• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Technical Article: Does Audio Cable Skin Effect Matter

Audiofire

Addicted to Fun and Learning
Forum Donor
Joined
Feb 8, 2022
Messages
625
Likes
347
Location
Denmark
I know that the way this thread was revived is sort of like flogging a dead horse, yet it still helped me to iron out every last detail before I make my purchase of cables (and can potentially help others to do the same).

This kind of topic just makes me want to learn about electrical engineering:) I have also donated money to this great forum.
 
OP
DonH56

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,834
Likes
16,496
Location
Monument, CO
To tie it off, below is the phase shift of 10' of 12 AWG speaker cable (defined in the earlier post) using a simple circuit simulation (could have hand-calculated it easily enough but lazy, use arctan(1/(2*pi*f*R*C), but this way there are pictures :) ). As is typical, there is a difference in phase, but it is way (WAY) below audibility.

Phase shift of just the cable at 10 Hz = -17.28e-9 degrees (-17.89 x 10^-9 = -17.89 n-deg)
Phase shift of just the cable at 20 kHz = -49.89e-6 degrees (-49.89 x 10^-6 = -49.89 u-deg)

1648300942099.png


1648300584434.png
1648300773516.png
 
Last edited:

Ingenieur

Addicted to Fun and Learning
Joined
Apr 23, 2021
Messages
938
Likes
747
Location
PA
Also interesting, but not on really on-topic...

e
V and I coexist with E and H
No magnetic field H without I
No I without V therefore no E
You need all for power flow in a conductor or the energy in the surrounding fields.
Normally 99% of the energy is within 105% of the cable radius.

imo, nonsense and a lack of understanding on the YouTuber's part. Click bait.
 

Ingenieur

Addicted to Fun and Learning
Joined
Apr 23, 2021
Messages
938
Likes
747
Location
PA
As noted phase shift
= arctan(X/R)
X = net Xc and Xl
Xc = 1/(2 Pi f C) and Xl = 2 Pi f L
For a cable Xl // Xc
Using data from attached, per unit 1'
R always 0.00242 Ohm

At 20 Hz
Xc = 274 MOhm
Xl = 21.36 uOhm
Since in // and Xc>>>Xl, Cl prevails
Calculated net X = 21.36 uOhm
ie, Xc can be ignored
Phase = arctan (21.36e-6/0.00242)
= 0.5 deg

At 20,000 Hz
Xc = 274 kOhm
Xl = 0.0214 Ohm
net X = 0.0214 still lol
arctan(0.0214/0.00242 ) = 83.5 deg
This is why L is critical

Does it matter?
0.5 deg at 20 Hz = 0.07 mSec
83.5 deg at 20 kHz = 0.01 mSec
Moot
And the shift is along the length
10,000 Hz
77 deg
0.02 sec



C497628F-763F-4730-AF9D-6F4B19F62191.jpeg
 

Ingenieur

Addicted to Fun and Learning
Joined
Apr 23, 2021
Messages
938
Likes
747
Location
PA
Here's what counts:
Using my example
10,000 Hz, 10' cable
Z = 0.05 + 0.21j = 0.216/76 deg

We forgot about the speakers lol
Mine at 10 kHz
Z = 7.5/-35 deg = 6.15 - 4.3j
total Z = 6.20 - 4.09j = 7.42/-33.4 deg

The cable only changed the speaker phase:
-33.4 - (-35) = 1.6 deg
Moot compared to the total net
 
OP
DonH56

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,834
Likes
16,496
Location
Monument, CO
The only parameter I considered was skin effect, the topic of this thread, per what the questioner asked. I (intentionally) did not consider cable inductance (and yes I am aware it exists). In any event as you have shown it is irrelevant, not really "critical".

For marketing we could include eddy current, surface roughness, dielectric absorption, and myriad other effects only relevant at frequencies far above audio. Quantum traps, I forgot about quantum traps...
As noted phase shift
= arctan(X/R)
X = net Xc and Xl
Xc = 1/(2 Pi f C) and Xl = 2 Pi f L
For a cable Xl // Xc
Using data from attached, per unit 1'
R always 0.00242 Ohm

At 20 Hz
Xc = 274 MOhm
Xl = 21.36 uOhm
Since in // and Xc>>>Xl, Cl prevails
Calculated net X = 21.36 uOhm
ie, Xc can be ignored
Phase = arctan (21.36e-6/0.00242)
= 0.5 deg

At 20,000 Hz
Xc = 274 kOhm
Xl = 0.0214 Ohm
net X = 0.0214 still lol
arctan(0.0214/0.00242 ) = 83.5 deg
This is why L is critical

Does it matter?
0.5 deg at 20 Hz = 0.07 mSec
83.5 deg at 20 kHz = 0.01 mSec
Moot
And the shift is along the length
10,000 Hz
77 deg
0.02 sec



View attachment 195533
 
Last edited:

EB1000

Senior Member
Joined
Jan 8, 2020
Messages
484
Likes
579
Location
Israel
As an ex-power system engineer, both proximity and skin effect did matter at 50 and 60Hz, but of course, we dealt with isolated thick aluminum/steel conductors running hundreds of miles...
 

Ingenieur

Addicted to Fun and Learning
Joined
Apr 23, 2021
Messages
938
Likes
747
Location
PA
Here's something else to consider, another form of phase angle is power factor
Cos(ph ang) = pf
At 20 Hz pf = 1.00, in other words all power absorbed by the cable is active or real power in Watts. Basically no reactive power in VAr.

At 20 kHz 0.11 real, 0.99 reactive, meaning it is exchanging power with the load or source, but little real power, little volume attenuation.

Real or active power does work
Imaginary or reactive does not, it is electrical and magnetic charging power in crude terms.
 
OP
DonH56

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,834
Likes
16,496
Location
Monument, CO
As an ex-power system engineer, both proximity and skin effect did matter at 50 and 60Hz, but of course, we dealt with isolated thick aluminum/steel conductors running hundreds of miles...
Hopefully most of us do not have speaker cables quite that long. ;) Transmission line effects, normally ignored at that frequency, are also significant for long lines.

Note the skin effect gets worse (in square-root fashion) as conductance decreases; my equations are all for copper.
 

mhardy6647

Grand Contributor
Joined
Dec 12, 2019
Messages
11,211
Likes
24,170
For marketing we could include eddy current, surface roughness, dielectric absorption, and a myriad other effects only relevant at frequencies far above audio. Quantum traps, I forgot about quantum traps...
Isn't that how Schroedinger caught that wascawwy wabbit wascawwy cat?
 

Audiofire

Addicted to Fun and Learning
Forum Donor
Joined
Feb 8, 2022
Messages
625
Likes
347
Location
Denmark
Thought it would be good to share this, since it's relevant and someone could also be interested. There's a reputable source that it's actually okay to use litz wire. And another of the manufacturers that thinks skin effect is actually good. QED does have their cheapest speaker cables that is just zip cord, but then again they contradict their own technical documents by using silver-plating in RCA cables and the like?!

"Braided cables utilize multiple individually insulated conductors (usually 22-26AWG) twisted in a pattern that helps to reduce series mutual inductance but usually at the expense of increased capacitance. Remember anytime you bring conductors closer together, the inductance gets reduced but capacitance goes up. I do appreciate these type of cable designs if they aren't overly blown out of proportion with high capacitance needing a zobel network termination to prevent amplifier oscillation for long runs. Braided cables are a fine choice provided that you aren't sacrificing resistance for the design."

"The result of skin effect is a rise in cable impedance at very high frequencies, due to the shrinking effective conducting area increasing the total impedance. (Curiously, unlike inductance, skin effect does not introduce phase shift but does give rise to increased power loss in the cable).

"In radio-frequency applications (way above the audio range) skin effect can be a serious problem, overcome by silver plating to reduce resistance at the surface, where the bulk of the high-frequency current flows. In audio cables, the assumption that skin effect is worth tackling has resulted in loudspeaker cables with single strands of diameter equal to or less than twice the effective signal penetration depth (the depth at which current density is reduced to 63% of its normal value) at some high audio frequency. The idea being that the cable will operating at all frequencies in reduced- current-density mode. By doing so, the symptoms of skin effect are swamped (though it has not been beaten) while the overall impedance of the cable has been increased at all frequencies.

"Interestingly, measured values of phase shift are generally lower than the theoretical predictions because skin effect, which is resistive in nature, increases the AC impedance of cable without introducing phase shift of its own. Curiously, skin effect actually reduces phase shift by countering the inductive reactance of the cable. (Those who wish to take this further are referred to text books on AC phasor theory)."
 
Last edited:

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,383
Likes
24,749
Location
Alfred, NY
Thought it would be good to share this, since it's relevant and someone could also be interested. There's a reputable source that it's actually okay to use litz wire. And another of the manufacturers that thinks skin effect is actually good. QED does have their cheapest speaker cables that is just zip cord, but then again they contradict their own technical documents by using silver-plating in RCA cables and the like?!

"Braided cables utilize multiple individually insulated conductors (usually 22-26AWG) twisted in a pattern that helps to reduce series mutual inductance but usually at the expense of increased capacitance. Remember anytime you bring conductors closer together, the inductance gets reduced but capacitance goes up. I do appreciate these type of cable designs if they aren't overly blown out of proportion with high capacitance needing a zobel network termination to prevent amplifier oscillation for long runs. Braided cables are a fine choice provided that you aren't sacrificing resistance for the design."

"The result of skin effect is a rise in cable impedance at very high frequencies, due to the shrinking effective conducting area increasing the total impedance. (Curiously, unlike inductance, skin effect does not introduce phase shift but does give rise to increased power loss in the cable).

"In radio-frequency applications (way above the audio range) skin effect can be a serious problem, overcome by silver plating to reduce resistance at the surface, where the bulk of the high-frequency current flows. In audio cables, the assumption that skin effect is worth tackling has resulted in loudspeaker cables with single strands of diameter equal to or less than twice the effective signal penetration depth (the depth at which current density is reduced to 63% of its normal value) at some high audio frequency. The idea being that the cable will operating at all frequencies in reduced- current-density mode. By doing so, the symptoms of skin effect are swamped (though it has not been beaten) while the overall impedance of the cable has been increased at all frequencies.

"Interestingly, measured values of phase shift are generally lower than the theoretical predictions because skin effect, which is resistive in nature, increases the AC impedance of cable without introducing phase shift of its own. Curiously, skin effect actually reduces phase shift by countering the inductive reactance of the cable. (Those who wish to take this further are referred to text books on AC phasor theory)."
Run, don’t walk, away.
 

Urubamba

Member
Joined
Dec 9, 2021
Messages
91
Likes
41
These are NOT OFC cables, obviously. The wire that has oxidized is 40 years old, and it connects my tube amplifier to the main speakers. (2.5 meters each )
The "new" looking cable is three years old and links my SS amplifier to the subwoofers. ( 3 meters each and spliced and welded because I cut "in half" before measuring ) yes, the years do not come alone .... :facepalm:
I change the old cables for new ones? The new one also because it has been spliced?
Will the rust be superficial or will it be present on all internal stranded conductors?
I would like to know what you would do if you were in my place.
Thank you ! ;)
 

Attachments

  • IMG_20230310_172939.jpg
    IMG_20230310_172939.jpg
    412.6 KB · Views: 63

fpitas

Master Contributor
Forum Donor
Joined
Jul 7, 2022
Messages
9,885
Likes
14,191
Location
Northern Virginia, USA

AdamG

Proving your point makes it “Science”.
Moderator
Forum Donor
Joined
Jan 3, 2021
Messages
4,636
Likes
14,918
Location
Reality

fpitas

Master Contributor
Forum Donor
Joined
Jul 7, 2022
Messages
9,885
Likes
14,191
Location
Northern Virginia, USA
:facepalm: Insult to injury the Gold one (most expensive) is SOLD OUT!

Adds level of smoothness and organic sound not achieved in our other purifiers
Well, it's on sale. You won't find a better deal if you're looking for quantum purification.

/Whatever that is
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,383
Likes
24,749
Location
Alfred, NY
These are NOT OFC cables, obviously. The wire that has oxidized is 40 years old, and it connects my tube amplifier to the main speakers. (2.5 meters each )
The "new" looking cable is three years old and links my SS amplifier to the subwoofers. ( 3 meters each and spliced and welded because I cut "in half" before measuring ) yes, the years do not come alone .... :facepalm:
I change the old cables for new ones? The new one also because it has been spliced?
Will the rust be superficial or will it be present on all internal stranded conductors?
I would like to know what you would do if you were in my place.
Thank you ! ;)
Clear vinyl is a problematic insulator. It's probably OK, but just for peace of mind, get thee to Home Depot and get some 12x2 zip cord with opaque insulation.
 

Urubamba

Member
Joined
Dec 9, 2021
Messages
91
Likes
41
Clear vinyl is a problematic insulator. It's probably OK, but just for peace of mind, get thee to Home Depot and get some 12x2 zip cord with opaque insulation.


Why is PVC "a problematic insulator"? I never read anything about it, we are talking about very low voltages / currents, I don't understand why you say that. Home Depot is too far away for me, and what would be the advantage of those 12 AWG x 2 over my cables? , are thinner, 2,053 mm. :confused:

 
Top Bottom