[sdiy] Roland DCOs

[Florian Anwander]

Hi Tom

> Thanks for your help, but I'm afraid I don't understand this. The  ramp 
> wave is produced by linearly charging a capacitor, which is then  
> discharged by a reset pulse from the counter. What can be  assymetrical 
> about this? How can the waveshape be altered?
This is not a real digital controlled oscillator, but half voltage 
controlled, half counter controlled. The voltage determines how fast the 
capacitor of the VCO is charged. In a normal VCO a comparator would 
check the load of the capacitor and if it reaches a certain voltage the 
comparators output would switch the transistor that shortens the 
capacitor. This part is missing here. So if there is no signal from the 
counter, the voltage at the capacitor will rise until the maximum and 
stay there.
The trigger from the counter acts like a sync signal in a normal VCO, 
where it unloads the capacitor. Now if the cap in the DCO is charged 
"too fast" it will keep the maximum and the voltage will stay there 
"flat" until the reset from the counter comes.

So the waveform will be more like a spike than a saw, if the cv is wrong.

> I'd also be interested to know more about the Cross Mod, as this is  the 
> part of the circuit that I understand least. The input marked   _METAL 
> is a mystery to me. The cross mod seems to use DCO1's output  (which 
> could be noise) to amplitude modulate the output of DCO2. This  will 
> give an FM-like effect, but with fewer sidebands. Is there more  to it 
> than this?
I have to admit, I never really understood the metal crossmod.

Florian