This video shows a mirror made up of moving wooden pieces. Please have a look. He achieves different intensity levels by motion of the wooden pieces and placing different light sources on the top.

But how did he achieve this motion? Did he really used 1 motor for every pixel? In any case how much energy does this thing consume?

  • \$\begingroup\$ I think it indeed uses one motor (servo) per pixel, I don't see how this can be done otherwise. How much energy it consumes can't be said without part numbers, of course :) \$\endgroup\$
    – user17592
    May 3, 2013 at 13:55
  • \$\begingroup\$ Still can you give any speculation? We can assume number of pixels say a screen of 100x100 where each pixel is controlled by a motor... Can we compare this energy with conventional screens? \$\endgroup\$ May 5, 2013 at 6:25
  • \$\begingroup\$ Yes, that's 100×100×I_m, where I_m is the current one motor uses. This, plus the current of the computer behind it of course. Really, nothing more can be said without specifications. \$\endgroup\$
    – user17592
    May 5, 2013 at 6:35

1 Answer 1


Rather than speculating how a particular designer built something (the video specifically states that individual servo motors were used), here is a method that will work:

Each "pixel" is controlled by a separate stepper motor (for minimum power use) or servo motor (for simple angular control).

Stepper motor pro & con:

  • Each pixel can be moved through a defined number of steps for a particular angle. After that, no further power is needed to maintain the position. A stepper motor provides a constant holding torque without any need for power.
  • This takes different numbers of computing cycles and / or time depending on the angle
  • Absolute positioning control is not guaranteed, i.e. if a particular motor misses a step, it can only be reset to its absolute steps-to-angle mapping when it hits an end stop position.
  • A stepper motor will be silent once a desired angle is achieved.
  • Stepper motors are generally cheaper than servo motors.
  • Since great precision is not essential, this is the most technically preferable approach but not one the designer used.

Servo motor pro & con:

  • Each pixel can be moved to a specified angle simply by changing the servo control Pulse Duration Modulation signal as needed. Power is continuously needed for the servo to maintain position.
  • This is instantaneous from a processor point of view, and the servo's inbuilt logic takes care of actually achieving and maintaining the desired angle over time
  • Absolute positioning control is available with reasonable precision, and each pixel will recover defined position even if it is manually tilted away
  • Servo motors will occasionally make a sound due to minor position corrections needed, even after achieving desired position: This could add up to quite a racket with so many motors.
  • This is a less ideal approach due to the complication of constantly supplying a large number of servo signal lines with the appropriate PDM signals, and the noise factor. However, the programming simplicity would probably have made this the designer's preferred choice.

As the question observes, using a servo motor approach will need a lot of power to be continuously consumed.

  • \$\begingroup\$ Thank you. I guess it satisfies by curiosity. Anyway is this possible to do the same with somewhat less motors? Since there will be a large correlation amongst the pixels, what if a motor controls more than 1 pixel? \$\endgroup\$ May 5, 2013 at 6:28

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.