Twisted or Straight?


I searched Audiogon for info on inductance and capacitance. From an excellent post by Sean on March 24, he explains that inductance increases with wire spacing ant that capacitance increases as wires move closer together. Therefore, a twisted pair raises capacitance as contact between wires is increased. On the other hand, I'm a bit confused in that I thought winding wire would increase inductance.
Here is my question: For an ac power cable running from the panel box to the outlet, would it be better to run twisted wire or straight (i.e., parallel) wire? Specifically, I'm referring to twisting the hot, neutral and ground vs. having them run parallel? I've read strong preferences for both. Per Subaruguru's post, straight romex increases inductance and allows unwanted high frequencies to roll off. Other posts suggest that twisted is better. Please help me sort this out since I am running dedicated lines to my stereo. Thanks in advance.
ozfly
Ozfly, increased inductance on a power cord has its' ups and downs. As inductance is increased, you obtain a natural amount of high frequency roll-off. Kind of like a "low pass filter" that lets the low frequencies "pass through" and attenuates "hash" somewhere above that point. Obviously, the amount of attenuation, cut-off frequency and rate that it slopes off at will vary with the total amount of inductance for the entire run of wire.

While this may be a good thing for power cords due to their limited bandwidth, it is obviously not so good for speaker cables or interconnects. That is why most of the "high end" cable designs are of limited inductance. It should be pointed out that the "roll-off" of an inductive cable can be put to good use if you have an overly bright system. That is walking a tight-rope though, as too much inductance will dull the system out and not enough will still leave you with too much brightness. Better to fix the problem than try to band-aid it. Sometimes though, band-aids are all one can afford at that specific time : )

As to the bad part, wires that are higher in inductance act as a more efficient antenna to RF based signals. It is easier to "induce" stray voltage into them. Obviously, this can be a bad thing if you live in a densely populated area with a lot of transmitters ( cell and cordless phones, AM / FM / TV broadcast stations, CB's, Ham Radio's, etc... ).

The method that Audioengr suggests should theoretically offer benefits over some of the other methods mentioned. However, it would be a lot more work than what Bob suggested and tough to predict whether the results would be worth the extra effort under real world conditions. If one were interested in putting forth that much effort, there are ways to achieve even better results. I don't want to go into details here for specific reasons though : ) Once again though, whether the results would be actually audible or simply more "icing on the cake / increased peace of mind" would be anyone's guess.

If one wanted to get REALLY serious, they could actually measure the total impedance of the run, break down the measurements into specific LCR ( L = inductance, C = capacitance, R = resistance ) characteristics and then create a tuned trap or bandpass filter by adding additonal parts. Something like that would not really suffer from current limitations ( if done properly ) and would benefit everything plugged into that line. Obviously, this would be the optimum way to go but also ( by far ) the most involved.

The bottom line on all of this would simply be a "blacker background" and "increased liquidity" from my experience. Sean
>
Glen,

You forgot the 27 vestal virgins singing the Benjamin Britten "Hymn to the Sacred Electrical Particles."

Incense will help, too. I've had good results with pure gum frankincense, although the addition of a bit of myrrh can help to tame a hot tweeter.

Let's smile, folks. I respect those who believe with all their hearts that these things make a difference and I respect those who think it is all blue smoke and mirrors. What a pity they so seldom respect each other....

will
Thanks all. I'll go with the 27 vestal virgins approach. In lieu of that, the twisted 10 gauge in conduit seems like a plan. Not too expensive and likely to gain good results. Thanks again.

P.S. Per suggestions in earlier posts and above, I will be using the "star" pattern in that all grounds will first go directly to a separate ground rod (next to the one for the house) and then be tied together at the box to avoid ground loop noise. A preferred ground "star" approach. Thanks.
Marakanetz - The di/dt currents in the power cord are what matters. IF it is too inductive, the drops can be in the Volts, not mVolts. I have one of the lowest inductance cords on the market and it is getting rave reviews when used with power amps. If you read my description on my poer-cord page, it describes how the voltage drops can occur. Poor power cords can have the same effect that plugging into multiple power strips has, except on transients. If your amplifier has switching supply, the impact of a low-inductance power cord may be much less, because of the regulation.

BTW - twisting the two current-carrying wires has the additional benefit of radiating a smaller magnetic field that can impinge on nearby interconnects compared to two parallel wires.