Solid Core Wiring

[ebslo]

I was going to watch that video. Someone else posted it.

Did you mean nanometers at 20kHz? mm seems quite large based on what I’ve previously read.

Source, from the wikipedia article on skin effect.

 

[Promit]

I was going to watch that video. Someone else posted it.

Did you mean nanometers at 20kHz? mm seems quite large based on what I’ve previously read.

Here you go -

Skin effect depth Calculator

Skin Depth/Skin Effect and Calculator - Ness Engineering Inc.

Skin depth or skin effect causes RF current to flow in a thin layer near the surface of the electrical conductor. Equations and a calculator are provided.

www.nessengr.com

 

[brimble]

In addition to the model panthers that show the quality of various pieces of equipment, we need model trolls that show the quality of threads like this.

 

[Newman]

IMO this is an honest thread and a teaching opportunity.

 

[arpinnurmela]

Here you go -

Skin effect depth Calculator

Skin Depth/Skin Effect and Calculator - Ness Engineering Inc.

Skin depth or skin effect causes RF current to flow in a thin layer near the surface of the electrical conductor. Equations and a calculator are provided.

www.nessengr.com

Thanks Promit and ebslo for the links. I was only off by a factor of 1000 in my mind, but that explains why everyone was saying that the skin effect shouldn't matter on frequency. I'd like to know the distribution of current within the skin, but largely the point is taken with a bit of salt and embarrassment on being off by three orders of magnitude. The skin effect could still be relevant since the majority (>63%) is still relatively traveling at the surface at 20kHz. Is this amount not relevant and subject to stating the surface flow is principally occurring on the surface. I realize that I was incorrect dramatically about the size of the skin effect, but it is still a valid statement that >63% is mostly happening on the surface of the conductor, which is where coatings are.

I appreciate the correction and should have vetted my thoughts better prior to writing them.

 

[arpinnurmela]

IMO this is an honest thread and a teaching opportunity.

I pulled this from Wikipedia: "The skin depth, δ, is defined as the depth where the current density is just 1/e (about 37%) of the value at the surface; it depends on the frequency of the current and the electrical and magnetic properties of the conductor."

This does indicate that there is a fairly substantial current flow happening at the surface. So why would a coating applied directly to the surface not effect the current flow and create a drag on the current.

 

[ebslo]

On stranded wires, the electrons are forced to jump the boundaries, which requires quantum tunneling, and this quantum tunneling causes a phase shift in the signal, or a very slight but still existent time delay.

A quick search of tunneling delay times reveals they are on the order of 10's of attoseconds, or 10^-17 seconds. Here's one of several references that corroborate this: Ultrafast resolution of tunneling delay time.

 

[arpinnurmela]

I pulled this from Wikipedia: "The skin depth, δ, is defined as the depth where the current density is just 1/e (about 37%) of the value at the surface; it depends on the frequency of the current and the electrical and magnetic properties of the conductor."

This does indicate that there is a fairly substantial current flow happening at the surface. So why would a coating applied directly to the surface not effect the current flow and create a drag on the current.

My whole point was that air-gapping the wire eliminates any potential interaction with a coating that would have some non-negligible effect.

A quick search of tunneling delay times reveals they are on the order of 10's of attoseconds, or 10^-17 seconds. Here's one of several references that corroborate this: Ultrafast resolution of tunneling delay time.

Yes. And how many tunneling events might there be? Consider that a mol of electrons is on the order of 10^23. I would imagine that that just might offset things in my favor. Especially taking into account distance traveled.

 

[Mnyb]

I voted unimpressed as I in my younger audiophools days was the proud owner of several expensive audioquest products .

They employees sevral variants of solid core wire topology .

When I came to my senses I was so embarrassed I could not sell them second hand that would be taking advantage of some other misinformed sod .

So I went to the recycling bin with all of it so it would not apear in any audio system ever again .

This is a kind of troll tread that goes against the ethos of this forum . There is flawed assumption that’s been debunked for decades at the outset , that there are audible differences to be had with properly designed wires.

Oh and the old chestnut skin effect is dragged out to the parade it sure exist buts is of no consequence in audio . For DC and low frequency applications L , C and R is enough for you.

However analog video coax and digital coax are often cables with a solid core but here the frequency is high enough to make a solid core desired

 

[ebslo]

Yes. And how many tunneling events might there be? Consider that a mol of electrons is on the order of 10^23. I would imagine that that just might offset things in my favor. Especially taking into account distance traveled.

Current flow is mostly random at a molecular scale. Here's an article that gives some sense of what is actually happening. The free electrons move at around 0.5 to 0.9 of the speed of light, but in random directions. They constantly collide with atoms in the conductor and bounce off in some other random direction. The current is due to the drift (average) velocity of the electrons, which is surprisingly slow. From the article:

The drift velocity of electrons through a copper wire of cross-sectional area 3.00 x 10-6 m2, carrying a 10A current, is approximately 2.5 x 10-4 m/s, or one-fourth of a millimeter per second!

So when considering these attosecond scale delays you're focusing on, also consider the context. For every electron that incurs a slight tunneling delay along it's mostly random trajectory, a bunch more electrons collided with an atom within the conductor and shot off in a completely different direction.

Edit: formatting

 

[arpinnurmela]

Current flow is mostly random at a molecular scale. Here's an article that gives some sense of what is actually happening. The free electrons move at around 0.5 to 0.9 of the speed of light, but in random directions. They constantly collide with atoms in the conductor and bounce off in some other random direction. The current is due to the drift (average) velocity of the electrons, which is surprisingly slow. From the article:
[Quote}
The drift velocity of electrons through a copper wire of cross-sectional area 3.00 x 10-6 m2, carrying a 10A current, is approximately 2.5 x 10-4 m/s, or one-fourth of a millimeter per second!

So when considering these attosecond scale delays you're focusing on, also consider the context. For every electron that incurs a slight tunneling delay along it's mostly random trajectory, a bunch more electrons collided with an atom within the conductor and shot off in a completely different direction.
[/QUOTE]

Okay, I'm tracking you. So let's take just 1amp. in an ampere there are 6.25x10^18 electrons, which would be about how many electrons might be going through a speaker wire. Possibly less unless we are talking about peak power output. So taking this as a low end, and considering that the speed is an average speed over a distribution of speeds, it would not be unreasonable to see speeds upwards of 2.5x10^-2m/s or higher. That's 2.5cm per second. Plus, even taking the average as you point out, being .25mm. Well that's a bit larger than most of the strands that are put into stranded copper wires. Most of the strands that are placed inside are either 32ga or 36ga, both of which are smaller than "one-fourth" a millimeter.

Now, I'm not claiming that all the electrons are quantum tunneling. I seriously doubt that they are. But given the random directions, a statistical distribution of speeds, no one can refute that some portion of the electrons are going to tunnel across to another strand. The question is not whether they do, the question is how many, and is it enough to alter the signal.

I am not claiming to know all the answers (insert zinger). But this is a pretty cool discussion if you think about it. @ebslo, seriously, that is good information that you are throwing out. I'm enjoying the discussion and I find it interesting. So I appreciate those who are arguing with me and taking me to task on things. I am going to do a double blind ABX test just because of you guys. And an Audacity. Also just for myself. It's important to know if you've gone crazy. (insert zinger) When I do my next speakers, I'll take some recordings with a stereo mic I have, put up a poll somewhere here, and just see if people can reliably distinguish the two speakers. Open poll results. It won't say what speaker is better, but it would be a clear demonstration that the same speaker, has a different sound. I have been taking pictures of my other internal speaker rewires (just so that I can be exactly sure that I'm putting them back together correctly). So I can do that for a separate thread that would include the poll with audio samples. Maybe I could make four or five of the same recording and see if people can match the remainder to the first. That could be interesting?

Plus, I'm still curious if other people are doing DIY solid core wires or ribbons and what their experiences have been.

 

[DSJR]

Going back to domestic amps and speakers, I think some of it was like the following for us... Properly stable amps didn't 'reproduce' changes in cables other than gauge so 'we' thought they were bland and not able to reproduce said differences. Amps that DID reproduce differences often lacked output inductors and were running on the verge of oscillation (Naims were classic for this back then with no output inductors and heavy use of non shielded cables in the system [preamp to supply cable had a shield but it wasn't/isn't connected], but we all liked the added 'excitement' in the sound). ANY change on LCR of commected wires made a sonic difference to us back then.

As for the Goertz, Litz and similar speaker cables, I may have it wrong, but weren't these highly capacitive?

 

[SIY]

So besides not understanding QM (which is entirely irrelevant), we also don't understand the difference between AC and DC.

There's so much willful non-understanding of basic physics (clearly gleaned from advertising) and basic perceptual psychology, coupled with some outrageously made-up claims ("I calculated all this, but it's in a box somewhere" "I did ABX tests but won't tell you anything about procedures and controls") that it is difficult to class this as anything but trolling.

Please stop feeding.

 

[arpinnurmela]

Going back to domestic amps and speakers, I think some of it was like the following for us... Properly stable amps didn't 'reproduce' changes in cables other than gauge so 'we' thought they were bland and not able to reproduce said differences. Amps that DID reproduce differences often lacked output inductors and were running on the verge of oscillation (Naims were classic for this back then with no output inductors and heavy use of non shielded cables in the system [preamp to supply cable had a shield but it wasn't/isn't connected], but we all liked the added 'excitement' in the sound). ANY change on LCR of commected wires made a sonic difference to us back then.

As for the Goertz, Litz and similar speaker cables, I may have it wrong, but weren't these highly capacitive?

Yes to these other speaker cable issues. In fact it was exactly the issue of capacitance that made me completely uninterested in trying out the Litz wire.

Where Litz people are concerned about skin effect and seek to internalize all of the current flow inside the wire, the idea behind solid core is sort of the opposite. The point of air-gapping is to remove anything on the surface where <63% of the current flows. The point of a solid core of wire is to simply eliminate the issue of quantum tunneling between strands.

I feel that we’ve been able to agree that <63% of the current does flow on the surface. But we haven’t agreed that quantum tunneling is happening between strands.

One poster said that tunneling only occurs if the spacing is greater than a few nanometers. So I pointed out that the spacing between strands easily satisfies this requirement.

These two effects, <63% surface flow, and tunneling between strands, are the basis of my guess as to WTF is causing a difference in sound.

But I haven’t proven to anyone yet that there is a difference in sound to begin with. So I must concede on that and focus my energy there.

 

[Killingbeans]

One poster said that tunneling only occurs if the spacing is greater than a few nanometers. So I pointed out that the spacing between strands easily satisfies this requirement.

Well, no? Quantum tunneling takes place when the barrier is a few nanometers or less. Meaning that the natural film of contaminants you get on any strand of copper, that wasn't made in a cleanroom and kept in an inert atmosphere, most likely will be enough to make tunneling events nearly nonexistent.

These two effects, <63% surface flow, and tunneling between strands, are the basis of my guess as to WTF is causing a difference in sound.

What would the result of those two effects be? Phase shift? Treble roll off? Both would be child's play to measure.

 

[arpinnurmela]

So besides not understanding QM (which is entirely irrelevant), we also don't understand the difference between AC and DC.

There's so much willful non-understanding of basic physics (clearly gleaned from advertising) and basic perceptual psychology, coupled with some outrageously made-up claims ("I calculated all this, but it's in a box somewhere" "I did ABX tests but won't tell you anything about procedures and controls") that it is difficult to class this as anything but trolling.

Please stop feeding.

SIY. My blind test was with my wife, but it wasn’t double blind. I played her the speaker with my regular speaker wire, then played it with the new speaker wire. Then I wanted to see if she could ID the wire, so I put it back to the original and asked what she thought. She said it wasn’t the new wire. Then I repeated the entire test again. She was able to ID the difference the second time. But for me personally, this isn’t evidence of anything which is precisely why I’m agreeing with your request to do a better test.

To be very clear to everyone, I never claimed to have calculated the phenomenon that I am hearing. I calculated two things for homework in grad school and had subsequent class discussions: current distribution in a solid wire (where most flows on the surface) and quantum tunneling between two pieces of metal with a small delta between them.

In this thread we’ve concluded that <63% of the current flows on the surface.

I am asking any physicist or engineer who knows about tunneling between two pieces of metal to back me up.

What I don’t specifically know is whether the tunneling amounts to enough to be reliably heard. It’s a hypothesis that I hope will be trumped by an overall sound difference.

So I’ll do a better ABX test, I’ll do an Audacity, and I’ll do a recording match poll. Again, not for you, but for me. It’s a good idea.

 

[arpinnurmela]

Well, no? Quantum tunneling takes place when the barrier is a few nanometers or less. Meaning that the natural film of contaminants you get on any strand of copper, that wasn't made in a cleanroom and kept in an inert atmosphere, most likely will be enough to make tunneling events nearly nonexistent.



What would the result of those two effects be? Phase shift? Treble roll off? Both would be child's play to measure.

Thanks Killingbeans. Those are very good points on the tunneling that I didn’t think about. So perhaps it isn’t happening at all due to the spacings and/or dirt and film? Or perhaps that mitigates the number of tunneling events down to a percentage that it doesn’t matter. This could very well be the case. Short of doing lab research I don’t know that we could know the right answer. I would be surprised to discover that no tunneling is occurring, that seems highly unlikely. But it could be highly mitigated in the way you are saying.

So let me suggest that we wait until I have better ABX, an Audacity, and a listening poll.

Also, to test overall outcome, I will make a air-gapped stranded wire. So I will test three different wires on Audacity. Then we will have stranded with coated, air-gapped stranded, and air-gapped solid core.

This will be a great thing for me to test as I’m very curious about what I believe I and seven other people are hearing. Depending on how the air-gapped stranded turns out I think it could remove or bolster the tunneling aspect that I am hypothesizing.

 

[arpinnurmela]

Thanks Killingbeans. Those are very good points on the tunneling that I didn’t think about. So perhaps it isn’t happening at all due to the spacings and/or dirt and film? Or perhaps that mitigates the number of tunneling events down to a percentage that it doesn’t matter. This could very well be the case. Short of doing lab research I don’t know that we could know the right answer. I would be surprised to discover that no tunneling is occurring, that seems highly unlikely. But it could be highly mitigated in the way you are saying.

So let me suggest that we wait until I have better ABX, an Audacity, and a listening poll.

Also, to test overall outcome, I will make a air-gapped stranded wire. So I will test three different wires on Audacity. Then we will have stranded with coated, air-gapped stranded, and air-gapped solid core.

This will be a great thing for me to test as I’m very curious about what I believe I and seven other people are hearing. Depending on how the air-gapped stranded turns out I think it could remove or bolster the tunneling aspect that I am hypothesizing.

I’ll also get some coated solid core. That’s like a couple of bucks at a hardware store. So I’ll do four recordings and post all four. I’d like to pin down precisely what makes a sonic difference if any at all. It could still turn out that I and seven other people are way off course on what we think we’re hearing. It will at least be interesting to me personally.

 

[arpinnurmela]

As an aside:

This is a different kind of forum than what I thought. It’s very data centric and there are some clearly smart people on board. So I am getting that more clearly now and will try to rise to the occasion or fail doing so.

 

[mansr]

I am asking any physicist or engineer who knows about tunneling between two pieces of metal to back me up.

Good luck with that.