Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!uwvax!oddjob!gargoyle!ihnp4!upba!eecae!crlt!russ
From: russ@crlt.UUCP (Russ Cage)
Newsgroups: rec.audio,sci.physics,sci.electronics
Subject: Re: Mercury Filled Speaker Wire
Message-ID: <771@crlt.UUCP>
Date: Wed, 16-Sep-87 10:02:51 EDT
Article-I.D.: crlt.771
Posted: Wed Sep 16 10:02:51 1987
Date-Received: Sat, 19-Sep-87 15:08:21 EDT
References: <3816@watdcsu.waterloo.edu> <578@uthub.toronto.edu> <1709@crash.CTS.COM>
Followup-To: rec.audio
Distribution: rec
Organization: CRLT , Ann Arbor, MI
Lines: 50
Keywords: uninformed bunk masquerading as audiophile gospel
Summary: A few numbers for Mr. Bill Blue
Xref: mnetor rec.audio:3299 sci.physics:2193 sci.electronics:1331

In article <1709@crash.CTS.COM>, bblue@crash.CTS.COM (Bill Blue) writes:
>In my view, the big culprit that kills the 'hardware store' variety
>cables is the varying time of arrival of different frequencies at the
>other end of the cable due to skin effect propagation and other factors.
>This simply can not be made up for, or controlled by simply paralleling
>average cables together.  While they'll have a low DC resistance, and
>will allow good control (damping) of the speakers (assuming the amp has
>good control in the first place), they will also have a very confused
>representation of inner detail.  They most certainly will not pass the
>test on any aware audiophile's ears.

Okay, to put the "differences in arrival time" argument to rest,
*permanently*:

Let us assume that we have a cable which is *extremely* dispersive.
At 20 Hz, signals propogate at .6c (velocity factor of .6); on the
other hand, at 20 KHz, they propagate at .4c.  (Typical variations
in a *real* cable over much larger ranges might be a few percent.)
Assume for a moment that our cable has:

1.)	Insignificant non-linear effects (not too unlikely), and
2.)	A ten-meter length.

What happens to a mix of signals from 20 Hz to 20 KHz, going from
one end to the other?

The 20 Hz signals move off at .6 of the speed of light.  They cover
the ten meters in 5.56e-8 seconds, or 56 nanoseconds.  The 20 KHz
signals arrive more slowly; they take 8.33e-8 seconds, or 83 nanoseconds,
to arrive.  Thus, they get to the speaker 28 nanoseconds later.

This 28 nanosecond delay is about one two-thousandth of a cycle
time at 20 KHz, or about two-tenths of a degree of phase delay
at the *maximum* frequency; it will be much less at the frequencies
where phase perception is important.  Is there *anyone* who can hear
such a phase delay?  I doubt it.

Note that dispersion in any *real* cable is going to be less by
orders of magnitude, and you'll see how ridiculous this discussion
about cables is.  Dispersion is simply not a factor; resistance,
on the other hand, affects the damping of the speaker by the
amplifier, and thus has a *large* effect on sound quality.  Skin
effect can be dealt with by using flat conductors or Litz wire,
or (again) *several parallel strands*.

Followups to rec.audio.technobabbling; keep this out of sci.*.
-- 
  The above are the official opinions and figures of Robust Software, Inc.
HASA, "A" division.                      Go ahead, flame.  I bought Dow stock!
Russ Cage, Robust Software Inc.		    ihnp4!itivax![m-net!rsi,crlt!russ]