"Isaac Wingfield" wrote in message
...
In article ,
"Trevor Wilson" wrote:

"Stewart Pinkerton" wrote in message
...
On Thu, 23 Sep 2004 01:01:44 GMT, TonyP
wrote:

Stewart Pinkerton wrote:

On Tue, 21 Sep 2004 22:32:42 GMT, TonyP
wrote:

I remember reading about speaker cables that caused some amps major
grief. I believe it was the Polk speaker cables. I never understood
what
happened and why. And, if you used those speaker cables today,
would
the
results be the same.

Highly capacitive cables triggered HF oscillation in amplifiers
which
were only marginally stable, most notably the Naim NAP250. Avoid
Naim
amps, and no modern amplfier should have a problem in this regard.
OTOH, you don't need such weird cables anyway, except in the most
extreme case of say an electrostat speaker driven by more than 30
feet
of cable.

Thanks for the reply. What was the "advantage" of high capacitive
cables?

In general, a cable which exhibits high capacitance will also exhibit
low inductance. Low inductance is *theoretically* desirable in a
speaker cable, as it reduces the cable reactance at high frequencies.
In practice, even driving a 3-ohm load over thirty feet of cable with
Naim NACA5 (probably the *highest* inductance cable commonly
available) will result in a treble droop of less than 1dB at 20kHz.

**Except that, with some speakers (notably electrostatics), low
inductance
cables may well be desirable. Here is the impedance curve of just such a
speaker:

www.rageaudio.com.au/accu.jpg

In this situation, NAIM cables (unless the speaker is to used with a
NAIM
amplifier) would be the very worst choice imaginable. Standard Figure 8
(Zip
cable) would be a slightly less worse choice. High power coax, or Goertz
MI-1 would be the best choices.

If you actually do the calculations to determine the "characteristic
impedance" of a transmission line in the audio frequency range, you
might be more than a little surprised.

**Nope. I won't be surprised in the slightest. The ONLY important
characteristics of speaker cables are (per unit length): Resistance,
Inductance. Capacitance is largely unimportant. Characteristic impedance is
not a point of issue, in any practical length speaker cable.

The line doesn't have one; it is
frequency dependent, unlike the situation at "RF".

**No argument from me.


See the "Schaum's Outline Series" tutorial on Transmission Lines to
learn how to do the calculations correctly. Most "transmission line"
texts do not cover how "characteristic impedance" is calculated at low
frequencies; that one does .The calculation is nowhere near the same as
at higher frequencies.

**OK. I studied this stuff 30 years ago. It is exactly as relevant to audio
today, as it was then. IOW: Not at all. Your strawman is duly noted,
however.

When you have time, do some calculations, using some regular 'zip' cable, in
(say) 10 Metre lengths, with the speakers whose impedance I posted. Get back
to me then.

BTW: Here are some figures pertaining to various speaker cables (thanks to
Fred Davis):

---

12 AWG (Manhattan 39059) 0.25 uH/ft 24 pF/ft 0.0036 ohm/ft

12 AWG (Belden 8477) 0.28 uH/ft 23 pF/ft 0.004 ohm/ft

12 AWG (Belden 9718) 0.23 uH/ft 22 pF/ft 0.0033 ohm/ft
(lamp cord construction - similar to RS MegaCable)

MIT CVT 0.22 uH/ft 151 pF/ft 0.004 ohm/ft
(approx. 13 AWG)

Dunlavy Z6 0.025 uH/ft 645 pF/ft 0.0028 ohm/ft
(11 AWG)

Goertz MI-1 (13 AWG) 0.004 uH/ft 480 pF/ft 0.004 ohm/ft
(Alpha-Core) (about 0.1uH for 25 feet)

Kimber 16LPC 0.07 uH/ft 61 pF/ft 0.0025 ohm/ft
(11 AWG)

Twisted pair flat cable 0.04 uH/ft 685 pF/ft 0.0026 ohm/ft
(Amphenol Spectra-Twist 843-138-2601-064)
(11 AWG)
---

12 AWG (Manhattan 39059) 0.25 uH/ft 24 pF/ft 0.0036 ohm/ft

12 AWG (Belden 8477) 0.28 uH/ft 23 pF/ft 0.004 ohm/ft

12 AWG (Belden 9718) 0.23 uH/ft 22 pF/ft 0.0033 ohm/ft
(lamp cord construction - similar to RS MegaCable)

MIT CVT 0.22 uH/ft 151 pF/ft 0.004 ohm/ft
(approx. 13 AWG)

Dunlavy Z6 0.025 uH/ft 645 pF/ft 0.0028 ohm/ft
(11 AWG)

Goertz MI-1 (13 AWG) 0.004 uH/ft 480 pF/ft 0.004 ohm/ft
(Alpha-Core) (about 0.1uH for 25 feet)

Kimber 16LPC 0.07 uH/ft 61 pF/ft 0.0025 ohm/ft
(11 AWG)

Twisted pair flat cable 0.04 uH/ft 685 pF/ft 0.0026 ohm/ft
(Amphenol Spectra-Twist 843-138-2601-064)
(11 AWG)

On Mon, 27 Sep 2004 00:16:10 GMT, "Trevor Wilson"
wrote:


"Stewart Pinkerton" wrote in message
.. .
On Thu, 23 Sep 2004 01:01:44 GMT, TonyP
wrote:

Thanks for the reply. What was the "advantage" of high capacitive cables?

In general, a cable which exhibits high capacitance will also exhibit
low inductance. Low inductance is *theoretically* desirable in a
speaker cable, as it reduces the cable reactance at high frequencies.
In practice, even driving a 3-ohm load over thirty feet of cable with
Naim NACA5 (probably the *highest* inductance cable commonly
available) will result in a treble droop of less than 1dB at 20kHz.

**Except that, with some speakers (notably electrostatics), low inductance
cables may well be desirable. Here is the impedance curve of just such a
speaker:

www.rageaudio.com.au/accu.jpg

In this situation, NAIM cables (unless the speaker is to used with a NAIM
amplifier) would be the very worst choice imaginable. Standard Figure 8 (Zip
cable) would be a slightly less worse choice. High power coax, or Goertz
MI-1 would be the best choices.

However, while that is *theoretically* true, even in this particularly
extreme case (and assuming the amplifier itself has no problem), I
would be surprised if there was any *audible* difference with the
usual 10-15 feet of cable.
--

Stewart Pinkerton | Music is Art - Audio is Engineering

"Stewart Pinkerton" wrote in message
...
On Mon, 27 Sep 2004 00:16:10 GMT, "Trevor Wilson"
wrote:


"Stewart Pinkerton" wrote in message
.. .
On Thu, 23 Sep 2004 01:01:44 GMT, TonyP
wrote:

Thanks for the reply. What was the "advantage" of high capacitive
cables?

In general, a cable which exhibits high capacitance will also exhibit
low inductance. Low inductance is *theoretically* desirable in a
speaker cable, as it reduces the cable reactance at high frequencies.
In practice, even driving a 3-ohm load over thirty feet of cable with
Naim NACA5 (probably the *highest* inductance cable commonly
available) will result in a treble droop of less than 1dB at 20kHz.

**Except that, with some speakers (notably electrostatics), low
inductance
cables may well be desirable. Here is the impedance curve of just such a
speaker:

www.rageaudio.com.au/accu.jpg

In this situation, NAIM cables (unless the speaker is to used with a NAIM
amplifier) would be the very worst choice imaginable. Standard Figure 8
(Zip
cable) would be a slightly less worse choice. High power coax, or Goertz
MI-1 would be the best choices.

However, while that is *theoretically* true, even in this particularly
extreme case (and assuming the amplifier itself has no problem), I
would be surprised if there was any *audible* difference with the
usual 10-15 feet of cable.

**You DID mention 30 feet (let's get with the programme, Stewart: It's
almost 10 Metres, not 30 feet). My comments were couched, with that length
under consideration.


--
Trevor Wilson
www.rageaudio.com.au

"Pooh Bear" wrote in message
...
Isaac Wingfield wrote:

If you actually do the calculations to determine the "characteristic
impedance" of a transmission line in the audio frequency range, you
might be more than a little surprised. The line doesn't have one; it is
frequency dependent, unlike the situation at "RF".

That's because audio frequencies have very long wavelengths.

Only when the cable length approaches the signal wavelength does a
characteristic
impedance become a relevant issue.


Has anybody done an appropriate ABX double-blind test on the effect of
hi-zoop speaker cables on lamps ?

Do the words spring out of one's book with more depth, is the room bright,
less glary ? Is the spectral balance more even (none of those stressful
ultra-violet undertones ) ????

geoff

In article ,
Pooh Bear wrote:

Isaac Wingfield wrote:

If you actually do the calculations to determine the "characteristic
impedance" of a transmission line in the audio frequency range, you
might be more than a little surprised. The line doesn't have one; it is
frequency dependent, unlike the situation at "RF".

That's because audio frequencies have very long wavelengths.

There's actually a bit more to it than that. The propagation velocity is
a function of frequency, too. As frequency drops without limit,
propagation velocity drops without limit too. Bass notes *really do*
take longer to get to the speaker.

But nowhere near enough to matter, unless your speakers are in the next
state.

Isaac

On Tue, 28 Sep 2004 06:04:22 GMT, Isaac Wingfield
wrote:

In article ,
Pooh Bear wrote:

Isaac Wingfield wrote:

If you actually do the calculations to determine the "characteristic
impedance" of a transmission line in the audio frequency range, you
might be more than a little surprised. The line doesn't have one; it is
frequency dependent, unlike the situation at "RF".

That's because audio frequencies have very long wavelengths.

There's actually a bit more to it than that.

No, I don't think so. The transmission line theory calculations is
*based on* the fact that the length of the line is of the same order
of magnitude (or greater) than the wavelength. If you go on and
calculate anyway, your calculations has no value at all. Check your
textbooks, please.


But nowhere near enough to matter, unless your speakers are in the next
state.

True, but that's what Pooh Bear said.


Isaac

Per.

"Geoff Wood" -nospam wrote in message

"John Krieger" wrote in message
...
Not actually nonesense, since the current runs mostly in the surface
of the
iwr. Finer strands = more surface are = less resistance.

Yep, at 200KHz or so. But these strands are touching, so do not act
as individual conductors.

Even if the strands were individually insulated, it wouldn't matter.

In article ,
Per Stromgren wrote:

On Tue, 28 Sep 2004 06:04:22 GMT, Isaac Wingfield
wrote:

In article ,
Pooh Bear wrote:

Isaac Wingfield wrote:

If you actually do the calculations to determine the "characteristic
impedance" of a transmission line in the audio frequency range, you
might be more than a little surprised. The line doesn't have one; it is
frequency dependent, unlike the situation at "RF".

That's because audio frequencies have very long wavelengths.

There's actually a bit more to it than that.

No, I don't think so. The transmission line theory calculations is
*based on* the fact that the length of the line is of the same order
of magnitude (or greater) than the wavelength. If you go on and
calculate anyway, your calculations has no value at all. Check your
textbooks, please.

The calculations use *different equations* in the low frequency domain.
Things that don't matter at higher frequencies predominate here. For any
given construction of a transmission line, there is a "crossover
frequency" below which it doesn't have a "characteristic impedance".
Study the treatment in the Schaum's Outline book "Transmission Lines",
and then we'll talk.

But nowhere near enough to matter, unless your speakers are in the next
state.

That difference in velocity between low and high audio frequencies is
exactly why the telephone company added all those "loading coils" to
their long lines before they developed carrier telephony.

The loading coils increased the value of inductance per unit length of
the lines; that tended to equalize the velocity over frequency. Without
them, voices were unintelligible because the different frequencies
arrived at different times.

Then they developed the carrier system, which used single sideband
modulation in the hundreds of kilohertz range. That eliminated the
velocity difference, but the system wouldn't work with all those loading
coils stuck up all those poles. So they had to take them all out.
Millions of them.

Isaac

In article ,
"Arny Krueger" wrote:

"Geoff Wood" -nospam wrote in message

"John Krieger" wrote in message
...
Not actually nonesense, since the current runs mostly in the surface
of the
iwr. Finer strands = more surface are = less resistance.

Yep, at 200KHz or so. But these strands are touching, so do not act
as individual conductors.

Even if the strands were individually insulated, it wouldn't matter.

If each individual strand was larger than a #18 wire, it would 8^}

But not much.

Isaac