10
\$\begingroup\$

I need to detect motion on a nylon (transparent) stringed instrument.

When the string is plucked, it would vibrate and I'd like to know what kind of sensor to use in order to detect the action.

Maybe a reflective optical sensor (visual detection) like the ones here? http://www.vishay.com/optical-sensors/reflective-outputisnot-16/

Or a tiny piezo element (detection by sound)? https://www.sparkfun.com/products/9199

  • I don't think piezo will work because there are more than one string to detect. The strings are close to each other, I guess the sounds will interfere with each other. Maybe using a treshold and computing/choosing the loudest emitter would be a solution. But I would prefer the cleanest way to achieve this.

  • At best, I do not want to paint the string thinking that it will damage the sustain and / or the color of the sound.

\$\endgroup\$
  • \$\begingroup\$ Have you considered a microphone? \$\endgroup\$ – Steve Robillard Sep 15 '13 at 21:32
  • \$\begingroup\$ :) I am basically trying to convert the action to a digital signal so that I can use it elsewhere. Using a mic and detecting pitch is not an option. \$\endgroup\$ – cenk Sep 15 '13 at 21:52
  • 1
    \$\begingroup\$ Is the goal to reproduce sound, or to detect string-plucking events? What sort of instrument - guitar, piano, harp, mandolin? \$\endgroup\$ – pjc50 Sep 16 '13 at 8:51
  • 1
    \$\begingroup\$ Do you want to detect the movement of just one string or do you want a composite signal from all strings? I get the impression you want to convert a single string to a digital signal but more info would help. \$\endgroup\$ – Andy aka Sep 16 '13 at 8:54
  • 1
    \$\begingroup\$ @user26129 In his last comment he says that it has 16+ strings. Doesn't seem like a guitar to me... \$\endgroup\$ – varesa Sep 16 '13 at 13:58
6
\$\begingroup\$

Something similar was designed for a percussion instrument, and can be converted for use with nylon stringed instruments as well:

enter image description here

  • A line-type low power laser is directed at the strings at an angle
  • Each of the string reflects narrow maxima in the corresponding reflection direction (the thin beige arrows shown)
  • An image sensor (CCD etc) is used to capture the individual reflected LASER points corresponding to each string
  • A suitable image processing software, perhaps OpenCV, is used to convert all motion of those spots of light into vibration information.

This strategy requires the LASER's line of incidence on the strings to be close to the bridge: The further the beam moves from the bridge towards the middle of the strings, the greater the contamination of the strong movement by parasitic and sympathetic vibrations from the instrument's body and sounding board. Too much such contamination, and it becomes impossible (or very difficult) to discern the actual string vibrations, from the "noise".

\$\endgroup\$
  • \$\begingroup\$ Very nice drawing +1 \$\endgroup\$ – Andy aka Sep 16 '13 at 9:21
  • \$\begingroup\$ @AnindoGhosh very interesting answer. The instrument is a qanoun Do you think that the musician's hands will block the laser using this technique? Maybe putting the laser/image sensor inside the instrument (below strings) would do it. \$\endgroup\$ – cenk Sep 16 '13 at 9:38
  • \$\begingroup\$ @cenk Amazing, that instrument is almost identical to a vintage German musical instrument for which I designed the sensors: Similar to the Hammered Dulcimer, but with 4 strings per note. Yes, I would put the laser and the optical frame sensor both on the body of the instrument, under the strings. Also, I would make the laser impinge the strings at a very sharp angle, almost parallel to the strings. Note that the laser positioning needs some thought, so that it does not distract the musician. \$\endgroup\$ – Anindo Ghosh Sep 16 '13 at 9:46
  • \$\begingroup\$ @AnindoGhosh do you have any documentation/url for your designs? I would love to see what kind of equipment you have used and also examining the building process would be a tremendous help. How did you adjust the laser and the receiver/reader etc. Thank you. \$\endgroup\$ – cenk Sep 16 '13 at 10:07
  • \$\begingroup\$ @cenk "Work-for-hire" contracts are covered by non-disclosure agreements, hence unable to share anything except conceptual inputs as I have done above. Adjustment of laser is easy: 3 tiny screws on a salvaged plastic frame of struts, for declination / ang;e / rotation. In my prior project, the camera and OpenCV work was by someone else, I just had to get the little reflected laser spots dancing on the bridge on the far side from the laser source. \$\endgroup\$ – Anindo Ghosh Sep 16 '13 at 10:31
3
\$\begingroup\$

I don't think you're going to get away without doing any signal processing. Once you do, you might find that pitch detection would actually work quite well.

Electric harp piezo pickups definitely exist: http://www.kortier.com/pickups.htm . You might be able to use or adapt those for your instrument. However, they're going to output an audio signal and have a small amount of crosstalk.

Optical pickups are a possibility, but the strings are a very small target and the environment may be optically noisy. If the strings are translucent, you could try aiming a narrow red laser beam through them and mounting a small detector in the refraction pattern behind it. Small movements of the string should produce bigger changes in the pattern.

If you can mount slot-type sensors close to the middle of the string, that might work; although you'll have to "tune" them into position every time you move the instrument.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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