I'm trying to find frequency response data on speaker cables in different price classes. Basically, I want to see how much difference in frequency response is there between a 10$/meter and a 1000$/meter cable.

This is a joke, right? As far as I know, the response of even the cheapest speaker cable is flat to a fraction of a db over the audible range. About the only characteristic of speaker cables that has any significant impact on sound quality is resistance.

Capacitance or inductance can matter too, but only at long distances (10+ m).

Really, at audio frequencies, whatever measurement you get will be equivalent to the theoretical values for an equivalent RLC network, so just get the RLC values for the cable, get the source and load impedances for your amp/speaker, grab SPICE and go to town.

Since the RLC is distributed I don't think it is quite that simple. You would need to treat it as a transmission line if the speaker impedance is anywhere near the characteristic impedance of the cable, which it is probably not. Even ignoring the C and treating it as a RL circuit is tricky because the inductances of the two wires are coupled.

The frequencies involved are way too low to be worrying about treating the cable like a transmission line.

My rough calculations give appriximately one microhenry per meter. That equates to 0.125 ohms per meter of inductive reactance per meter at 20 kHz.


Also, the inductance of the $10 per meter cable and the $1000 per meter cable are going to be almost identical.

This isn't enough information to compel one not to use a lump-element model.

I agree that you won't be able to hear any difference... You might be able to hear a difference with 1000 feet of cable.

You can't know the frequency response without knowing the characteristics of the amplifier and speaker. For example, if you design a crossover network to work with an 8-ohm drivers, it won't work properly with 4-ohm drivers.

Even with an amp and speakers connected, the difference is probably too small to measure. If it could be measured, the high-end speaker-cable manufactureres would publish it. Mostly, you just read about "openness" or "transparancy", or some other audiphile marketing hype.

And, if you've ever looked at a (passive) crossover network, you will see large capacitors and inductors. These are required due to the low resistance of speakers. You don't get inductance and capacitance anywhere near that "accidently" from a speaker cable. The combination of low resistance and low (audio) frequencies means that you need high series inductance and high parallel capacitance to have an effect. When you get into high frequencies (megahertz), small inductance & capacitance values can have an effect.

BTW - Capacitance is created by two insulated conductors near each other. You can greatly decrease the capacitance by simply "un-zipping" and separating the conductors. It follows the "inverse square law"... If you double the distance, the capacitance will drop by a factor of 4... If you increase the distance by a factor of 10, you decrease the capacitance by a factor of 100, etc.

There are no similar simple way to decrease inductance. And as pdq noted, there is no simple way to the cable manufactures to do it either. Inductance is created by the magnetic field generated by current flowing through a conductor. You can multiply inductance by forming the wire into a coil, and increase it further by winding the coil around a magnetic core.

While I'm on the subject, the third factor is resistance, and that's determined by the wire material, the diameter (bigger diameter = lower resistance), and the length (longer = higher resistance). You can get slightly lower resistance by using "ultra pure" copper, but its far more effective to increase the wire diameter (use a lower gauge).

True, for cylindrical radiation patterns. With planes (i.e., sheets of metal adjacent to one another) the rule is simply 1/r, where r=distance.

Not that it matters one whit, and I wouldn't dream of unzipping my speaker wires.. All that matters for speaker cables shorter than a half mile is the gauge of wire and hence the resistance per unit length. All the hype about expensive cables is absolute, unadulterated trash.

While low current (higher impedance) interconnects are somewhat more sensitive to potential interaction, I believe that all the expensive cables in that arena are just as useless and demonstrate no measurable superiority whatsoever. That market, IMHO, is the opportunistic selling to the gullible.

If you can't measure something, you can't engineer it, either. That does NOT mean that your ears may not detect a difference; it only means that if they do, you cannot say what or why that difference truly is and thus cannot "design" anything to accomplish your goal.

Based upon my last post, I really think you should consider values closer to $1/meter. Seriously.

h(t) = delta(t)

It's worth noting that when you look at high-frequency stuff (Mhz-Ghz) you will see differences. This is what PS Audio has been harping about recently. But they never actually proved that it's audible, nor that it significantly effects anything at audio frequencies - especially frequency response.

Actually, inductance is a logarithmic function of the ratio of the conductor spacing to the conductor diameter. This means that if you "un-zip" the conductors you are actually increasing the inductance, but by a far smaller amount than the inverse square change in capacitance.

Here is a link to a graph and a comparison that might help:

http://www.empiricalaudio.com/frclarity7_e...erformance.html

To me, that page is a good argument for why the standard 11 gauge cable is perfect. In the frequency domain, it's 3dB down at 238kHz - which is 210kHz above any frequency which could be considered "audio". Even choosing the 1dB point, the cheap cable is flat to 70KHz.

The time domain measurements they present just reflect the frequency domain effects. Phase response is similarly comical - 5 degrees at 20Khz is approximately equal to moving the speaker or your head by 0.23mm. Unless you are re-arrange your listening room when the temperature of your house changes, you probably don't need to worry about that. As for their resonance measurements, the first product is at 18MHz!

I also wonder why they did all of this with simulation, and not measurements. The test equipment needed to do these types of measurements is not particularly expensive.

The lossless cable to connect the test equipment might have been too expensive.

Did you catch how they specified the output impedance of the amplifier (0.2 ohms) and not the impedance of the speaker? The effect of reactive impedance when R = 0.2 is vastly greater than when R = 4 or R = 8. As small and significant as the claimed differences are, they are also completely bogus!


Obviously I meant INsignificant (why can't I edit my post)?

Well, actually it wasn't a joke. A dumb question maybe

It's just that I a couple weeks back I saw a frequency response graph for different guitar cables, and the cheapest ones did have fluctuations of about 1-2 dB. Because I only LISTEN to music, I wanted to know if there is a same kind of difference between speaker cables.

And I'm not an audiophile. If the 1$/meter cable has 99% same frequency response with the 1000$/meter cable, I'm happy. Placebo effect is real but I'm not paying extra for it

Placebo effect is real, but any particular effect is only inside your head. Control that and you can have what ever you want from a $10,000 expenditure, or a $1 purchase. Let someone else control it and you make them a $9,999 profit.

Guitar cables are different for a number of reasons - the most obvious is that small changes in cable impedence can change the sound of the guitar because the output impedence of the guitar behaves completely differently from an amplifier, and has a high gain amplifier at the end. For a good example, try picking up a guitar cable and dropping it on the ground while the amp is on - you'll notice a pretty loud sound (and get kicked in the crotch by the owner of the cable). Try the same with speaker cables and you will hear very little. This microphonic effect is only one of the reasons that guitar cables can be harder to choose than speaker cables.