[LAU] turning a consumer soundcard into "prosumer" w/ quasi-balanced outs

[Monty Montgomery]

> I think this would be an excellent area to investigate with your
> tools. Almost a better use of the scientific method than whole
> soundcard testing because you can focus on a single variable that is
> known to be problematic.

...*reputed* to be problematic.  The issues people have with
capacitors tend to be the big, low-ESR power caps that are driven hard
their entire lives.  The conditions for audio caps are very different.

> And unfortunately, electrolytic capacitors are the one component that
> are about as far from their theoretical ideal of any component.

No, not even close.  Cored inductors are *way* worse.  Transistors
themselves are far less linear then caps.

> In
> fact, i think it's totally braindamaged to be using electrolytic
> capacitor for audio coupling...

Also false, and I've proven it to myself experimentally.
Electrolytics used to couple small signals (such as line in/line out)
affect audio unmeasurably, unless the cap is dead or the circuit is
incorrectly designed.

Caps can make a difference in filter circuits where they're operating
near the active area of the filter, but I've never been able to
measure any electrolytic contribution to an audio circuit in a strict
coupling function.

At one point I had intended to write up the differences between
various capacitors used in output stage couplig via direct
experimental measurement, and found that I couldn't actually measure a
damned thing.  No measurable contribution whatsoever (down to -110dB,
my measurement limit).   None.  Zero.  Nada.  The analyzers that were
far more sensitive than my very good ears couldn't tell the caps from
a wire.

Monty