In article , Bob Latham
wrote:
In article , Serge Auckland
wrote:

[Snip]

All you are doing with bi or tri wiring is reducing the resistance
from already low values, increasing the capacitance from normally low
values to still low values, and reducing inductance. Any
amplifier/'speaker combination for which this makes ANY difference is
only fit to be used as a boat anchor.

I've seen these sort of comments before and simply don't understand, I
would be grateful for some clarification.

The amplifier will see the capacitance of two speaker cables and in this
sense they are in parallel which will add the capacitance of each.

Yes.

The cables are not connected together at the speaker end and don't cover
the same frequency spectrum so in no sense are they in parallel for
current flow to the speaker.

In essence, correct. When bi-wired, the individual speaker unit won't have
any way of sensing what currents flow in the 'other cable' - i.e. the one
not connect to it.

Therefore, to my mind, resistance is not changed anywhere. Where could I
connect my fictitious superbly accurate meter to measure this change in
resistance?

The problem is that what Serge wrote only really explains the behaviour if
the cables are connected in parallel at both ends (and we ignore intercable
field interactions). My understanding is that is *not* what people normally
do when 'bi-wiring' loudspeakers. Hence your puzzlement is quite
understandable. :-) The webpages which I referred to in an earlier posting
outline the situation in more detail.

Briefly: The network when bi-wired has a topologically different
arrangement of nodes/paths to when conventionally wired. This means the
affects of the impedances have to be modelled in a suitably different
manner. However unless the bi-wires are connected together at *both* ends
(amp and speakers) the bi-wiring isn't simply described just by saying that
the series resistance halves and the shunt capacitance doubles. (Assuming
two nominal lengths of cable.)

If bi-wiring is used with two identical speaker cables then I can see no
possibility that this will make any difference what so ever.

See the analysis I refer to above. In principle, it can alter the frequency
response by an amount that would be quite measurable. What is unclear is
if, in practice, this happens to a degree that would be audible. On that
point I have my doubts as I've never seen a single reliable test whose
results showed this.

Bi-wiring opens up the possibility of using two different cables and
that, in my experience can make a difference even if it would require
one of the cables to be less than perfect in an engineering sense.

Yes. If you wished to deliberately use cables with wildly different series
resistances, etc, then they could then interact with the load impedances of
the speaker units and alter the overall response accordingly. However the
levels of resistance required for this would probably more cheaply and
controllably be obtained by deliberately adding low-value series resistors
and shunt capacitors, etc, at the speakers. This is usually called
"modifying the crossover networks"... ;-

Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc http://www.st-and.demon.co.uk/AudioMisc/index.html
Armstrong Audio http://www.st-and.demon.co.uk/Audio/armstrong.html
Barbirolli Soc. http://www.st-and.demon.co.uk/JBSoc/JBSoc.html