What is 'Dielectric Propagation Velocity'?

Q.) Dear George, what is 'Dielectric propagation velocity'? To our common knowledge, no signal goes in dielectric material, meaning that, there would not be propagation velocity in there.

Please explain.

Thank you,

Sheen

A.) Hi Sheen - Great question! This is a key to why Clear is a break through cable.

There is no current flow in dielectric materials, but the change in voltage is passed along in the dielectric - much as the dielectric charges and discharges in a capacitor.

The rate of this charging and discharging is what determines the velocity of propagation in a cable. Solid dielectric materials cannot pass the charge as fast as the conductor can. This is why the VPROP rating of cables is always a percentage of the Speed of Light, rather than the Speed of Light, which is the propagation velocity of copper or silver.

In other cables, the discharge of the dielectric lags the changes in the conductor, causing a smearing of low level information in the cable.

Other Cables do not compensate for this differential in the cable itself. Some try to compensate with networks but the damage has already been done. Once the low level information in the signal has been smeared, it is lost.

Cardas matched propagation conductors maintain low level information, and natural transients like no other cable can. Matched propagation conductors do not exaggerate the leading edge of the signals or smear low level information as conventional conductors do.

Thank you,

George

Q.) In a standard Golden Reference Speaker Cable, how many discrete conductors are there? Frank

A.) There are 68 X 12 conductors or 816 in the cable and 408 per polarity. - George

Q.) Dear Mr. Cardas, may I ask a question? Quadlink interconnect capacitance is higher than Golden Cross, but in the case of loudspeaker cable the Golden Cross capacitance is higher. How do you explain? Omer

A.) What you say is true. It is because the cables are designed as a matched set. Which ever storage medium is dominant establishes the "Q" of the cable. In interconnects, the capacitance is the most relative because signal is transferred as voltage fluctuation with little current flow. Speaker cables transfer signal as current and the dominant storage is inductive. Capacitance tends to mirror inductance so in a set you may see capacitance drop in the interconnect as it rises in the speaker cable. George

Q.) I'm thinking of upgrading some of my Cardas interconnects from Quadlink to Neutral Reference. I was wondering what the cable impedances are for the Quadlink interconnect, Neutral Reference interconnect, and the Neutral Reference video. -Eric

A.) Nominally Quadlink is 50 ohms and the Neutral Reference cables are 75 ohms. More to the point, because audio is not setup around matched impedance (it is a rising impedance system) and home audio cables are far to short to be included in a transmission line loop, these cables are optimized for constant impedance direct transfer. - George

Q.) I noticed you are one of the few high-end audio, cable companies to use ferrite in your cables to deal with interference. I've read a few times, on different forums, that the use of ferrite rings and ferrite snap-ons has a negative affect on the way an interconnect sounds or the way a cable sounds for that matter.

I own a pair of your IC's (without ferrite rings) and have listened to a few others. I must admit, you have some of the finer cables out there. What's your take on ferrite? Does it change the way cables sound, or is it a certain kind of ferrite that is negative? - A. Wynn

A.) The use of ferrite is usually limited to power applications. We do, however, use a specialized ferrite to control shield resonance in the Quadlink interconnect and it has proven effective. In general I avoid using ferrite signal applications. - George

Q.) Is there anything you do differently in the design i.e. geometry, materials, gauge, etc. of your phono interconnects versus line level interconnects that would address the fact that the cable is seeing a much smaller and more delicate signal from the cartridge? Should I gather from your website that your Reference Series interconnects are designed to perform equally well without compromise to these low level signals as they would with the CD player I've hooked up to balanced Neutral Reference? Sincerely, Shawn H.

A.) Hi Shawn. The Phono cable is for a spacific application (DIN to two RCA) and is in fact diffrent than the line level in many ways; they are lower capacitance and contain both channels in the same shield to reduce RF and noise and are terminated in a DIN plug that connects directly to the tone arm. However, in the Reference series specifically I believe in using two cables. Standard pair is the best setup if it is possible. - George

Q.) Hi, I own a pair of Crosslink interconnects, Crosslink speaker cables, and a pair of Quadlink 5c interconnects.  They all are broken in and they sound fantastic with my Heresy II's.  I took a course in Audio Engineering so I have a keen interest in audio tech stuff.  Would you please explain what "Constant Q" means in your cables?  Does it relate to the ratio of reactance to resistance.  Does it relate to the Q (bandwidth) in graphic and parameteric equalizers, or do you have a different meaning for "Constant Q"? Thanks, Brian

A.) In a solid conductor the center has a lower “Q” than the surface. The signal is transmitted on the surface first.  In a constant “Q” conductor we wind and scale the conductor strands to smooth out the difference between the surface and core of the conductor. Constant “Q” conductors have a very constant transfer impedance (change with frequency is extremely low) and the over all transfer rate of the constant "Q" conductor can be adjusted to match the propagation limitations of the better dielectrics. The Constant "Q" conductor is far more constant in the audio range and inherently cancels it's own out of band resonance. Thanks, George

Q.) Would you be kind enough to provide Capacitance (pF/m) and Loop Inductance (micro H.) figures for your interconnect and speaker cables?

I would like to give you an example: Van Den Hul the Second can work properly as an interconnect even though its measured resistance and inductance (1110 ) would be unsuitably high for a speaker cable.

I am seeking interconnects with low Capacitance and speaker cable with minimum inductance. The old Kimber PBJ interconnects had 51.4 pF/m Capacitance as compared with 217.1 pF/m for Wireworld's Eclipse interconnects. Nordost reports 70pF and 0.17 H. for their Valhalla interconnect.

Thank you in advance for your time and attention in this matter. Sincerely, Christopher

A.) No Problem Christopher. The cap and inductance figures are on the site at http://www.cardas.com/content.php?area=products

For interconnects, Neutral Reference an Golden Reference (7pF/ft for XLR) are about as low as it gets.

Neutral Reference ( .034 uh/ft),  Golden Reference and SE9 are all similar in inductance. George

Q.) I've never tried changing the power cords and I'm pretty sceptical about it. Could you please try to explain in engineering terms how changing a power cord to a source component (CD player) which has an internally regulated power supply, running in a fully balanced system can change the sonic character of the system?

I could understand if there were circulating ground currents between the individual components in an unbalanced system, but in a balanced system ground currents shouldn't be an issue unless the system had insufficient common mode rejection or wasn't truely balanced in and out. Thanks and regards, Sam F.

A.) Sam, You are on the right track with the ground currents. Most cords and power conditioners choke the line current which is useless because the power supply almost always has a transformer on the input that limits bandwidth quite well. Further slowing of the current input is counter productive. Grounds on the other hand are a real issue. The house ground is well too far away to provide any real ground as related to RF or digital frequencies or sharp transients circulating in the system. The wave length of these frequencies is far to short to find a proper ground path so they find paths to circulate in the system as a whole, bypassing transformers because of their inherent asymmetry. This is a multi path condition that cannot be common modded because it is the interaction of multiple path lengths, like a ground loop, but at much higher frequency. This problem can be so bad in shops and studios where there is a lot of digital equipment, that the equipment actually malfunctions entirely. My chords have a substantial cancellation of trans audio frequencies in the ground plane, they are also effectively longer than the path to house ground and in most cases this essentially stops the components from talking to each other if they each have a cord.

I did substantial testing using microphones and microphone power supplies that are very sensitive to this problem. Replacing the power cords is not a total solution but short of battery isolation it beats most everything I have tried, short of lifting all the grounds, which can have side effects far worse. The amount that a power cord can help varies with prevaling conditions. Very simple, single point ground, tube analogue systems are a far better situation to begin with than a broadbanded solidstate digital system. Hope this helps. - George

Q.) Someone, on one of the forums, suggested you are working on a new IC which they termed Musical Reference, which will be between NR and GR. Any truth to this? - Jay

A.) I am working on a new interconnect, but it is to serve the pro musicians. I am not sure how it will fit into the home audio situation. Certainly it will work very well in some applications, but it will not be a balanced cable and I don think it exactly fits between NR and GR. - George

Q.) Is Skin Effect, resistance, a substantial factor in audio cable?

A.) At audio frequencies the effect is quite small. The effect of cables relates far more to resonance than resistance in any form. It is interesting to note that self inductance actually goes down with frequency in conductors, because of "skin effect". At video frequencies, solid core conductors begin to behave like a tube, because skin effect attenuates core flow. Silver plated copper works in video cables, basically because of skin effect. Video problems pertain more to loss than with resonance. Skin effect makes video conductors behave like thin walled tubes, thus they have less self inductance/resonance and more loss.

I know that this can seem convoluted at times, but it is interesting to note that solid core, round, OFHC is the conductor of choice for radio transmitters. In audio systems the whole cross section of the  conductor is in play at all times, and system impedance scaling is optimised for low loss. Here skin effect resistance is hardly a factor. Self inductance and ring (resonance) are the problem as I see it.

Q.) How are your standard XLR/RCA adaptors wired? - Tanaka

A.) Standard XLR/RCA adaptors are wired with pin 2 hot, pins 1 and 3 grounded. We can make custom adaptors with any pin configuration if your amplifier requires something different. - Brian

Q.) George, How is a constant impedance achieved with Neutral Reference interconnect? Brian

A.) There is no such thing as a constant impedance cable. Inherently, the impedance of all cables varies with length and rate of change. The rising impedance standard in audio exaggerates the effects of cable impedance. To compensate for this effect we make the characteristic impedance of our Reference cables quite high and wind the conductors in the cable to match the propagation velocity of the dielectric. This combination smoothes the transfer of information in the cable and reduces the effects changing impedance. It dramatically reduces out of band resonance. George