FWIW, I would disagree with the statement in the Audioquest paper that was referenced by Tony (Tls49). That statement being:
When BiWiring, the two (bass and treble) cables must either be identical, or have essentially identical geometries. If the cables have different geometries they will have different capacitance and inductance. Capacitance and inductance are the values used to create a loudspeakers low-pass and high-pass filter networks, together making a crossover. Having different values in the two cables effectively redesigns the crossover...not a good thing! The integrity and coherence of the speaker will be compromised.
As is often the case in cable marketing literature, effects are claimed as being audibly significant without being put into quantitative perspective.
The value of the capacitors in speaker crossovers are typically in the area of several microfarads (uf) or more, and generally have tolerances (sample-to-sample variations) in the area of +/- 3% to 5% or even more. The capacitance of most speaker cables of reasonable length will be in the rough vicinity of 0.01% of several uf. Which is not only a miniscule fraction of the value of the crossover capacitors, but is even a miniscule fraction of the variations in capacitance between supposedly identical capacitors.
In the case of a few speaker cables having extremely high capacitance, such as Goertz, the 0.01% figure may increase to the point of being a significant fraction of 1%, but it will still be much lower than variations among supposedly identical capacitors.
Likewise for inductance. Although the inductance of different speaker cables can differ considerably, as a rough order of magnitude it can be assumed that the inductance of a 10 foot length of speaker cable will often fall in the rough vicinity of 1 to 3 microHenries (uH) or thereabouts, and in some cases will be much less than that. That is a negligible fraction of the inductance of nearly all crossover inductors.
Which is not to say that cable inductance is necessarily insignificant. If speaker impedance is low at high frequencies, as it especially tends to be with electrostatic speakers for example, the impedance presented by the inductance of some cables at those high frequencies may be audibly significant (as the impedance presented by an inductance increases in proportion to frequency). However, that would have nothing to do with whether the low frequency cable has the same inductance or a different inductance than the high frequency cable.
The bottom line: **If** in fact it is preferable to use identical high frequency and low frequency cables when biwiring, it is not because of the reason stated in the Audioquest paper.
IMO. Regards,
-- Al
