biwire trick

Shadorne and Riley804, I'm not so sure.

Keep in mind that the swap Tholt is describing will result in different frequency components flowing in the + leg and the - leg of each two-conductor pair of wires. It seems to me that would in effect amount to a change in the inductance of the cables, because the magnetic fields associated with current flow through adjacent conductors would no longer be equal and opposite.

As you probably realize, depending on cable length, cable type, and the impedance of the speaker at high frequencies, speaker cable inductance can sometimes have a significant effect on upper treble response. And upper treble response would seem to be something that can subjectively correlate with soundstaging.

The connection swap would also reduce the degree to which any noise pickup that may occur is common mode. I would not expect that to generally be significant in an amplifier-to-speaker connection, because of the low impedances and relatively high signal levels that are involved, but I would be hesitant to totally rule out the possibility.

As you know, I am certainly one who tends to be skeptical about a lot of tweaks that are reported, but in this case I would not rule out the possibility that the differences were real.

Best regards,
-- Al

Thanks very much for the nice words, gentlemen.

Rcrerar, no, what Tholt is describing is a conventional biwire arrangement but with the + conductors of the two cables interchanged at the speaker terminals. It may not have been clear to you that the statement he quoted from the MD site is addressing something different, a non-biwire arrangement in which jumpers are used but the connections of the + and - conductors of the single speaker cable are made to "diagonal" speaker terminals, rather than adjacent speaker terminals. I have no idea, btw, why the latter would make a difference sonically, although a number of people have claimed in past threads that it did for them.

Shadorne, neat calculator that you linked to. I agree that if the two sets of cables are within the same jacket, the effects I described would probably be reduced to insignificance. However, if they are physically separated by some number of inches or feet over a significant fraction of their run length, I think that the following factors would make those effects more significant than the calculator would seem to indicate:

1)The calculator assumes the speaker is purely resistive. Dynamic speakers will tend to have an impedance at high frequencies that is somewhat inductive, due to tweeter voicecoil inductance. Cable inductance will have a greater effect on bandwidth when it is connected to an inductive impedance than when it is connected to a resistive impedance of the same magnitude.

2)The speaker impedance at high frequencies may be less than the 8 ohms that is assumed in the calculator.

3)While the effect may amount to only a db or less at 20kHz, which I agree is insignificant in terms of steady-state frequency response, as you certainly know our hearing mechanisms give increased emphasis to the leading edges of rapidly changing transients. Fast leading edges, of course, correspond to high frequency spectral components.

The bottom line is that I'm not asserting that the effects I've described are necessarily the explanation of what Tholt perceived, but that there is technical rationale that is sufficiently plausible for this tweak to not be dismissed as placebo effect.

Best regards,
-- Al

Unless there is a downside electrically, which I haven't read yet ....

I think that the only conceivable downside would be a philosophical one, along the lines of Shadorne's last comment. It could be argued that you are raising the effective cable inductance to compensate for an imperfection elsewhere in the system or the room. But given that nothing is perfect, and that the adjustment is extremely small in electrical terms, as a practical matter I don't see any issues. I would think of it simply as fine tuning the system.

Best regards,
-- Al