[jms007bnd007]

I am wondering what kind of speaker cables I should buy for my system. I don't think I need to spend much, since my components aren't the best, but I don't want to spend all this money on the cd player and speakers and have it go to waste because I didn't spend enough on speaker wire. So anyway here is what I have, or is in the mail:

---Denon DCM-370 CD player

---My parents' old Sony integrated, but I'm planning on getting a NAD or Rotel integrated some time this year

---Axiom Millennia M3ti bookshelves

Oh, also I was just gonna use Radio Shack IC's with this as well...so if you think that would be a mistake let me know.

Thanks for any help

 

[Sol_Zhen]

Kimber Kable 4PR wire is pretty cheap, and it improved my setup's sound.

http://www.kimber.com/4pr.htm

 

[TaffyGuy]

try to find cable that has 4 insulated stranded wires inside the same sleeve. its kind of hard to find, we got some on a tip from a no-******** pro audio guy and did testing on it, and in all seriousness it outperforms all but the most expensive wires. ($800 range) for about 50-75cents a foot.


.___ . ___
|O O| |+ -|
|O O| |- +|
.--- . ---


EDIT: jesus ascii art is hard when formatting changes between screens. oh i see they take out extra spaces. obnoxious.

looks like that on the end. then you wire opposite wires together to make + and - . what this does is make tons of + against - surfaces which maximizes capacitance in the cable. speaker cable is a big deal because good amps have like 0 impedance output, so any small impedance on the cable can screw up the load essentially screw up the "speed" or "control" the amp has on the drivers. impedance of transmission lines are something like (L/C)^(1/2), so the idea is to maximize C (capacitance) to make the impedance 0.

sounds crazy but its true. we got a 200 foot spool of the stuff for 100 bucks or something and have never been able to find it again, but i'm sure its out there. probably applications are for wiring 2 speakers in one sleeve, but it makes fantastic speaker cable that happens to be far less expensive than marketed stuff.

sounds crazy perhaps, and surely no one will listen to me, but its true. =)

 

[qwerty870]

Right now a component upgrade is much, much more important than speaker cable in your system. For now get some 18ga Radioshack speaker cable and run two lengths to each speaker to reduce resistance. That should be more than fine for now.

The Kimber 4PR is a little better, but will cost you more although its still extremely inexpensive compared to most speaker cables. Since I just got a pair of Harmonic Technology Pro-11s I will be selling my Kimber 4PRs. If you are interested I have an 8ft pair I would sell for $15 plus shipping.

If I were you I would get the Radioshack cable for now though, because it'll cost about $8 for a 50 foot roll and sounds passable.

 

[jms007bnd007]

Thanks for the suggestions. I think I'm gonna go with the 4pr. With that cable I will be able to upgrade the amp and other components and not have to upgrade the cable as soon as I do.

But what about the IC's? Do you guys think the Radio Shack one's will do alright? Or should I look into some Outlaw or Kimber IC's?

 

[carlo]

personally i think 4pr blows, but thats just me. qwerty's price offered is a steal but if money is that tight you may as well save for your amplification stage and not look at cabling until you can get it right.

for the best value speaker cable it's chris vh's cat5 receipe. in no way hard to make but nonetheless time consuming. worth it though, i've had a lot of speaker cables come through here and the only stuff thats beaten it across the board are empirical's stuff and kimber's select (we're talking expensive)... haven't built jon risch's stuff but its said to improve on the cat 5 sonics. search aa for info if you're interested. if you don't want to use your hands or just want a cheap comercial cable the place to start would be audioquest's type 4 or 6 (avoid the rest of their line though).

for interconnects once again diy is where its at. if you want to go radioshack gold series you'll benefit from the shielded design and will sound better than the freebie cables, but there's some wireworld ics for under $100 thats almost decent, kimber pbj w/ wbt0147 if you can jump up to a bit over 3 figures (avoid the ultraplate terminated stuff). or you could just go outlaw - i haven't heard them and not planning on it, but of course some here seem to love them (note: outside of head-fi i haven't heard anyone tempted by them in the least, so keep the grains of salt nearby).

 

[Sol_Zhen]

I like the 4pr, but it will probably depend on the gear. The place I got the 4pr from would accept a return anytime within a month if I decided I didn't like it. And, it was cheap.

Whatever you get, at least get 14 gauge. 18 gauge is ridiculously small.

 

[Helmut]

TaffyGuy, are you talking about CAT 5 cable? Looking at your diagram I think this cable has the same structure. Yes,no.

 

[TaffyGuy]

cat5 is just ethernet cable, no? very good engineering for the cost, which would make it good speaker wire. heard small mentions of that, but never tried it out myself. this is much heavier duty.

what i'm talking about would be like taking 2 monster cables and putting them in a big sleeve, then... doing what i said before. if anyone is interested i can put a picture of some or something. it *really* is good speaker cable, much heavier duty than cat5, although that would be neat in making diy speakers, just plug it in like ethernet. i can dig it.

 

[yage]

Isn't the wire you're describing of the Litz variety? In those designs, the four conductors are also braided to reduce inductance. AudioQuest Type 4 is a good example of this kind of design.

As for suggestions on speaker wire, why not try some Radio Shack 12-gauge 'megacable'? You might be surprised at how well it performs against the 4PR or the AQ Type 4... I certainly was...

 

[TaffyGuy]

looks like there is some wire called litz with similar design with 4 cables, but it doesn't have to be braided around a solid silver core or any voodoo like that. just 4 stranded coper conductors that you wire diagonally to make + and - which creates 4 capacitors instead of 1 with regular lamp wire style. therefore you get 4x the capacitance, which pretty much does the trick, for a fraction of the cost.

really fancy wire like coax style with huge core and shielding outperforms it for a $100 a foot or so. if you can get your hands on some wire like this, it could really save a bundle. you can still ask and i'll post pics if you'd like. you could probably even diy this kind of thing easily, and even improve on the design if you wanted to use fancy braids or something, but doubleing up the wire on each speaker gives you so much more capacitance which takes the impedance of the line down to 0, and that is really -all- speaker cable needs to do.

i agree that expensive, pretty wire would sound better to me too, cuz i'm quite partial to pretty stuff. kimber for one is very stylish. this solution is just for practicality/cost performance.

 

[yage]

Yes, my mistake, the Litz wire is entirely different from what you were describing. I was confused because AudioQuest calls their design 'Hyperlitz'. However, the AQ Type 4 cable does use four insulated solid core conductors inside a PVC jacket, which equates (according to them) to 14 AWG.

About capacitance - I'm not sure that it plays a large role in the sonics of a cable. The difference between capacitance and inductance in speaker wire is usually about three orders of magnitude. So even a quadrupling of capacitance will have a miniscule effect on total impedance, assuming that the wire can be treated in terms of lumped circuit elements. The assertion that speaker wire is somehow a transmission line (which, I believe, is how you obtained your formula) is entirely baseless. The transmission line model would be valid in the case of a 10 ft. length of wire when you're dealing with very high frequencies - in the RF range and above - such that the reactance dominates the impedance and the wavelength of the signal is much shorter than the length of the line.

 

[TaffyGuy]

umm, well i'm not sure what else it is other than a transmission line, and you can clump all passive componants in terms of resistance, capacitance, and inductance; these componants interacting in different ways that can have different effects on signal.

i'm also not sure there is much else to go for in this model besides high capacitance, assuming you have a good connection. i don't think interference has as much play in a line running many volts as it does in lines before the amplification stage, and since phase change across the audible spectrum has... well no effect on sound, the main worry in transmission is any impedance added which will significantly distort the signal. this impedance is miniscule to start with, especially since the wires are short, as you said, but if someone can hear the difference between 2 cap's in the power supply, i bet they can hear the difference when you half the line impedance, by using this method.

cat5 wire is similar i think becuase you have 8 wires, or 4 twisted pairs, that i assume you wire all together, alternating + and -, which, when all in a row, would also create 4 capacitors in the line. the only difference is that the wire i speak of, where rather inexpensive, is much thicker and more sturdy, and would be better considering the high voltage output to speakers.

this is all i know about the subject. if you know something else, in all seriousness, i am very curious about it, so please tell me. this isn't meant to be arrogant or anything; i am a student in engineering so obviously this is of interest.

 

[yage]

You're an engineering student? Cool... let's fire up some brain cells..

Alright, let's first agree on a model of a transmission line. Most are only a series inductance with a shunt capacitance, correct?

Not entirely.

In the real world, there's also a series resistance (the resistance of the wire itself) and a shunt conductance as well (the current flow within the insulation, I believe). So a more accurate formula for the characteristic impedance of a transmission line would be:

Z = sqrt[ (R + jwL) / (G +jwC) ]

If you have any EE profs you can get a hold of, ask them about what order of magnitude these values would be (most likely, per foot) for a transmission line.

Then we can really start an analysis...