I am attempting to measure the amplitude of a small AC signal (40-200mv) from an electric guitar pickup. I am not concerned about the frequency just the amplitude of the signal. I have done some research and know that I can't put a negative voltage into the analog input, and that this signal is pretty weak, so I'm looking for a circuit that would give me a a quick response <10ms (preferably less) that I can connect to an analog input.
1 Like
Assuming you use an Uno...
Use a voltage divider with two 2.2Megohm resistors (1.1Megohm load), between 5volt and ground, with the centre connected to an analogue pin. Connect the guitar also to that pin (and ground), but via a 100n cap in series. The A/D is now biased on ~2.5volt and returns about 512 without sound.
Reading the A/D continuously will return the deviation from 512 (with a response time of ~0.1ms).
Leo..
Here is a circuit that will "kill" the negative voltage:
For a guitar change the resistors to about 1M and 10M.
Actually, you can probably leave-out the diode because the Arduino already has "small" built-in protection diodes and a guitar pick-up can't put-out much current and the series resistor also limits current.
Then, assuming you're using a "regular" Arduino you can use the optional 1.1V ADC reference. That way, each "step" is about 1mV and you can read 40mV. (Of course, with the 2.5V bias method you can't have a 1.1V reference, and if you need frequency information you can't use the half-wave rectification method because it adds severe distortion.)
The circuit is no problem. 10ms is only 100Hz, But you'll want to sample as fast as possible. And, remember you are reading a waveform so the readings will "look random. And your case with the negative-half chopped-off, half of the readings will be zero. So, you'll have to find the peaks or take an average.
Or with an op-amp you can make a Peak Detector which will hold the peaks, depending on the resistor and capacitor values. For my sound activated lighting effects I typically read & update about 10 times per second so I have a time-constant of about 100ms.
1 Like
And give the peak-to-peak voltage rather than just the amplitude.
But the Schottky is better than the protection diode which is silicon.
If OP wants exact peak/peak amplitude of a guitar pickup, then the circuit from post#3 is unsuitable.
A guitar signal does not peak the same positive and negative, which the above circuit does not measure. And the signal could easily be more than 2volt peak/peak, which influences peak/peak value when the clamping diode starts to work. A load of about 1Megohm is also required, and so is the included capacitance (length) of a guitar lead.
Leo..
I would like to add some suggestions once you have explained more about your application.
This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.