Effects of Inductance and Skin-effect in a speaker wire.

[Tony_Montana]

As far as sonic effects of inductance and Skin effect is concern, there are alot of debate about this issue that will make your head to spin. But how about effect of above issues theoretically-such as phase shift and Group (frequency) Delay distortion?

Here is sample picture of an audio wave:

<img src="http://www.vcdhelp.com/forum/images/guides/117248/goldwave2.png">

As one can see, these type of signals have alot of [upper frequency] harmonics, accompanied with lower frequency. The question is since higher frequency(s) will be more subject to cable's Inductance* and Skin effects*, then wouldn't that cause the higher frequency to shift/attenuate more from its original place on the signal (as compared with lower frequency) causing distortion or signal deformation?

Shouldn't we look for cables [beyond zip cord] that will address these issues also such as twisted cables (inductance), and make cable insulated strands less than 20 gauge size (skin effect)?
Or should we say that since these effects are not audible, then it should be ignored?

*Inductance:cable inductance is frequency dependable. Its effect will be higher as frequency increases causing higher frequency to attenuate and shift more than lower frequency(s).
*Skin Effect:This is also frequency dependable. As frequency increases, it will occupy less [area] space in a conductor causing higher frequency to be more subject to cable's resistance than lower frequency are.

[mtrycraft]

As far as sonic effects of inductance and Skin effect is concern, there are alot of debate about this issue that will make your head to spin. But how about effect of above issues theoretically-such as phase shift and Group (frequency) Delay distortion?

Here is sample picture of an audio wave:

<img src="http://www.vcdhelp.com/forum/images/guides/117248/goldwave2.png">

As one can see, these type of signals have alot of [upper frequency] harmonics, accompanied with lower frequency. The question is since higher frequency(s) will be more subject to cable's Inductance* and Skin effects*, then wouldn't that cause the higher frequency to shift/attenuate more from its original place on the signal (as compared with lower frequency) causing distortion or signal deformation?

Shouldn't we look for cables [beyond zip cord] that will address these issues also such as twisted cables (inductance), and make cable insulated strands less than 20 gauge size (skin effect)?
Or should we say that since these effects are not audible, then it should be ignored?

*Inductance:cable inductance is frequency dependable. Its effect will be higher as frequency increases causing higher frequency to attenuate and shift more than lower frequency(s).
*Skin Effect:This is also frequency dependable. As frequency increases, it will occupy less [area] space in a conductor causing higher frequency to be more subject to cable's resistance than lower frequency are.


"Amplifier-Loudspeaker Interfacing", Greiner, R.A., JAES vol. 28, no. 5 May 80, page 310-315.


RA Greiner discusses the issue of group delay due to skin effects between 100Hz and 10kHz. I would not worry about a 1x 10 to -8 power sec.

Phase angle of 3 degrees at 20kHz if you can hear 20kHz is nothing.

"Effects of cable, Loudspeaker and Amplifier Interactions", Davis, Fred E., JAES, vol. 39, no. 6 Jun 91,

Davis calculated a .04dB loss due to skin effect at 20kHz.

Again, one must be able to detect 20kHz and second, one must have a Just noticable threshold this low on top of it.
But, JND has been measured at 16kHz to be about 3dB on average. This is of course much more than one would detect closer to the midbands where out acuity is much better.

Again, th egolden ears make much about nothing as they have too much time on hand to worry about nothing. That is not their fault mostly as the high end audio industry marketing has done a good job of voodoo magic and human gullibility has jumped on it. No different that anything else in consumer land.

Just reading an article on immunization for whooping cough over the past number of decades. Some raised alarms over the negative effects of immunization. Gullible people followed and the rate of whooping cough skyrocketed again.

[FLZapped]

"Amplifier-Loudspeaker Interfacing", Greiner, R.A., JAES vol. 28, no. 5 May 80, page 310-315.

RA Greiner discusses the issue of group delay due to skin effects between 100Hz and 10kHz. I would not worry about a 1x 10 to -8 power sec.

Phase angle of 3 degrees at 20kHz if you can hear 20kHz is nothing.

"Effects of cable, Loudspeaker and Amplifier Interactions", Davis, Fred E., JAES, vol. 39, no. 6 Jun 91,

Davis calculated a .04dB loss due to skin effect at 20kHz.


I calculated 0.03dB additional loss due to skin effect for a 10 foot run of 12 AWG cable at 20kHz.

As for phase shift, power amplifiers have more than the 3 degrees and no one ever says boo about that! (Although outside of pro amps, this is usually not a published spec)

Here is a refrence that talks about group delay in loudspeakers. Keep in mind that your speaker cable has a group delay in the hundreds of nano-seconds (10E-09)

http://www.trueaudio.com/post_010.htm

Here is a reference where the phase shift and group delay are talked about in wire and where the number I used above came from:

http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/cableshift/cp.html

-Bruce

[mtrycraft]

I calculated 0.03dB additional loss due to skin effect for a 10 foot run of 12 AWG cable at 20kHz.

As for phase shift, power amplifiers have more than the 3 degrees and no one ever says boo about that! (Although outside of pro amps, this is usually not a published spec)

Here is a refrence that talks about group delay in loudspeakers. Keep in mind that your speaker cable has a group delay in the hundreds of nano-seconds (10E-09)

http://www.trueaudio.com/post_010.htm

Here is a reference where the phase shift and group delay are talked about in wire and where the number I used above came from:

http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/cableshift/cp.html

-Bruce

Oh, sucks. I don't know if we can accept that .01dB difference:)
Maybe a different cable gauge? :)

[markw]

With all this talk about what wires and cables does to an electrical signal, hoccum nobody ever brings up what a transducer (such as a speaker) does to it?

That's kinda like refusing to go to a hospital 'cause you're wearing yesterday's underpants when you've just suffered several gunshots to the torso.

[FLZapped]

Oh, sucks. I don't know if we can accept that .01dB difference:)
Maybe a different cable gauge? :)


Oh, I forgot, JR can hear the difference....-b

[jneutron]

A #12 zip, at .2 microhenries per foot, will store 2.5 uJoules per foot under a 100 wrms signal into 8 ohms. The same zip, with total and complete skinning, will store 2.0 uJoules per foot.

A 10 foot cable, will be capable of shedding 5 ujoules (absolute maximum) if the audio changes from lf to hf content during a waveform peak.

At 10Khz, that is .4% of the load energy of one cycle lobe. That is also a .2 uSec, or 200 nanosecond, leading edge delay and trailing edge advance. (simplistic calculation of integral of energy from one half cycle)

What does an amplifier do if wire stored energy tries to dump in during high slew rate signals?

Transducers will have an effect on what we hear (duh), but if the wires can cause different delays because they see different signals, the soundstage will shift, or blur, or whatever..

Question is...are the energy numbers large enough to be significant? Can the amp have a problem with 5 uJoules being dumped into a 2 millihenry line in series with the x-ducer?

Beats me..I'm not at the level to measure that.

But, you shoulda seen my condo upgrades...sweet.. Now, restoring an 1873 built house. Test setup is backburner.

Cheers, John

[E-Stat]

Here is a reference where the phase shift and group delay are talked about in wire and where the number I used above came from:

The conclusions of that study using dummy speaker loads are worth considering:

"The fact that the reported observations can be modelled so easily does not, of course, constitute ‘proof’ that no new and unknown cable properties can possibly exist. Nor does it ‘prove’ that cables cannot affect the perceived quality of audio signals."

Indeed.

rw

[jneutron]

The conclusions of that study using dummy speaker loads are worth considering:

"The fact that the reported observations can be modelled so easily does not, of course, constitute ‘proof’ that no new and unknown cable properties can possibly exist. Nor does it ‘prove’ that cables cannot affect the perceived quality of audio signals."

Indeed.

rw

Jim does show he is open with those comments..I'd have been all over him if he said otherwise..course, the word "perceived" within that context is rather interesting..it leaves open the possible explanation of visual cueing as a possible reason.

The fact that the observations can be explained using standard theory just tosses the incorrect explanations out. That was the reason for the analysis.

John

[mtrycraft]

With all this talk about what wires and cables does to an electrical signal, hoccum nobody ever brings up what a transducer (such as a speaker) does to it?
.


Because then they would have nothing to talk about or worry about :)
Then, all their pet voodoo nonsense would go uo in smoke and they would have to eat crow, lots of it.

[mtrycraft]

Oh, I forgot, JR can hear the difference....-b


Hell, he can hear the differences insulation causes, at the .00001dB level.

[mtrycraft]

Question is...are the energy numbers large enough to be significant? Can the amp have a problem with 5 uJoules being dumped into a 2 millihenry line in series with the x-ducer?

Beats me..I'm not at the level to measure that.

But, you shoulda seen my condo upgrades...sweet.. Now, restoring an 1873 built house. Test setup is backburner.

Cheers, John

Good to see you posting again :)

That much energy? Should I worry? ")

1873 house restore? WOW. I guess too late to back out? Is this the first restoration job it sees?

[mtrycraft]

Jim does show he is open with those comments..I'd have been all over him if he said otherwise..course, the word "perceived" within that context is rather interesting..it leaves open the possible explanation of visual cueing as a possible reason.

The fact that the observations can be explained using standard theory just tosses the incorrect explanations out. That was the reason for the analysis.

John


Perceived observations? That is what all the glowing reports of cable differences are based on, perceptions only. And, we know perceptions can be unreliable and fooled so easy.

[E-Stat]

With all this talk about what wires and cables does to an electrical signal, hoccum nobody ever brings up what a transducer (such as a speaker) does to it?

Look near the top of your computer screen. This is the Cables Forum. You will also find a Speaker Forum.

rw

[jneutron]

Good to see you posting again :)

That much energy? Should I worry? ")

1873 house restore? WOW. I guess too late to back out? Is this the first restoration job it sees?

When I did the calcs, it did seem to be really low energy numbers. But, in reality, it does represent a good fraction of the total, good fraction being .1% to .5%. How that energy bandies about, I don't know yet..could just be innocuous, or it could be some kind of nefarious off harmonic garbage..my guess is just slight temporal shifts, but I've been wrong before..

Back out? you gotta be kiddin me...using my hands with the tools is the second best therapy I can think of..

It's been worked and modified through the years. Plaster yes, but romex throughout, some sheetrock, plumbing meets present code. It's not on the historic register, so I can use current materials.

Tin roof..12/12 pitch..just starting to make an AV room up there, back half HT, front computer room, 3 systems networked. Got a pair of 18 inchers I think I'll be building in along the sides where the low kick wall is gonna be, so that wall will be 2 by 6 construction.

Then, the basement ele workshop. Looks like my home testing desires will be delayed about 4 months...oh well, priorities prevail..

Cheers, John

[skeptic]

The question is not merely one of magnitude of frequency response change but of whether this is the ONLY difference in wires. Linear distortion (frequency response deviation) is easily correctable through pre-equalization to compensate for it. Non linear distortion (harmonic and intermodulation distortion) isn't. If the best the high end cable advocates can come up with is lower linear distortion, they have a very poor case. And so far they haven't even been able to do that. Both forms are easily measurable electrically and their auditory significance can be determined by DBTs. Where are their results? After 25 years I'm still waiting.

[FLZapped]

A #12 zip, at .2 microhenries per foot, will store 2.5 uJoules per foot under a 100 wrms signal into 8 ohms. The same zip, with total and complete skinning, will store 2.0 uJoules per foot.

A 10 foot cable, will be capable of shedding 5 ujoules (absolute maximum) if the audio changes from lf to hf content during a waveform peak.

At 10Khz, that is .4% of the load energy of one cycle lobe. That is also a .2 uSec, or 200 nanosecond, leading edge delay and trailing edge advance. (simplistic calculation of integral of energy from one half cycle)

What does an amplifier do if wire stored energy tries to dump in during high slew rate signals?

Transducers will have an effect on what we hear (duh), but if the wires can cause different delays because they see different signals, the soundstage will shift, or blur, or whatever..

Question is...are the energy numbers large enough to be significant? Can the amp have a problem with 5 uJoules being dumped into a 2 millihenry line in series with the x-ducer?

Beats me..I'm not at the level to measure that.

But, you shoulda seen my condo upgrades...sweet.. Now, restoring an 1873 built house. Test setup is backburner.

Cheers, John


'Bout bloody time you got yer behind back here :p

Now then, what about the other end of the termination, the amp....how will the damping factor(output impedance) affect where this energy goes considering an SS amp is several orders of magnitude lower in impedance than the speaker they are driving......

-Bruce

[FLZapped]

A #12 zip, at .2 microhenries per foot, will store 2.5 uJoules per foot under a 100 wrms signal into 8 ohms. The same zip, with total and complete skinning, will store 2.0 uJoules per foot.

A 10 foot cable, will be capable of shedding 5 ujoules (absolute maximum) if the audio changes from lf to hf content during a waveform peak.

At 10Khz, that is .4% of the load energy of one cycle lobe. That is also a .2 uSec, or 200 nanosecond, leading edge delay and trailing edge advance. (simplistic calculation of integral of energy from one half cycle)


Considering the group delay of the typical loudspeaker is in the millisecond range, I'd say you're safe in assuming this won't be audible.

Another way of looking at it is suddenly displacing your hearing point by 2 microinches(0.05 micrometers).....your heartbeat probably does that...or blinking...or clenching your teeth....or the remodelling dust settling on the furniture.....or.....


-Bruce

[jneutron]

'Bout bloody time you got yer behind back here :p

Now then, what about the other end of the termination, the amp....how will the damping factor(output impedance) affect where this energy goes considering an SS amp is several orders of magnitude lower in impedance than the speaker they are driving......

-Bruce

First order...The amp (most, anyway) is not a current device, but a voltage one. If skinning of the current and shedding scenario is actually happening (don't forget, I'm still talking about hypothetical stuff), if the wire response to storage or shedding is to change the current of the loop, the amp will see nothing to correct..the error would simply be dissipated at the load. The amp will only guarantee that it's terminal voltage is what it expects it to be, we rely on the speaker to draw the appropriate current from that correct voltage to accelerate the voice coil.

That is why I'm so anal about the b dot error of my load resistors...I really, really, need to have the current read of the resistor reflect the actual current..not current plus a factor times the rate of change of current..as well as the voltage read at the load not including b dot errors. Because the rate of change stuff(hate typing that, so I'll use b dot) is orthogonal to the current, it will defy FFT analysis, and any skinning effect that is simply orthogonal to slew will be lost in the load inaccuracies.

I also worry that the damping factor of an amp, at higher frequencies, will vary based on the quadrant of operation involved..it the output transistors are heavily into absorbing reactive speaker energy return, then, is the loop they are in still capable of the same damping factor across the spectrum..I can't answer that particular question.

Skep...

B dot error and b dot shedding will be very difficult to measure correctly..as to whether it has anything to do with audible? Darned if I know...I'll walk first, then see if I'm walking long on a short bridge..

I don't know if what I look for is harmonically related to the drive signal, or if it's just time shifting..I do know that the FFT stuff I have used in the past was unable to discern relative time shifting of the various components of a complex waveform..It's along the line of using an FFT to display a square wave...then, anytime you see that spectra, voila..it's a square wave...course, there are many different waveforms that can produce the same spectra as a square wave..
Leaving me with the ol' dual IA setup/difference hardware method.

Cheers, John

[jneutron]

Considering the group delay of the typical loudspeaker is in the millisecond range, I'd say you're safe in assuming this won't be audible.

Another way of looking at it is suddenly displacing your hearing point by 2 microinches(0.05 micrometers).....your heartbeat probably does that...or blinking...or clenching your teeth....or the remodelling dust settling on the furniture.....or.....


-Bruce

Group delay of a speaker will only be important for single source fidelity..and I'm not considering that at all.

I am considering only the possibility that the left-right signals will temporally shift independently of each other, causing temporal relative shifts in the 10 to 20 microsecond range, as that is the shift magnitudes needed for a one foot shift of apparent source at ten feet distance directly in front...ears 6 inches apart (as if it's adjustable). If I find 1 usec shifts from skin effect....that ain't audible.

Dust? get this...

My ex father-in-law is a real prominent bridge/tunnel engineer in manhattan. When the first plane hit...MTA called him to get there fast and eval the structural integrity including the subway system. He got within four blocks when they collapsed (there are times when traffic delays are a blessing). He got home covered head to toe..left dust everywhere.

When I was spackling my condo kitchen, I made huge amounts of dust...My six year old daughter saw the dust on my coffee table, and asked me if I had been in the city..

Cheers, John

[Tony_Montana]

"Amplifier-Loudspeaker Interfacing", Greiner, R.A., JAES vol. 28, no. 5 May 80, page 310-315.


RA Greiner discusses the issue of group delay due to skin effects between 100Hz and 10kHz. I would not worry about a 1x 10 to -8 power sec.

Phase angle of 3 degrees at 20kHz if you can hear 20kHz is nothing.

"Effects of cable, Loudspeaker and Amplifier Interactions", Davis, Fred E., JAES, vol. 39, no. 6 Jun 91,

Davis calculated a .04dB loss due to skin effect at 20kHz.

Thank Mtry.

The references you provide seem to be alot of good reading. Do you know how can I get a copy of those reports....free be best :D?

You know, I got so bugged down on details about inductance and skin effects of the cable that I forgot to take into consideration the effects of speaker and amplifier on the signal (as others have mentioned). That does bring this issue to whole new light as far as cable's effect on the system [as a whole] is concern.

S I guess the question now becomes...will be worth to fix few nanosecond/dB of signal shift/attenuation caused by cables, as compared to greater magnitude of phase-shift/attenuation caused by speaker/amplifier?

Back to the drawing board again :)

[markw]

Look near the top of your computer screen. This is the Cables Forum. You will also find a Speaker Forum.

Since this entire discussion is about the audiability of differences in cables, by ignoring the speakers contribution to this, you render any further discussion on your part null, void and silly unless you can tell me how you can hear your widely touted differences without a speaker, or transducer?

Measurments alone? Be serious.

If such is the case, the next step would be to ascertain at which point the results of those measurements are audiable. Nobody on "that side" seems interested in that. They "just know".

What you are trying to do is arrest a hunter for urinating in a river and causing the death of said river's ecosystem when a chemical plant 100 feet downstream is dumping thousands of gallons of toxic waste daily.

Again, be serious. nice trolling

[E-Stat]

Without a speaker, how can you tell if cables make a difference?

I'm pretty new to this site, but it would seem to me that the purpose of having specialized forums is to focus attention on that individual aspect of the playback chain.

rw

[markw]

I'm pretty new to this site, but it would seem to me that the purpose of having specialized forums is to focus attention on that individual aspect of the playback chain.

What's that clicking sound? Out of ammo, eh?

heh heh heh... since when can one listen to only one part of that chain without taking into account any others, up to and including the ear? heh heh heh...

Perhaps a recap of the first sentence in this thread is called for...
"As far as sonic effects of inductance and Skin effect is concern..."

Key word here is "sonic", which refers to sound, which is generally sense by using ones ears to listen.

.... or have we removed listening itself from the table?

[Bobby Blacklight]

Just reading through the thread but wouldn't any stored energy in the cables be insignificant or secondary to the back EMF from the drivers?? Or is this only an issue with the woofer as the midrange/tweeter would be an easier load?? I would think the drivers would be the real issue the amp has to deal with.

Hey John good luck on the house!!

[E-Stat]

Perhaps a recap of the first sentence in this thread is called for...
"As far as sonic effects of inductance and Skin effect is concern..."

Don't stop there with the ellipsis. The last four words in the topic are:

in a speaker wire


.... or have we removed listening itself from the table?

Why would you do that?

rw

[markw]

Don't stop there with the ellipsis. The last three words in the topic are:

in a speaker wire



Why would you do that?

rw


I bow to your superior knowledge.

You've certainly proved to me that some people are so smart that the only people who can outsmart them are they themselves.

[jneutron]

Just reading through the thread but wouldn't any stored energy in the cables be insignificant or secondary to the back EMF from the drivers?? Or is this only an issue with the woofer as the midrange/tweeter would be an easier load?? I would think the drivers would be the real issue the amp has to deal with.

Hey John good luck on the house!!

If it were only a mono system, then it would be a horse of a different color. In a mono system, I'm not sure if shifting a vocal by 20 useconds would be distinguishable.

I was thinking about soundstage imaging only..Our ability to discern the direction of a sound source by the left to right ear delay. If any musical content (like a vocal) is time shifted 20 usec by some nefarious skin effect thingy brought on by a bass guitar (for example) on the left channel only, the human brain (mine too) would discern that time shift as a source shift if the leading and trailing transients shifted the same way, and I don't know what it sounds like if the shift is dependent on the amplitude or polarity of the bass current.

With a stereo pair, the actual response of the speakers, both phase and amplitude, is not in the least bit important with respect to localization of the image of the sounds. (this statement neglects ENTIRELY the science of the human ear, auditory canal, a huge amount of very good research...I believe Chuck (and JJ) has a very good grasp of that stuff if you really want to find out more about it. But all that will only confound what it is I am looking for..skin based time shifting of some component of the signal. My discussion is but a small and specific subset of all that other research...Ask Chuck how we can distinguish between a noise in front of us as opposed to behind us...)

I have experienced source shifting in headphones, one record in each ear, and varying the time relationship between the two...the image sounds like it is sweeping through the head. I can envision how confusing it would be if only a part of the musical content shifted that way.

Course..this all assumes a skin shift is even real..

Thanks...I'm chompin at the bit to start the attic. Cause that's where I'll put the computers..right now, I am cpu-less, and that is where the e/m analysis program I'm working on sits..waiting for more code.


Cheers, John

[E-Stat]

I'm chompin at the bit to start the attic. Cause that's where I'll put the computers..right now, I am cpu-less, and that is where the e/m analysis program I'm working on sits..waiting for more code.

When the wife and I built our home seven years ago, I tried to plan as best I could for various wiring needs. I have two dedicated AC lines in the basement intended for the main audio system. All rooms accomodate four telephone lines. Most rooms have a cable TV outlet. One thing, however, I didn't do was wire for CAT5. At first I thought that a problem, but now with fast and inexpensive WiFi, the situation has changed. I say all of this to suggest that you may want to place a wireless router in a more central area of the house to allow access virtually anywhere. It is handy indeed to be able to take my laptop to any room or outside on the deck so that I can spend far too much time on audio boards! :)

rw

[jneutron]

When the wife and I built our home seven years ago, I tried to plan as best I could for various wiring needs. I have two dedicated AC lines in the basement intended for the main audio system. All rooms accomodate four telephone lines. Most rooms have a cable TV outlet. One thing, however, I didn't do was wire for CAT5. At first I thought that a problem, but now with fast and inexpensive WiFi, the situation has changed. I say all of this to suggest that you may want to place a wireless router in a more central area of the house to allow access virtually anywhere. It is handy indeed to be able to take my laptop to any room or outside on the deck so that I can spend far too much time on audio boards! :)

rw

Thanks for the suggestion. I am concerned with security and wireless networks, though. At work, they are a pita for the it people.

When I built a house a long time ago, I went in before the sheetrock was placed, and took a lot of pictures of the stud locations, so that I could more easily add wires in the future should the need arise (which it did). I was quite proud of my ability to think ahead in this regard..


Film in the camera would have been nice, though...

Cheers, John

[E-Stat]

I am concerned with security and wireless networks, though.

Absolutely. I use WEP both at home and at work. (I work for a software developer) Configuring a good firewall and router security is a bit of a PITA, but once set, usually requires little maintenance.


Film in the camera would have been nice, though...

Naw, I've never done anything like that... :)

rw

[Tony_Montana]

If it were only a mono system, then it would be a horse of a different color. In a mono system, I'm not sure if shifting a vocal by 20 useconds would be distinguishable.

With a stereo pair, the actual response of the speakers, both phase and amplitude, is not in the least bit important with respect to localization of the image of the sounds.

Hey John

I agree with you that in mono, phase shift of signal (at least for 20 usecond) will not matter that much. But i don't understands yourstatment about the stereo pair.

Since for stereo pair, phase shift property of "stereo" signal is used to project imaging to side or even to the back of head (for example the simulated surround capability of two speakers used for TV), then wouldn't phase shift be a concern in stereo pair?

[mtrycraft]

I was thinking about soundstage imaging only..Our ability to discern the direction of a sound source by the left to right ear delay. If any musical content (like a vocal) is time shifted 20 usec by some nefarious skin effect thingy brought on by a bass guitar (for example) on the left channel only, the human brain (mine too) would discern that time shift as a source shift if the leading and trailing transients shifted the same way, and I don't know what it sounds like if the shift is dependent on the amplitude or polarity of the bass current.

But, that 20usec needs to be the difference between the two cables, in that example to cause the same shift. R.A.Grainer and the other referenced article could not come even close to those time periods, by several order of magnitudes.


I have experienced source shifting in headphones, one record in each ear, and varying the time relationship between the two...the image sounds like it is sweeping through the head. I can envision how confusing it would be if only a part of the musical content shifted that way.

Varying it by how much?

Course..this all assumes a skin shift is even real..

Why would it be not real? All the present day texts claim it real and calculate it to be real. The audibility of the magnitude is the question.
No such data is available.

[jneutron]

But, that 20usec needs to be the difference between the two cables, in that example to cause the same shift. R.A.Grainer and the other referenced article could not come even close to those time periods, by several order of magnitudes.

Yes, I agree, the diff has to be between cables..that is why I gave the example of the bass lines on one channel only, somehow causing a nefarious shift of the vocals on one channel only..

I have experienced source shifting in headphones, one record in each ear, and varying the time relationship between the two...the image sounds like it is sweeping through the head. I can envision how confusing it would be if only a part of the musical content shifted that way.


Varying it by how much?.

I don't know, but it varied from 100 milliseconds (obvious echo type) to zero. When I played a nightclub back in the late 70's, one of the effects I used was to play two identical songs at the same time, at equal levels. The phasing/flanging effect was neat, making a comb filter (inna gadda da vita style). But as the DJ, I had each turntable feeding one ear only, the perfect cancellation the dancers heard when shift=zero was when the image in my head was dead center. I could tell by the image location within my head, how the sound effect outside was going.

Course..this all assumes a skin shift is even real..


Why would it be not real? All the present day texts claim it real and calculate it to be real. The audibility of the magnitude is the question.
No such data is available.

My apologies..when I say skin shift, I am not talking about the group delay stuff, 250 nano type..that is unquestionably beyond human hearing. I am talking about a complex musical content waveform being capable of time shifting some transient information by AT LEAST 20 microseconds (my arbitrary, simplistic threshold for image shifting). 250 nano is a waste of time w/r to audibility, IMHO. I must confess that I've not been able to peruse any of your links yet, I'm concentrating on putting boxes into and out of my minivan as the movers come Sat...but I'd be surprised to see that anyone has even examined their test setups for accuracy issues I point out. (I also assume some of the links include testing, and not just math.) But the present day texts I've seen still treat skin effect as a simple sine analysis, using the frequency domain without regard to energy balance equations..If I am correct, they will have to re-write the textbooks for the inclusion of slew rate based skinning. (not asking much, huh?) I still discuss my stuff with the guys here, and nobody has been able to shoot it down yet..(doesn't mean I'm right either).

In all my research, I've not seen anybody concentrating on the capability of the load resistor with it's slew errors...a very confounding issue for high speed high current low impedance measurement accuracy, and the very first item I hit upon in my skin test endeavors. I will (once I'm back up) be applying a rigorous testing regime to my load and current viewing resistors, to make sure the I/V phase errors remain very low up to at least 1 to 2 Mhz. That way, I can assure myself that the errors at audio freq's are negligable.


Tony: ""then wouldn't phase shift be a concern in stereo pair?""

Yes, it is..but, I'm concentrating only on left to right difference related to a skin phenomena...overall phase shift of each speaker is way, way over my simplistic approach, and something more suited to a speaker type forum..

Cheers, John

[mtrycraft]

Thanks. Maybe you will come up with a new textbook :)

[mtrycraft]

Thank Mtry.

The references you provide seem to be alot of good reading. Do you know how can I get a copy of those reports....free be best :D?

You know, I got so bugged down on details about inductance and skin effects of the cable that I forgot to take into consideration the effects of speaker and amplifier on the signal (as others have mentioned). That does bring this issue to whole new light as far as cable's effect on the system [as a whole] is concern.

S I guess the question now becomes...will be worth to fix few nanosecond/dB of signal shift/attenuation caused by cables, as compared to greater magnitude of phase-shift/attenuation caused by speaker/amplifier?

Back to the drawing board again :)

Yes, free is best:) but I don't know. AES is not that expensive, or a nearby University library with EE course?


I would worry about speakers first, today. nothing second, really. Considering the complexity of properly converting the electrical signal to acoustic properly, everything else is insignificant today. Radiation pattern uniformity, enery output uniformity with off axis performances, speaker compression, distortion, etc.

[Tony_Montana]

I live on university campus so I see if I can check out those AES papers. They probably got them on microfilms :)

[mtrycraft]

I live on university campus so I see if I can check out those AES papers. They probably got them on microfilms :)

Maybe, but I am not aware that any has been filmed :) I think they are only $10 for Journal papers.