can be handled. i used an
      old mouse, not the optical one (maybe optical is better as no direct
      contact is needed), it was a bit sensitive. all in all, a sensor type
      setup will be more robust.
      are the bike stationery?
      On 5/20/07, bragatel rmd74 wrote:
      >
      >
      > hello and thanks to everyone for the quick answers.
      >
      > Julien, you were right. There is indeed a thread that started in April 2006
      > called "Bicycle as a Max/MSP interactive controller ", wich seems to have
      > escaped my first searches.
      >
      > The mouse hacking > hi object solution would of course be the cheapest and
      > easiest to execute, and I'll certainly try it first, but I'm guessing it
      > won't be as robust and effective as, probably, the hall effect sensors plus
      > magnets.
      >
      > Has someone really tried with the mouse? Did it work?
      >
      > rd
      >
      >
    • May 20 2007 | 10:01 pm
    • May 20 2007 | 10:22 pm
      or use a camera looking at the pedals, analyse its image and measure
      with Jitter the alternance of changes (e.g. luminance average level)
      when the pedal is in front of the camera;
      see Davina project of one of my student on
      --
      Yves Bernard yb@imal.org
      asbl iMAL vzw
      30-34 Quai des Charbonnages
      1080 Bruxelles/Brussel
      tel 32 2 410 30 93
      http://www.imal.org
      http://www.erg.be/blogs/artNumeur
    • May 20 2007 | 10:25 pm
      one more suggestion. may or may not be helpful to you.
      i created a patch for a sound installation that used a stationary bicycle wheel to control the sounds based on the speed and direction of rotation of the wheel.
      i did this by pointing a camera at the wheel and motion tracking a bright red reflector on the wheel.
      it was quite effective and responsive and easy to setup.
      if you want, i can dig up the old patch that i used for this and post it. however, i don't think there was anything too brilliant about it, just color tracking in four quadrants. don't think i used any cv objects either, just jitter.
      -rob
    • May 21 2007 | 8:02 am
      Or attach in some way a rotary encoder to the wheel - this would solve the
      "twisted wire" problem and be really easy to interface to max - just hack a
      cheap midi controller that has a rotary encoder. You can then track the
      speed of the rotation of the encoder in max easily. This is probably the
      cheapest and simplest option?
      Jonathan.
      On 20/05/07, Yves Bernard wrote:
      >
      > or use a camera looking at the pedals, analyse its image and measure
      > with Jitter the alternance of changes (e.g. luminance average level)
      > when the pedal is in front of the camera;
      > see Davina project of one of my student on
      > http://www.erg.be/erg/spip.php?article263#
      >
      > --
      > Yves Bernard yb@imal.org
      > asbl iMAL vzw
      > 30-34 Quai des Charbonnages
      > 1080 Bruxelles/Brussel
      > tel 32 2 410 30 93
      >
      > http://www.imal.org
      > http://www.erg.be/blogs/artNumeur
      > http://www.i-cult.be
      >
      --
      Jonathan Green
      0777 1680 497
      jonathan@jg1983.co.uk
    • May 21 2007 | 8:43 am
      What about audio.
      You should put something on your bycicle, touching the wheel rays,
      producing a rithmic sound (a card for example).
      You then analyze this sound...
      Just an idea ;-)
      All the best
      Alessandro Fogar
      2007/5/21, Jonathan Green :
      > Or attach in some way a rotary encoder to the wheel - this would solve the
      > "twisted wire" problem and be really easy to interface to max - just hack a
      > cheap midi controller that has a rotary encoder. You can then track the
      > speed of the rotation of the encoder in max easily. This is probably the
      > cheapest and simplest option?
      >
      > Jonathan.
      >
      >
      > On 20/05/07, Yves Bernard wrote:
      > > or use a camera looking at the pedals, analyse its image and measure
      > > with Jitter the alternance of changes (e.g. luminance average level)
      > > when the pedal is in front of the camera;
      > > see Davina project of one of my student on
      > > http://www.erg.be/erg/spip.php?article263#
      > >
      > > --
      > > Yves Bernard yb@imal.org
      > > asbl iMAL vzw
      > > 30-34 Quai des Charbonnages
      > > 1080 Bruxelles/Brussel
      > > tel 32 2 410 30 93
      > >
      > > http://www.imal.org
      > > http://www.erg.be/blogs/artNumeur
      > > http://www.i-cult.be
      > >
      >
      >
      >
      > --
      > Jonathan Green
      > 0777 1680 497
      > jonathan@jg1983.co.uk
      >
      >
      --
      Alessandro Fogar
    • May 21 2007 | 9:12 am
      In a similar project i used 1 ultra-cheap reed-switch and a few magnets, and measure the time interval between the magnet(s). very simple, very cheap and (at least for my project) very useful.
      Jeroen.
    • May 21 2007 | 9:56 am
      Jeroen Groot skrev:
      > In a similar project i used 1 ultra-cheap reed-switch and a few magnets, and measure the time interval between the magnet(s). very simple, very cheap and (at least for my project) very useful.
      >
      > http://images.google.com/images?svnum=10&hl=nl&gbv=2&q=reed+switch
      Very nice! I didn't know about reed switches until now. Great product!
      To me this looks like a winner.
      Andreas.
    • May 21 2007 | 11:20 pm
      Peter McCulloch schrieb:
      > I'd look into a Hall effect sensor paired with a magnet or magnets. You
      > really need something hi-speed, and these should be quite accurate.
      > There was a cool project at the NYU ITP department that used that to
      > drive a persistence of vision LED display in the bike wheel spokes.
      I would hack one of these cheap tachometers, they use hall sensors, some
      of them are even wireless....
      Stefan
      --
      Stefan Tiedje------------x-------
      --_____-----------|--------------
      --(_|_ ----|-----|-----()-------
      -- _|_)----|-----()--------------
      ----------()--------www.ccmix.com