[sdiy] Idea - Triangle wave DCO core

[Magnus Danielson]

On 03/14/2012 10:07 PM, Tom Wiltshire wrote:
>
> On 14 Mar 2012, at 20:19, David G Dixon wrote:
>
>>> You'd need some sort of servo mechanism around the integrator
>>> to deal with any DC drift, otherwise you'll end up hitting
>>> one of the rails sooner or later as DC terms get integrated over time.
>>
>> I concur.  I don't see any way around the schmitt trigger.  However, it
>> could simply fire an electronic switch between the CV inverter + input and
>> ground to change the polarity of the CV into the integrator.  Essentially,
>> then, you'd have a digital expo, so you'd need some pretty high resolution
>> for tuning.  Expo FM would simply be counting in binary up and down.  Could
>> be kewl.
>>
>
> Couldn't we use a leaky integrator? The DC drift problem is likely to occur over longer periods (seconds) and the lowest frequency is going to be in the tenths of seconds, say. So highpass the integrator to lose any DC. Or is that nonsense? I've only ever done it in the digital domain.

This is what I proposed a few emails back.

> The idea was to do the vast majority of the work in the software and have almost no hardware. If I start adding comparators and switches and such like, it loses its charm and I might as well build a sensible VCO circuit.

Indeed, But essentially, what you are trying to do is a form of 
sigma-delta modulation.

Just recall that instead of just interpolating cycles to on average 
reach the right rate, you can interpolate the step such that two samples 
take their "share" of the step and hence fractional step per sample can 
be achieved for better time correctness.

Cheers,
Magnus