Paul's Posts — 01 February 2012
By Paul McGowan
The debate over long interconnects and short speaker cables vs. the opposite rages on. My opinion is unwavering and counter to the rule of thumb: long speakers cables and short interconnects are best as power amps have a better ability to drive a complex load than does a preamp.
Your mileage may vary and most cable manufacturers will tell you exactly the opposite.
Here’s the deal: what works best in your system? Despite my first statement, I actually do the oppositie in my reference system: long balanced interconnects and short speaker cables. Why? Because that’s the most practical application in my room.
There’s optimum and there’s practical. Either can work, just make sure you optimize the components to fit the situation. A rule of thumb is just something we use as a starting point, not an end dictate.
Do what works.
Forward to a friend and help us engage more readers
Related Articles
Share
About Author
(9) Readers Comments
Leave a Reply
You must be logged in to post a comment.
Gannon
But…if it WORKS, Paul…what will we tweak?!
heh
Sorry, couldn’t resist.
I have only balanced output preamps in my two main systems…hopefully they can drive more than a meter of interconnect. Now I have to try it both ways to hear any difference. Damn.
Cheers!
hahax
I use short interconnects, long speaker cables too. In my case it’s partly because I use Fulton Golds given to me a long time ago by Bob Fulton and they were always long cables because Bob believed in modules of just over 14 feet for reduced reflection of something I can’t recall. Mine are over 28 feet. But they are arrainged in a way to reduce the high inductance of the Golds.
Because of the length I have my speakers in one room and electronics in another. And because of the massive size of the Golds I don’t need to worry about high resistance with that length which could mess up frequency response especially with tube amps. The only thing I need to worry about is tripping over the monsters.
Soundminded
I think one difference between sound systems assembled by audiophiles and sound systems assembled by engineers is that engineers understand the functional purpose of each element in the system and its role in the overall context of what the system is intended to do. Wire had never been considered a control element of a sound system until around the 1980s when Monster Cable realized there was a potential market that could be convinced and tapped at substantial profit. Soon after many others followed. Electrical engineers may have forgotten that wire’s characteristics and its behavior are very well explained by the Telegrapher’s equation. This equation has been used for well over 100 years to predict how various wire constructions will affect a network with excellent correlation between theory and measurement. The mainstream wire and cable industry is neither stupid nor insensitive to the needs of audiophiles and sound engineers (Belden is the world leader and world standard in the wire industry.) The industry has devised low cost solutions that are so effective they’d been taken for granted for decades until a market was found for a solution looking for a problem to solve, one that largely didn’t exist except in the minds of those who were sold on it.
Wire is intended as a passive means to connect one point in a circuit to another with as little effect on whatever is being passed through it as possible. Therefore the valid test of a wire judged by that criterion is to compare it to a shunt, an ideal wire. It’s small wonder then that people who manufacture, sell, and buy audiophile wire reject this notion, they have to convince a market or themselves that it is otherwise.
When “using what works” is taken out of context, it can lead to solutions that are expensive, unpredictable, ineffective, and a poor choice. Engineers look at a problem and know that there may be more than one path to the same goal. Their objective is to provide not only solutions to defined problems that they know will work but the ones that are most reliable and cost effective. For that reason, the optimal way to alter the frequency response of a sound system is by means designed for that specific purpose, not by experimenting with wires. The sources I use for wire for my sound systems are The Home Depot, Radio Shack, and more recently The Dollar Store. For their purpose these products perform their function at a more than satisfactory level. They are in lengths I use them indistinguishable from a shunt. I also use relatively short signal cables and long speaker wires since it is the low level signal cables that are most prone to induced hum. The low source and load impedances of amplifier output stages and loudspeaker systems makes them less prone to induced noise than higher impedance low level signal wire.
Gannon
So, if the particular shunt you use induces some odd behavior…in the complex interaction between devices…then the test will be invalid.
The assumption only holds if you merely consider the static resistance, capacitance, and inductance measurements, right? What of dynamic performance, it is assumed to follow those parameters?
Not sure that is the case.
I’ve been around a lot of wires…and it seems that the dialectric material chosen for the ‘simple’ task of insulating the copper makes the MOST difference in accurate sonic transfer, especially regarding transient behavior. Beyond wire purity, winding/braiding, and added electrical circuits in funky boxes that make the wires nearly unmanageable.
What are the applicable static measurements that could explain the variance in performance of these dialectrics?!
The shunt test would ignore too much to be of any use, I fear. But I’d still try it…
Cheers,
John
Soundminded
“So, if the particular shunt you use induces some odd behavior…in the complex interaction between devices…then the test will be invalid.”
Since a true shunt bypasses the cable under test, any odd behavior eliminating the cable’s effect is inherent in the equipment itself and was “cured” by the cable. In that unlikely case the incompatibility between the two connected units should be identified and a suitable inexpensive LCR network used as a substitute. In more drastic cases an isolation transformer or buffer stage may be more suitable than a mere wire.
“The assumption only holds if you merely consider the static resistance, capacitance, and inductance measurements, right? What of dynamic performance”
I have never heard of “dynamic” resistance, capacitance and inductance,” Could you cite a reference giving examples and how that is determined? Analysis of dynamic performance of a system including its network parameters customarily relies on fixed values of impedance parameters. If the parameters change as a function of current and voltage level, the system is not linear and all bets are off. Some argue the point that large capacitors are not linear for signals at very low voltages across them and rationalize using small bypass capacitors in shunt. I’m not aware of any evidence for this but I suppose it is possible.
“I’ve been around a lot of wires”
So have I. I’ve been directly responsible for the purchase and installation of well over one million dollars worth of wire of all types for projects I’ve worked on, possibly over two million. These include large communications networks and power distribution networks. My views stem in part from advice from the factory reps and distributors of many major wire manufacturers including Belden I dealt with.
“I’ve been around a lot of wires…and it seems that the dialectric material chosen for the ‘simple’ task of insulating the copper makes the MOST difference in accurate sonic transfer, especially regarding transient behavior.”
Wire is manufactured to generic IEEE standards. Those standards began many decades ago when it was called the IRE (Institute of Radio Engineers.) Large manufacturers specify their products in great detail, both in their electrical characteristics including lumped sum equivalent parameters and physical characteristics. In the 1980s when the National Electrical Code changed requiring low smoke emitting insulation products to be used in installations where wire and other materials are located in the return air plenums of HVAC systems, materials like halar and teflon began to replace PVC. Professional installers invariably use Teflon because they don’t want to make a mistake that would cost them a great deal of labor to remove the wrong material and replace it. I don’t have comparable cost data now but in the 1980s teflon was 3x as expensive as PVC. I’m not aware of any differences in electrical performance of equivalent wire manufactured using one dielectric versus another although some say Teflon is superior to PVC as a dialectric. Have you got any data for that?
Audiophiles often advance their arguments in qualitative terms rather than quantitative terms. For example they cite digital jitter induced by some wire but never take into consideration that even at its worst it may be below the threshold of hearing or that a digital signal may be re-clocked in a subsequent stage eliminating jitter altogether. I’ve heard countless rationalizations for exotic wires and they may be constructed so strangely compared to their common substitutes that they will sound different, which does not necessarily mean better. In at least one famous case, a manufacturer’s speaker wire was so bizarre with presumably high shunt capacitance that it actually sometimes formed a tank circuit on the output of some power amplifiers resulting in blowing them up. If you know some repair technicians who worked in the 1980s and early 1990s they could probably tell you some good horror stories about it.
omniclassic
When I’m out on location doing a recording, the rule rather than exception is; I’m using long microphone cables. 25 ft minimum, sometimes 100 to 150ft. Micorphones put out a very small signal, ususally less robust than the output of an audio pre-amp feeding a power amp. That being said, for the few feet of interconnect, or speaker cable in our systems, I don’t think it makes a lot of difference as to length. Granted balanced may offer a little more protection that simple shielded two conductor, but signal integrity will seldom suffer. Some of the best amps and pre-amps made don’t even offer balanced in/out as an option, negating the use of such cables, yet manage to sound wonderful.
Soundminded
As I understand it some recording studios install preamplifiers right at the microphone output and some even D/A converters. That should eliminate all such cable related problems. The cherished vinyl phonograph records made decades ago were produced without such benefits or what would have been for that time extreme measures.
Gannon
I never said there was such a thing as dynamic resistance, capacitance, or inductance. You read that into my words. I said those values were STATIC measurements, and if you don’t consider that you’re flowing DYNAMIC alternating current with extreme transients and subtle interactions of musical and non-musical harmonics, then you’ve wasted MILLIONS of dollars buying wire.
Simple as that. Congratulations!
You are just like the other half-minded engineers I’ve known in my life.
If you cannot hear the difference between similar wire with various insulations, it is only because you never tried. Or, you simply might not have the capacity to perceive it fully. Or, you listen to other incorrect engineers and have speakers which will not ever resolve it.
Sorry. Trying to give you the biggest lattitude for doubt.
It is too obvious for me to argue, these are my experiences actually listening closely to real-world applications.
I know of the wire that induced instabilities in amplifiers…it was a complex braid…cannot remember the brand. I never sold it…never had much chance to audition it. Big deal, one stupid design doesn’t discount the entire genre. Unless your mind is already closed and you are merely scrapping for things to add to your mental barricade!
Cheers!
John
Detroit
Soundminded
A course in the basiic theory of AC and DC electricity strikes me a good investment of time and money.
It’s one thing to install Rube Goldberg speaker wire that blows up some amplifiers and interconnects that create their own peculiar distortions, those are in power limited circuits under 50 volts and are judged as not particularly dangerous. But I’ve seen and read postings of after market and diy home brew power cords made from coaxial cable, cat 5 telephone wire, unconnected grounds, and ferrite rings around ground wires that are not only not UL listed but present an electrocution hazard to those who come in contact with them when they are plugged into a wall outlet. When someone does that they assume the full onus of responsibilty for any resulting injury or accident. Whatever the supposed audible merits of these projects they seem to me to be a very foolish risk.