Speaker cables and their frequency responses

Did i hear you say "WHAAAAAT ?!?!?!" when you read the title of this post ?

Take a look at an article on Nelson Pass's website. Do yourself a favor though and open up the link below in a seperate "window". This way, you can switch back and forth between that article and what you're reading now.


Bare in mind that this article was researched and probably written back in 1979 and published in early 1980, so some of the names and models have changed.

In specific, take a look at what Nelson has labled as "figure 4" on the second page. If you study that chart, you will see that it has both a frequency range across the bottom and a resistance scale across the side.

There are two important deductions that can be derived from this chart. One is that some cables are notably lower ( the old Fulton Gold and Monster Cable ) than others ( like the 24 gauge zip cord ) when it comes to "series resitance". A cable that is higher in series resistance ( like the 24 gauge wire ) will act as a "buffer" between the amp and speaker ALL the time. This means that the amplifier will have less control over the drivers but at the same time, the amplifer will see less reflected EMF ( electromotive force or "voltage" ) if the speaker is highly reactive. The longer the run of this cable, the bigger the "buffer zone" i.e the less control of the drivers due to even greater signal loss going out to the speaker and / or reflected back down the line. This can be either a grave disadvantage or benefit depending on what one is trying to achieve. Most folks would consider this "bad", especially when trying to reproduce something accurately and using speakers that are not highly reactive.

The other important "deduction" one can make from this chart is that signal transfer is NOT linear from cable to cable across the frequency range. As can be seen, some are FAR flatter than others. While some have a relatively straight line with a rise at the top end, others are only "flat" within the bass range and then flare up drastically as frequency rises. However, these "curves" are NOT the frequency response of these cables.

If you take the results of the series resistance curves and invert them, THEN you'll have a basic idea of what the frequency response / power transfer characteristics of that specific cable is. In other words, instead of a rising high end, you get high end roll-off. Keep in mind that this is with a relatively benign ( non reactive ) load applied. Using a typical dynamic driver speaker, the results could be far more exagerated. Under extreme conditions, some cables with very specific electrical characteristics coupled with highly reactive loads can create "bad" sound and even hazardous electrical conditions for the amp that is trying to load into them.

As can be seen, the Fulton and Monster are basically only good for woofer or subwoofer operation whereas some of the others ( Polk, Mogami, etc..) are quite excellent up to appr 10 KHz or so. THOSE cables and cables of similar design are the ones that you want to investigate. That is, IF you want "accuracy" over a wide frequency range.

If you think back, didn't your ears tell you that the bass was a LOT more solid when you switched to "Monster" ??? The fact that you lost detail and high frequency information didn't seem quite as noticeable compared to how much more impact, "warmth" and "smoothness" the system picked up. While a cable of this type might be good in a system that is overtly bright, hard and glaring i.e. using low quality digital recordings and cheap cd players, it is of little to no use in a full range HI FIDELITY system.

On the other hand, some of those "wide band" cables achieve their phenomenal results at the expense of simplicity. Some have very complicated geometries ( Polk ) and require stabilizing networks ( zobels ). Due to the cables high capacitance and extended high frequency capabilities, some amplifiers ( typically high quality, wide bandwidth models ) can go into oscillation and destroy themselves. This was a BIG problem back then. Goertz cables are prime examples of this type of design in a modern day product. Fortunately, the previous problems with high capacitance speaker cables was recognized and we now have the know-how to correct such situations.

Of course, there are current designs that try to achieve the best of both worlds i.e. low inductance and "reasonable" capacitance. Some that come to mind are various designs by Kimber, Audioquest, Axon, XLO, etc... This is not to say that all of these cables are created equal or use similar designs. Some offer excellent electrical characteristics but may be able to be improved in terms of their physical geometry and how they come terminated from the factory.

The point that i was trying to make out of all of this is that:

A) The audible effects of cables was well documented over twenty years ago. Both Nelson Pass and J. Peter Moncrieff /IAR published VERY similar reports and findings way back then.

B) The differences in cables ARE real. As such, one should strive for finding cables that are the most "benign" or "neutral" in terms of electrical characteristics and frequency response while offering the lowest series resistance possible.

C) Choosing cables to "correct" problems within a system should be a last resort. Get the basics ( selection of components and matching of the speakers to the room ) right and then go from there. In plain English, the components must work together and not leave massive gaps between them. Fine tuning a system with cables might be able to fill in "small gaps" but should not be done to cover "gaping holes" or "canyons".

Hope this explains some things and helps someone out. It can be quite confusing trying to understand the situation that you may have been dealing with. Sean
There were bunch of articles/discussions on that topic and I hope you're familiar with it. It is well known that most of the companies do not use engineering and measurements to design cables rather than seeking how to make an attractive view to hook up rich and illiterate home-theatre or stereo hunter.
No matter what price you've paid for the cable -- still longer cable will have larger capacity, inductance and active resistance(which you can neglect in the most cases).
In most cases today's amps can drive any capacity loads which means that only Inductance is our enemy?
Not at all!

The main component in implementing a good quality speaker/interconnect cable is a conductor.
Every conductor more or less has inner and surface impurities. Inner impurities are "born inside" crhrystalic malstructures that tended to have a semi-conductor structure. Surface impurities are simply oxides. Silver conductors have a little-less impurities per unit than cooper. The minimal number of impurities has gold, lead, platinum and other color metals.

The first objective is to use the conductor with minimal number of inner impurities. If there were no inner impurities on our widely used silver and cooper conductors, they would be a goldworth. There is no way to get rid of all of them.

The surface impurities are only concearn when we want to establish connection. That can be avoided by gold plating and soldering peeled conductors into gold-plated connectors.

The displayed ideal cable equivalent circuit is not quite real since somewhere within you can plug in an equivalent of diode(one way conductivity) which can represent a conductor's impurity.

Cardas manufactures it's own cooper or silver conductors and tests them. Most of the cable companies do not manufacture their conductors and ordering them from the other vendors. Do you think they check the conductor's purity?

I wish I could test myself to compare 2 conductors one from RadioShack and the other one is stripped from Transparent audio cable! I do believe that RadioShack wire will be either the same or even better purity...
So what kind and lengths of speaker cables and ICs do you use, Sean?
I have TONS of different cables within the multiple systems that i'm running. As a general rule, i try to keep both the IC's and speaker cables as short as possible within reason. Many of my IC's are "homebrew". Some of the "commercial" models are made by Magnan, FMS, Tara, Cardas, Goertz, Kimber, etc... Some of my speaker cables are "homebrew" along with commercial models from Goertz, Kimber, XLO, Axon, etc... Power cords range from ( once again ) "homebrew" to commercial models from TG Audio, CPCC, Cardas, etc... Several of my components have been rewired internally with Cardas, some with Kimber, some with Axon, some with Belden, etc.... Some of these were chosen for very specific reasons whereas some of them went in simply because that is what i had on hand and seemed to work the best there.

As you can see, i have a real hodge-podge of cables and wiring. Then again, every system and interface from component to component is slightly different, so i adjust accordingly. I am not a believer in using one brand of cable throughout a system as i think it tends to flavour it too strongly in one direction. While finding what works best where is a LOT of hassle, quite time consuming and requires one to have a LOT of cables on hand to choose from, it is what i have found to work best. Like anything else though, things change and so do the components and cables within these systems. One of these days though, i'll stop "playing around" and just enjoy the music.... Sean

Electrical properties and their relationship to treble rolloff are fairly obvious, and well-elucidated again, Sean (thanks); indeed Monster sounds bass-heavy only because one increases the gain to average out the rolled highs...hence a "big bottom"! What I find more mysterious are the cables that seem to have a slightly RISING treble response AND a laid back midrange (like Pro-Silway MkII in my system).
Any explanation for a "u" shaped curve?
I imagine that so-called "uptilted" cables like Nordost are simply flat and neutral, but they of course sound leaner than other good cables that simply offer slightly more treble roll. And maybe that the relative cleanliness of this unrolled treble is what distinguishes the best flat cables from the worst.
Is it that we all simply have differing preferences for spectral tilt?
Was it in Alton's tome that I saw that the general public prefers a -2dB per octave rolloff; -1dB for musicians and engineers; and 0-0.5 dB/octave for audiophiles?!
Why do I prefer the fuller mids of Discovery Essence over Red Dawn and Pro-Silway? Are the Essence warmer up top? Don't think so. Could the RD and Silways actually be "u"-shaped?...or is the Essence (being totally copper) SUBTLY
rolled off?
Whaddaya say, Sean? Ern
Ernie, i too have experienced this with various cables. I think a LOT of this has to do with specific cable / component interfaces. While the sonic traits of a specific cable seem pretty consistent, i've heard them do a "chameleon" and change their color under certain circumstances with different components. How to explain this, i don't know. I guess that's why we keep buyin' and sellin' : )

Honestly though, ( and i'm striclty guessing ), i'm almost wondering if the combination of the input impedance of one component, the output impedance of the other component and the various electrical characteristics of cables doesn't form some type of "notch filter". As your probably aware, filters can be narrow or sharp depending on their Q. If the cable has several "impedance bumps", it may even produce several smaller, narrow bandwidth "filters". The various cable and component combos would offer different "filters" due to multiple impedance variations. This might explain the inconsistent reviews and attitudes that some people develop towards specific cables.

As to cables that sound lean or bright, If you've never checked out a GOOD cable burner ( commercial models or homebrew ), i would HIGHLY recommend it. The difference that this makes on most cables ( especially those that you consider to be hard, bright or glaring ) is rather immense. I had some cables that were being used in a system for about two months at about 4 - 6 hours a day. They always sounded splashy, tilted up, etc... After appr two days on my burner, i could not believe the difference. The "experts" that i've talked to about the burner that i have told me "30 days does wonders". While that may be true, i've always been too "excited" to let any burn that long. Technical explanations or not, i AM a "believer" when it comes to cable burn in. Sean