[sdiy] Diode Matching

[René Schmitz]

Hi Mattias,

On 11.04.2020 13:31, Mattias Rickardsson wrote:
> On Sat, 11 Apr 2020 at 11:51, René Schmitz <synth at schmitzbits.de 

> It's true that a passive transformer ring mod has quite a lot of 
> distortion-like "artefacts", but there's a good chance that smaller 
> inconsistencies due to bad matching do reveal themselves. A bit like 
> when you run a sound into a clipping distortion stage - the output is 
> primarily the characteristic sound of clipping, but it still matters 
> what input signal you give it and what DC offset it has, etc.

I'm not talking about distortion. I mean "unanticipated effects".

One example: the DC winding resistances of the half-winding in the 
transformer will likely not be matched to the same degree. They might 
have the same number of turns, but are wound on top of each other. That 
means one has a longer wire then the other. (True bifilar windings would 
be expensive.)

My point is: you always need to look at the whole application, you don't 
gain much using one low tolerance part when everything else around it is 
all around the shop tolerance wise.

> In the case of a ringmod, even a fuzzy one, I'd expect badly matched 
> diodes to be able to ruin the suppression of carrier harmonics just 
> slightly - which (even if it's a small difference in theory) I'd expect 
> to be a but very audible difference even if you'd run it through all 
> your dist pedals in series. :-)

A mismatch in the winding resistance will give you a bad carrier 
suppression as well. (Granted, you can balance this with additional 
circuitry, but so can you with the diode mismatch.)
The carrier suppression in these things is not exceptionally good to 
begin with, matched diodes or not.

Best,
  René

--
synth at schmitzbits.de
http://schmitzbits.de