Sound detector to detect hammer strikes


Senior Member
Yes, vibration detectors are freely available. I just thought that it would be a simple "quick and dirty" means of testing whether monitoring sound board vibrations is a viable technique for detecting hammer strikes (assuming that you have access to a record player). Furthermore it would be non-invasive (to the dulcimer) insofar as it would not affect the soundboard vibration, nor require any fancy mounting (at least for testing the concept).


Technical Support
Staff member
While all the suggestions are useful ideas, I can't help but think the simple microphone and amplifier solution would be cheapest and easiest to get working if one isn't going to use the equivalent of a guitar pickup.

The bear

Senior Member
That led me to wonder what the difference is between gramophone & phonograph - I had no clue; think I'll stick with "record player" ;-0

Gramophone: Any sound-recording device, or device for playing previously-recorded sounds, especially if it uses a flat spinning disk.

Phonograph: Any sound-recording device, or device for playing previously-recorded sounds, especially if it uses a spinning cylinder.


Senior Member
Here's an update.
We cleaned the strings on the dulcimer with fine sandpaper and connected each string pair between the pin and ground. Using a hammer with fine stranded wire taped on the edge we were getting 99 to 100% positive sequencing.
My wife said I cannot use her gramophone for any experiments so I bought a couple of these:


Senior Member

Hmm, IMHO that looks like the wrong type of "vibration" sensor. There are references to earthquake, motorcycle/car theft and washine machine (drum unbalanced ?). The "sensor" looks a little like a mercury tilt swich (but probably using something cheaper and safer than mercury).

The circuit diagram lower down in this link shows the sensor shunted by a 100nF capacitor and with a 10k pullup, so it's unlikely to respond to any "audio" frequencies.

As hippy and I have suggested several times now; if you're trying to detect a string of a musical instrument being struck, then the "obvious" method is an audio detector. The waveform shown by hippy in post #106 looks eminently detectable.

Cheers, Alan.


Senior Member
Well this could go on forever... however 100nF x 10K = 10^-3 so probably detects lower audio ok... I think a bandpass filter for the instrument audio range and amplifier put into a clipping shaper circuit should give a nice pulse that could be measured by PULSIN of a PICAXE running high clock speed - and pulse width can be related to pitch... however this circuitry is apparently beyond Gramps' knowledge and I don't know any ready made circuit to suggest...


Senior Member
Last edited: