[sdiy] "Time Winding" in Audio Cables ???

[Magnus Danielson]

From: Glen <mclilith at charter.net>
Subject: Re: [sdiy] "Time Winding" in Audio Cables ???
Date: Mon, 11 Jul 2005 14:54:51 -0400
Message-ID: <4.3.2.7.2.20050711143758.03ecc6f8 at mail.charter.net>

Glen,

> Thanks everyone, for confirming my original belief that this time-aligned 
> cable stuff was just so much worry over nothing noticeable. I knew about 
> the capacitive low-pass filtering action of cables, and I knew about 
> reflections that can occur (especially at RF frequencies), and I knew about 
> signal losses due to resistance and other factors. I had just forgotten 
> about the difference in propagations speeds for different frequencies. Even 
> if there is a difference, I just couldn't imagine it being enough to be a 
> noticeable problem for a 1 or 2 meter shielded patch cable.  :)
> 
> Out of curiosity, does anyone have any idea what amount of arrival time 
> difference there might be, between a 20Hz signal and a 20 kHz signal, over 
> a "typical" mediocre-quality 2 meter shielded patch cable? I imagine it's a 
> very small difference, but does anyone have any actual stats regarding how 
> small?
> 
> This might be a good question for Magnus. He knows everything.  :)

I don't know everything, but I do keep to wish that I knew more and
occasionally do something about it, and then some stuff rub off and get stuck
in memory. At rare occasions I do but the duplo-pieces together, and everyone
seems to enjoy seeing it. :P

I did however measure the phase difference between 600 Hz and 20 kHz of a
2 meter cable (actually two 1m cables with XLRs, they where again on armlengths
away, as any highly scientifically study would use). The extremely flat phase
curve had the huge difference of -144 m degrees of phase from the 600 Hz and
20 kHz reading. The group delay stays fairly flat too (sorry, forgot the
readings!). I could make better measurements, but that would consume time which
I don't have at the moment. It takes time to make calibrations etc. Also, I am
not too happy about using the analyser at those frequencies, I need to make a
better R/T-kits for those frequencies.

Anyway, there are measurable differences (at least with the right equipment,
which the average buyer certainly does not have) over frequency, but they are
really down at levels which are ridiculously low to be considered.

If I where to toss one of my speakers with a mike into the measurement loop
oh my gooly what a ride we would have!

So, seriously, even when doing measurements with real tools instead of only
toy with words on abstract theory the differences are veeery tiny. I would not
waste much more time on finding the differences there. For short cable runs you
only need to care about sufficient cross-area for sufficiently low resistance.
For a little longer cable runs you start to care more about it, but then only
slightly.

Cheers,
Magnus