[Dick Pierce]

On Jul 9, 11:35*am, Ed Seedhouse wrote:
On Jul 8, 9:28*pm, ScottW2 wrote:

There are occacionally reasons where cables may make an
audible difference. * High source impedance devices requiring low
capacitance
cables or noise due to ground impedances (ground loops) are two that
quickly come to mind. *But none of these require hi-dollar cables to
address.

These effects are only applicable in cables that exceed one half
wavelength. *

Nope, wrong.

That the wavelengths are MUCH longer makes the
lumped parameters important (why they're longer
you also got wrong, but we'll deal with that separately)

Take a high source impedance, oh 10kOhms.
Hook a cable to that whose capacitance is,
say, 200 pF. And let's say the load resistance
is 100kOhms. That arrangement, all by itself, will
result in a frequency response that looks like:

20 -0.8
... ...
2030 -0.8
2560 -0.9
3230 -0.9
4060 -0.9
5120 -0.9
6450 -1
8130 -1.1
10200 -1.2
12900 -1.4
16300 -1.7
20500 -2.1

This is the simple attenuation due to the source
impedance, distributed capacitance and load
impedance. And the wavelengths are MUCH
longer than this 10 foot cable used in this test.

If your assertion were carried to it's logical
extreme, you could assert that simple resistors
and capacitors couldn't filter frequencies whose
wavelength is bigger than the individual
components.

The half wavelength of an electrical impulse
oscillating at 100K

An impulse doesn't oscillate at such a frequency.
An impulse, by its definition, is a broad-band
phenomenon.

will be 3000 meters at most. *

Speed of light is roughly 3x10^8 meters/second.
Since:

c = f l

and thus:

l = c/f

then if c=3x10^8 and f=1x10^5, then

l = 3x10^8 / 1*10^5

thus l = 3x10^3, or 3000 for a FULL wavelength.
There's no way that a half wavelength could
ever be more than 1500 meters.

If electrons in a
wire travel at 1/10 of the speed of light in a
vacuum then the half wavelength of a 20Khz
tone will be 750 meters. *

Wrong. Completely. The signal propogation
velocity is NOT related to the electron drift
velocity in a conductor. The drift velocity is,
literally, a snail's pace: for currents like
what we're talking about, it's measured in
feet per HOUR.

If your assertion were correct: that the
electrons in the wire were zipping along
at some 30,000,000 (3x10^7) meters per
second, your assertion has some issues.
For one, at that sort of velocity, relativistic
effects are already significant. For another,
an electron witn that sort of velocity has
a kinetic energy of around 3000 eV: where
did the 3000 volt accelerating potential that
got them going that fast come from?

The SIGNAL PROPOGATION velocity is
something else entirely. Yes, it's related to
the speed of light in a vacuum by something
called the velocity factor for the conductor,
which for caox cable is on the order of 0.5
or so. Thus, the signal velocity in a cable is
on the the order of 150,000,000 meters per
second.

That wavelength at 20 kHz is not, as you
suggested, 750 meters, but 7500 meters,
or on the order of a mile.

It becomes pertinant at radio frequences,
where a 14 mhz signal is assumed to
have a wavelength of about 20 meters in
air.

It's not ASSUMED to be anything: it is or it isn't.

Sorry, I don't think your gonna have any 750
meter cables in your home
audio setup.

Irrelevant, since your assumptions are wrong.