[sdiy] Deep thoughts on Old drum machines: the "VCA"

Richie Burnett rburnett at richieburnett.co.uk
Sat Jan 29 13:34:12 CET 2022

> From: Adam Inglis (synthDIY)
> ...When I’m editing recordings at the waveform level in Cubase, I’ve 
> noticed that my most, er… ‘characterful' synths have the most waveform 
> distortion about the zero axis (sorry, can’t think of the technical term!) 
> i.e. the waveform will be skewed more above or below the line (Korg 700, 
> TB-303, SH-2000) whereas more polite sounding synths are perfectly 
> symmetrical about this axis. I’m assuming there is some kind of DC offset 
> going on there, but that may not be the correct explanation
> A

It could be a DC offset...  However, any DC offset inside the synth would 
fairly quickly get converted to a decaying DC transient by the numerous AC 
coupling capacitors in the audio chain before it gets digitised.  So if by 
"characterful" you mean it has a thump or "key click" at the onset, it could 
well be a DC transient thing that you're seeing.

The other thing that can make waveforms *look* asymmetrical about the time 
axis, is short pulse waveforms...  If you dial up a nasal sounding narrow 
pulse waveform from your synth and then record that and look at the waveform 
you will probably see that it tends to have a greater excursion in one 
direction than the other.  (Whether this is the positive or negative 
direction depends on how many times the signal gets inverted before it gets 
digitised and whether the original duty ratio was like 10% or like 90%!) 
This is because the DC component of the pulse waveform doesn't make it 
through to the recording...  Therefore the average value over one cycle must 
be zero and the waveform moves up or down accordingly so that the areas 
enclosed under the wide part and the narrow part are equal.  Conversely, 
waveforms like Square, Saw, Triangle, Sine tend to look more symmetrical 
about the time axis with roughly equal excursions in the positive and 
negative directions, because they don't have to move up or down in order to 
have an average of zero and no DC offset.

However, in the case where a low-pitch sawtooth is being passed through lots 
of AC coupling (high-pass filter action like that in the TB-303) it can 
start to look less like a textbook sawtooth and more like a step in one 
direction followed by an exponential decay after it.  More like the blue or 
red waveforms, and less like the "textbook" green sawtooth in this pic that 
I rather conveniently found online:


This happens not due to DC offset, but because the high-pass filter action 
starts to mess with the phases of the harmonics that make up the sawtooth if 
you play sufficiently low pitched notes.

I wouldn't like to say how much influence this has on the character of the 
pure sawtooth alone, but it definitely has a profound effect on how it 
sounds when you apply lots of distortion!!!


This email has been checked for viruses by AVG.

More information about the Synth-diy mailing list