A-B testing of cables

Gorquin - since, and I quote....

"I'm open to learn something proven by science"

I'll share some of my experiences....
- I was installing a new light fixture in the hallway that employs a two-way switch - i.e. you can turn on/off from either end

- to verify the wiring was correct I tested the circuit with a digital multi-meter
- With the light in the on position I checked the circuit
- the wire that carried the power to the light registered 120 volts - as it should
- to my amazement the other wire i.e. the one that carries voltage when both switches are in the other position and should be at ZERO volts - but actually registered a voltage around 43 volts.
- to make sure I was not measuring incorrectly I double checked with a second analogue meter - it recorded a voltage of 13.5 volts.

What I was actually measuring was a voltage that had been "induced" into the second wire...

The reason for the disparity between the two meters was due to the internal circuitry - the analogue meter draws more power than a digital meter, therefore the voltage recorded was less.

This is actually a well documented phenomena - simply google...

"two way light switch register phantom voltage" for some articles about it.

So what does this have to do with audio?

Inductance takes place ANY TIME that two conductors (i.e. wires) are
1. parallel to each other and
2. very close together

The induced signal, or noise is directly related to the length of the two conductors and how close they are.

If you look at the most common geometry used for cables i.e. IC's, Speaker Cables and Power cables, you have two conductors encased inside a sleeve in very close proximity.

They may have a slight twist in them, which makers state is a "noise rejecting geometry' (noise rejecting my A**).

Why are they made this way? - it's much cheaper to produce!

However...

Companies like Kimber Kable, figured this out many years ago and started using braided conductors for a superior result.

Some companies figured out that the induction process could be reduced significantly by tightly twisting two conductors together - such as the geometry used in Stager Silver Solids IC's

Others utilized a loose spiral wound around a spacer between the two conductors.

AntiCable (amongst others) found a very good solution was to wind the neutral conductor around the signal conductor in their IC's for even better performance - but they are hand made and a little more expensive to produce

Now - If you cross two conductors at an acute angle - generally between 50-80 degrees, such as that employed in Anticables spiral IC's - the induction effect is significantly reduced and this translated into a quieter background and better clarity. It also facilitates a more spacious and wider/deeper image.

Won't having two wires of differing length effect cable performance?

NO - see *** below

Going back to your "No Moving parts" - brings me to my own "atomic theory for beginners" understanding of electricity.

Electrically Conductive materials have loosely coupled electrons (i.e. moving parts) that allow them to conduct electricity - they are called valence electrons ( for the sake of brevity look it up on the web)

The reason silver is a better conductor than copper is that silver has more valence electrons. But more importantly, the electrons are in a more "spacious orbit", so it takes less electrical potential to get them to conduct, i.e. compared to other materials.

Net result - silver cables (in general) tend to exhibit a more dynamic presentation - because they are simply - faster

Some people reported silver cables as being "too bright", but this is probably due to the cable geometry used and the other materials used in it's construction - it has little to do with the conductive qualities of Silver.

So, couple the better conductive properties of Silver with the advanced geometries now being explored and you have a cable that WILL perform better many other cables out there.

One last thing to expand your thought process even further.

*** There is a common belief that
- one conductor in a cable is the for the signal
- the other is the return path

To challenge that belief - if you think about the function the cable is actually performing from the perspective of the attached circuits you may come away with a very different conclusion.

Please read the lower section of this blog post:
http://www.image99.net/blog/files/category-cable-architecture.html

As you can see - the neutral conductor simply maintains a potential of ZERO VOLTS between the two components and has little to do with the "return path"

There should be no signal in the neutral side of either component!

And so - Their lengths really are immaterial

But it is imperative that no noise gets "induced" into the neutral sides of the cable, resulting in noise in the neutral sides of the connected components - resulting in a degraded system performance.

Which is just one reason why cables DO make your system "sound different"

Power cables also experience similar issues and cable geometry is probably the most important thing to consider when trying to remedy power related issues. But that's for another day :-)

That's a very brief summation of some of the cable related thing's I've personally encountered and tried in the last 4-5 years

Hope you found it informative :-)

Reards...

Steve (Williewonka), re the blog post you referenced, keep in mind (as I'm sure you realize) that for current to flow a complete circuit has to be present, from source to destination and back to source.

When the output of one component provides a signal to the input of another component, there indeed has to be a path for the corresponding current to return to the source of the signal. And the quality of that return path will matter just as much as the quality of the path for the "signal." In fact, it will often matter more, as explained in the next paragraph.

Depending on the internal grounding configuration of the two components, even very small amounts of resistance in the return conductor may contribute significantly to ground loop issues, which can in turn result in issues involving high frequency noise as well as low frequency hum. While the resistance of the signal conductor in a line-level interconnect that is conducting an analog audio signal will only matter if that resistance is a significant fraction of the impedances of the connected components, especially the input impedance of the destination component. And for line-level analog interconnections, that resistance will be a totally miniscule fraction of that input impedance in nearly all home audio systems.

Regards,
-- Al

Thanks Al - Nicely put

Absolutely, the circuit has to be "complete" for it to work.

I guess my problem is the "vision" people tend to get in their minds that an actual return signal flows in the neutral side.

But if the neutral side of all components is grounded at zero volts(as depicted in the diagram), at least in all good designs, there should be no signal as such in the neutral side of the component or cable - is there?

In support of your statement about miniscule amounts of resistance in the neutral side of cables - I found with various versions my own Spiral IC's - performance improved as I added more conductors to the neutral line - but the signal conductor was better with a single 24 gauge solid core silver conductor.

I think were are saying the same thing, just a little differently - but I know we agree on the importance of a "zero volt" neutral side of the audio pathway.

Cheers

Hi Steve,

Agreed for the most part, and the experience described in your second from last paragraph makes a lot of sense. However, a current will indeed flow in the return conductor of an unbalanced line-level interconnect, that is equal to the current in the signal conductor. Or, at least, the two currents SHOULD be equal. Any slight differences that may exist between them would mean that some fraction of the return current is finding an alternate path, such as the return conductors of other cables that may be connected between the same components, or through the AC safety ground connections of the two components (which would constitute a ground loop issue, to some degree that may or may not have audible consequences).

But if the resistance and impedance of that return conductor is 0 ohms at all relevant frequencies (to a very close approximation), then the voltage drop from one end of that conductor to the other which results from the current flowing in it will be, per Ohm's Law, 0 volts (to a very close approximation). Which in turn will mean that the return conductor and the points in each component to which it is connected will all be at 0 volts, relative to the circuit grounds of both components. (Keep in mind that any voltage number, including 0, is only meaningful if a reference point is defined, at least implicitly. In this case the reference point(s) would be the circuit grounds of the two components).

In any event, thanks for sharing your experiences, which I know are particularly extensive in these kinds of matters.

Best regards,
-- Al

ZD you don't know much about me to make a statement like that. Oh wait, so that means I couldn't possibly be correct in anything I say because you've already convicted me and YOU must be correct. LOL Works both ways doesn't it.