[sdiy] How does a DCO work?

[Theo]

Many variations on multiple methods used. Just 3 examples.

1) A counter is preloaded with a value and down counted to zero at a fixed
rate.
When the counter reaches zero it reloads and gives out a pulse. A lower
preload value results in a higher frequency.
The pulse from the counter is then converted in to a saw and square wave.
The saw wave is a bit tricky. Usually the pulse is used to quickly discharge
a capacitor that is otherwise continuously charging.
However the amplitude of the resulting saw wave is depended on frequency. To
correct the amplitude the most significant bits of the preload value are fed
into a DA (might be 3 or 5 bits only). DA output then controls a VCA.

2) Same as method 1 but with a different way of correction for the saw
amplitude.
Again a DA (with a low bit count, 3 to 8 bits) converts part of the preload
value for the counter into a voltage.
However this time the DA is used to charge a capacitor.
Basically you may see this as a crude VCO where the frequency is "corrected"
using the digital counter.

Both method 1 and 2 produce a pure "analogue" wave form however a rather
high rate (Mhz) for the counter is needed to offer good frequency
resolution.
Therefore:
3) A value is added to a modulo counter at a fixed (sample) rate.
Larger value => higher frequency. The counter output might feed directly
into a DA converter to produce a saw wave.
Or alternatively index a ROM/RAM containing a stored waveform. As the output
is always at full amplitude DA can be 8 bit (or less) but in modern times
one would use 16 bit or better anyways??.
Downside, digital waveforms, therefore: quantizing noise, aliassing and or
band-limiting (high frequency loss at lower pitches).

HTH
Theo

----- Original Message -----
From: Johannes Öberg <johannes.oberg at gmail.com>
To: <rdd at rddavis.org>; <synth-diy at dropmix.xs4all.nl>
Sent: Thursday, December 30, 2004 6:51 AM
Subject: Re: [sdiy] How does a DCO work?


> > Digital signals rather than analog signals, i.e. a control voltages,
> > control the frequency of the oscillators.
>
> Yes obviously, but the early DCO's weren't 'digital oscillators' (that
> is, real-time rendering via D/A), were they? How did they control the
> frequency of the (I assume) VCO 'inside' ?