12 simultaneous & unique frequencies (PWM)

Status
Not open for further replies.
This is how it is working:

Code:
12 PWM out -> 12 lm386 amplifier -> 12 coils (not speakers) just below the strings -> vibration -> guitar pick-up -> final out

So we don't actually hear the PWM, instead what we hear is the strings vibrating recorded by the pick-up. The noise in the video is actually cause by the strings colliding with the coils (will need to make more space between the strings and the coils). The PWM frequencies are precise enough for my needs.

Put some volume in your headphone and listen to the very end of the video. There's a tic tac tic tac. I think it's cause by the fact that I am scanning all ADC (for capacitive touch sensing). Also when connecting the USB cable I hear my mouse moving on my screen.
 
So we don't actually hear the PWM, instead what we hear is the strings vibrating recorded by the pick-up.

But you need to tune it up so that the frquency of the string would be the same as PWM.

The noise in the video is actually cause by the strings colliding with the coils (will need to make more space between the strings and the coils).

That's what it is! I thought that was electronic caused by off-frequencies.

You probably also need some stops to prevent the string from vibrating when you turn it off.

Also when connecting the USB cable I hear my mouse moving on my screen.

I see. This might be a grounding problem. If USB ground is somehow connected to your system ground, this creates a ground loop so that you can hear all the currents flowing through it. You can try an isolating USB connection, which may fix the problem.
 
But you need to tune it up so that the frquency of the string would be the same as PWM.

Yep and this is the fun part, you get all kinds of harmonics when you're off.


Wow, 100$ for buying an USB isolator!?!
Is there an alternative?
 
Next time you do this, consider using an FPGA to generate the PWM waveforms: one channel per pin and a hundred or so pins, nanosecond resolution.
(Especially if you have from March to August to get it running.)
 
Next time you do this, consider using an FPGA to generate the PWM waveforms: one channel per pin and a hundred or so pins, nanosecond resolution.

I don't know much about FPGA and I think there's a reason why Using a FPGA for PWM output is overkill AFAIK (instead next time I could use a dedicated PWM driver or even a 555). Also my goal was to make a single "brain" for everything: USB (midi implementation), capacitive touch sensing and PWM.
 
Last edited:
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…