Human Oscilloscope

I needed a way to compare two signals on an engine control unit project I’m working on. I’ve only got a very cheap and basic single channel digital oscilloscope so I couldn’t use that to compare them.

So, I though I could probably use a pair of headphones to feed a signal into each ear, and I would perhaps be able to hear the difference. This actually worked pretty well. I suppose if you’ve got some of those mad 7.1 channel headphones then you can have 7.1 oscilloscope channels!

This will probably only work well for fairly low frequency signals, I’m in the low kHz range at the moment. You can’t really take measurements or see the shape of the signal (although different shapes will have different sounds), and it probably only works for comparing two signals to hear if they’re the same. I don’t think you’ll be able to detect phase difference either. Also, I’m not sure that driving the headphones directly from the microcontrollers this way is completely safe.

Here’s a guide to this highly complex procedure

  1. Chop off a few centimeters of single core wire
  2. Pull the insulation off of it
  3. Wrap it around your headphone jack
  4. Repeat three times and stick each loop into some breadboard

Pull the insulation off of some single core wire

Pull the insulation off of some single core wire

Wrap single core wire around headphone plug

Wrap single core wire around headphone plug

Use one ear per channel

Use one ear per channel

5 thoughts on “Human Oscilloscope

  1. Pingback: Use your ears as an oscilloscope

  2. Scott Harden

    Like it! Perfect solution for the application. Kudos on not making it overly complex :)

    For more accurate results, feed the audio into the microphone jack of a PC. You’ll be able to measure frequency to at least 20 or 40khz accurately!

    A lot of volt meters (especially automotive ones) measure frequency specifically for this purpose. If you’re liking this stuff and think you might keep doing it, I’d recommend one! Lately I’ve been measuring frequency with my http://www.swharden.com/blog/2013-04-17-tenma-72-7750-multimeter-excellent-for-rf-engineering/

    Reply
    1. Scott Snowden Post author

      Yeah – I was less concerned about the frequency than actually making sure the signals matched. I’m making an engine simulator and an ECU to control it, and I needed to confirm that the ECU was reading the signal from the engine simulator correctly. It’s a square tooth wave with two teeth missing in every 60, hence it was easy to hear when those missing teeth were aligned.

      Reply
  3. Changojon

    I see you are using Launchpads ┬┐are the Stellaris ones? May be this proyect could help: http://www.fischl.de/arm/sllogiclogger_logic_analyser_for_stellaris_launchpad/
    8 digital channels, running at 10Mhz… but i don’t know if a 1.6 (more or less) seconds in sampling time helps. Be aware, PB0 & PB1 aren’t 5 volt tolerant in the Stellaris Launchpad
    I agree, the sound card is the best in the audible frequency range. Audacity (http://audacity.sourceforge.net/?lang=en) could help to make a register of your signals
    Great work..!

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>