Maximum wire length is a function of... what?

[August Karlstrom]

I have read more than one article where the recommended maximum length
of loudspeaker wire is presented as a function of the wire gauge and
the impedance of the speakers, e.g. in the Wikipedia article

http://en.wikipedia.org/wiki/Speaker_wire#Wire_gauge

I have also heard that output power and damping factor should be
considered as well - a high power amplifier needs a thicker cable.

Are there any studies made about the wire length contra output power and
damping factor?


/August

[Arny Krueger]

"August Karlstrom" > wrote in message


> I have read more than one article where the recommended
> maximum length of loudspeaker wire is presented as a
> function of the wire gauge and the impedance of the
> speakers, e.g. in the Wikipedia article
>
> http://en.wikipedia.org/wiki/Speaker_wire#Wire_gauge

I chased the footnotes for this paper, and found that it was heavily based
on this paper:

http://procosound.com/download/whitepapers/Understanding%20Speaker%20Cables.pdf

Proco Sound is a well-known supplier to the professional audio and in
particular the live sound industry. In general the acccuracy requirements
for live sound are somewhat relaxed as compared to high-accuracy home or
mastering/mixing room audio. The Wikipedia article wire length
recommendations seem to be based on maintaining a damping factor of about 20
which again is somewhat relaxed as compared to the requirements for high
performance audio. I would recomment using a criteria of maintaining the
damping factor between 50 and 100.

> I have also heard that output power and damping factor
> should be considered as well - a high power amplifier
> needs a thicker cable.

Generally speaking, sizing wire for a good damping factor will result in
adequate capacitor for the kinds of power amps generally used for home
audio.

> Are there any studies made about the wire length contra
> output power and damping factor?

Yes. The Greiner JAES paper is a classic and covers other issues. If the
speakers being used have unusually variable and low impecance curves then
the resistance and inductance of the speaker cable become more signficant.

[August Karlstrom]

On 2010-10-15 21:41, Arny Krueger wrote:
> In general the acccuracy requirements
> for live sound are somewhat relaxed as compared to high-accuracy home or
> mastering/mixing room audio. The Wikipedia article wire length
> recommendations seem to be based on maintaining a damping factor of about 20
> which again is somewhat relaxed as compared to the requirements for high
> performance audio. I would recomment using a criteria of maintaining the
> damping factor between 50 and 100.

That's interesting. I read the article with the impression that they
were talking about home audio. The Wikipedia article probably needs to
be updated then. If we would make a similar table applicable to high
performance audio, what would it look like?

>> I have also heard that output power and damping factor
>> should be considered as well - a high power amplifier
>> needs a thicker cable.
>
> Generally speaking, sizing wire for a good damping factor will result in
> adequate capacitor for the kinds of power amps generally used for home
> audio.

Sorry, I'm not sure what that means in a practical sense. To take a
concrete example, let's say I have two 2x1.5 mm^2 (approx.15 AWG) OFC
cables, each of length four meter. Is there a high performance system in
which these cables would be insufficient?

>> Are there any studies made about the wire length contra
>> output power and damping factor?
>
> Yes. The Greiner JAES paper is a classic and covers other issues. If the
> speakers being used have unusually variable and low impecance curves then
> the resistance and inductance of the speaker cable become more signficant.

OK, thanks for the reference Arny.


/August

[Trevor Wilson[_3_]]

"August Karlstrom" > wrote in message
...
>I have read more than one article where the recommended maximum length
> of loudspeaker wire is presented as a function of the wire gauge and
> the impedance of the speakers, e.g. in the Wikipedia article
>
> http://en.wikipedia.org/wiki/Speaker_wire#Wire_gauge

**Like all such cliams, the Wiki article over-simplifies things. They ignore
the very real effects of inductance, when using long cable runs and speakers
which exhibit a falling impedance at HF (like ESLs). In those cases, the
inductive effects of cable can become quite significant.

>
> I have also heard that output power and damping factor should be
> considered as well - a high power amplifier needs a thicker cable.

**Again, it depends on the speakers and the length of the cable.

>
> Are there any studies made about the wire length contra output power and
> damping factor?

**It's just maths. There's no real mystery here. The problem with the term
'damping factor' is that it relates to (usually) an 8 Ohm impedance. Hardly
any loudspeakers present an 8 Ohm impedance. A far better term for amplifier
manufacturers to use is output impedance vs. frequency. IOW: The output
impedance should always be specified from 20Hz ~ 20kHz. Many SET and Class D
amplifier manufacturers carefully avoid citing this data for some very good,
albeit cynical, reasons. The output impedance of such amplifiers can easily
reach several Ohms at 20kHz.


--
Trevor Wilson
www.rageaudio.com.au

[August Karlstrom]

On 2010-10-17 03:17, Trevor Wilson wrote:
> "August > wrote in message
>> http://en.wikipedia.org/wiki/Speaker_wire#Wire_gauge
>
> **Like all such cliams, the Wiki article over-simplifies things. They ignore
> the very real effects of inductance, when using long cable runs and speakers
> which exhibit a falling impedance at HF (like ESLs). In those cases, the
> inductive effects of cable can become quite significant.

To take my example again, let's say I need a pair of cables, each four
meters long. Is there a system in which a 2x1.5 mm2 (approx.15 AWG) OFC
wire would be insufficient?

Several people in this newsgroup believe that all wire sounds the same
provided it is of the "correct type". What I'm trying to find out here
is what the correct type is. I am not so interested in general claims
that cannot be used in practice to make decisions.

Thanks for the input.


/August

[Audio Empire]

On Sun, 17 Oct 2010 06:54:25 -0700, August Karlstrom wrote
(in article >):

> On 2010-10-17 03:17, Trevor Wilson wrote:
>> "August > wrote in message
>>> http://en.wikipedia.org/wiki/Speaker_wire#Wire_gauge
>>
>> **Like all such cliams, the Wiki article over-simplifies things. They ignore
>> the very real effects of inductance, when using long cable runs and speakers
>> which exhibit a falling impedance at HF (like ESLs). In those cases, the
>> inductive effects of cable can become quite significant.
>
> To take my example again, let's say I need a pair of cables, each four
> meters long. Is there a system in which a 2x1.5 mm2 (approx.15 AWG) OFC
> wire would be insufficient?
>
> Several people in this newsgroup believe that all wire sounds the same
> provided it is of the "correct type". What I'm trying to find out here
> is what the correct type is. I am not so interested in general claims
> that cannot be used in practice to make decisions.
>
> Thanks for the input.
>
>
> /August

All wire sounds exactly the same. Copper zip cord, 14 Gauge or bigger should
be sufficient for any 4-meter run for any domestic amplifier or speaker
setup. If you have a really big amp (say 700 watts/channel or more), you
might want to increase that size to something like the bulk cable used in
industrial-sized extension cords (12 -10 Gauge). Forget everything else, it
doesn't matter.

[Trevor Wilson[_3_]]

"August Karlstrom" > wrote in message
...
> On 2010-10-17 03:17, Trevor Wilson wrote:
>> "August > wrote in message
>>> http://en.wikipedia.org/wiki/Speaker_wire#Wire_gauge
>>
>> **Like all such cliams, the Wiki article over-simplifies things. They
>> ignore
>> the very real effects of inductance, when using long cable runs and
>> speakers
>> which exhibit a falling impedance at HF (like ESLs). In those cases, the
>> inductive effects of cable can become quite significant.
>
> To take my example again, let's say I need a pair of cables, each four
> meters long. Is there a system in which a 2x1.5 mm2 (approx.15 AWG) OFC
> wire would be insufficient?

**Without knowing the impedance curve (20Hz ~ 20kHz) of your speakers, I
cannot say. In all probability, yes, it would be OK.

>
> Several people in this newsgroup believe that all wire sounds the same
> provided it is of the "correct type". What I'm trying to find out here is
> what the correct type is. I am not so interested in general claims that
> cannot be used in practice to make decisions.

**There are three characteristics of speaker cable that are of interest.
They are:

* Resistance. The lower the better.
* Inductance. The lower the better.
* Capacitance. Largely unimportant, when using a competently designed
amplifier.

That leaves resistance (R) and inductance (L). The significance of the two
paramaters depends on the impedance of the speaker and the length of the
cable. 4 Metres is a modest cable run and SHOULD present few problems in any
system.

Resistance is governed by the amount of conductive material in the cable.
Indutance is much more complicated. Standard 'zip' type speaker cables
(basically any parallel pair of conductors) exhibits pretty much the highest
possible inductance. The wider the spacing of the cable conductors (see:
Naim speaker cable), the higher the inductance and less desirable the cable.
As the conductor spacing is made smaller, the inductance falls. Some
manufacturers use multiple conductors, closely bound together to reduce
inductance still further. Others (Goertz, et al) use flat conductors, in
intimate proximity to reduce inductance to extremely low levels. For my
part, I suggest the use of RG213/U coax cable, which exhibits quite low
resistance and very low inductance, at reasonable cost.


--
Trevor Wilson
www.rageaudio.com.au

[Audio Empire]

On Sun, 17 Oct 2010 16:08:37 -0700, Trevor Wilson wrote
(in article >):

> "August Karlstrom" > wrote in message
> ...
>> On 2010-10-17 03:17, Trevor Wilson wrote:
>>> "August > wrote in message
>>>> http://en.wikipedia.org/wiki/Speaker_wire#Wire_gauge
>>>
>>> **Like all such cliams, the Wiki article over-simplifies things. They
>>> ignore
>>> the very real effects of inductance, when using long cable runs and
>>> speakers
>>> which exhibit a falling impedance at HF (like ESLs). In those cases, the
>>> inductive effects of cable can become quite significant.
>>
>> To take my example again, let's say I need a pair of cables, each four
>> meters long. Is there a system in which a 2x1.5 mm2 (approx.15 AWG) OFC
>> wire would be insufficient?
>
> **Without knowing the impedance curve (20Hz ~ 20kHz) of your speakers, I
> cannot say. In all probability, yes, it would be OK.
>
>>
>> Several people in this newsgroup believe that all wire sounds the same
>> provided it is of the "correct type". What I'm trying to find out here is
>> what the correct type is. I am not so interested in general claims that
>> cannot be used in practice to make decisions.
>
> **There are three characteristics of speaker cable that are of interest.
> They are:
>
> * Resistance. The lower the better.
> * Inductance. The lower the better.
> * Capacitance. Largely unimportant, when using a competently designed
> amplifier.

I practice, none of these things have the slightest effect on the sound.
Resistence/foot is less than a 0.006 Ohm for 14 Ga. zip.

> That leaves resistance (R) and inductance (L). The significance of the two
> paramaters depends on the impedance of the speaker and the length of the
> cable. 4 Metres is a modest cable run and SHOULD present few problems in any
> system.

Capacitance is around 12pf/foot or about 144pf for 4 meters of 14 Ga zip
cord. It has been found that capacitive loads of up to 20,000 pf (.02 mfd)
have no audible or visible effect on audio signals. That means that 4 meters
of 14 Ga zip cord would need 100 times more capacitance/foot to have any
effect whatsoever on a an audio signal passing through it. Inductance is
likewise negligible.

One can run 80 ft of 14 Ga zip as speaker cable for 8 Ohm speakers before it
is recommended that one increase the wire size to 12 Ga.

IOW, as long as the zip cord is 14 Ga or larger and your runs are less than
80ft, don't sweat it. just buy what you like and cheap hardware store zip
cord is FINE.

[isw]

In article >,
Audio Empire > wrote:

> On Sun, 17 Oct 2010 06:54:25 -0700, August Karlstrom wrote
> (in article >):
>
> > On 2010-10-17 03:17, Trevor Wilson wrote:
> >> "August > wrote in message
> >>> http://en.wikipedia.org/wiki/Speaker_wire#Wire_gauge
> >>
> >> **Like all such cliams, the Wiki article over-simplifies things. They
> >> ignore
> >> the very real effects of inductance, when using long cable runs and
> >> speakers
> >> which exhibit a falling impedance at HF (like ESLs). In those cases, the
> >> inductive effects of cable can become quite significant.
> >
> > To take my example again, let's say I need a pair of cables, each four
> > meters long. Is there a system in which a 2x1.5 mm2 (approx.15 AWG) OFC
> > wire would be insufficient?
> >
> > Several people in this newsgroup believe that all wire sounds the same
> > provided it is of the "correct type". What I'm trying to find out here
> > is what the correct type is. I am not so interested in general claims
> > that cannot be used in practice to make decisions.
> >
> > Thanks for the input.
> >
> >
> > /August
>
> All wire sounds exactly the same. Copper zip cord, 14 Gauge or bigger should
> be sufficient for any 4-meter run for any domestic amplifier or speaker
> setup. If you have a really big amp (say 700 watts/channel or more), you
> might want to increase that size to something like the bulk cable used in
> industrial-sized extension cords (12 -10 Gauge). Forget everything else, it
> doesn't matter.

Why is the wire size used to feed the speakers so much more important
than the wire used *in* the speaker -- for crossovers and more
importantly, for the voice coil? Clearly, the *length* of the latter two
far exceeds the former.

Isaac

[Sebastian Kaliszewski]

Trevor Wilson wrote:
> "August Karlstrom" > wrote in message
> ...
>> On 2010-10-17 03:17, Trevor Wilson wrote:
>>> "August > wrote in message
>>>> http://en.wikipedia.org/wiki/Speaker_wire#Wire_gauge
>>> **Like all such cliams, the Wiki article over-simplifies things. They
>>> ignore
>>> the very real effects of inductance, when using long cable runs and
>>> speakers
>>> which exhibit a falling impedance at HF (like ESLs). In those cases, the
>>> inductive effects of cable can become quite significant.
>> To take my example again, let's say I need a pair of cables, each four
>> meters long. Is there a system in which a 2x1.5 mm2 (approx.15 AWG) OFC
>> wire would be insufficient?
>
> **Without knowing the impedance curve (20Hz ~ 20kHz) of your speakers, I
> cannot say. In all probability, yes, it would be OK.

4m of zip cable will do on any home-use speakres, Kappa 9 included.

>
>> Several people in this newsgroup believe that all wire sounds the same
>> provided it is of the "correct type". What I'm trying to find out here is
>> what the correct type is. I am not so interested in general claims that
>> cannot be used in practice to make decisions.
>
> **There are three characteristics of speaker cable that are of interest.
> They are:
>
> * Resistance. The lower the better.
> * Inductance. The lower the better.

At 4m cable length inductance is unimportant.

> * Capacitance. Largely unimportant, when using a competently designed
> amplifier.
>
> That leaves resistance (R) and inductance (L). The significance of the two
> paramaters depends on the impedance of the speaker and the length of the
> cable. 4 Metres is a modest cable run and SHOULD present few problems in any
> system.

I would say no problems in any home system.

> Resistance is governed by the amount of conductive material in the cable.
> Indutance is much more complicated. Standard 'zip' type speaker cables
> (basically any parallel pair of conductors) exhibits pretty much the highest
> possible inductance. The wider the spacing of the cable conductors (see:
> Naim speaker cable), the higher the inductance and less desirable the cable.
> As the conductor spacing is made smaller, the inductance falls. Some
> manufacturers use multiple conductors, closely bound together to reduce
> inductance still further. Others (Goertz, et al) use flat conductors, in
> intimate proximity to reduce inductance to extremely low levels. For my
> part, I suggest the use of RG213/U coax cable, which exhibits quite low
> resistance and very low inductance, at reasonable cost.

It's an overkill unless you use >20m cable together with extremely low
(broken by design?) impedance (<1.5ohm) speakers.


rgds
\SK
--
"Never underestimate the power of human stupidity" -- L. Lang
--
http://www.tajga.org -- (some photos from my travels)

[August Karlstrom]

On 2010-10-17 23:53, Audio Empire wrote:
> All wire sounds exactly the same. Copper zip cord, 14 Gauge or bigger should
> be sufficient for any 4-meter run for any domestic amplifier or speaker
> setup.

OK, so you mean that the resistance in a four meter wire of gauge 15 or
more may be too high. Do you base this claim on some resistance
percentage similar to the Wikipedia article where 5% is mentioned?

According to the Wikipedia article I could even use a wire as thin as 20
AWG for a four meter run. On the other hand Arny Krueger suspects that
the 5% value is a recommendation for live sound rather than for critical
listening in a domestic environment. So what is the maximum resistance
percentage for critical listening?


/August

[Audio Empire]

On Mon, 18 Oct 2010 07:12:18 -0700, isw wrote
(in article >):

> In article >,
> Audio Empire > wrote:
>
>> On Sun, 17 Oct 2010 06:54:25 -0700, August Karlstrom wrote
>> (in article >):
>>
>>> On 2010-10-17 03:17, Trevor Wilson wrote:
>>>> "August > wrote in message
>>>>> http://en.wikipedia.org/wiki/Speaker_wire#Wire_gauge
>>>>
>>>> **Like all such cliams, the Wiki article over-simplifies things. They
>>>> ignore
>>>> the very real effects of inductance, when using long cable runs and
>>>> speakers
>>>> which exhibit a falling impedance at HF (like ESLs). In those cases, the
>>>> inductive effects of cable can become quite significant.
>>>
>>> To take my example again, let's say I need a pair of cables, each four
>>> meters long. Is there a system in which a 2x1.5 mm2 (approx.15 AWG) OFC
>>> wire would be insufficient?
>>>
>>> Several people in this newsgroup believe that all wire sounds the same
>>> provided it is of the "correct type". What I'm trying to find out here
>>> is what the correct type is. I am not so interested in general claims
>>> that cannot be used in practice to make decisions.
>>>
>>> Thanks for the input.
>>>
>>>
>>> /August
>>
>> All wire sounds exactly the same. Copper zip cord, 14 Gauge or bigger
>> should
>> be sufficient for any 4-meter run for any domestic amplifier or speaker
>> setup. If you have a really big amp (say 700 watts/channel or more), you
>> might want to increase that size to something like the bulk cable used in
>> industrial-sized extension cords (12 -10 Gauge). Forget everything else, it
>> doesn't matter.
>
> Why is the wire size used to feed the speakers so much more important
> than the wire used *in* the speaker -- for crossovers and more
> importantly, for the voice coil? Clearly, the *length* of the latter two
> far exceeds the former.
>
> Isaac
>

It isn't. In any case, "more than sufficient for the task" is just being
pretentious.

[Audio Empire]

On Mon, 18 Oct 2010 07:47:42 -0700, August Karlstrom wrote
(in article >):

> On 2010-10-17 23:53, Audio Empire wrote:
>> All wire sounds exactly the same. Copper zip cord, 14 Gauge or bigger should
>> be sufficient for any 4-meter run for any domestic amplifier or speaker
>> setup.
>
> OK, so you mean that the resistance in a four meter wire of gauge 15 or
> more may be too high. Do you base this claim on some resistance
> percentage similar to the Wikipedia article where 5% is mentioned?

I base it on the fact that 14-Ga Zip cord is less than 0.006 Ohm/foot. A
meter is a hair over three feet (3.37 inches over to be exact). You figure it
out.
>
> According to the Wikipedia article I could even use a wire as thin as 20
> AWG for a four meter run. On the other hand Arny Krueger suspects that
> the 5% value is a recommendation for live sound rather than for critical
> listening in a domestic environment. So what is the maximum resistance
> percentage for critical listening?

Just buy 14-Ga zip cord and forget about it. Your are counting angels on the
heads of pins here.

[August Karlstrom]

On 2010-10-18 21:05, Audio Empire wrote:
> On Mon, 18 Oct 2010 07:47:42 -0700, August Karlstrom wrote
>> According to the Wikipedia article I could even use a wire as thin as 20
>> AWG for a four meter run. On the other hand Arny Krueger suspects that
>> the 5% value is a recommendation for live sound rather than for critical
>> listening in a domestic environment. So what is the maximum resistance
>> percentage for critical listening?
>
> Just buy 14-Ga zip cord and forget about it. Your are counting angels on the
> heads of pins here.

The question remains: In the case of a four meter cord, why is 14 AWG
better than 15 AWG? What's magic about 14 in this case? It seems to me
that you base your arguments on things you have heard but not really
understood. If you don't know, pleas say so.


Thanks,

August

[Arny Krueger]

"August Karlstrom" > wrote in message

> On 2010-10-18 21:05, Audio Empire wrote:
>> On Mon, 18 Oct 2010 07:47:42 -0700, August Karlstrom
>> wrote
>>> According to the Wikipedia article I could even use a
>>> wire as thin as 20 AWG for a four meter run. On the
>>> other hand Arny Krueger suspects that the 5% value is a
>>> recommendation for live sound rather than for critical
>>> listening in a domestic environment. So what is the
>>> maximum resistance percentage for critical listening?
>>
>> Just buy 14-Ga zip cord and forget about it. Your are
>> counting angels on the heads of pins here.
>
> The question remains: In the case of a four meter cord,
> why is 14 AWG better than 15 AWG? What's magic about 14
> in this case? It seems to me that you base your arguments
> on things you have heard but not really understood. If
> you don't know, pleas say so.

Wire gauges are numbered in reverse by size. So, 14 gauge wire has more
copper per foot and higher conductivity than 15 gauge wire.

This is because wire was initially numbered by the number of times it was
drawn through a tapered sizing die. A smaller die was used each time. of
course.

14 gauge copper wire has the same conductivity as 15 gauge silver wire,
since silver is slightly more conductive than copper.

[Trevor Wilson[_3_]]

August Karlstrom wrote:
> On 2010-10-18 21:05, Audio Empire wrote:
>> On Mon, 18 Oct 2010 07:47:42 -0700, August Karlstrom wrote
>>> According to the Wikipedia article I could even use a wire as thin
>>> as 20 AWG for a four meter run. On the other hand Arny Krueger
>>> suspects that the 5% value is a recommendation for live sound
>>> rather than for critical listening in a domestic environment. So
>>> what is the maximum resistance percentage for critical listening?
>>
>> Just buy 14-Ga zip cord and forget about it. Your are counting
>> angels on the heads of pins here.
>
> The question remains: In the case of a four meter cord, why is 14 AWG
> better than 15 AWG? What's magic about 14 in this case? It seems to me
> that you base your arguments on things you have heard but not really
> understood. If you don't know, pleas say so.

**I have no idea what 14 AWG or 15 AWG is. I wish Americans would learn SI
units when discussing units of measurement. That said, the amount of copper
in a cable (as typified by it's size) is far less important than the
construction of the cable. The amount of copper affects the resistance of
the cable and pretty much nothing else. The cable construction affects
inductance and capacitance. Depending on the cable length and the speaker
impedance, inductance may be quite relevant. Capacitance is irrelevant to
any properly designed amplifier and may be disregarded.

Bottom line: Depending on what the impedance of your speakers is, the
difference between a 4 Metre run of 14 AWG or 15 AWG (whatever that is)
will be essentially zero.


--
Trevor Wilson
www.rageaudio.com.au

[Arny Krueger]

"August Karlstrom" > wrote in message

> On 2010-10-17 23:53, Audio Empire wrote:
>> All wire sounds exactly the same. Copper zip cord, 14
>> Gauge or bigger should be sufficient for any 4-meter run
>> for any domestic amplifier or speaker setup.
>
> OK, so you mean that the resistance in a four meter wire
> of gauge 15 or more may be too high. Do you base this
> claim on some resistance percentage similar to the
> Wikipedia article where 5% is mentioned?
>
> According to the Wikipedia article I could even use a
> wire as thin as 20 AWG for a four meter run. On the other
> hand Arny Krueger suspects that the 5% value is a
> recommendation for live sound rather than for critical
> listening in a domestic environment. So what is the
> maximum resistance percentage for critical listening?

The maximum resistance percentage for critical listening is between 1 and 2%
of the lowest impedance of the speaker. That ensures that the effect of the
wire is less than from 0.1 to 0.2 dB, and thus inaudible in all cases.

[Audio Empire]

On Mon, 18 Oct 2010 12:53:03 -0700, August Karlstrom wrote
(in article >):

> On 2010-10-18 21:05, Audio Empire wrote:
>> On Mon, 18 Oct 2010 07:47:42 -0700, August Karlstrom wrote
>>> According to the Wikipedia article I could even use a wire as thin as 20
>>> AWG for a four meter run. On the other hand Arny Krueger suspects that
>>> the 5% value is a recommendation for live sound rather than for critical
>>> listening in a domestic environment. So what is the maximum resistance
>>> percentage for critical listening?
>>
>> Just buy 14-Ga zip cord and forget about it. Your are counting angels on the
>> heads of pins here.
>
> The question remains: In the case of a four meter cord, why is 14 AWG
> better than 15 AWG?

It's probably not. But it is easier to find. Heck, at 4-meters 16-Ga is
probably fine.

> What's magic about 14 in this case?

Easy to find. Most hardware stores in the USA carry it in bulk and cheap. You
can even generally choose white, brown or black.

> It seems to me
> that you base your arguments on things you have heard but not really
> understood. If you don't know, pleas say so.

What I do know is that you're making a mountain out of a molehill. This is
NOT critical at all, and certainly doesn't warrant all the discussion it's
generated. BTW, I have a BSEE and worked in the Cable Laboratory at Lockheed
Missile and Space Company for three years during which time I tested and
measured every kind of wire you can imagine and under every imaginable
condition including vacuum and simulated high gravity at frequencies pretty
much from DC to daylight. Cable I know.

[Trevor Wilson[_3_]]

Audio Empire wrote:
> On Sun, 17 Oct 2010 16:08:37 -0700, Trevor Wilson wrote
> (in article >):
>
>> "August Karlstrom" > wrote in message
>> ...
>>> On 2010-10-17 03:17, Trevor Wilson wrote:
>>>> "August > wrote in message
>>>>> http://en.wikipedia.org/wiki/Speaker_wire#Wire_gauge
>>>>
>>>> **Like all such cliams, the Wiki article over-simplifies things.
>>>> They ignore
>>>> the very real effects of inductance, when using long cable runs and
>>>> speakers
>>>> which exhibit a falling impedance at HF (like ESLs). In those
>>>> cases, the inductive effects of cable can become quite significant.
>>>
>>> To take my example again, let's say I need a pair of cables, each
>>> four meters long. Is there a system in which a 2x1.5 mm2 (approx.15
>>> AWG) OFC wire would be insufficient?
>>
>> **Without knowing the impedance curve (20Hz ~ 20kHz) of your
>> speakers, I cannot say. In all probability, yes, it would be OK.
>>
>>>
>>> Several people in this newsgroup believe that all wire sounds the
>>> same provided it is of the "correct type". What I'm trying to find
>>> out here is what the correct type is. I am not so interested in
>>> general claims that cannot be used in practice to make decisions.
>>
>> **There are three characteristics of speaker cable that are of
>> interest. They are:
>>
>> * Resistance. The lower the better.
>> * Inductance. The lower the better.
>> * Capacitance. Largely unimportant, when using a competently designed
>> amplifier.
>
> I practice, none of these things have the slightest effect on the
> sound. Resistence/foot is less than a 0.006 Ohm for 14 Ga. zip.

**Incorrect. Both resistance and inductance can affect sound quality,
depending on the length of the cable and the impedance of the speaker.
Capacitance, as I have already stated many times, is irrelevant with any
properly designed amplifier.

>
>> That leaves resistance (R) and inductance (L). The significance of
>> the two paramaters depends on the impedance of the speaker and the
>> length of the cable. 4 Metres is a modest cable run and SHOULD
>> present few problems in any system.
>
> Capacitance is around 12pf/foot or about 144pf for 4 meters of 14 Ga
> zip cord. It has been found that capacitive loads of up to 20,000 pf
> (.02 mfd) have no audible or visible effect on audio signals. That
> means that 4 meters of 14 Ga zip cord would need 100 times more
> capacitance/foot to have any effect whatsoever on a an audio signal
> passing through it.

**As I carefully pointed out, capacitance is irrelevant.

> Inductance is likewise negligible.

**No. Inductance is not negligible. It *may* be irrelevant, depending on the
length of the cable and the speakers used, however.
>
> One can run 80 ft of 14 Ga zip as speaker cable for 8 Ohm speakers
> before it is recommended that one increase the wire size to 12 Ga.

**Just a reminder: There is no such thing as an "8 Ohm speaker". All
speakers, with the possible exception of Maggies, exhibit varying
impedances. SOME speakers, like ESLs, which may present a relatively high
impedance over most of the audible range, can often present very low
impedances at high frequencies. Here is one such example:

http://www.rageaudio.com.au/index.php?p=1_12

Click on 'Accustat'.

With a 25 Metre length of standard 'zip' type speaker cable, a serious and
very audible problem at HF can be expected. Low resistance and low
inductance cables are absolutely essential with such speakers and such long
cable runs.

>
> IOW, as long as the zip cord is 14 Ga or larger and your runs are
> less than 80ft, don't sweat it. just buy what you like and cheap
> hardware store zip cord is FINE.

**I have no issue with cheap, but misleading people by suggesting that they
can connect very long cables to speakers and expecting no problems is just
wrong.


--
Trevor Wilson
www.rageaudio.com.au

[isw]

In article >,
"Arny Krueger" > wrote:

> "August Karlstrom" > wrote in message
>
> > On 2010-10-17 23:53, Audio Empire wrote:
> >> All wire sounds exactly the same. Copper zip cord, 14
> >> Gauge or bigger should be sufficient for any 4-meter run
> >> for any domestic amplifier or speaker setup.
> >
> > OK, so you mean that the resistance in a four meter wire
> > of gauge 15 or more may be too high. Do you base this
> > claim on some resistance percentage similar to the
> > Wikipedia article where 5% is mentioned?
> >
> > According to the Wikipedia article I could even use a
> > wire as thin as 20 AWG for a four meter run. On the other
> > hand Arny Krueger suspects that the 5% value is a
> > recommendation for live sound rather than for critical
> > listening in a domestic environment. So what is the
> > maximum resistance percentage for critical listening?
>
> The maximum resistance percentage for critical listening is between 1 and 2%
> of the lowest impedance of the speaker. That ensures that the effect of the
> wire is less than from 0.1 to 0.2 dB, and thus inaudible in all cases.

If you're being *that* picky, you might want to take skin effect into
account, and use nothing larger than about #18 AWG, paralleling to get
the resistance down to what you think you need.

Isaac

[August Karlstrom]

On 2010-10-19 01:03, Dick Pierce wrote:
> In stores in the US, you can easily find 22 gauge, 20 gauge, 18
> gauge, 16 gauge, 14 gauge, 12 gauge, 10 gauge stranded insulated
> two-conductor wire, in other words even number AWG gauge sizes.

If you have followed the thread you should know that I refer to a 2x1.5
mm^2 wire which is very much buyable here in Sweden where we use SI units.


/August

[Audio Empire]

On Mon, 18 Oct 2010 16:03:41 -0700, Trevor Wilson wrote
(in article >):

> August Karlstrom wrote:
>> On 2010-10-18 21:05, Audio Empire wrote:
>>> On Mon, 18 Oct 2010 07:47:42 -0700, August Karlstrom wrote
>>>> According to the Wikipedia article I could even use a wire as thin
>>>> as 20 AWG for a four meter run. On the other hand Arny Krueger
>>>> suspects that the 5% value is a recommendation for live sound
>>>> rather than for critical listening in a domestic environment. So
>>>> what is the maximum resistance percentage for critical listening?
>>>
>>> Just buy 14-Ga zip cord and forget about it. Your are counting
>>> angels on the heads of pins here.
>>
>> The question remains: In the case of a four meter cord, why is 14 AWG
>> better than 15 AWG? What's magic about 14 in this case? It seems to me
>> that you base your arguments on things you have heard but not really
>> understood. If you don't know, pleas say so.
>
> **I have no idea what 14 AWG or 15 AWG is. I wish Americans would learn SI
> units when discussing units of measurement. That said, the amount of copper
> in a cable (as typified by it's size) is far less important than the
> construction of the cable. The amount of copper affects the resistance of
> the cable and pretty much nothing else. The cable construction affects
> inductance and capacitance. Depending on the cable length and the speaker
> impedance, inductance may be quite relevant. Capacitance is irrelevant to
> any properly designed amplifier and may be disregarded.
>
> Bottom line: Depending on what the impedance of your speakers is, the
> difference between a 4 Metre run of 14 AWG or 15 AWG (whatever that is)
> will be essentially zero.
>
>
>

Actually, the operational difference will be EXACTLY zero. Sorry about the
American Wire Gauge (AWG) problem. Not our fault and all that. Just remember
that the smaller the number, the bigger the conductor. This goes for solid as
well as stranded wire. The strands might all be the same size (or not) but as
the number gets smaller, there are generally more of the the thin ones or
fewer of thicker ones. Clear as mud? Same here.

[August Karlstrom]

On 2010-10-19 01:03, Trevor Wilson wrote:
> **I have no idea what 14 AWG or 15 AWG is. I wish Americans would learn SI
> units when discussing units of measurement.

I agree. I tried to convert my 2x1.5 mm^2 measurement to AWG in order to
make myself understood but then the Americans here say that you can't
buy a 15 AWG wire in the US. Jeez!

> That said, the amount of copper
> in a cable (as typified by it's size) is far less important than the
> construction of the cable. The amount of copper affects the resistance of
> the cable and pretty much nothing else. The cable construction affects
> inductance and capacitance. Depending on the cable length and the speaker
> impedance, inductance may be quite relevant. Capacitance is irrelevant to
> any properly designed amplifier and may be disregarded.

OK, so you do believe that there can be an audible difference between
different kinds of wire with the same resistance. Then we have a
discussion after all.

> Bottom line: Depending on what the impedance of your speakers is, the
> difference between a 4 Metre run of 14 AWG or 15 AWG (whatever that is)
> will be essentially zero.

For me the statement "depending on x the difference will be essentially
zero" does not make sense. Did you mean "idependent"?


/August

[August Karlstrom]

On 2010-10-19 01:55, Audio Empire wrote:
> On Mon, 18 Oct 2010 12:53:03 -0700, August Karlstrom wrote
>> The question remains: In the case of a four meter cord, why is 14 AWG
>> better than 15 AWG?
>
> It's probably not. But it is easier to find. Heck, at 4-meters 16-Ga is
> probably fine.

Only probably? ;-) Here in Sweden common wire thicknesses in mm^2 are
2x0.75, 2x1.5, 2x2.5, 2x4 and 2x6.

> What I do know is that you're making a mountain out of a molehill. This is
> NOT critical at all, and certainly doesn't warrant all the discussion it's
> generated. BTW, I have a BSEE and worked in the Cable Laboratory at Lockheed
> Missile and Space Company for three years during which time I tested and
> measured every kind of wire you can imagine and under every imaginable
> condition including vacuum and simulated high gravity at frequencies pretty
> much from DC to daylight. Cable I know.

That's great. Then we have real expertise here. What I'm/we're trying to
find out here is how thin a wire can be before it will make an audible
difference in the best possible system - a table like the one in the
Wikipedia article but for critical applications.


/August

[August Karlstrom]

On 2010-10-19 01:55, Arny Krueger wrote:
> The maximum resistance percentage for critical listening is between 1 and 2%
> of the lowest impedance of the speaker. That ensures that the effect of the
> wire is less than from 0.1 to 0.2 dB, and thus inaudible in all cases.

This is valuable information indeed.


Thanks,

August

[Arny Krueger]

"isw" > wrote in message

> In article >,
> "Arny Krueger" > wrote:

>> The maximum resistance percentage for critical listening
>> is between 1 and 2% of the lowest impedance of the
>> speaker. That ensures that the effect of the wire is
>> less than from 0.1 to 0.2 dB, and thus inaudible in all
>> cases.

> If you're being *that* picky, you might want to take skin
> effect into account, and use nothing larger than about
> #18 AWG, paralleling to get the resistance down to what
> you think you need.

Skin effect has fairly benign effects at audio frequencies and for wire
sizes commonly used in home audio.

Figure 3 at
http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/skineffect/page2.html
shows that the loss due to skin effect at 20 KHz is about 0.15 dB for 12
gauge (2 mm) wire and an 8 ohm speaker.

The previous page of the same document compares the skin effect of a bundle
of wires as compared to a solid wire with the same cross-section:

http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/skineffect/page1.html
says:

"...a bundle of small packed strands that are in electrical contact can be
regarded as similar to a solid but which has some air inclusions which mean
that overall wire cross section is only partly filled with conductor."

The reason for this is that skin effect is due to magnetic flux which flows
unimpeded by wire boundaries or even thin layers of insulation. If you want
to reduce skin effect, you do what cable companies and TV stations do, you
use conductors that are magnetically speaking, hollow tubes. This includes
both actual hollow tubes, silver plated large diameter aluminum wire or
tubing, or strands of wire wrapped around a foam plastic core. These are all
practical examples, some of which may be as near to your home as the nearest
cable company trunk line.

[August Karlstrom]

On 2010-10-15 20:38, August Karlstrom wrote:
> I have read more than one article where the recommended maximum length
> of loudspeaker wire is presented as a function of the wire gauge and
> the impedance of the speakers, e.g. in the Wikipedia article
>
> http://en.wikipedia.org/wiki/Speaker_wire#Wire_gauge
>
> I have also heard that output power and damping factor should be
> considered as well - a high power amplifier needs a thicker cable.
>
> Are there any studies made about the wire length contra output power and
> damping factor?

Here is my calculation of the minimum cross-sectional area of speaker
wire. Please tell me if you spot any errors.

***

The resistance R of a wire can be computed as

R = rho * l / A

where l is the length of the conductor in meters, A is the
cross-sectional area in square meters and rho is the electrical
resistivity in ohm meters.

Now, if we want the resistance of a wire of a specified length to be no
more than a percentage p of the lowest impedance of the speaker R_L we
have the relation

rho * l / A <= p * R_L

which is equivalent to

A >= rho * l / (p * R_L)

For copper wire the resistivity is 1.68E−8 ohm meters. If we choose p to
be one percent, which according to Arny Krueger ensures that the effect
of the wire is less than 0.1 dB and thus inaudible, we have the relation

A >= 1.68E−10 * l / R_L (1)


Example:

Let's say we need two runs of four meter speaker wire. The lowest
impedance of our speaker is 8 Ω. Which cross-sectional area should a
copper wire have for maximum performance? By applying formula (1) we
have that the total area should be at least 8.4E−07 m^2 so a 2x0.50 mm^2
wire should be enough.


/August

[August Karlstrom]

On 2010-10-19 17:23, August Karlstrom wrote:
> Here is my calculation of the minimum cross-sectional area of speaker
> wire. Please tell me if you spot any errors.
>
> ***

Unfortunately, some strange characters seem to have creped into the
text. Here is a corrected version:

***

The resistance R of a wire can be computed as

R = rho * l / A

where l is the length of the conductor in meters, A is the
cross-sectional area in square meters and rho is the electrical
resistivity in ohm meters.

Now, if we want the resistance of a wire of a specified length to be no
more than a percentage p of the lowest impedance of the speaker R_L we
have the relation

rho * l / A <= p * R_L

which is equivalent to

A >= rho * l / (p * R_L)

For copper wire the resistivity is 1.68E-8 ohm meters. If we choose p to
be one percent, which according to Arny Krueger ensures that the effect
of the wire is less than 0.1 dB and thus inaudible, we have the relation

A >= 1.68E-10 * l / R_L (1)


Example:

Let's say we need two runs of four meter speaker wire. The lowest
impedance of our speaker is 8 ohm. Which cross-sectional area should a
copper wire have for maximum performance? By applying formula (1) we
have that the total area should be at least 8.4E-07 m^2 so a 2x0.50 mm^2
wire should be enough.


/August

[Trevor Wilson[_3_]]

August Karlstrom wrote:
> On 2010-10-19 01:03, Trevor Wilson wrote:
>> **I have no idea what 14 AWG or 15 AWG is. I wish Americans would
>> learn SI units when discussing units of measurement.
>
> I agree. I tried to convert my 2x1.5 mm^2 measurement to AWG in order
> to make myself understood but then the Americans here say that you
> can't buy a 15 AWG wire in the US. Jeez!

**It is, indeed, unfortunate that America resists adopting the widespread
and logical measurement system that everyone else uses.

>
>> That said, the amount of copper
>> in a cable (as typified by it's size) is far less important than the
>> construction of the cable. The amount of copper affects the
>> resistance of the cable and pretty much nothing else. The cable
>> construction affects inductance and capacitance. Depending on the
>> cable length and the speaker impedance, inductance may be quite
>> relevant. Capacitance is irrelevant to any properly designed
>> amplifier and may be disregarded.
>
> OK, so you do believe that there can be an audible difference between
> different kinds of wire with the same resistance. Then we have a
> discussion after all.

**Indeed. This is what I have been banging on about. The amount of copper in
a speaker cable is a distraction (mostly). With standard, 'zip' type (two
parallel conductor, PVC insulated) cable, the only significant parameter
that is varied with different sized conductors is resistance. For the most
part, this does not significantly affect sound quality, over short cable
runs and when using relatively benign loads. HOWEVER, when using long cable
runs and/or 'difficult' speaker loads, both resistance and inductance may
play a significantly audible effect on sound. Inductance can be reduced by
using certain types of cable construction. My favourite is RG213/U, simply
because it is cheap, readily available and relatively easy to use.

>
>> Bottom line: Depending on what the impedance of your speakers is, the
>> difference between a 4 Metre run of 14 AWG or 15 AWG (whatever that
>> is) will be essentially zero.
>
> For me the statement "depending on x the difference will be
> essentially zero" does not make sense. Did you mean "idependent"?

**Nope. I meant that the difference (in audible terms) will be very, very
small. Moving from one size conductor elicits very small, possibly
insignificant differences in sound.


--
Trevor Wilson
www.rageaudio.com.au

[Audio Empire]

On Tue, 19 Oct 2010 06:47:43 -0700, isw wrote
(in article >):
>
> If you're being *that* picky, you might want to take skin effect into
> account, and use nothing larger than about #18 AWG, paralleling to get
> the resistance down to what you think you need.
>
> Isaac
>

"Skin effect" at audio frequencies? You're joking, right?

[August Karlstrom]

On 2010-10-19 19:33, Trevor Wilson wrote:
> August Karlstrom wrote:
>> OK, so you do believe that there can be an audible difference between
>> different kinds of wire with the same resistance. Then we have a
>> discussion after all.
>
> **Indeed. This is what I have been banging on about. The amount of copper in
> a speaker cable is a distraction (mostly). With standard, 'zip' type (two
> parallel conductor, PVC insulated) cable, the only significant parameter
> that is varied with different sized conductors is resistance. For the most
> part, this does not significantly affect sound quality, over short cable
> runs and when using relatively benign loads. HOWEVER, when using long cable
> runs and/or 'difficult' speaker loads, both resistance and inductance may
> play a significantly audible effect on sound. Inductance can be reduced by
> using certain types of cable construction. My favourite is RG213/U, simply
> because it is cheap, readily available and relatively easy to use.

The Wikipedia article says the following about capacitance and inductance:

"Speaker wire capacitance and inductance normally have no effect on
audio quality, though extreme examples using unusually low-impedance
speakers and exceptionally long wire runs can show a small effect."

It says "small effect" rather than "significantly audible effect". Maybe
it depends on the ears of the listener.


/August

[Audio Empire]

On Tue, 19 Oct 2010 06:48:24 -0700, August Karlstrom wrote
(in article >):

> On 2010-10-19 01:55, Audio Empire wrote:
>> On Mon, 18 Oct 2010 12:53:03 -0700, August Karlstrom wrote
>>> The question remains: In the case of a four meter cord, why is 14 AWG
>>> better than 15 AWG?
>>
>> It's probably not. But it is easier to find. Heck, at 4-meters 16-Ga is
>> probably fine.
>
> Only probably? ;-) Here in Sweden common wire thicknesses in mm^2 are
> 2x0.75, 2x1.5, 2x2.5, 2x4 and 2x6.
>
>> What I do know is that you're making a mountain out of a molehill. This is
>> NOT critical at all, and certainly doesn't warrant all the discussion it's
>> generated. BTW, I have a BSEE and worked in the Cable Laboratory at Lockheed
>> Missile and Space Company for three years during which time I tested and
>> measured every kind of wire you can imagine and under every imaginable
>> condition including vacuum and simulated high gravity at frequencies pretty
>> much from DC to daylight. Cable I know.
>
> That's great. Then we have real expertise here. What I'm/we're trying to
> find out here is how thin a wire can be before it will make an audible
> difference in the best possible system - a table like the one in the
> Wikipedia article but for critical applications.
>
>
> /August
>

I don't understand the purpose of your query, I guess. Bigger/less resistance
than necessary and required will not result in any improvement in sound
whatsoever. You could buy speaker cable as big as a baby's arm that cost
thousands of Euros per meter and it will not sound any different than common
lamp cord for a 4-meter run. The load impedance is pretty irrelevant, even in
highly reactive speaker loads such as those commonly found in electrostatic
panels.

I realize that in Europe, you use 220-240 volts for your mains, and therefore
your lamp cord PROBABLY tends to be smaller than ours (higher voltage = less
amps the cable has to carry for the same Watt light-bulb. Less current,
smaller wire). As has been pointed out, here in the USA, we still use SAE
measurements instead of DIN and outgrowths of those standards used in Europe
and SAE measurements for wire is listed in American Wire Gauge. Never having
had to deal with wire in European measurements, I really can't help you with
the conversion, but I can't believe that such conversion info (maybe even
tables) doesn't exist on the Internet somewhere.

Here's an idea. I suspect that they sell Monster Cable in Scandinavia, yes?
Then go take a look at some Monster Cable Clear Jacket Speaker Cable. It's
essentially 12-Gauge jacketed copper wire and sells here in the states for
less than about $6/meter in bulk. Either take a caliper with you and measure
it or buy about 8 meters for your two 4-meter runs. It is more than
sufficient for ANY speaker system likely to be found in any home stereo.

[Audio Empire]

On Tue, 19 Oct 2010 06:47:33 -0700, Trevor Wilson wrote
(in article >):

> Audio Empire wrote:
>> On Sun, 17 Oct 2010 16:08:37 -0700, Trevor Wilson wrote
>> (in article >):
>>
>>> "August Karlstrom" > wrote in message
>>> ...
>>>> On 2010-10-17 03:17, Trevor Wilson wrote:
>>>>> "August > wrote in message
>>>>>> http://en.wikipedia.org/wiki/Speaker_wire#Wire_gauge
>>>>>
>>>>> **Like all such cliams, the Wiki article over-simplifies things.
>>>>> They ignore
>>>>> the very real effects of inductance, when using long cable runs and
>>>>> speakers
>>>>> which exhibit a falling impedance at HF (like ESLs). In those
>>>>> cases, the inductive effects of cable can become quite significant.
>>>>
>>>> To take my example again, let's say I need a pair of cables, each
>>>> four meters long. Is there a system in which a 2x1.5 mm2 (approx.15
>>>> AWG) OFC wire would be insufficient?
>>>
>>> **Without knowing the impedance curve (20Hz ~ 20kHz) of your
>>> speakers, I cannot say. In all probability, yes, it would be OK.
>>>
>>>>
>>>> Several people in this newsgroup believe that all wire sounds the
>>>> same provided it is of the "correct type". What I'm trying to find
>>>> out here is what the correct type is. I am not so interested in
>>>> general claims that cannot be used in practice to make decisions.
>>>
>>> **There are three characteristics of speaker cable that are of
>>> interest. They are:
>>>
>>> * Resistance. The lower the better.
>>> * Inductance. The lower the better.
>>> * Capacitance. Largely unimportant, when using a competently designed
>>> amplifier.
>>
>> I practice, none of these things have the slightest effect on the
>> sound. Resistence/foot is less than a 0.006 Ohm for 14 Ga. zip.
>
> **Incorrect. Both resistance and inductance can affect sound quality,
> depending on the length of the cable and the impedance of the speaker.
> Capacitance, as I have already stated many times, is irrelevant with any
> properly designed amplifier.

No, not incorrect. You are obfuscating the issue with hypotheticals that are
irrelevant. Resistance and inductance of properly sized cable is
insignificant at normal lengths and don't become even slightly significant
'till you get past 15 meter runs., and even then, for a speaker that has an
impedance of 8 Ohms or less, I doubt seriously if one would hear any
difference.

>>
>>> That leaves resistance (R) and inductance (L). The significance of
>>> the two paramaters depends on the impedance of the speaker and the
>>> length of the cable. 4 Metres is a modest cable run and SHOULD
>>> present few problems in any system.
>>
>> Capacitance is around 12pf/foot or about 144pf for 4 meters of 14 Ga
>> zip cord. It has been found that capacitive loads of up to 20,000 pf
>> (.02 mfd) have no audible or visible effect on audio signals. That
>> means that 4 meters of 14 Ga zip cord would need 100 times more
>> capacitance/foot to have any effect whatsoever on a an audio signal
>> passing through it.
>
> **As I carefully pointed out, capacitance is irrelevant.
>
>> Inductance is likewise negligible.
>
> **No. Inductance is not negligible. It *may* be irrelevant, depending on the
> length of the cable and the speakers used, however.

It is negligible for all practical purposes. Were it not, don't you think
speaker cable would be prominently marked for its inductance/meter? It's not
generally, because a difference that makes no difference is no difference at
all. The OP is interested in two 4-meter runs. at that length inductance is
irrelevant and your saying that it is is merely confusing the poster who is
here trying to get some practical guidelines for buying speaker cable.

>> One can run 80 ft of 14 Ga zip as speaker cable for 8 Ohm speakers
>> before it is recommended that one increase the wire size to 12 Ga.
>
> **Just a reminder: There is no such thing as an "8 Ohm speaker". All
> speakers, with the possible exception of Maggies, exhibit varying
> impedances. SOME speakers, like ESLs, which may present a relatively high
> impedance over most of the audible range, can often present very low
> impedances at high frequencies. Here is one such example:
>
> http://www.rageaudio.com.au/index.php?p=1_12
>
> Click on 'Accustat'.
>
> With a 25 Metre length of standard 'zip' type speaker cable, a serious and
> very audible problem at HF can be expected. Low resistance and low
> inductance cables are absolutely essential with such speakers and such long
> cable runs.

I'm aware of that. But this pedanticism that you insist on bringing to this
discussion is not at all constructive.
>
>>
>> IOW, as long as the zip cord is 14 Ga or larger and your runs are
>> less than 80ft, don't sweat it. just buy what you like and cheap
>> hardware store zip cord is FINE.
>
> **I have no issue with cheap, but misleading people by suggesting that they
> can connect very long cables to speakers and expecting no problems is just
> wrong.

Published speaker cable tables say otherwise. The math says otherwise, common
sense says otherwise.

[Trevor Wilson[_3_]]

Audio Empire wrote:
> On Tue, 19 Oct 2010 06:47:33 -0700, Trevor Wilson wrote
> (in article >):
>
>> Audio Empire wrote:
>>> On Sun, 17 Oct 2010 16:08:37 -0700, Trevor Wilson wrote
>>> (in article >):
>>>
>>>> "August Karlstrom" > wrote in message
>>>> ...
>>>>> On 2010-10-17 03:17, Trevor Wilson wrote:
>>>>>> "August > wrote in message
>>>>>>> http://en.wikipedia.org/wiki/Speaker_wire#Wire_gauge
>>>>>>
>>>>>> **Like all such cliams, the Wiki article over-simplifies things.
>>>>>> They ignore
>>>>>> the very real effects of inductance, when using long cable runs
>>>>>> and speakers
>>>>>> which exhibit a falling impedance at HF (like ESLs). In those
>>>>>> cases, the inductive effects of cable can become quite
>>>>>> significant.
>>>>>
>>>>> To take my example again, let's say I need a pair of cables, each
>>>>> four meters long. Is there a system in which a 2x1.5 mm2
>>>>> (approx.15 AWG) OFC wire would be insufficient?
>>>>
>>>> **Without knowing the impedance curve (20Hz ~ 20kHz) of your
>>>> speakers, I cannot say. In all probability, yes, it would be OK.
>>>>
>>>>>
>>>>> Several people in this newsgroup believe that all wire sounds the
>>>>> same provided it is of the "correct type". What I'm trying to find
>>>>> out here is what the correct type is. I am not so interested in
>>>>> general claims that cannot be used in practice to make decisions.
>>>>
>>>> **There are three characteristics of speaker cable that are of
>>>> interest. They are:
>>>>
>>>> * Resistance. The lower the better.
>>>> * Inductance. The lower the better.
>>>> * Capacitance. Largely unimportant, when using a competently
>>>> designed amplifier.
>>>
>>> I practice, none of these things have the slightest effect on the
>>> sound. Resistence/foot is less than a 0.006 Ohm for 14 Ga. zip.
>>
>> **Incorrect. Both resistance and inductance can affect sound quality,
>> depending on the length of the cable and the impedance of the
>> speaker. Capacitance, as I have already stated many times, is
>> irrelevant with any properly designed amplifier.
>
> No, not incorrect.

**I am correct. I have very carefully maintained that capacitance plays no
part in the sound of a speaker cable, unless the amplifier is faulty. I have
also stated, very carefully, that inductance (of 'zip' cable) can play a
part under certain conditions.

You are obfuscating the issue with hypotheticals
> that are irrelevant.

**No. I am merely citing real-world examples, that I have actually
encountered and dealt with. I readily acknowledge that such instances are
not common, but they do exist and should not be ignored.

Resistance and inductance of properly sized
> cable is insignificant at normal lengths and don't become even
> slightly significant 'till you get past 15 meter runs.

**You recently cited an example of a 25 Metre (aka: 80 feet) cable run.

, and even
> then, for a speaker that has an impedance of 8 Ohms or less, I doubt
> seriously if one would hear any difference.

**Indeed. Many speakers do possess impedances of less than 8 Ohms.
Particularly at HF. ESLs are a prime example of such a speaker system.

>
>>>
>>>> That leaves resistance (R) and inductance (L). The significance of
>>>> the two paramaters depends on the impedance of the speaker and the
>>>> length of the cable. 4 Metres is a modest cable run and SHOULD
>>>> present few problems in any system.
>>>
>>> Capacitance is around 12pf/foot or about 144pf for 4 meters of 14 Ga
>>> zip cord. It has been found that capacitive loads of up to 20,000 pf
>>> (.02 mfd) have no audible or visible effect on audio signals. That
>>> means that 4 meters of 14 Ga zip cord would need 100 times more
>>> capacitance/foot to have any effect whatsoever on a an audio signal
>>> passing through it.
>>
>> **As I carefully pointed out, capacitance is irrelevant.
>>
>>> Inductance is likewise negligible.
>>
>> **No. Inductance is not negligible. It *may* be irrelevant,
>> depending on the length of the cable and the speakers used, however.
>
> It is negligible for all practical purposes.

**It *may* be negligible for most instances. That is not "all".

Were it not, don't you
> think speaker cable would be prominently marked for its
> inductance/meter?

**No. For several reasons:

1) Very few people understand the significance of such things. You,
apparently do not, despite my patient explanations.
2) The vast majority of speaker cables possess very similar levels of
inductance/Metre. There is no advantage for a manufacturer to cite data
which is similar to the competition. In marketing, manufacturers usually
highlight figures which allow them to stand out from the competition. Since
the vast majority of speaker cables possess the highest inductance possible,
there is no value in mentioning it.

It's not generally, because a difference that makes
> no difference is no difference at all. The OP is interested in two
> 4-meter runs. at that length inductance is irrelevant and your saying
> that it is is merely confusing the poster who is here trying to get
> some practical guidelines for buying speaker cable.

**Neither you nor I know if the inductance is relevant in teh system, since
we do not know the impedance characteristics of the speaker system. I
readily admit that it is highly likely that, given a 4 Metre cable run,
almost any cable will work, it is appropriate to point out that speaker
impedance should be determined before a recommendation can be given.


>
>>> One can run 80 ft of 14 Ga zip as speaker cable for 8 Ohm speakers
>>> before it is recommended that one increase the wire size to 12 Ga.
>>
>> **Just a reminder: There is no such thing as an "8 Ohm speaker". All
>> speakers, with the possible exception of Maggies, exhibit varying
>> impedances. SOME speakers, like ESLs, which may present a relatively
>> high impedance over most of the audible range, can often present
>> very low impedances at high frequencies. Here is one such example:
>>
>> http://www.rageaudio.com.au/index.php?p=1_12
>>
>> Click on 'Accustat'.
>>
>> With a 25 Metre length of standard 'zip' type speaker cable, a
>> serious and very audible problem at HF can be expected. Low
>> resistance and low inductance cables are absolutely essential with
>> such speakers and such long cable runs.
>
> I'm aware of that. But this pedanticism that you insist on bringing
> to this discussion is not at all constructive.

**I merely deal with reality. I've dealt with difficult installations many
times, where low inductance speaker cable was extremely helpful. I could
take the other tack and suggest that your own faulty tests (using an almost
resistive speaker load) is extremely unhelpful in uncovering the truth.

>>
>>>
>>> IOW, as long as the zip cord is 14 Ga or larger and your runs are
>>> less than 80ft, don't sweat it. just buy what you like and cheap
>>> hardware store zip cord is FINE.
>>
>> **I have no issue with cheap, but misleading people by suggesting
>> that they can connect very long cables to speakers and expecting no
>> problems is just wrong.
>
> Published speaker cable tables say otherwise. The math says
> otherwise, common sense says otherwise.

**Let's deal with the math then. Using your own, 25 Metre cable run and
using these speakers:

http://www.rageaudio.com.au/index.php?p=1_12


--
Trevor Wilson
www.rageaudio.com.au

[Trevor Wilson[_3_]]

August Karlstrom wrote:
> On 2010-10-19 19:33, Trevor Wilson wrote:
>> August Karlstrom wrote:
>>> OK, so you do believe that there can be an audible difference
>>> between different kinds of wire with the same resistance. Then we
>>> have a discussion after all.
>>
>> **Indeed. This is what I have been banging on about. The amount of
>> copper in a speaker cable is a distraction (mostly). With standard,
>> 'zip' type (two parallel conductor, PVC insulated) cable, the only
>> significant parameter that is varied with different sized conductors
>> is resistance. For the most part, this does not significantly affect
>> sound quality, over short cable runs and when using relatively
>> benign loads. HOWEVER, when using long cable runs and/or 'difficult'
>> speaker loads, both resistance and inductance may play a
>> significantly audible effect on sound. Inductance can be reduced by
>> using certain types of cable construction. My favourite is RG213/U,
>> simply because it is cheap, readily available and relatively easy to
>> use.
>
> The Wikipedia article says the following about capacitance and
> inductance:
>
> "Speaker wire capacitance and inductance normally have no effect on
> audio quality, though extreme examples using unusually low-impedance
> speakers and exceptionally long wire runs can show a small effect."
>
> It says "small effect" rather than "significantly audible effect".
> Maybe it depends on the ears of the listener.

**As I have stated, ad nauseum, it depends on the length of the cable and
the impedance characteristics of the speakers. And, of course, the user's
listening abilities.


--
Trevor Wilson
www.rageaudio.com.au

[Trevor Wilson[_3_]]

Audio Empire wrote:
> On Tue, 19 Oct 2010 06:47:43 -0700, isw wrote
> (in article >):
>>
>> If you're being *that* picky, you might want to take skin effect into
>> account, and use nothing larger than about #18 AWG, paralleling to
>> get the resistance down to what you think you need.
>>
>> Isaac
>>
>
> "Skin effect" at audio frequencies? You're joking, right?

**The skin effect occurs at all frequencies above DC. It is one reason why
power utilities are moving to DC transmission of power. That said, for all
practical purposes, in the vast majority of normal sound systems, skin
effect plays no part. BTW: Skin depth at 20kHz is approximately 0.5mm.


--
Trevor Wilson
www.rageaudio.com.au

[Audio Empire]

On Tue, 19 Oct 2010 12:37:35 -0700, August Karlstrom wrote
(in article >):

> On 2010-10-19 19:33, Trevor Wilson wrote:
>> August Karlstrom wrote:
>>> OK, so you do believe that there can be an audible difference between
>>> different kinds of wire with the same resistance. Then we have a
>>> discussion after all.
>>
>> **Indeed. This is what I have been banging on about. The amount of copper in
>> a speaker cable is a distraction (mostly). With standard, 'zip' type (two
>> parallel conductor, PVC insulated) cable, the only significant parameter
>> that is varied with different sized conductors is resistance. For the most
>> part, this does not significantly affect sound quality, over short cable
>> runs and when using relatively benign loads. HOWEVER, when using long cable
>> runs and/or 'difficult' speaker loads, both resistance and inductance may
>> play a significantly audible effect on sound. Inductance can be reduced by
>> using certain types of cable construction. My favourite is RG213/U, simply
>> because it is cheap, readily available and relatively easy to use.
>
> The Wikipedia article says the following about capacitance and inductance:
>
> "Speaker wire capacitance and inductance normally have no effect on
> audio quality, though extreme examples using unusually low-impedance
> speakers and exceptionally long wire runs can show a small effect."
>
> It says "small effect" rather than "significantly audible effect". Maybe
> it depends on the ears of the listener.
>
>
> /August
>

If you work the math. you'll find that both inductance and capacitance are
vanishingly low for adequate sized speaker wire and for lengths shorter than
15 meters (around 50 ft). By adequately sized, I mean any zip cord or speaker
cable equalling or larger than AWG 14-Ga (roughly 30/0.25 metric or larger).
These this inductance and capacitance would have to be several orders of
magnitude greater than they are (for the total run) in order to have any even
slightly audible effect.

[August Karlstrom]

On 2010-10-20 01:06, Trevor Wilson wrote:
> **Neither you nor I know if the inductance is relevant in teh system, since
> we do not know the impedance characteristics of the speaker system. I
> readily admit that it is highly likely that, given a 4 Metre cable run,
> almost any cable will work, it is appropriate to point out that speaker
> impedance should be determined before a recommendation can be given.

In my case I use a pair of Amphion Helium II.

"its plot of impedance magnitude and electrical phase angle reveals the
speaker to be very easy to drive, the impedance remaining above 8 ohms
for almost the entire audioband" -- John Atkinson at Stereophile

http://www.stereophile.com/budgetcomponents/105amphion/index4.html


/August

[August Karlstrom]

On 2010-10-19 21:57, Audio Empire wrote:
> On Tue, 19 Oct 2010 06:48:24 -0700, August Karlstrom wrote
>> That's great. Then we have real expertise here. What I'm/we're trying to
>> find out here is how thin a wire can be before it will make an audible
>> difference in the best possible system - a table like the one in the
>> Wikipedia article but for critical applications.
>
> I don't understand the purpose of your query, I guess.
[...]

If I can use an even thinner and cheaper cable with the same excellent
result why shouldn't I choose that one? I find this optimization problem
theoretically interesting.


/August

[August Karlstrom]

On 2010-10-19 18:41, August Karlstrom wrote:
> On 2010-10-19 17:23, August Karlstrom wrote:
>> Here is my calculation of the minimum cross-sectional area of speaker
>> wire. Please tell me if you spot any errors.

Well, I spotted an error myself. In equation (1) the constant should of
course be 1.68E-06 rather than 1.68E-10. The rest of the calculation
should be correct however. Sorry about that.

Here is the corrected version:

***

The resistance R of a wire can be computed as

R = rho * l / A

where l is the length of the conductor in meters, A is the
cross-sectional area in square meters and rho is the electrical
resistivity in ohm meters.

Now, if we want the resistance of a wire of a specified length to be no
more than a percentage p of the lowest impedance of the speaker R_L we
have the relation

rho * l / A <= p * R_L

which is equivalent to

A >= rho * l / (p * R_L)

For copper wire the resistivity is 1.68E-8 ohm meters. If we choose p to
be one percent, which according to Arny Krueger ensures that the effect
of the wire is less than 0.1 dB and thus inaudible, we have the relation

A >= 1.68E-06 * l / R_L (1)


Example:

Let's say we need two runs of four meter speaker wire. The lowest
impedance of our speaker is 8 ohm. Which cross-sectional area should a
copper wire have for maximum performance? By applying formula (1) we
have that the total area should be at least 8.4E-07 m2 so a 2x0.50 mm2
wire should be enough.


/August

[Audio Empire]

On Tue, 19 Oct 2010 19:49:02 -0700, August Karlstrom wrote
(in article >):

> On 2010-10-19 21:57, Audio Empire wrote:
>> On Tue, 19 Oct 2010 06:48:24 -0700, August Karlstrom wrote
>>> That's great. Then we have real expertise here. What I'm/we're trying to
>>> find out here is how thin a wire can be before it will make an audible
>>> difference in the best possible system - a table like the one in the
>>> Wikipedia article but for critical applications.
>>
>> I don't understand the purpose of your query, I guess.
> [...]
>
> If I can use an even thinner and cheaper cable with the same excellent
> result why shouldn't I choose that one? I find this optimization problem
> theoretically interesting.
>
>
> /August

OK, as an intellectual exercise, maybe, but practically speaking, we're
talking pennies of price difference between thinner and thicker. I feel that
a little overkill is the lesser of two evils, I guess.

[isw]

In article >,
Audio Empire > wrote:

> On Tue, 19 Oct 2010 06:47:43 -0700, isw wrote
> (in article >):
> >
> > If you're being *that* picky, you might want to take skin effect into
> > account, and use nothing larger than about #18 AWG, paralleling to get
> > the resistance down to what you think you need.
> >
> > Isaac
> >
>
> "Skin effect" at audio frequencies? You're joking, right?

Do the numbers. The skin depth of "ordinary" copper wire is about equal
to the radius of #18 wire at 20 kHz. Using larger diameter wire will
give you greater resistive losses at high than at low frequencies; using
smaller will give you more-or-less equal losses over the audio range.

To be sure, the differences are totally inconsequential, but so are many
other things that some people insist on spending a lot of money on in
their audio systems. If you're going to be picking nits, you should not
ignore any...

Isaac

[Audio Empire]

On Tue, 19 Oct 2010 18:57:30 -0700, Trevor Wilson wrote
(in article >):

> Audio Empire wrote:
>> On Tue, 19 Oct 2010 06:47:43 -0700, isw wrote
>> (in article >):
>>>
>>> If you're being *that* picky, you might want to take skin effect into
>>> account, and use nothing larger than about #18 AWG, paralleling to
>>> get the resistance down to what you think you need.
>>>
>>> Isaac
>>>
>>
>> "Skin effect" at audio frequencies? You're joking, right?
>
> **The skin effect occurs at all frequencies above DC. It is one reason why
> power utilities are moving to DC transmission of power. That said, for all
> practical purposes, in the vast majority of normal sound systems, skin
> effect plays no part. BTW: Skin depth at 20kHz is approximately 0.5mm.
>
>
>

You remind me of a guy I used to know who was heavy into DIY audio. He used
to build all his own equipment, both tubed and solid-state. He would even buy
vintage pieces like Harman-Kardon Citation I preamps and Dynaco PAS-3s and
Leak and Quad and McIntosh and Marantz stuff and replace all the old paper
and ceramic caps with modern, polypropylene and Polystyrene units and he'd
replace all the old carbon composite resistors with modern metal film units.
All laudable tasks, indeed. But whether he was rebuilding an old classic or
building something from scratch, he insisted on using Mil-Spec ICs and
transistors and replacing 10% resistors and capacitors with precision, 1%
resistors. I and others tried to tell him that these precision parts were
overkill because nothing in these circuits was that critical. In fact, the
vintage stuff was only designed to "slide-rule" accuracy, and then rounded up
or back to the nearest standard component value. For instance, the maths
might give the designer a value of 45,500 Ohms for a certain resistor in the
circuit. Well, there is no standard 45,500 Ohm resistor, so the designer will
specify 47K because that's a standard value. Even the modern stuff, designed
with digital calculators and computers, uses parts rounded to the nearest
standard value. But he insisted on a 1% 47K Ohm resistor instead of a 10% 47K
Ohm resistor. It made no difference. 10% tolerance parts or 1% tolerance, it
all sounded the same. Now an RIAA preamp might have had a more accurate curve
using precision parts, but not precision to standard values, but rather
precision to the RIAA turnover and rolloff. This guy couldn't see it. I
guess using precision parts made him feel better about his creations, because
you certainly couldn't hear the difference, or even measure it.

If being hung-up on the minutia of speaker cable makes you feel better about
your system, then go for it, but when discussing these matters with people
less technically proficient than you are, you merely muddy the waters for
them. By adding ifs, thens, and buts that simply have little or no meaning to
the guy who just wants an adequate pair of 3-5 meter speaker cables for a
domestic (nominally) 8 -Ohm speaker system, you just confuse them.

[Sebastian Kaliszewski]

August Karlstrom wrote:
> On 2010-10-20 01:06, Trevor Wilson wrote:
>> **Neither you nor I know if the inductance is relevant in teh system, since
>> we do not know the impedance characteristics of the speaker system. I
>> readily admit that it is highly likely that, given a 4 Metre cable run,
>> almost any cable will work, it is appropriate to point out that speaker
>> impedance should be determined before a recommendation can be given.
>
> In my case I use a pair of Amphion Helium II.
>
> "its plot of impedance magnitude and electrical phase angle reveals the
> speaker to be very easy to drive, the impedance remaining above 8 ohms
> for almost the entire audioband" -- John Atkinson at Stereophile
>
> http://www.stereophile.com/budgetcomponents/105amphion/index4.html

This is 7ohm load overall minimum and nearly 8ohm at 20KHz. This is
benign load, any cable will do for significant (50m!) lengths.

rgds
\SK
--
"Never underestimate the power of human stupidity" -- L. Lang
--
http://www.tajga.org -- (some photos from my travels)

[Chuckster]

> "Never underestimate the power of human stupidity" -- L. Lang

I personally drive two pair of 'difficult' speakers: Carver Amazing
Silver MKll's and Infinity WTLC's, and have done a fair bit of
research. Given what's out there, I put my trust in one of the giants
of audio - McIntosh's Roger Russell. This can be found online, and I
consider it the 'bible' of speaker wire info - it has served myself
and my friends well:

http://www.roger-russell.com/wire/wire.htm

You simply can't go wrong with this information, and it's easy to
understand and apply.

Chuckster

[Trevor Wilson[_3_]]

"August Karlstrom" > wrote in message
...
> On 2010-10-20 01:06, Trevor Wilson wrote:
>> **Neither you nor I know if the inductance is relevant in teh system,
>> since
>> we do not know the impedance characteristics of the speaker system. I
>> readily admit that it is highly likely that, given a 4 Metre cable run,
>> almost any cable will work, it is appropriate to point out that speaker
>> impedance should be determined before a recommendation can be given.
>
> In my case I use a pair of Amphion Helium II.
>
> "its plot of impedance magnitude and electrical phase angle reveals the
> speaker to be very easy to drive, the impedance remaining above 8 ohms
> for almost the entire audioband" -- John Atkinson at Stereophile
>
> http://www.stereophile.com/budgetcomponents/105amphion/index4.html
>

**With such a benign load, you are not likely to experience any problems
with almost any sensible cable over a 4 Metre length.


--
Trevor Wilson
www.rageaudio.com.au

[Audio Empire]

On Wed, 20 Oct 2010 05:00:29 -0700, isw wrote
(in article >):

> In article >,
> Audio Empire > wrote:
>
>> On Tue, 19 Oct 2010 06:47:43 -0700, isw wrote
>> (in article >):
>>>
>>> If you're being *that* picky, you might want to take skin effect into
>>> account, and use nothing larger than about #18 AWG, paralleling to get
>>> the resistance down to what you think you need.
>>>
>>> Isaac
>>>
>>
>> "Skin effect" at audio frequencies? You're joking, right?
>
> Do the numbers. The skin depth of "ordinary" copper wire is about equal
> to the radius of #18 wire at 20 kHz. Using larger diameter wire will
> give you greater resistive losses at high than at low frequencies; using
> smaller will give you more-or-less equal losses over the audio range.
>
> To be sure, the differences are totally inconsequential, but so are many
> other things that some people insist on spending a lot of money on in
> their audio systems. If you're going to be picking nits, you should not
> ignore any...
>
> Isaac
>

"To be sure, the differences are totally inconsequential". BTW, I probably
(due to my cable lab experience) know more about skin effect than anyone
here, my incredulity was prompted by the very idea that someone would
actually mention skin effect in the context of a speaker wire discussion. You
might as well bring-up the Miller effect and it's influence on amplifier
performance. It too is totally irrelevant to the user.

[isw]

In article >,
Audio Empire > wrote:

> On Wed, 20 Oct 2010 05:00:29 -0700, isw wrote
> (in article >):
>
> > In article >,
> > Audio Empire > wrote:
> >
> >> On Tue, 19 Oct 2010 06:47:43 -0700, isw wrote
> >> (in article >):
> >>>
> >>> If you're being *that* picky, you might want to take skin effect into
> >>> account, and use nothing larger than about #18 AWG, paralleling to get
> >>> the resistance down to what you think you need.
> >>>
> >>> Isaac
> >>>
> >>
> >> "Skin effect" at audio frequencies? You're joking, right?
> >
> > Do the numbers. The skin depth of "ordinary" copper wire is about equal
> > to the radius of #18 wire at 20 kHz. Using larger diameter wire will
> > give you greater resistive losses at high than at low frequencies; using
> > smaller will give you more-or-less equal losses over the audio range.
> >
> > To be sure, the differences are totally inconsequential, but so are many
> > other things that some people insist on spending a lot of money on in
> > their audio systems. If you're going to be picking nits, you should not
> > ignore any...
> >
> > Isaac
> >
>
> "To be sure, the differences are totally inconsequential". BTW, I probably
> (due to my cable lab experience) know more about skin effect than anyone
> here, my incredulity was prompted by the very idea that someone would
> actually mention skin effect in the context of a speaker wire discussion. You
> might as well bring-up the Miller effect and it's influence on amplifier
> performance. It too is totally irrelevant to the user.

But not to the designer, while skin effect is inconsequential to both
camps *in the context of home audio systems*. As are a lot of other
characteristics that get thrown about in r.a.h-e.

Interestingly, the electrical power guys have to take skin effect into
account too, if their busbars get big enough; it begins to matter at
about 4 inches at 60 Hz, AFAIR.

Do you mean cable lab experience as in "CableLabs" in Boulder? I've had
some interaction with them...

Isaac

[August Karlstrom]

On 2010-10-20 22:15, Trevor Wilson wrote:
> "August > wrote in message
>> In my case I use a pair of Amphion Helium II.
>>
>> "its plot of impedance magnitude and electrical phase angle reveals the
>> speaker to be very easy to drive, the impedance remaining above 8 ohms
>> for almost the entire audioband" -- John Atkinson at Stereophile
>>
>> http://www.stereophile.com/budgetcomponents/105amphion/index4.html
>>
>
> **With such a benign load, you are not likely to experience any problems
> with almost any sensible cable over a 4 Metre length.

Yes, and this is most likely true even if I had a speaker with minimum
impedance of only four ohms - even a wire as thin as 2x0.75 mm^2 would do.


/August

[Sebastian Kaliszewski]

Trevor Wilson wrote:
> "August Karlstrom" > wrote in message
> ...
>> On 2010-10-20 01:06, Trevor Wilson wrote:
>>> **Neither you nor I know if the inductance is relevant in teh system,
>>> since
>>> we do not know the impedance characteristics of the speaker system. I
>>> readily admit that it is highly likely that, given a 4 Metre cable run,
>>> almost any cable will work, it is appropriate to point out that speaker
>>> impedance should be determined before a recommendation can be given.
>> In my case I use a pair of Amphion Helium II.
>>
>> "its plot of impedance magnitude and electrical phase angle reveals the
>> speaker to be very easy to drive, the impedance remaining above 8 ohms
>> for almost the entire audioband" -- John Atkinson at Stereophile
>>
>> http://www.stereophile.com/budgetcomponents/105amphion/index4.html
>>
>
> **With such a benign load, you are not likely to experience any problems
> with almost any sensible cable over a 4 Metre length.

You could substitute 40 for 4 and the statement will be true as well.

rgds
\SK
--
"Never underestimate the power of human stupidity" -- L. Lang
--
http://www.tajga.org -- (some photos from my travels)

[Audio Empire]

On Thu, 21 Oct 2010 07:18:13 -0700, isw wrote
(in article >):

> In article >,
> Audio Empire > wrote:
>
>> On Wed, 20 Oct 2010 05:00:29 -0700, isw wrote
>> (in article >):
>>
>>> In article >,
>>> Audio Empire > wrote:
>>>
>>>> On Tue, 19 Oct 2010 06:47:43 -0700, isw wrote
>>>> (in article >):
>>>>>
>>>>> If you're being *that* picky, you might want to take skin effect into
>>>>> account, and use nothing larger than about #18 AWG, paralleling to get
>>>>> the resistance down to what you think you need.
>>>>>
>>>>> Isaac
>>>>>
>>>>
>>>> "Skin effect" at audio frequencies? You're joking, right?
>>>
>>> Do the numbers. The skin depth of "ordinary" copper wire is about equal
>>> to the radius of #18 wire at 20 kHz. Using larger diameter wire will
>>> give you greater resistive losses at high than at low frequencies; using
>>> smaller will give you more-or-less equal losses over the audio range.
>>>
>>> To be sure, the differences are totally inconsequential, but so are many
>>> other things that some people insist on spending a lot of money on in
>>> their audio systems. If you're going to be picking nits, you should not
>>> ignore any...
>>>
>>> Isaac
>>>
>>
>> "To be sure, the differences are totally inconsequential". BTW, I probably
>> (due to my cable lab experience) know more about skin effect than anyone
>> here, my incredulity was prompted by the very idea that someone would
>> actually mention skin effect in the context of a speaker wire discussion.
>> You
>> might as well bring-up the Miller effect and it's influence on amplifier
>> performance. It too is totally irrelevant to the user.
>
> But not to the designer,

Sigh! Of course Miller effect is not irrelevant to a designer, that's why I
said "to the user". But it is something that is not only inconsequential to
an amplifier's end user (the amp either HAS the bandwidth required, or the
input capacitance is so high that it doesn't), it's also beyond his control.
The same with skin effect of speaker cable. It's not only inconsequential for
the user, but beyond his control as well. Must we be so pedantic?

> while skin effect is inconsequential to both
> camps *in the context of home audio systems*. As are a lot of other
> characteristics that get thrown about in r.a.h-e.

It's called obfuscation and whether done maliciously or just to "show off",
these irrelevancies don't help people like the OP who ask for help here at
all.

> Interestingly, the electrical power guys have to take skin effect into
> account too, if their busbars get big enough; it begins to matter at
> about 4 inches at 60 Hz, AFAIR.
>
> Do you mean cable lab experience as in "CableLabs" in Boulder? I've had
> some interaction with them...

No, I mean like the Cable Laboratory at Lockheed Missiles and Space Company,
Inc. Where I spent three years doing everything to wire and cable and
connectors that it is possible to do. We evaluated wire and cable types for
bandwidth, skin effect, insulation integrity, capacitance, inductance,
impedance at frequencies from DC to daylight. We also tested cable in a
vacuum, cable (and connectors) in a pure oxygen environment, Cable and
connectors under high acceleration, high vibration, high temperature, low
temperature etc. We tested connector contact resistance (where I learned
about Stabilant (Tweek)), connector mating frequency, and reliability, pin
contact area, hermetic integrity, etc., etc, etc., ad nauseum.