Technical discussion on loudspeaker cable

[MikeG]

I don't want to open up yet another thread of unsubstantiated opinions on whether this or that cable sounds better. There are already pages and pages of posts like that, and clearly some people are in the 14 gauge zipwire camp and others are convinced they got good value with their pricey stuff.

I'll admit I'm in the former camp, based on measurement. Before hooking up my speakers I tested a bunch of cables from bell wire (yes, solid core 0.6mm) through to QED OFC cable (sorry don't have a part number, borrowed it) on a Wayne Kerr Bridge Analyser, which is good to the last nH and mOhm. The results suggest that the only impedance large enough to have a significant effect is resistance, and that the resistance is exactly what you'd expect it to be given the cross sectional area of the conductor, from d.c. to 30kHz.

I have a degree in Electronics and work in Radio, and hence am well aquainted with impedance matching, capacitace, inductance, cross coupling etc. Skin effect is real and measurable - at radio frequencies. But I can't see how any of these are large enough at audio to impact any half decent amp. If there is nothing measureable at 300kHz (the limit of the Analyser) then it is difficult to argue it makes a difference.


What I'd like is for the *engineers* out there to explain what it is about their expensive cable, in electrical terms, that made it worth the money? How does 10's pF of loading have an effect when the speaker crossover contains 10uF? What circuit simulation could I run that even shows a theoretical advantage?


p.s. The measurements matched the listening tests. High resistance thin cable killed the bass, but QED and 15A ring mains Twin and Earth sounded the same. I used latter, £10 total from Wickes. :-)

[pinkmouse]

Ok folks, before this gets messy, the rules have been set down by the threadstarter. Measurement based opinions only please.

Thanks.

[soongsc]

I keep getting logged off in the middle of typing. Anyone know why?

I got the AlphaCore Python for the following reasons:

1. Solid core minimizes conductor length differences associated with multi-conductor cables.

2. The conductors are closer together than other cables, thus minimizes the magnetic induced current interaction caused by difference in induced magnetic fields caused by opposit current direction.

3. Taking into consideration skin depth at different frequencies, the conductor thicknes is such that impedance is constant throughout the audion range.

I would have gotton the silver version, but did not want to spend such $$$ until I confirm the effects are audible. I did not have very good speakers at the time.

Hope this discussion does not get emotional.

[SY]

The crossover "contains" 10uF, but it's not shunted directly across the amp terminals.

Do not forget that many amplifiers out there which are popular with the audiophile set have distinctly twitchy characteristics regarding source impedance and stability. In the early days of "high end" speaker cable, I remember seeing certain amp/speaker cable combinations show a definite measurable difference involving bangs and smoke. Amplifier nonideality has to be factored into any simulation.

jneutron has some hypotheses about how certain measurable variables can influence localization; I expect he'll be popping in to this thread at some point.

[soongsc]

I don't think any circuit simulation would properly simulate the situation because the models are not perfect, and we always measure electrical properties in a macro level. You would probably need s super computer to simulate at a the micro level if you can get it modelled correctly.

[pinkmouse]

Originally posted by soongsc
I keep getting logged off in the middle of typing. Anyone know why?

Make sure the "Use Cookies" option is turned on in your user control panel.

[macboy]

MikeG,
If you haven't already, read Rod Elliot's article on speaker cable impedance. Lots of technical info and simulation results in that one. Also check out his The Truth About Cables, Interconnects and Audio in General and More Truth About Cables and his other articles. I think that many people look at the projects part of his website and ignore the very informative articles.

[soongsc]

Originally posted by SY
The crossover "contains" 10uF, but it's not shunted directly across the amp terminals.

Do not forget that many amplifiers out there which are popular with the audiophile set have distinctly twitchy characteristics regarding source impedance and stability. In the early days of "high end" speaker cable, I remember seeing certain amp/speaker cable combinations show a definite measurable difference involving bangs and smoke. Amplifier nonideality has to be factored into any simulation.

jneutron has some hypotheses about how certain measurable variables can influence localization; I expect he'll be popping in to this thread at some point.

I think the starter wanted more engineering terms and why we purchase expesive cables. It would be interesting to know specifically which combinations yielded measurable differences so that engineers can try to figure out what's going on that creats such differences.

[soongsc]

This is a site that goes into some mathematics.
http://www.st-andrews.ac.uk/~www_pa/...ect/page1.html

[soongsc]

Originally posted by MikeG


I'll admit I'm in the former camp, based on measurement. Before hooking up my speakers I tested a bunch of cables from bell wire (yes, solid core 0.6mm) through to QED OFC cable (sorry don't have a part number, borrowed it) on a Wayne Kerr Bridge Analyser, which is good to the last nH and mOhm. The results suggest that the only impedance large enough to have a significant effect is resistance, and that the resistance is exactly what you'd expect it to be given the cross sectional area of the conductor, from d.c. to 30kHz.

I have a degree in Electronics and work in Radio, and hence am well aquainted with impedance matching, capacitace, inductance, cross coupling etc. Skin effect is real and measurable - at radio frequencies. But I can't see how any of these are large enough at audio to impact any half decent amp. If there is nothing measureable at 300kHz (the limit of the Analyser) then it is difficult to argue it makes a difference.


What I'd like is for the *engineers* out there to explain what it is about their expensive cable, in electrical terms, that made it worth the money? How does 10's pF of loading have an effect when the speaker crossover contains 10uF? What circuit simulation could I run that even shows a theoretical advantage?


p.s. The measurements matched the listening tests. High resistance thin cable killed the bass, but QED and 15A ring mains Twin and Earth sounded the same. I used latter, ?0 total from Wickes. :-)

MikeG,

I am interested in knowing the following about your tests:

1. What signal amplitude did you use during the test?
2. What kind of load and driving source you used?
3. Did you do both phase and amplitude tests?
4. What kind of test signal did you use?