wrote in message
oups.com...
You need to look at it from the viewpoint of what does the lower
resistance buy you, and it is not insignificant. Low end amplifiers
have damping factors of a few hundred. A high end product (say a Jeff
Rowland design) can have a damping factor of 1000. So your high end amp
has an output impedance of 8 milliohms, and maybe 30 milliohms for the
average amp.

http://www.epanorama.net/documents/w...esistance.html

Let's ignore the skin effect for the sake of simplicity, i.e. will will
deal with DC resistance. [The skin effect at audio isn't that big of a
deal.] Using cheesy 18 guage wire, you have 7.5milliohms per foot,
while 12 guage is about 2 milliohms per foot. For twenty foot runs
(i.e both sides of the cable), this yields 150 milliohms and 40
milliohms. Thus you spent all that money on a amp to control the
speaker, but the wire resistance is at least 5 times greater than the
output impedance of the amp.

Sure, but is there really much difference between a dampning factor of 1000
or 100 or whatever? What if we could arrange an infinite dampning factor?

An infinite damping factor should control the voice coil as much as it can
be controlled. Imagine, for the purpose of illustration, a speaker with
it's terminals dead zero ohms shorted. Can the voice coil be moved without
any electrical input? Yep. Back emf (or current) or not, it can be moved
pretty easily. It can be moved at any frequency. Sure, with the speaker
shorted, the voice coil will try to act as an electrical brake, but it's a
pretty lousy brake. I doubt an infinitely damped speaker would control
unwanted motion any more completely than a shorted mechanical meter
movement.

I imagine the voice coil would be a much better electrical brake if it
didn't have any electrical resistance. But it does have electrical
resistance. The majority of most speaker's impedance is electrical
resistance.

So, as I see it, voice coil resistance is the limiting factor in all these
wire size/dampning factor considerations.

I can think of one way to get around the effects wire size has on dampning
factor. Just run another pair of thin sensor wires from the speaker
terminals back to the amplifiers feedback network. Voice coil resistance
effects could be greatly reduced if somebody would wind a sense coil on the
voice coil form and run sensor wires back to the amp.

I'm not convinced such extraordinary damping factor designs would be worth
the effort. After all, tube amps have a comparatively crummy dampning
factor, yet still can sound pretty good.


Is heavy wire a good idea? Yes, Does it have to be fancy stuff? No.

There is another school of thought called biwiring. Here you wire the
drivers (or more corectly the input to the crossover filter)
individually. This is related to the damping factor as well. Since the
wire has resistance, the back EMF of the woofer (which is significant)
can find it's way back to the tweeter since the amplifier, no matter
how well damped, is connected to the speaker with a resistor (aka
wires). If you bi-wire, the woofer back EMF, which can contain higher
order harmonics, gets damped at the amp.

There is good science behind many audio tweaks, but often the science
gets lost in the argument. For instance, Kiwa suggesting to yank
ceramic bypass caps is just plain stupid. There may be arguments to
yank them in the audio chain, but low inductance ceramic caps are
really a requirement for good power supply bypass.

Keeping this remotely on topic, I use that bad ass speaker cable wire
you get from the hardware store as an "extension" cord from my
cigarette lighter outlet to a power distribution breakout box. This way
when monitoring out in the field, I can set up my gear outside the SUV
and have a little elbow roon. Speaker cable is quite flexible, so the
setup packs up nicely. Eventually I'm going to set this up so that I
can have gel cells to run the gear, but use the car battery to supply
some of the power. What you need to do is prevent the situation where
the gel cell power flows back to the car. I'll probably end up using
power schotky diodes unless some clever linear circuit comes to mind.