wjm@whuxk.UUCP (10/01/83)
There are several problems with Dave Martindale's approach of using 75 ohm coax
terminated with a 75 ohm resistor at the amplifier as speaker wire. First,
(I'm assuming the resistor is in parallel with the amp output - if it were in
series almost all the power would go into the resistor rather than the speakers)
the coax is going to see the combination of the resistor and the amp's output
impedance (about 1/4 ohm for most modern solid-state amps) as its load and
that is about the same as the amp's output impedance - 1/4 ohm - so much for
an impedance match. Second, the resistance per foot of RG-59U is about the
same as #24 wire, since the center conductor diameter is the same .02"
(RG-59 is the standard coax used for TV and FM tuner leadins and cable TV drops)
Admittedly, one could use RG-11U (the heavier stuff - about .4" in diameter)
but even that only has the resistance of #18 which is marginal for all but the
lowest power amplifiers with short speaker runs. At audio frequencies, the
most important circuit parameter is the RESISTANCE of the wire, which is why
I use #14 SJ cord (available from your friendly hardware store or electrical
supply house) since I have fairly short speaker lines. I agree with rabbit!jj
that #12 should be used on longer lines.
This is why the results of the "Stereo Review" speaker tests make sense.
The concept of putting a mono amp next to the speakers is not a bad idea, and
has been used by many manufacturers. Its only problem is that you have to
run high impedance lines to the speakers, where the capacitance of the line
is more significant than with the low impedance power-amp speaker lines and
which are more susceptable to hum.
Bill Mitchell
(whuxk!wjm)michaelk@tekmdp.UUCP (Michael Kersenbrock) (10/04/83)
Your speaker cable does not start to show transmission line effects until the the length of the cable approaches 0.05 wavelength (more or less) of the highest frequencies transmitted down it. The cable's "characteristic impedance (e.g. 75 ohms for RG-59)" is of no significance until the length is an appreciable fraction of the wavelength. The wavelength of 20Khz in coaxial cable (assuming a velocity factor of 0.66 which is typical) is about ** 6 MILES **. If your speaker cables are an apreciable fraction of this length, you have other problems. The reason preamp leads use "coax" cable is for shielding of low-level high-impedance circuits. Speakers are high-level low-impedance circuits (that are driven by a balanced cable anyway). Mike Kersenbrock Tektronix Microcomputer Development Products Aloha, Oregon
dmmartindale@watcgl.UUCP (Dave Martindale) (10/06/83)
Sorry if my original article wasn't clear. I was suggesting using 75 ohm coax between the preamp and the amp (which is located at the speaker), NOT as cabling between the amp and the speaker - the latter is clearly silly. The idea of using properly-terminated 75 ohm cable driven by a 75 ohm source at the preamp was to get around the problems due to sending the signal from the preamp to the amp on high-impedance coax. This would solve cable capacitance problems, wouldn't it, as well as hum pickup problems? (I'm not an EE, so I don't have enough knowledge to tell if there are holes in this argument.) Dave Martindale