PCRECORDING.COM - - Cables Matter Wally
I have had numerous discussions with people about the effect different cables have on the recording audio chain. As one might imagine, opinions vary as do particular loyalties. I contacted Phil Tennison at Marshall Electronics, distributors of Mogami cables for North America, and told them I wanted to do a non-scientific article on perceptible audio quality differences due to cable changes. They arranged for a shipment of some Mogami Neglex quad-wire microphone cables and a set of 1/4" TRS connector cables for me to test.
Of particular interest to me was the effect of the microphone cables on the signal chain. As I indicated above this was not to be a scientific analysis of cables, rather, it is one of perception. My system is a Celeron 366, Win98SE, 128meg of RAM, a 20Gig Maxtor Harddrive, a set of Marshall 2003 microphones, Folio Notepad mixer, and Vergence Audio M-00/S-00 powered monitors. For these tests, I decided to use my Digital Audio CardDeluxe, the quietest card I have in my studio.
My first challenge was to set up a system where I knew I could get consistent recordings when comparing recordings. I figured that the only fair way to compare the effect of the cables was to setup everything as consistently as possible, with a known source signal. In addition, I wanted to do some live recording with the same instrument but with different cables. I also wanted to keep the recordings simple so I could more easily keep track of changes in tone. I settled on a simple recording of a preset actual drum track sample from the Drag and Drop Drummer and then my guitar and flute.
I had a few expectations about where I might hear more significant differences, particularly with the microphone cables. So, this is where I began. I have general duty microphone cables I have purchased over the years. Since microphones operate at such low signal levels, I expected that a change in cables might have a more significant effect. I ran a drum track through my monitors and miked it into the soundcard with my general duty mike cables. Then, I did the same thing, at the same settings but with the Mogami cables. This was my first set of test tracks. Next, I recorded my Gibson Gospel and bamboo flute with the general duty cables and then the Mogami cables. I took special care to play at the same levels and effort on each track. I now had my second set of test tracks.
I then did A/B tests on the results. With respect to the drum tracks, the Mogami recordings seemed louder with greater presence particularly in the high frequencies, cymbals etc. I was able to detect a slightly tighter bass drum and a longer more subtle tail on the resolution of the cymbal tones. The snare drum was crisper, with more of the "bite" of the drum stick striking the surface of the drum. The guitar and flute recordings were less dramatic but had more "space" between the tones. There was more air in the signal and a greater presence in the treble portions of the guitar and an almost overcrispness to the aperture noise from the flute. Keep in mind, these differences were not "OH MY GOD" different but definitely noticeable and appreciable. Of course, perception is never objective either.
My next test was changing out the cables between the monitors. I used the same recordings I mention above and switched out the Mogami cables with my standard issue speaker wires. The change out was not initially obvious to me as I was not able to detect much difference between the cables. After repeated listenings over time though, I began to detect very subtle differences in tone. Overall the Mogami cables were more pleasing, less tiring and shared greater subtleties of the underlying recording.
Gee Wally, better cables sound better! Truth is, I could hear differences between the cables I used, particularly with the microphone cables. Huge differences? No, but noticeable and worth the testing I did. I am no wire technician and so I asked Phil Tennison to write an article to help explain why cables matter. His excellent article follows below:
Thoughts on cable
I guess you could call me a wire-head. I have also spent countless hours over many years listening to music over fine music systems, only to change the cables around in the system and listen again, to the same music, to ferret out the differences in sound that different brands and models of signal cable caused in the system. Again and again. (Ask my wife. She is a patient woman, often wondering why I care so much about such things. Mind you, she can hear the differences immediately; most women hear better than men anyway, especially in the highs. But virtually all the other people who have done these comparisons with me have been men. A "guy" thing?) But there is a reason sane people would spend valuable time in such an apparently strange pursuit - often these differences we are listening for are not at all subtle.
What do we hear? Often most obvious is an effect like unto an equalizer. One cable causes a system to have more highs, deeper lows, softer, brighter, fuller, the type of thing you could do with a good octave-band E/Q. Then again, sometimes a product seems to bring out more detail, where you can get a better sense of the harmonic envelope of an instrument, or hear more of the reverberation off the walls of the recording booth (or the synthetic recreations of the same). These effects are not particularly difficult to hear, No "golden ears" are needed. Really. Just listen to the music.
More difficult to identify than some specific frequency balance difference is what I call a sense of "coherence" in the sound, that the instruments somehow sound more natural. This can be caused by a variety of aspects of cable design. For example, sometimes lowering the background noise will cause listeners to report that they hear more clarity to the sound. (Substituting a quad conductor for a twisted-pair cable can do this. More about this later). A person may not be aware of the noise itself, but its removal causes the music to sound more defined or clear. A high inductance can cause a cable to be reactive to the load, to roll off some frequencies more than others.
Recording professionals should really care about cabling differences. After all, the cables that are used in every step of the recording chain affect the results of their work. While it seems signal processing is ever more sophisticated by the day, I would like to make the case that keeping the original signal intact for as long as possible in the chain makes manipulations less necessary when you are trying to re-create, and more dramatic when you are creating with effects.
It amazes me that people still occasionally approach me to say that all "premium" cable is a rip-off, and that all cable sounds the same. Often when I ask if they have tried listening tests for themselves, they reply that they donít need to, there is no possible audible difference, and that is that! I must concede that I see explanations of sonic advantages ascribed to some products that strain credibility, which utilize "alternative physics" to explain how some cable product makes better sound. I donít believe in trained electrons that cause different musical notes to travel down separate paths within a cable, or that know which direction to travel in a conductor. (Example: Some cables claim to improve sound by using small conductors for high frequencies, larger conductors for lower mid and bass frequencies. So how do you keep the bass frequencies from traveling down the small "high frequency" conductors? This kind of thinking works great for woofers and tweeters, but not for copper conductors at audio frequencies.)
There are, however, a number of genuine reasons why different cables should sound different in a system, based on genuine established physical principles.
In the first place, a cable comprises a simple electrical circuit. It has inherent resistance, capacitance, and inductance, the same elements that designers of our favorite electronics juggle to obtain the final sound of their creations. These can and do affect the sound of electrical equipment in varying degrees, depending on which frequencies are traveling through the cable and the electrical characteristics of the equipment being connected. The range of signal amplitude varies by a factor of at least 120 dB which makes the system very sensitive to the most minute interactions with the electrical and even mechanical environment.
Choose thy Noise carefully
Lets talk about shielding. Shields keep noise out of a cable, and also keep noise from radiation from a cable. Poor or damaged shields cause buzz, hum, radio frequency breakthrough, etc. The increasingly noisy electrical environment we live in has added to the traditional problems of noises from lighting, appliances, dimmers and the like. Effective shielding is more important than ever. However, the shield can change the sound of the system more than simply blocking noise. Rather than causing electrical interference to somehow bounce off the cable, shielding channels energy to ground before it has a chance to interact with the conductors underneath. Most cables will specify the degree of shielding as a % of coverage, from mid 50% to 100% coverage. This tells us how completely a cable is covered by the material (s), but there is more the story. Materials used have a sonic consequence.
Lots of cables use foil as a shield. By nature foil will be 100% coverage. Foil is a very nice shield against rf (radio frequency) noise, so it is great at keeping the local radio station from breaking through into your system. It also opens up easily when flexed, and is not very effective at preventing magnetic coupling that causes hum in a system. It is used by itself in inexpensive cables, or ones that will be permanently installed and thus not moved. If you want to get a bit exotic (and maybe controversial), foil shields are also poor with respect to eddy current effects. These subtle distortions are due to the fact that electrons have no clearly defined path through the material - they can move in much more complex paths than electrons within a small diameter single conductor. The name is descriptive of eddies which appear on the surface of moving water due to the swirling of the waterís current. Commonly recognized in electro-magnetic behavior in loudspeaker design, eddy current issues are a subtle but significant cause of distortion in signal cables.
Most people are familiar with braided shields, many using "braided" to refer to any all-copper shield. These are also effective at preventing radio frequency noise, and are very rugged when cables will be flexed, stepped on, rolled over, spindled, folded and generally mutilated. These are often combined with foil in coaxial video cables. There is a problem with braid, though - it cannot achieve 100% coverage by its very nature. The best braids have about 95% coverage. Even double layer braids spec at 98%. There are always gaps, no matter how tight the braid. Eddy current problems are inherent here too because of the geometry of the conductors forming the braid.
The last common shield type is spiral shielding. Due to experience with low cost cables with poor coverage, spiral is incorrectly considered by many to be a poorer shield than braid. A quality spiral can achieve 100% coverage, and is the best shield available against hum noise. Hum is probably the most audible type of cable-related noise in a system, so reducing it often makes the most gratifying improvements. Spirals are also the best against eddy current problems. They are made up of many individual conductors. Current flows down the path of least resistance, so even slight oxidation on the surface of the shield conductors causes the lapped conductors to behave as individual paths and prevent electrical crossing across conductors. Spiral shielded cables tend to be quite flexible and maintain shielding in flex. However, nothing in life is perfect, and even the best spiral shields tolerate less abuse than braided shields, and donít reduce radio frequency as well.
A less often seen "shield" is carbon impregnated plastic, seen as a black coating around the dielectric of the wire. This material is not a particularly good shield, but is used in virtually all guitar cables to prevent noise when the cable is flexed. Guitars are high impedance circuits (before the preamp) where voltage is significant but very little current flows. Any movement of the cable actually generates a small signal within the cable itself that is heard as static when amplified by the system. While necessary in these circuits, this material unavoidably adds capacitance to a cable. The added capacitance results in loss of some of the signal we are trying to transmit!
Dielectric - - the plastic stuff that keeps the cable from shorting - - also affects the signals traveling through conductors, and between closely spaced conductors. They are not completely passive, and can store and release small amounts of energy like a capacitor. Some dielectrics perform better than others. Stated as a dielectric constant, this is a ratio of how well the material compares to the performance of air alone. (It turns out that air by itself is the best dielectric, but it is a bit hard to mold into permanent shapes). Teflon is about the best, followed by various polyethelenes, polypropylenes, and pvcís (in that order).
Even if you donít believe they change the sound a whit, the good ones are also much easier to work with because they donít shrink back when you solder them.
This electrical characteristic is a measure of a materialís ability to store an electrical charge. Too much of it in a cable causes loss of high frequencies. The higher the capacitance, the more highs will be lost in the cable. This is most apparent in high impedance circuits (like guitar cables) but affects all circuits to some degree. Less capacitance is always a good thing if truly accurate signal transfer is the goal. Unless you are trying to roll off the highs in the cable. Some people are actually attracted to a "fat" sound in particular cables caused by high capacitance. This kind of system tuning comes at a price. Since this cable - induced equalization will vary depending on the equipment being hooked up, using cable as an equalizer is unpredictable. What may sound OK on one configuration will be completely different in another. You may disregard a terrific piece of equipment because the interconnect caused it to misbehave! Donít put the cart before the horse. Once again, I recommend using processors for frequency changes, and aim for a neutral response in the cable.
Here is one that people sometimes pay too much attention to. The size of the cross section of wire is expressed in gauge. Lower gauge means fatter wire and less resistance, with every three numbers essentially doubling the area of the wire. (So a 13 awg cable is twice as large as a 16 awg). How important gauge is depends on the type of cabling.
Most microphone and line cables will be around 22 to 26 awg. In quality line and microphone cables, gauge has very little to do with quality of sound. Smaller gauge results in a higher decibel loss, but in line and mic cables this loss is linear with respect to frequency (i.e. all tones from high to low are effected equally). It just means that a slider might have to be pushed up a tiny bit more. The worst effect of a bit too small a gauge here is a slightly higher noise floor. If you are going more than 300 feet, look for 22 AWG, otherwise I wouldnít sweat it. Other factors of construction are more important.
Speaker cables are a different story, but people still get carried away by size. You want to minimize the resistance in a speaker cable to keep the damping factor of the amplifier high. The more resistance in the cable, the less the amplifier can control the motion of the drivers. We want the woofers under control so the bass stays tight. But bigger and bigger gauge cable does not keep improving the sound once resistance is low enough. (It has been demonstrated by some that too much conductor size starts to make the sound worse. This is controversial, but easy to verify for oneself by stacking cables with banana plugs). Even if you have 4 ohm speakers, you donít need much more than 14 AWG cable for a 25 foot run. Double that length for an 8 ohm system. Quality and design is far more important.
This is a measure of resistance to current in a cable. Inductance is sneaky, because it causes changes in current to lag behind voltage changes. It can cause a cable to behave like a smaller cable at some frequencies, or to act like a comb filter. Too much can cause disproportionate losses of the midrange and highs. Low inductance is better! (Yet another digression, co-axial and quad wired speaker cables have less inductance than typical parallel or twisted pair type.).
(Very) Basic Wiring
In unbalanced wiring the signal is carried in a single conductor, with the shield carrying the ground/return. The problem with this is that any noise that gets past the shield will be picked up and amplified by the system. In a balanced system, two conductors carry the signal, one positive and the other negative. If a noise is introduced, it affects both conductors equally. The positive and negative portions are then added together at the equipment end and therefore cancel out. (Mathematically N+ (-N)=0).
Instrument and guitar cables are generally unbalanced, so can only stretch a relatively short distance (typically 20 feet or so) before noise starts being audible.
The longer the cable, the more important balanced lines are. The benefits in an electrically noisy environment are often very audible even in short lengths. In professional equipment, always try to go balanced.
Cables matter and do affect the quality of your audio. I would recommend using balanced cables whenever possible. If noise reduction is the goal, use quad cables over twisted pair. Quality spiral shielding is great if there is little likelihood of substantial wear and tear on the cable. (Keep in mind that the rules for line cables and speaker cables are entirely different). Finally, look for hard engineering information derived from actual tests/data before spending a fortune on exotic cables. This is not to say that all more expensive cables should be rejected, only that you be sure of what you are buying before you give up the green.