Arenith,
I’ve already given you everything you need to know about this topic in the tech notes, but you STILL REFUSE to read them it seems.
A shield between two coaxial cables in parallel does NOT alter the dielectric to be, “not there”. Two 75-ohm cables in parallel are HALF the impedance of one cable. Coaxial in series they are DOUBLE the impedance. If the dielectric was “gone” how could we arrange cables like this and derive the impedance? We are saying we have zero capacitance and self inductance with no impedance. Nonsense. Analog has impedance but not at all like RF, and analog does have L and C, always.
A UTP design twinaxial cable doubles the distance between signal wires so the RF impedance is TWICE the individual wires. Design each wire for 50-ohm and twist or bond them together and it is a 100-ohm UTP balanced line.
Here is the basic schematic for series parallel coaxial cable.
http://on5au.be/content/a10/trans/spcoax.html
CHARACTERISTIC IMPEDANCE implies a steady state L and C relationship at RF. Impedance is SQRT (L/C). The RATIO of L and C is the same over LENGTH, so the impedance is always the same no matter the length. But this ratio assumes we have reach the limiting speed of the dielectric. That happens at RF, not analog frequencies. Notice that the VP is not steady state until 1 MHz or higher in the tables below. But, L and C are pretty flat across the spectrum. At least we have that on our side.
Characteristic Impedance has ZERO to do with audio because the Vp of the dielectric changes with frequency, there is no “characteristic” anymore. Again, I’ve shown all the data on this to figure it all out. There are no secrets.
Matching impedance at analog is impossible as the cable impedance keeps changing. It isn’t an issue because the wavelengths are so long, even 20 KHz is many, many times longer than the cable.
And no, the concept of RF impedance was way after oceans transmission line cable. The theory wasn’t defined until way after those ocean lines by almost thirty years.
For audio, you need to do an open-short impedance at EACH frequency and the test cable length needs to be as long as you can test because the wavelength is so long. Shorter is OK, as it does show the general impedance through low frequency but do compare nearly like lengths only if you can. Impedance will RISE at low frequency as Vp drops, so the frequencies where the most power is located is at the worst “impedance” to match the speaker load. This is always the case with passive cable.
The telephone analog lines were called 600-ohm at 1 KHz, as an example. Why? That’s what the Vp provided in LONG analog lines. Look at the traces of modern cable, they are also about 600-ohm at 1 KHz.
Best,
Galen
