Speaker Cable Length Differences: Do they matter?

[jneutron]

The velocity of propagation within a cable is not the speed of light, but rather, the spare root of the dielectric coefficient times the relative permeability of the media.

For cables, that is roughly 1/2 lightspeed.

However, the only thing that moves at the inherent propagation velocity of the cable is signal WHICH HAS THE VOLTAGE TO CURRENT RELATIONSHIP WHICH IS CONSISTENT WITH THE CHARACTERISTIC IMPEDANCE OF THE CABLE.

This means that for a standard zip cord with approximately 150 ohms of impedance, only signal with 150 volts per ampere will travel at the prop velocity. Since the load has an impedance of about 8 ohms, this means that a 150 volt signal will hit the load, but only 8 will get to the load, the remanining 142 volts are reflected to the source.

The settling time for this system is the important thing, not the length vs prop velocity..

Coiling a zip cord does not introduce further inductance. It is only when the two opposing conductors are seperated will coiling alter the inductance and therefore the coupling to external time varying magnetic fields.

Cheers, John

 

[JerryLove]

The velocity of propagation within a cable is not the speed of light, but rather, the spare root of the dielectric coefficient times the relative permeability of the media.

For cables, that is roughly 1/2 lightspeed.

That seems to be choosing on the low end (http://en.wikipedia.org/wiki/Velocity_of_propagation)

 

[jneutron]

That seems to be choosing on the low end (http://en.wikipedia.org/wiki/Velocity_of_propagation)

Actually, no.

Oh, and yes I forgot to say it's 1/sqr...sorry bout that..

Here's a chart of some cables.. half lightspeed is an effective DC of 4, note that most of the cables exceed that quite a bit.

Cheers, John

ps...oops, I forgot that I am no longer able to provide any jpegs to this website. The highest EDC was 14, most were over 4.

 

[JerryLove]

Actually, no.

Do you have a cite?

 

[jneutron]

Do you have a cite?

Yes, I do. The information was gleaned from either the first or the second speaker cable faceoff within this website (edit:I checked, it was the raw data from the second face-off article.). Gene had sent me the information to review, and I did some additional analysis of the measurements to determine the effective dielectric coefficient (relative permeability times relative permittivity). There were 11 cables that Gene tested.

To correctly determine the effective dielectric coefficient, use the equation:

LC = 1034 times EDC.

L in nanohenries per foot, C in pf per foot.

Note that this equation breaks down to LC = 1034 DC when the cable is fully constrained, such as a coaxial cable or a parallel plane T-line with very high aspect ratio (wide and not very thick).

Also note that if you use this equation to look at belden coaxial cables, you will find that about 5% of their datasheets on the web (as of about 4 years ago) contain errors in either L or C. I ran 34 of their cables to determine the dielectric coefficients of their inner core insulation, and found the errors.

ps..this equation can be arrived at by taking the equation for capacitance within a coax, the equation for inductance within a coax. equating the ln(A/B) term within each, and re-arranging the variables. It boils down to the relation I give..

Cheers, John

 

[mtrycrafts]

The velocity of propagation within a cable is not the speed of light, but rather, the spare root of the dielectric coefficient times the relative permeability of the media.

For cables, that is roughly 1/2 lightspeed.

However, the only thing that moves at the inherent propagation velocity of the cable is signal WHICH HAS THE VOLTAGE TO CURRENT RELATIONSHIP WHICH IS CONSISTENT WITH THE CHARACTERISTIC IMPEDANCE OF THE CABLE.

This means that for a standard zip cord with approximately 150 ohms of impedance, only signal with 150 volts per ampere will travel at the prop velocity. Since the load has an impedance of about 8 ohms, this means that a 150 volt signal will hit the load, but only 8 will get to the load, the remanining 142 volts are reflected to the source.

The settling time for this system is the important thing, not the length vs prop velocity..

Coiling a zip cord does not introduce further inductance. It is only when the two opposing conductors are seperated will coiling alter the inductance and therefore the coupling to external time varying magnetic fields.

Cheers, John

Ah, so you are hanging out here

I have a question for you since I called you someplace else and you missed it
Stranded cables that have a tube in the center, non conductive, insulator type, allowing some mag field in there, will this decrease inductance or no difference compared to no such tubing in the center?

 

[jneutron]

Ah, so you are hanging out here

I have a question for you since I called you someplace else and you missed it
Stranded cables that have a tube in the center, non conductive, insulator type, allowing some mag field in there, will this decrease inductance or no difference compared to no such tubing in the center?

Within a tubular conductor, there is no magnetic field created by the current which travels along the tubular conductor wall.

The internal inductance of a cylindrical wire is 15 nanohenries per foot. If you take say 100 strands of wire to make a tube, the basic internal inductance of the conductor will be 1/100th of the 15 nH, or 150 picohenries per foot. (note that this is a rough explanation, the lowering of inductance is a result of the reluctance path relative increase and proximity cancellation of the magnetic field between the many wires.)

The external magnetic field (hence the external inductance) will be unaffected by the geometry of the inner conductor as long as the current density is uniformly distributed among the wires. With reasonable insulation thicknesses, the wire pair will have somewhere between 150 and 180 nH per foot, so the 15 nH drop is not that much.

Using two tubular conductors arranged as a zip will not act the same as two full cylinders however. As proximity effects begin to occur, the current centroid will move together faster than a solid wire set, so at higher frequencies, the inductance will drop faster. And, the resistance will climb faster.

Cheers, John

 

[highfigh]

Looping cable actually doesn't change the properties (namely inductance) significantly unless you have a numerous loops tightly wound.

The point of the article was that cable lengths do not need to be kept identical so long as the lengths are reasonable to begin with. Obviously you don't want the left main speaker cable to be 10ft in length and the right one to be 500ft, but having one cable 2-3 times longer than the other makes little difference. If your cable runs become excessive (greater than 50'), it is recommended to use lower gauge cable (10AWG or less) to minimize losses.

If you can find out who gave the 2005 CEDIA Audio Measurement and Setup seminars in Indy, I think someone (maybe with the proper test equipment and his own website, like this one) should inform him that it doesn't make an appreciable difference if the speaker cables are the same length, or not. He made it very clear that it's his opinion they should all be the same length in a high-end system.

 

[jneutron]

That seems to be choosing on the low end (http://en.wikipedia.org/wiki/Velocity_of_propagation)

Oh, thanks for the link.

One of these days, I'll have to correct that wiki entry. It has Vprop as 1/sqr(dielectric constant), where really it should be 1/sqr(relative permittivity times relative permeability), but only for material which has no bulk conductivity. Lenz's law (a consequence of Faraday's law applied to a bulk conductive media) bucks the propagation.

I see that it does include the 1/sqr(LC) term...so if one desired, one could simply plug in the L and C of the measured cables in the faceoff 2, and eventually back into the effective dielectric constant...(edit: but this requires getting the units correct in the calculation.) But using the LC = 1034 EDC is far more easy to do either in the head, or on the back of an envelope,(edit: as it requires only the relative values of permittivity and permeability, and ignores the free space values.)

edit: coiling a coax or a zip will still not increase the bulk inductance. The conductors must be seperated for inductance to be altered in that fashion.

Cheers, John

ps..of course, prop velocity is being normalized to C = 1..

 

[teresa]

Cable guages

Thanks for the useful post gene. If 14/4 cable is equivalent to 11 AWG cable, what AWG is 16/4 cable equivalent to? And what would be the maximum recommended length for use in a moderately priced whole house audio system for 16/4 cable?

 

[Fedaykin]

Cable length is irrelevant to timing. Electrons in a wire are not like mail messages on the road. Individual electrons do not carry a message from your receiver to your speakers.

Information is passed along a wire by inducing a current (moving electrons) along the entire length of the wire simultaneously. Think of it like this: If you had two garden hoses in a big loop filled with water. On one side of the loop your have a pump. On the other side of the loop you have a wheel that is spun by water moving through the hose. When you turn on the pump, the spinning wheel will move instantly. It doesn't have to wait for water to be "processed" by the pump.

The actual speed of the current depends on the resistance of the wire. In a typical house current travels only a couple meters a second -- so if electrical signals were like mail on the road, it would take several seconds for your light to turn on when you flipped the switch -- which of course doesn't happen (barring lights that have to warm up first of course)

 

[jneutron]

Electron speed is proportional to current density. For room temp wires, speeds in the mm/sec area are closer. Meters per second would mean the wire is way too hot. The hyperphysics website has a calculator for this.

The water example is incorrect. For small loops it will indeed look instant, but it actually cannot exceed the velocity of sound within the water.

Same with cables, the speed at which the information flows can never exceed the velocity of propagation of the cable.

And, it is only information at the characteristic impedance of the cable which travels at the prop velocity. That is not the situation with a speaker cable.

jn

 

[Got Hi-Fi?]

as my amps have no inductance networks in the output, length definitely matters. Too short and my amps will overheat. For most mass market produced hi-fi, it really makes no difference. So it really comes down to what equipment you are running.

 

[highfigh]

Wow. Just wow.

 

[Speedskater]

as my amps have no inductance networks in the output, length definitely matters....................

I wouldn't talk about them, I would just quietly sell the amps on e-bay.
They are also probably ringing or even oscillating at some very high frequency. (up to 5MHz)

 

[highfigh]

Cable guages

Thanks for the useful post gene. If 14/4 cable is equivalent to 11 AWG cable, what AWG is 16/4 cable equivalent to? And what would be the maximum recommended length for use in a moderately priced whole house audio system for 16/4 cable?

You mean 14/4 or 16/4 connected as two conductor, right? When the cross-sectional area of the conductor is doubled, the gauge increases by three, so doubling 16 gauge becomes equivalent to 13 gauge.

According to most cable manufacturers, 16/2 is fine for about 50', barring excessive power/current applications.

 

[Otto Plyot]

http://www.roger-russell.com/wire/wire.htm

 

[clnconcpts]

I was just wondering about this today . I see both sides of the argument. One part of me wants to believe it doesn't matter if they are slightly different in lengths , but then another part of me says if they are not , there will be a time difference

 

[lovinthehd]

I was just wondering about this today . I see both sides of the argument. One part of me wants to believe it doesn't matter if they are slightly different in lengths , but then another part of me says if they are not , there will be a time difference

The time difference is undetectable, tho. Unless you're simply OCD, then it matters

 

[Pogre]

The time difference is undetectable, tho. Unless you're simply OCD, then it matters

Or if you can hear at near the speed of light, lol.

I believe it's meaningless, but I made the cables for my front two the same length anyway...