[zareliman]

Hi headfiers

I recently started wondering, why do people use measures related to electrical wires (cross section) as a measure for their audio cables ?

AWG and metric wire gauges are old standards used for electrical applications that audio equipment probably inherited since we use electric signals to send audio.
While on current you find single stranded wires, cable for "audio purposes" is always multi stranded. Even the most skeptic non-placebophile audio engineer will prefer a multistranded wire over a single stranded if they're the same gauge. So why?
The most feasible explanation I found is that, while electrons can use all the thickness of the wire, they prefer (see less resistance in their path) to travel on the surface of the wires. While a lamp probably won't notice the difference, audio equipment that's sensitive and stuff is supposed to benefit from having more PERIMETER. So what is the function of the perimeter in this case ? Is "audio conductivity" directly proportional to wire perimeter ?
 
At the same gauge, a multistranded wire will have much more wire perimeter. The conversion from diameter to perimeter is basic pitagorean math, you only need the total gauge and number of strands. And if the perimeter thing is true, you can even run the numbers to see what's the PERIMETER equivalent of a single 9 AWG in say 86 strands.

9 AWG = 2.91 mm diameter -> 9.14 mm perimeter.
9 AWG 86 strands = 6.63 mm2 / 86 strand = 0.077mm2
0.077 mm2 / pi = r2
0.025 mm2 = r2 -> 0.156 mm = r
2*pi*0.156 mm = 0.98 mm = perimeter / strand
or 84mm total perimeter.
Roughly 9.256 times the original perimeter.
 
What do you think ?
 

 

[pinnahertz]

  Hi headfiers

I recently started wondering, why do people use measures related to electrical wires (cross section) as a measure for their audio cables ?

AWG and metric wire gauges are old standards used for electrical applications that audio equipment probably inherited since we use electric signals to send audio.

Wire gauge is useful because it's been standardized, and many important formulae are based on standardized wire sizes. The intended use of something doesn't dictate a change in the standard measurement units....pretty much anywhere.  We measure speaker diameters with the same units as we measure the diameter of a manhole cover. 

While on current you find single stranded wires, cable for "audio purposes" is always multi stranded. Even the most skeptic non-placebophile audio engineer will prefer a multistranded wire over a single stranded if they're the same gauge. So why?  

Flexibility. Audio cables are usually out in the open, need to have smaller connectors attached, and get handled. Stranded conductors are more flexible. The more and smaller strands, the more flexible the conductor.

 

The most feasible explanation I found is that, while electrons can use all the thickness of the wire, they prefer (see less resistance in their path) to travel on the surface of the wires. While a lamp probably won't notice the difference, audio equipment that's sensitive and stuff is supposed to benefit from having more PERIMETER. So what is the function of the perimeter in this case ? Is "audio conductivity" directly proportional to wire perimeter ?

You need to study "skin effect".  It's impact on transmission is determined by frequency, conductor size, and load.  Short story: It's generally not an issue in audio cables, especially the smaller ones.  (see the second link)  
 
And we probably shouldn't anthropomorphize electrons, lamps, and audio equipment. 
 
https://en.wikipedia.org/wiki/Skin_effect
http://www.belden.com/blog/broadcastav/understanding-skin-effect-and-frequency.cfm


 

 

 

[spruce music]

  Hi headfiers

I recently started wondering, why do people use measures related to electrical wires (cross section) as a measure for their audio cables ?

AWG and metric wire gauges are old standards used for electrical applications that audio equipment probably inherited since we use electric signals to send audio.
While on current you find single stranded wires, cable for "audio purposes" is always multi stranded. Even the most skeptic non-placebophile audio engineer will prefer a multistranded wire over a single stranded if they're the same gauge. So why?
The most feasible explanation I found is that, while electrons can use all the thickness of the wire, they prefer (see less resistance in their path) to travel on the surface of the wires. While a lamp probably won't notice the difference, audio equipment that's sensitive and stuff is supposed to benefit from having more PERIMETER. So what is the function of the perimeter in this case ? Is "audio conductivity" directly proportional to wire perimeter ?
 
At the same gauge, a multistranded wire will have much more wire perimeter. The conversion from diameter to perimeter is basic pitagorean math, you only need the total gauge and number of strands. And if the perimeter thing is true, you can even run the numbers to see what's the PERIMETER equivalent of a single 9 AWG in say 86 strands.

9 AWG = 2.91 mm diameter -> 9.14 mm perimeter.
9 AWG 86 strands = 6.63 mm2 / 86 strand = 0.077mm2
0.077 mm2 / pi = r2
0.025 mm2 = r2 -> 0.156 mm = r
2*pi*0.156 mm = 0.98 mm = perimeter / strand
or 84mm total perimeter.
Roughly 9.256 times the original perimeter.
 
What do you think ?
 

You can answer your question by seeing if wires of the same gauge stranded have higher, lower or the same resistance as wires that are solid.  The answer is at audio frequencies no differences.  There is skin effect that might begin to become barely noticeable at the upper audio frequencies, but normally is a non-issue until you get well into the ultrasonic region. 

 

[zareliman]

 
You can answer your question by seeing if wires of the same gauge stranded have higher, lower or the same resistance as wires that are solid.  The answer is at audio frequencies no differences.  There is skin effect that might begin to become barely noticeable at the upper audio frequencies, but normally is a non-issue until you get well into the ultrasonic region. 

I looked into the skin effect.
It's around 3% at 20khz for 12 AWG wires. Smaller wires stop having a considerable skin effect.
This means around 13 AWG it's worth looking into multistranded and ditch the lamp wire.

 

[pinnahertz]

 
I looked into the skin effect.
It's around 3% at 20khz for 12 AWG wires. Smaller wires stop having a considerable skin effect.
This means around 13 AWG it's worth looking into multistranded and ditch the lamp wire.

That would be an incorrect analysis.  The 3% is an effective change in resistance at 20kHz.  The entire wire is part of a voltage divider (always) comprised of the driving device (amp) the wire, and the speaker (the load).  That 3% change is a change in an already ridiculously low resistance.  The net result, a voltage variance with frequency, is far below audibility.   However, a smaller wire, while not having nearly as much skin effect, has a much higher series resistance.  With a reactive load such as a speaker, the net frequency response change caused by that resistance would be much higher than larger conductors with skin effect.  Not by a little bit either.  Of course, all of this is wire-lenght dependant as well.
 
But honestly, who doesn't use stranded wire anyway?  13ga and bigger is very stiff and hard to work with, a bigger and more important reason than the skin effect.  Doesn't everyone use 12-14ga stranded wire for speakers? 
 
Why are we worried about this?

 

[Don Hills]

 
I looked into the skin effect.
It's around 3% at 20khz for 12 AWG wires. Smaller wires stop having a considerable skin effect.
This means around 13 AWG it's worth looking into multistranded and ditch the lamp wire.

 
Nope. You're forgetting a crucial point: In stranded wire, all the strands are in electrical contact with each other. At frequencies where skin effect is significant the bulk of the current flows in the outermost strands, just like it flows in the outermost part of a solid wire. To prevent this happening, you need to individually insulate each strand. Such wire is commonly called "Litz" wire, and has been used for many years in RF circuits. The effect is so slight at audio frequencies that no-one bothers to go to the trouble of using it, especially due to the hassle of cleaning the insulation off all the strand ends to make a good connection.

 

[zareliman]

   
Nope. You're forgetting a crucial point: In stranded wire, all the strands are in electrical contact with each other. At frequencies where skin effect is significant the bulk of the current flows in the outermost strands, just like it flows in the outermost part of a solid wire. To prevent this happening, you need to individually insulate each strand. Such wire is commonly called "Litz" wire, and has been used for many years in RF circuits. The effect is so slight at audio frequencies that no-one bothers to go to the trouble of using it, especially due to the hassle of cleaning the insulation off all the strand ends to make a good connection.

Plus, the cheapest "Litz" speaker wire is the Cardas 101, which is like 20 times more expensive than the old trusty 12 AWG Monoprice bulk cable.
If you know some good braids you can make your own " budget Litz" speaker cable with cheap 4x18AWG or something like that.
 
http://howdidyoumakethis.com/8-strand-flat-braid/