The Tachometer Inside Your Smartphone

It’s the latest in instrumentation for the well-appointed shop — an acoustically coupled fast Fourier transform tachometer. Sounds expensive, but it’s really just using a smartphone spectrum analyzer app to indirectly measure tool speeds. And it looks like it could be incredibly handy.

Normally, non-contact tachometers are optically coupled, using photoreceptors to measure light flashing off of a shaft or a tool. But that requires a clear view of the machine, often putting hands far too close to the danger zone. [Matthias Wandel]’s method doesn’t require line of sight because it relies on a cheap spectrum analyzer app to listen to a machine’s sound. The software displays peaks at various frequencies, and with a little analysis and some simple math, the shaft speed of the machine can be determined. [Matthias] explains how to exclude harmonics, where to find power line hum, isolating commutator artifacts, and how to do all the calculations. You’ll need to know a little about your tooling to get the right RPM, and obviously¬†you’ll be limited by the audio frequency response of your phone or tablet. But we think this is a great tip.

[Matthias] is no stranger to shop innovations and putting technology to work in simple but elegant ways. We wonder if spectrum analysis could be used to find harmonics and help with his vibration damping solution for a contractor table saw.

Thanks to [Itay Ramot] for the tip.

26 thoughts on “The Tachometer Inside Your Smartphone

  1. I downloaded a free app called Giri that does the same thing – except it actually gives you an RPM readout. No idea how accurate it is, haven’t put a physical tach on anything I’ve measured with it.

  2. From an experienced vibration analyst….yes, this *can* work, but a “simple, quick” measurement can very easily lead you astray. If you know how many teeth are on the saw blade, how many vanes are on the pump, how many rotor bars are in the motor, etc, picking out shaft rate may not be too difficult. But the less you know about a machine, the easier it is to be fooled. What you think is rotation rate may in fact be 2x rotation rate, caused by misalignment. Or, if a machine has journal bearings, that “rotation rate” may be 40 – 50% of running speed, caused by excessive clearance and oil whirl.

    Yes, the technique can work and it’s virtually free to try it…but use caution in what you think you know.

    1. This, absolutely.

      But on the other side of the coin, a digital recording at idle, plus the spectrometer in VLC helped me make the distinction between piston slap (at engine RPM; engine rebuild) and valve lifter noise (half engine RPM; flush & change the oil you idiot).

      What a happy oil change.

  3. For a while I have been “planning to think” about hacking together an optical tach using the IR sensor from a mouse wheel, an arduino, some code and a PC (to estimate the gear ratio I would need for another hack I’m planning to think about)
    And THIS is so obvious … (after someone else points it out)

  4. For measuring combustion engines rpm there should be a device to pick up the signal from the ignition coil, and send it to a cellphone over bluetooth.
    You would get a very exact reliable and responsive reading.

    1. You should be able to couple it directly to the microphone port via a capacitive clamp. Of course you need some input protection, a series resistor and two anti-parallel diodes should work.

  5. anything for less demanding ? Tape a flap to anything rotating . Say your electric drill , Put your finger against it or whatever. record sound > Produce approx rpm from that wave file ???

  6. This is goog app, I measure many trials on many applinces between 50 Hz to 22000 Hz and found correct. but i think commom man can not workout easely on spactrum analysis

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.