Guitar synth successful build

[xtian]

Wow! Would love to see this made into a PCB for DIY with a BOM, TIA!!!

[Mick Bailey]

Here's how I did it on Vero. It's a bit of a mess due to making changes after I breadboarded it. Also, the filter was added after the original build was finished - hence the second board. As a result of poor planning, the switches went on the side and the gain control was replaced by a fixed resistor due to lack of space - this needs reinstating as per the schematic. You may observe that the debounce resistors and caps aren't on the main board, again due to them being added later (they're mounted on the pots and switches). Also there's an additional trimmer to the bottom left which is left over from the pre-filter incarnation and no longer needed. The battery power seemed like a good idea but a fresh battery only lasts for a full day or so and isn't worth including.

It gives you some idea of how small and convenient the Seeed processor is - the USB is all included. I mounted mine so I could update the code without taking the back off.

[tedmich]

wow, vero-complex! It does sound great!

[Mick Bailey]

I showed a guitarist friend the clips and he asked would it not be better to demonstrate the pedal without the overdubs. Just to be clear in case anyone else thinks the same, the clips are recorded live - no overdubs. The harmonies and chords are created on the fly and notes held with the pedal to give some interactivity. The regular guitar is blended in with a simple mixing pot at the end of the signal chain.

[g1]

That's funny and a testament to your success.

[Mark Hammer]

So are the switches shown for octave, mode and waveform momentary or latching?
And since everything else seems to run off +5V, I wonder if, instead of requiring a separate 3.3V source for the controls, one could simply stick a suitable fixed resistor between +5V and the input to each pot so that the pot outputs never exceed 3.3V. I see a pin on the Seeed marked 3.3V. What I can't tell from a cursory read of the datasheet is whether that pin wants 3.3V or provides 3,3V, downregulated from the 5V supplied to the unit.

[Mick Bailey]

They're tactile momentary switches, though I suppose any momentary type would do but I liked the feel of tactile switches. I couldn't find any panel mount ones, so ended up with a laborious construction method with the prototype to make up the actual buttons and mount the switches on a piece of Vero that screws onto threaded mounting blocks JB-Welded onto the underside of the enclosure top. I've now found some 6mm-threaded click-action push buttons of the same type as mini toggle switches, so intend to order a few of those for future use. The octave, mode and waveform functions are also doubled up to provide hammer-on, sustain and glide if the hold switch is held down.

The 3.3v pin on the Seeed comes off its internal regulator output. The device is powered from 5v for USB compatibility and the 3.3v is a more of a 'convenience outlet' to power low-demand peripherals. I'm not sure whether the Seeed can be powered off 3.3v, but with some other processors supplying the main voltage to the regulator output will work, but can cause damage to the regulator.

I take your point about using a resistor on the input of each pot. You'd have to measure each pot individually, as the tolerance can be quite broad. The neat thing about digitizing pot rotation is that the value of the pot doesn't really matter too much, but the voltage does. Using the regulated 3.3v as a reference means you can use a 10K pot upwards - all the processor needs to know is that it goes from 0v to 3.3v. With resistive dividers you'd have to be more careful about values.