IXOS cable opinions?

[zumbo]

You mean over the cell phone while pulling up in the drive way right?

With a bucket of chicken, and some Red Stripe!

 

[Nick250]

With a bucket of chicken, and some Red Stripe!

And there was a thunder storm.

 

[mtrycrafts]

You mean over the cell phone while pulling up in the drive way right?

I don't know about that. Need to experiment and see if I can find volunteers

 

[cbevil]

Well, you could try it by manual switching of cables but that has memory drawbacks. By the time you switch, your memory is gone.
If you want to try. Have an interested party switch it while you are out of the room. Have them keep track of course, random order, or no switching at all for any of the trials , and the listener, you, other golden ear friends, listen and write down your guesses. You need 9 of 10, 12 of 16 correct guesses.

I don't consider myself a "golden ear", nor does anyone I know, though I do seem to be more picky about audio quality than any of my friends.

There's probably a way to set the experiment up without manually switching cables, but then you introduce variables such as different inputs or outputs, or switches, etc; you'd need to find a way to isolate the cables as the varying factor. And yes, you'd have to get a statistically significant result depending on your sample size. The smaller the number of trials, the more "hits" you have to have. And for there to be a clear winner I would think everyone involved would have to pick the same cable. For this to happen there would have to be a pretty big difference between cables. From what I've read here and other places it just ain't going to happen. But I still think it would be an interesting exercise, or at least a good excuse to drink beer afterwards.

Oh, and if my wife (who doesn't care about audio at all) was in the driveway, calling me from a cell phone in a thunderstorm, I'm sure it'd be to tell me to get off my butt and come help her bring in the Red Stripe and bucket of chicken, and to turn down that damn stereo already!

 

[jneutron]

Anyway, below is what Cobalt has to say;

""OK, so what happens when you run a signal whose frequency is too low to take advantage of skin effect? What happens is you have the signal using two different conductor materials for its transmission. You have one part of the signal (usually the highest frequency portion of the signal, typically the highest treble in audio) using the silver plating and the rest of the signal uses whatever the core is made out of. So, you have two different conductors handling different parts of the signal. Since silver is a better conductor than copper (most typically used as the core, but we have seen silver plated steel conductors), the portion of the signal that is using the silver will get an artificial boost.""

ARGHHHHHHHHH

Where do they get this stuff???

1. The conductivity difference is not that great.
2. Skin depth at audio frequencies far exceeds the wire diameters, there will be no difference in the radial distribution of conduction current at any audio frequency between copper and silver plated copper.

The explanation sounds valid to me but they did gloss over one little detail and that is that silver is only a marginally better conductor than copper. I assume that the reason there will be an artificial boost is because the portion of the signal travelling over the silver will propogate slightly faster. Emphasis on *slightly* - the signal is already travelling at approximately 70% of the speed of light

The explanation is invalid. The boost does not exist.

The velocity of propagation in ANY cable is V = 1/sqr(LC). For a cylindrical conductor or conductor set, L will be dependent on the external inductance as well as the internal inductance. The internal inductance of a cylindrical wire is 15 nH per foot. For a twisted pair, the external inductance will be approximately 150 nH per foot. So, for complete skinning, such as at a Ghz, the L in the velocity equation will be approximately 10% lower, producing a slightly higher velocity (about 5% higher)..
Mtry...because conductivity of the conductor is the reason skinning does occur, it will directly impact the internal inductance of the conductor, as a fully skinned conductor will have zero internal inductance. But plating is insignificant in this regard below meg frequencies.

For double braid coax, the prop is dependent on DC, as you state..but for non coaxial constructs, the L also comes into play, as well as for a solid inner wire in a coax.

Cheers, John

 

[mtrycrafts]

OK, and how much is that L will have an effect vs DC?

 

[mtrycrafts]

... From what I've read here and other places it just ain't going to happen. But I still think it would be an interesting exercise, or at least a good excuse to drink beer afterwards.

Oh, and if my wife (who doesn't care about audio at all) was in the driveway, calling me from a cell phone in a thunderstorm, I'm sure it'd be to tell me to get off my butt and come help her bring in the Red Stripe and bucket of chicken, and to turn down that damn stereo already!

You are killing me

Yes, that would be an interesting experience, a worthwhile one for sure. Just be careful where you tell that experience. Golden ears will tell you you just cannot hear, or your gear is crude, no resolution to it

 

[mtrycrafts]

I was going to edit my other post. Either ran out of time or it isn't working right now:

The velocity of propagation in ANY cable is V = 1/sqr(LC).

That would include the coax and its dependence on DC. Where does that enter the formula?

LC in what units since V is in a different unit.


So, for complete skinning,

Is that a one skin depth thickness wire?



For double braid coax, the prop is dependent on DC, as you state.

Only on DC?

 

[jneutron]

I was going to edit my other post. Either ran out of time or it isn't working right now:

The velocity of propagation in ANY cable is V = 1/sqr(LC).

That would include the coax and its dependence on DC. Where does that enter the formula?

LC in what units since V is in a different unit.





So, for complete skinning,

Is that a one skin depth thickness wire?



For double braid coax, the prop is dependent on DC, as you state.

Only on DC?

1. Units, shmoonits... gots that at woik, I aint there. If your interested, I'll post it all mon lunch. For a coax, L is strictly based on mu (permeability), and C is strictly based on epsilon (permittivity). both mu and epsilon are actually based on the real free space numbers multiplied by a relative one..a mu of 100 is actually 100 times mu space, and a dc of 3 is actually 3 times the free space permittivity.

2. Complete skinning is a hypothetical max, with the corresponding hypothetical min internal inductance. A wire can only approach it as freq tends toward infinity.

3. For a simple coax with a wire gut, the inductance of the cable is the sum of the inductance due to the magnetic field within the dielectric, and the inductance due to the inner wire. At very low freq, the prop velocity is the same equation, but L has 15 added nanohenries.. At good freq, like video and rf, the inner inductance is ignored as well it should be.

A double braid coax is one where the inner wire is itself a braid. So as a tube of conductor, it has virtually no internal inductance.

Gotta go, I'll be back mon..forgive any typos, I screamed this out without proofing..

Cheers, John.

 

[mtrycrafts]

I was going to edit my other post. Either ran out of time or it isn't working right now:


1. Units, shmoonits... gots that at woik, I aint there. If your interested, I'll post it all mon lunch. For a coax, L is strictly based on mu (permeability), and C is strictly based on epsilon (permittivity). both mu and epsilon are actually based on the real free space numbers multiplied by a relative one..a mu of 100 is actually 100 times mu space, and a dc of 3 is actually 3 times the free space permittivity.

2. Complete skinning is a hypothetical max, with the corresponding hypothetical min internal inductance. A wire can only approach it as freq tends toward infinity.

3. For a simple coax with a wire gut, the inductance of the cable is the sum of the inductance due to the magnetic field within the dielectric, and the inductance due to the inner wire. At very low freq, the prop velocity is the same equation, but L has 15 added nanohenries.. At good freq, like video and rf, the inner inductance is ignored as well it should be.

A double braid coax is one where the inner wire is itself a braid. So as a tube of conductor, it has virtually no internal inductance.

Gotta go, I'll be back mon..forgive any typos, I screamed this out without proofing..

Cheers, John.

Not to worry about typos, etc.
That LC looked like inductance and Capacitance at first so don't worry about units. Thanks for the rest.

 

[jneutron]

Not to worry about typos, etc.
That LC looked like inductance and Capacitance at first so don't worry about units. Thanks for the rest.

L and C are indeed inductance and capacitance. First, the easy method, then the more rigorous one...

We know that for a constrained system such as a coax, this equation holds:

LC = 1034 DC.... L in nanohenries/ft, C in pf /foot.

Now, lets look at belden type 9222...
L = 77 nh/ft
C = 30.8 pf/ft
published Z 50 ohms
published V 66% lightspeed.

First, calc of impedance.. Z = sqr(L/C)

L = 77,000 pH/ft, C = 30.8 pF/ft..(make both same power of ten)

77000/30.8 = 2500

sqr(2500) = 50. Ok, Z worked..they don't have a gross error in their specifications (I found 3 errors in belden's spec sheets out of 31 cables I looked at)

Now, V = 1/sqr(LC), and V=c/sqr(epsilon mu)...c being lightspeed.

V = c times (1/sqr (epsilon mu))..for coax, this boils down to:

V (% lightspeed)= 1 / sqr(DC)....for example, if DC = 4, V =50%...if DC=9, V = 33%

Here's the easy method: use second equation by calculating the DC. (assume mu =1)

LC = 1034 DC.... DC = LC/1034.

DC = (77 times 30.8)/1034

DC = 2371.6 / 1034 = 2.293617

sqr(2.293617) = 1.5144

V = c times (1/ 1.5144) = c times .66029

V is 66% lightspeed, just what belden posted..

((Edit: The significance? For ALL coaxes made using that dielectric will have the exact same prop velocity. This is absolutely independent on the impedance of the coax, or it's physical size. All coaxes which are made with an insulator which has that same dielectric coefficient, will also have the exact same velocity of propagation..

This equation is useful for ALL wire pairs...coax is special in that the magnetic field is confined to inside the outer conductor while a wire pair does not do that..so a slight mod is in order.. The DC in the LC=1034 DC must be considered as an Effective Dielectric Constant, or EDC, to reflect the spillage of magnetic field outside the working dielectric. Cable EDC's in the 5 to 10 range are common, and that indeed reflects the prop velocity of those wires.))


Hard way:

sqr(LC) = sqr(77* 10^-9 * 30.8 * 10^-12)

= sqr(2371.6 * 10^-21)
= sqr(2371.6) * sqr(10^-21) = 48.699 * sqr(10^-22) * sqr(10)
=48.699 * 3.162 * 10^-11
= 159.986 * 10^-11..now, this is the denominator of the first eq..

1/(159.986 * 10^-11) = 6.494 * 10^-3 * 10¹¹

6.494 * 10⁸ (remember, this is using feet as origional units, lets convert..

6.494 * 10⁸ ft/sec * 12 in/ft * 1 meter/39.4 inches =

6.494 * 12 / 39.4 * 10⁸ = 1.97786 * 10⁸ meters/sec

Lightspeed is 2.99792 * 10⁸ meters/sec.

so, fraction of lightspeed is:

1.97786 / 2.99792, or .6597, this is 65.97% of lightspeed.

A heck of a lot of calcs to do it the hard way, I prefer the easy method..

Cheers, John

ps..finally fixed all the eq typo's, sheesh..

 

[mtrycrafts]

This was great, thanks very much
When you eat, drink and sleep this stuff day in, day out, it is simple. When it is a small part of a hobby, a part to expand knowledge base, and not used much or at all but once in a while, you add time factor, memory is an issue

 

[jneutron]

This was great, thanks very much
When you eat, drink and sleep this stuff day in, day out, it is simple. When it is a small part of a hobby, a part to expand knowledge base, and not used much or at all but once in a while, you add time factor, memory is an issue

It raises my hackles whenever prop speed, skin effect, and some other stuff gets bandied about with audio wires. It's amazing how far this bogus stuff has gotten since the early days when some guru's first put the words together to explain what was heard. But quite honestly, the topic is NOT an easy one, nor is it understood by many...and the ones who really do know this stuff have no desire to be associated with high end audio.

The term "propagation velocity" applies ONLY to the energy that matches the characteristic impedance of the wire. If you apply a 10 volt square wave to a 50 ohm speaker cable, it instantly draws 10/50 amps, 200 mA. So only 2 watts of actual power travels down the wire at the propagation velocity, and will do so for an infinite time unless something else occurs..that something else is the load impedance. If there is a 5 ohm load at the other end, it will not get 10 volts times 5 ohms when the front gets there, because the charging current is not 2 amperes, but only 200 ma. This difference, for rf, is seen as a reflection which heads back towards the source. In the meantime, the load sees only 200 mA. It's only after this reflection has bounced back many times that the load actually is driven by 2 amperes, and during this timeframe the cable is being charged up to the 2 amperes the load expects.

At audio frequencies, we cannot distinguish this back and forth, but only see the end result. That being a slow rise consistent with the cable inductance, and a cable which has stored energy within itself consistent with the load current and not it's own impedance.

That is why I say the best cable is one which matches the load impedance. That is because there are no reflections back to the source per se, and the cable stores the least amount of energy as a result of no bounce.

Unfortunately most people have no idea how to make a cable which is at or close to the load impedance of a speaker, and many amps become unstable under the lumped capacitance.

Cheers, John

 

[mtrycrafts]

Unfortunately most people have no idea how to make a cable which is at or close to the load impedance of a speaker, and many amps become unstable under the lumped capacitance.

Cheers, John

Again, thanks.
So, by this, I could assume a high capacitance cable would meet that impedance matching at audio frequency?

 

[jneutron]

Again, thanks.
So, by this, I could assume a high capacitance cable would meet that impedance matching at audio frequency?

Yes, within the constraints of Z = sqr(L/C).

For example, I made a coax with a tefzel dielectric (DC = 2.7), it measured about 10 nh/ft and 288 pf/foot.

L C =1034 DC, 10 times 288 = 2880. 2880/1034 = 2.78, close enough.

The impedance :

sqr (L/C) = sqr(10,000/288) = sqr(34.7) = 5.89 ohms.

From the equation format, it's clear that higher cap lowers the impedance. Lets look at a zip cord and 8 ohms.

L = 160 nH per foot.

sqr(160,000/C) = 8

160,000/C = 64

C = 160,000/64

C = 2500

For a zip to make 8 ohms, you need 2500 pf per foot of capacitance..that's a lot.

For cat 5, it's a bit easier. They have 100 ohm impedance pairs already, and the overall impedance of multiples will be quite close to 100 divided by the number of pairs. 4 pairs will give 25 ohms, 8 pairs 12.5 ohms.

Cheers, John

 

[mlhm5]

Cannot Go Wrong With Blue Jeans Cable

First, coax is far better at conducting and shielding than non coax. Hardline is even better but difficult to work with.

Blue Jeans uses Belden 1694A and Canare RCAP-C53 a true 75 ohm RCA plug.

If you want flexible, the Belden 1505f is the same as 1694A but the center conductor is stranded.

Unless you make them yourselves, expect to pay around $20 for a 3' cable. You can get them here
Ram Electronics or here Blue Jeans Cable Belden 1694A or Belden 1505F both come with Canare RCA connectors.

 

[jneutron]

Canare RCAP-C53...a true 75 ohm RCA plug.

How it that considered 75 ohm? As far as I can see, that requires negative permittivity for that design.

I have a suspicion it is not, regardless of the ad copy.

Cheers, John

 

[mtrycrafts]

How it that considered 75 ohm? As far as I can see, that requires negative permittivity for that design.

I have a suspicion it is not, regardless of the ad copy.

Cheers, John

If I remember correctly, you found one that is 75 ohms. Which one? Just curious

 

[stratman]

Ahhhh, you guys make Mr.Spock so very proud (really, even for me, just a caveman, it's quite interesting)

 

[jneutron]

If I remember correctly, you found one that is 75 ohms. Which one? Just curious

WBT.

To raise the impedance of any line, you either raise the inductance, or lower the capacitance. With an RCA, the geometry is defined by history, and this determines both the inductance and the capacitance.. The RCA requires reducing the capacitance to below free space capacitance, requiring a relative dielectric coefficient less than 1. (said negative before, my bad.. )

WBT took a different approach (sadly, 31 days before I did). They raised the local inductance by splitting the outer shell into two parts. This has the side effect of spilling magnetic field outside the shield, compromising the shielding at the connector a tad. This should not be an issue, however.

It must be noted that this approach requires that a mating WBT connector also be used. If not, the shield is re-constituted by the mate connector, spoiling the spillage and returning the connector to its too low impedance.


Cheers, John