Measuring Cable Performance &
Correlating Results with the Listening Experience
technical article by TARA Labs designer Matthew Bond.
There is an increased awareness among audiophiles as to the importance of cables in the sound of an audio system. It is a subject that has been surrounded by controversy, in part because many feel the differences to be either too subtle to be audible, or too system-dependent to hold any universal truth for buyers of audio equipment.
In fact, it is possible to make measurements of different audio cable conductor designs that will correlate with audible differences in the cables' performance. Moreover, with these measurements as a learning tool, one can begin to distinguish conductor designs which are linear and accurate as opposed to designs which soften, brighten or otherwise color the sound.
In 1988 TARA Labs developed Constant Current Impedance Testing TM (CCZT), a testing method which has been used in advanced university engineering studies to measure cable performance. These measurements provide reliable predictions about the sound to be heard from the changes of cable conductor design and configuration. With CCZT, we have been able to reliably and repeatable correlate the listening experience to the test-bench experience CCZT measures impedance vs. frequency or linearity with frequency. This is both a necessary and important criterion of cable performance because it directly relates to rise time and phase coherency. These two elements, more than any other, correlate directly to one's perception of a cable's sound as either alive,; or reproduced.
In CCZT testing we use conductor runs of equal mass (i.e. same D.C. resistance) but varying conductor shape and arrangement. They are set up in a test jig having the same parallel configuration between the send and return lines. This methodology accurately compares the design qualities of the conductors themselves while keeping all other factors identical.
· Single 2mm (14 gauge) round conductor: Upper bass and mid-range are warm. Treble is soft and rolled
·· Two 1mm (14 gauge) round conductors: Upper bass and mid-range are cleaner, with better definition. Sound is more natural and coherent. Less roll-off in high frequencies.
= Two 1mm (14 gauge ) rectangular conductors: Upper bass and mid-range are more vivid, palpable and live sounding. The sound through the mid treble and upper frequencies is extremely coherent and natural. Overall, the natural Harmanic structure of the music is more accurately revealed.
With even a rudimentary under-standing of the principles of cable design it's possible to make good predictions about the sound of a cable just by examining its internal structure. In Part Two, we'll examine why various conductor configurations yield the differences in frequency linearity (and therefore, sound) demonstrated here, and what to look for when comparing cable designs.
The testing methodology for CCZT is relatively simple to duplicate. Contact SARA Labs for detailed instructions on seeing up a test jig and sample data from TARA Labs in-house testing.
In Part 1, we measured the frequency linearity of various cable designs using TARA Labs' Constant Current Impedance Testing (CCZT).
Why do different conductor types of the same mass yield such different results? In a few words: electromagnetic flux linkage.
Referring to the graph of the CCZT results, we see that the single 2 mm2 (14 gauge) conductor shows the least linearity with frequency. This is because in a larger single conductor there is more electromagnetic flux, which increases in density towards the center of the conductor. This crowding, or density of the electromagnetic lines of force at the center of the conductor effectively chokes off higher frequencies and forces them to travel towards the outside of the conductor.
Any compact or uniform shape increases the tendency of the whole conductor to have greater density in the coupling or linkage of electro-magnetic flux. In this diagram, a stranded conductor shows the same tendered to roll off high frequencies as a single solid conductor of the same mass.
An important note: this is true whether the conductor is a single solid-core or a stranded conductor of the same conductive mass or DC resistance. A large diameter conductor, whether solid-core or stranded, will have the same impedance vs. frequency curve for a given diameter and mass. In other words, the closely bundled small conductors in a multi-strand conductor approximate a single large solid-core conductor, so nothing is gained by stranding many smaller conductors.'
In the second trace, we have split the single conductor into two smaller ones. Combined, they have the same mass, but the frequency linearity is improved because of their smaller individual diameters and lower electromagnetic flux linkage. Although the conductors are subject to flux linkage because of proximity, they have the greater frequency linearity that goes with a smaller diameter. This is the principle behind many of TARA Labs' Prism TM Series solid-core cable designs.
In the third trace, the Rectangular Solid Core conductors still have the same mass but their frequency linearity is improved further. This is because the rectangular conductor has less coupling of electromagnetic flux at the center of the conductor. Due to its shape, there is effectively no "center" to speak of.
What to look for, then, when choosing cables? A design with thinner conductors in a more open configuration will yield cleaner, clearer and more frequency-linear sound. One with a single, large conductor or a bundle of smaller conductors will yield sound that is smoother and rolled off.
These guidelines hold true regardless of variations on these design themes and account for most of an audio cable's sound. Other elements, such as dielectric and conductor material and treatments, are the icing on the cake of cable design, having a lesser effect on cable performance than good, solid design principles. In the next article, we'll begin to examine those issues to shed some light on their relevance to audio cable performance.
All designs have the same conductive mass, but frequency linearity (i.e. a cleaner, clearer sound) will improve from left to right due to conductor size, shape S arrangement.
In Part 2, we discussed the conductor's own inductive reactance and its effect on the sound in an audio cable. In this installment, we'll examine conductor materials and treatments, as well as dielectric materials
and their effect on the sound. Although it's important to note that these factors have a lesser effect on the sound than the design of the conductors themselves, when the conductor is more linear with frequency, these minor differences in materials do become more apparent.
The two most common conductor materials today are copper and silver. Is one inherently better than the other? Not necessarily. So much depends on the purity and treatment of the raw conductor material. The treatment process known as annealing softens and purifies the conductor material, affecting its specific resistivity. Proper annealing of copper conductors increases conductivity (lowers specific resistivity) by increasing the length and size of the crystals within the material. This results in fewer electrical discontinuities in the conductor, removing the distortion, brightness, or hashiness from the sound.
Conductors must also be properly designed to deliver maximum frequency linearity with any given material. The mathematical formula shown above shows a direct relationship between the diameter of the conductor and the specific resistivity of the material. We see that, for a given conductor material, there are different frequency response curves and different linearity with frequency. The sound of a properly designed and Created Conductor is open, neutral and extended, yet smooth and without grain. Conductors which sound harsh or bright have not been properly designed, or treated, or both.
Insulating materials exposed to electric fields are called dielectrics." Dielectrics are necessary Components in any cable because they prevent oxidation and keep the conductors from touching one another.
In audio cables, relatively low voltage and current levels mean
that dielectric strength is not the most important factor. Far more significant
in its effect on the sound is a material s dielectric absorption. This characteristic
describes the way a dielectric may discharge a secondary signal into the conductor
out of phase with the audio signal.
Audio Signal creates an electro-magnetic filed around the conductor.
Dielectric material absorbs energy and releases it back into
the conductor out of phase with the audio signal.
As a current is passed through a conductor, an electromagnetic field is created which interacts with the dielectric material and temporarily displaces the molecular structure. If the dielectric material has good elasticity and can return quickly to its normal state, then the material is said to have low dielectric hysteresis or loss and will have little audible effect on the signal.
Dielectric materials, then, sound different because of the different rates that the materials store and release energy at different frequencies. PVC, a common dielectric material, causes distortion and coloration mostly audible in the mid-bass and mid-range frequencies, whereas Teflon causes distortion in she lower treble frequencies, making coloration less noticeable.
TARA Labs uses a proprietary dielectric material called "Aerospace Polyethylene TM" or "Aero-PE." This material is chemically treated to have low dielectric absorption and high dielectric elasticity. Therefore, it reacts less and returns more quickly to its neutral state, making it more sonically neutral than other materials. Aero-PE is also extruded at a lower temperature than other insulating materials. Copper conductors insulated with Aero-PE are not exposed to high heat and therefore retain their specially annealed qualities.
In comparing lesser quality cables, you may never hear the difference between PVC and PE insulation, or hyper-pure vs. low grade copper. The limitations of the design itself will obscure these subtler effects. However, with high-quality cable designs, one can more readily hear the differences in materials, proper annealing and good quality insulation.
In Part 4, we'll examine the cable as an interface having the properties of a second-order low-pass filter.
In the last three installments, we examined the aspects of cable design which have the greatest effect on performance: conductor design and arrangement, and dielectric design and materials. We also showed how the differences between cables can be measured and correlated with the listening experience.
In this installment we'll discuss the cable as a second-order low-pass filter and examine the subject of neutrality in an audio cable.
With its series inductance and parallel capacitance, an audio cable is a simple second-order low-pass filter. By reducing the inductance and Capacitance, we car, increase she bandwidth of the cable and extend the cable's frequency response.
Within its frequency response an audio cables' capacitance remains fairly constant with frequency, but the inductance in the cable varies with frequency. This is due to the inductance being dependent on the diameter (or shape) of the conductors and the configuration of the conductors within the cable. These changes in inductance cause audible differences which will be different with different component output/input impedances. When the electromagnetic field (inductance), which varies with frequency, interacts with the electrostatic field (capacitance), this causes different electrical resonances and filtering effects within the cable interface. Depending on the diameter and configuration of the conductors within the cable, the amount of inductance will vary considerably and the sound will be audibly affected.
In Part Two, we examined how to reduce the series inductance by having
smaller conductors, and having them in a more open arrangement. To reduce the parallel capacitance it is simply necessary to space the positive and negative conductor runs further apart. Increasing the cable's bandwidth in this way improves the cable's linearity with frequency and ensures less system interaction because of the reduced electrical characteristics of the cable.
The ideal audio cable then, has low series inductance (smaller conductors or rectangular ones with less inductive reactance) and low parallel capacitance. In this way it has very high bandwidth and is then less system dependent.
The reduced system interaction of the cable created by increased bandwidth will yield an important quality we call neutrality. Most would agree that the ultimate system is one which brings us as close to the experience of the original musical event - whether that was in a studio or a live concert setting. We want to hear the music as it was recorded: nothing left out, nothing added. Neutrality then, not alteration or coloration is an important quality in audio cable performance.
Cables are the only component within an audio system that can be designed to be completely neutral. Every other component, by its very nature, alters the signal in some way. The theoretical ideal of an audio cable is one with zero series inductance and zero parallel capacitance. In this way the cable has unlimited bandwidth and is also not system dependent. TARA Labs cables, designed against this theoretical ideal, are designed to have the lowest LCR specs and widest bandwidth on the market.
If neutrality in your audio system is important to you, begin to narrow down your cable choices by starting with those that have the lowest LCR specs. (Any reputable manufacturer should be able to supply these figures and to explain the method by which they were obtained.) Then use educated listening techniques to determine which of the cables sounds best in your system.
This philosophy, by the way, is not universally endorsed by cable manufacturers. There are other cables on the market which are designed to act as Tone Controls for the system. They use networks, filters or additional elements that are meant to somehow improve the audio signal by altering it. In fact, these do nothing more than impose upon the system someone else's idea about how the music should sound. The result is contrived and artificial, rather than transparent and neutral. Above all, it will most likely produce colorations that destroy the natural, musical reproduction you've tried so hard to create.
Achieving complete neutrality in an audio system may be an impossible dream. After all, we are reproducing a musical event, not experiencing the real thing." But the pursuit of this ideal is a worthy one, primarily because it puts the emphasis where it belongs: on the music.
Parts 1-3 of this series appear in Stereophile Vol. 19, Numbers 1, 3 & 5.
Extended Warranties Available!
We now offer 90 Days Same As Cash (SAC) 365 days a year !
Come in and ask us about it.
Our Business Hours are:
Monday - Friday 11am to 7pm Pacific Time
Saturday 12pm to 5pm.
We are closed on Sunday!
ORDERING BY PHONE
If there is no authorized dealer in your area you can order directly from Soundscape with a 10-day, satisfaction, money-back guarantee.
US and Canada, 1-800-450-HIFI (1-800-450-4434)
Outside the US and Canada, (707) 578-4434
Monday - Friday 11:00 AM - 7:00 PM (PST)
Fax: (707) 578-4436
Last Update: February 8, 2012
1044 Fourth Street, Santa Rosa, CA 95404
Phone/Fax 707-578-4434 or 800-450-HIFI