Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83 v7 ucbtopaz-1.8; site ucbtopaz.CC.Berkeley.ARPA
Path: utzoo!watmath!clyde!burl!mgnetp!ihnp4!zehntel!dual!amd!decwrl!decvax!ucbvax!ucbtopaz!newton2
From: newton2@ucbtopaz.CC.Berkeley.ARPA
Newsgroups: net.audio
Subject: Re: Re: Wiring Your House for Sound (pt 2 of 2)
Message-ID: <520@ucbtopaz.CC.Berkeley.ARPA>
Date: Fri, 10-Aug-84 01:26:40 EDT
Article-I.D.: ucbtopaz.520
Posted: Fri Aug 10 01:26:40 1984
Date-Received: Sun, 5-Aug-84 05:49:21 EDT
References: <90@whuxl.UUCP>, <522@clyde.UUCP>
Organization: Univ. of Calif., Berkeley CA USA
Lines: 53

I agree (since I was one of those who proposed it) that the most sensible
solution is to run low-level audio everywhere and have locally sited
power amps drive nearby speakers.

HOWEVER, well-intentioned mythology isn't limited to insisting on stranded
bus-bars (bifilar niobium-tin braid, helium-cooled?). While so-called 
low-impedance lines (low source impedance, that is) are a necessity,
this is largely to avoid high-frequency roll-off due to the shunt capacitance
of long lines. For interference rejection, low impedance is most resisant
to capacitively-coupled crosstalk. Balanced lines, however, are usually
overkill, although an indispensible marketing ploy, particularly in
semi-pro gear- likewise the use of "Cannon" or XLR-type connectors.
Believe me, no one needs or will long tolerate XLRs at home- they seem
neat when you're a hobbyist pretending to be Captain Video, but if you
really are a professional audio drudge and have a living room cluttered
with that fabled "professional" gear, it gets old fast. By and large,
consumer stuff is as good or better *in performance* as just about any
commercial or recording studio stuff- the difference is in ruggedness,
serviceability, conformance to interface standards and other costly
features you can't hear. 

And consider how one would distribute "balanced" audio: unless you
resort to transformers, which are *VERY* expensive at the quality levels
which bear mentioning on net.audio, you will have to set up some variety
of kludged differential input at each amp, and will also consider how
to derive an allegedly "balanced" *output* to drive the line.
I say
allegedly because the concept of "balanced" is ill-defined (or underdetermined)
in audio parlance- sometimes one means "floating", i.e. both signal
conductors exhibit a high impedance to ground; sometimes "balanced to
ground", meaning equal but not large impedances to ground. Sometimes the
source is single-ended but the load is differential, and so endlessly on.
Each configuration has its applications in interference-rejection or
particular measurement applications (consider the power-amp as measuring
the signal-plus-noise presnted to it and trying to derive the best
estimate of the signal alone). In most domestic installations, low source-
impedance single-ended circuits will prove adequately immune (if shielded!)
to *induced* noise; the more likely source of problems is from ground-loop
hum, and possibly RFI. As much thought will be necessary to eliminate
this with all but the floating transformer-to-floating transformer setup
regardless of whether single-ended or differential circuits are used.

I'm not disparaging differential approaches, just trying to keep
effort/benefit ratios in perspective. The first thing to try is
lifting the AC power grounds of one or more units of widely dispersed
systems.

By the way, I don't know what to make of the prescription "common
mode rejection should be 100 dB or more". Is this like saying
everyone should be healthy, wealthy and wise? Just the twisted-pair
*cable* is likely to have worse than 100 dB CMRR over DC-20 kHz,
and that's only a tiny slice of the spectrum of interest to
*real* audiophiles, eh?