Saturday 14 January 2012

Slow but steady...

Well, I've spent a couple of months scouring Yamaha patents for inspiration and purchasing components, supplies etc.

I now have a rudimentary +-15V power supply that runs from the mains. The regulators (generic 7815 & 7915 models) were getting pretty toasty on a 300 mA load from a bulb across each rail so I've bought two nicely sized heatsinks to keep things cool.

While ordering the heatsinks from Farnell I had to make sure my order amount was above £20 so I splashed out and bought a PICkit 3 for programming PIC microcontrollers. One of these will form the basis for my keyboard scanner and I initially chose the 16f1507 PIC for this purpose, buying a pack of 4 to try out. However, I realised that a different model (16f687) could open the door to a useful alternative feature: MIDI. My course last year involved a detailed discussion of the MIDI protocol and contained exhaustive information about the various commands and how they are generally implemented. If nothing else a simple note on/off so that the synth can be controlled from a PC sequencer would be very handy...

Going back to the first PICs I bought, the 16f1507 comes with an interesting feature: a Numerically Controlled Oscillator. From a cursory read of the datasheet it would seem that it could be feasible to use this to form the front end of a sawtooth DCO. This would provide superior tuning stability to that found in voltage-controlled oscillators which use temperature-sensitive exponential converters i.e. they go out of tune when they get warmer or cooler.

However there are of course caveats. Quantised frequency control is one issue, as is the slight imprecision in controlling frequency using an NCO. This is something I'll have to investigate further, as well as working out how to implement channel 2 detune, vibrato and portamento in such a setup. Certainly the Roland Juno-6 managed vibrato but not portamento. I'll have to look into how the former was achieved.



No comments:

Post a Comment