Braiding vs side by side?

I have a custom speaker cable (teflon/silver about 16 ga) that has separate cable for the + and another for the -. I can leave them separate, running along side each other, or twist them together. Any reasons for one being better than the other?

These are brand new and will try it both ways but curious if there is any science behind it.
Post removed 
Hi Harry,

Yes, there is some science behind it. Twisting the conductors together will reduce inductance and increase capacitance.

For either configuration, as you probably realize, both parameters (and resistance as well) are proportional to length.

Under most circumstances, including the cables you appear to be describing, capacitance in speaker cables is unimportant.

Inductance can be marginally significant, if the total inductance of the cable for the length being used is excessive in relation to the impedance of the speakers in the uppermost octave (10kHz to 20kHz). That would most commonly tend to be a concern with electrostatic speakers, which often have impedances that drop to very low values at high frequencies. In that situation excessive inductance will result in a slight loss of upper treble extension.

In the case of dynamic speakers, which commonly have impedances that rise or at least remain fairly flat at high frequencies, I would not expect that effect to be significant even for very long cable lengths.

If you'd like a more quantitative answer, let us know the length of the cables, and the make and model of the speakers and amplifier. Also, if possible, an approximate indication of the overall diameter of the insulation. Assuming an impedance vs. frequency curve is available for the speakers, I could then do some calculations that would provide a more quantitative perspective on it all.

Best regards,
-- Al
Sorry to butt in here but I can't help myself....As usual, Al, you are a wonderful person to have around. Your advice is always pertinent and detailed. I always learn something from your posts, including this one. Thanks.
I think leaving them separate would let them act as a radio antenna, and may pickup signal interference also.
Thanks for the responses.

Hifihvn, good point. So far the braided seems best but they are still pretty new.

Elizabeth, not sure how good my braiding was. I'll check for uniformity tomorrow and make it so if not.

Al, Cables are 8' in length, speakers are Magnepan 3.6's. I'm using 2 different amps, a pair of Pass XA100.5 monos and (hope you're sitting down) a pair of Cary 805 AE monos. The insulation is very thin teflon. I can try and measure it tomorrow, for now I'd guess 1/32" or less.

I've used this wire on my hi-eff/2 way 8 ohm speakers for years and really like it. It has just one wire per spade and have tried it that way on the Maggies but my friend had a little left over so I had him make up a set using 2 wires per spade wondering if the Maggies would like the extra current. These are so skinny to begin with, looks like it could be 16 ga, maybe even 18 ga. Don't know how much the silver factor fits in.

The Pass and Maggies are a nice match but the Cary has an output impedance of about 1.5 ohms giving it a damping factor of only 3 or 4 into the Maggies. There is a feedback control which if I switch in could reduce the impedance to around .5 but I'd rather not use it. Was wondering if the thicker wire would help the Cary out in the damping dept?

Here is a graph from Stereophiles measurements.
(Can't get the graph to appear but here are their comments)

The speaker's impedance (fig.1) approximates a resistive load of around 4 ohms over much of the audioband. However, there is a slight magnitude peak centered at 1.6kHz, due to the crossover between the ribbon and the midrange diaphragm. The minimum value is 3.3 ohms at 10kHz, which is not going to be problem for any good amplifier to drive, while the increasingly positive electrical phase angle at the top of the audioband is, I assume, due to the residual inductance of the ribbon driver. There is a small wrinkle in the trace between 50Hz and 60Hz, which is probably due to the tuning of the woofer diaphragm.

Fig.1 Magnepan MG3.6/R, electrical impedance (solid) and phase (dashed). (2 ohms/vertical div.)

This can be seen to the left of fig.2, as the big peak in the mid-bass. This is a nearfield measurement, which will exaggerate the behavior of the bass panel.
Tgrisham, thanks very much for the nice words.

Here is a link to the Stereophile measurements on the Maggie that Onemug referred to.

Data for 16 gauge wire can be obtained from this wire gauge table. Plugging the 1.29 mm conductor diameter that it indicates into this calculator yields an inductance of 5.95 uH (microHenries) for the 16 foot round-trip that the signal has to make, based on the worst case assumption that the two conductors are widely spaced.

Based on the formula for inductive reactance (the inductive form of impedance), 2 x pi x f x L, where f is frequency in Hertz and L is inductance in Henries, 5.95 uH corresponds to an inductive reactance of 0.75 ohms at the worst case frequency (20kHz). That verges on being negligible in relation to 4 ohms, taking into account that the phase angles of that reactance and the speaker's impedance differ considerably. (If the angles were identical, the resulting loss at 20kHz would correspond to 4/(4 + 0.75) = -1.5db).

Alternatively, this calculator, which was called to my attention by Shadorne in a thread a while back, can be used. For wide spacing of the conductors, it indicates a loss of about 0.1db at 20kHz into an 8 ohm resistive load (and less than that in relation to the slight loss that occurs at low frequencies, due to resistance). That would approximately double into 4 ohms, and increase a little more due to the somewhat inductive impedance characteristic of the speaker at high frequencies.

As that calculator makes clear, the loss numbers would diminish to complete insignificance if the wires were twisted, especially given that the thin insulation makes it possible for the conductors to be in close proximity.

So the bottom line for the 8 foot run length and the particular speakers appears to be, to the extent that cable effects are explainable by generally recognized science, that it probably won't make much difference either way, but twisting can't hurt (assuming you have confidence in the integrity of the insulation, so that shorts won't occur) and may help slightly. That may also be true for the reason Hifihvn cited, if either of the amplifiers are sensitive to rfi that may be presented to their output terminals.
Was wondering if the thicker wire would help the Cary out in the damping dept?
I doubt it. The 16 foot round-trip length of 16 gauge wire corresponds to a resistance of 0.064 ohms, based on the wire gauge table I linked to above. Paralleling two conductors for each leg would halve that to 0.032 ohms. Adding either number to the 1.5 ohm amplifier output impedance would not change the damping factor significantly.

Best regards,
-- Al
Al, thank you for doing all that research, I'm learning a lot.

Other than a limited SPL, do you see any reason a SET amp would be a mismatch for the 3.6's? My thoughts are/were: The Maggies are pretty resistive in nature (tubes like that) and, being a line source, will sound louder at the listening position than a point source of the same specs (so less power would be needed).
I second Tgrisham, good advice Al. Now wouldn't it be nice if Audiogon was organized so that you did not have to answer similar requests every other week?

Or is there a deliberate design in the way that Audiogon repeats the same threads endlessly? XLR vs RCA. Damping factor.

Audiogon = The Matrix ??

Anyone else notice the wrinkle in The Matrix (The "I've seen that thread so many times before" moment)?
braiding vs parallel runs is only one consideration when designing a cable.

have you considered solid core vs stranded ? have you considered multi gauage vs single guage, have you considered shielding, have you considered dielectric, have you considered connectors, have you considered method of connection, have you considered metal ?, have you considered alloy ?, have you considered combining methals, not as an alloy ?, have you considered plating ?

there are probably other variables.

i would think that capacitance is more significant for interconnects and inductance is more significant for speaker cables.
Very nice posting again Almarg. I too am very appreciative that you take the time out to make such in depth replies.

To Shadorne's point about deja vu/The Matrix moments. I've been coming here frequently in recent years and I get that "I've seen that thread so many times before" moment all the time. For you members that have been here for years I'm surprised you haven't gone nuts already with all the repeat questions.

I think it is part A'gon issue in design that repeat questions come up so often. The other is on the part of the user. I think a lot of posters just don't search the archives before posting. I think I just saw yet another power cord question thread.

Again thanks to Al and other veteran members who take the time out to answer all the duplicate questions in such detail.

Also my apologies to Onemug for the thread hi-jack.
04-25-11: Onemug
Other than a limited SPL, do you see any reason a SET amp would be a mismatch for the 3.6's? My thoughts are/were: The Maggies are pretty resistive in nature (tubes like that) and, being a line source, will sound louder at the listening position than a point source of the same specs (so less power would be needed).
If you haven't seen it, check out this older thread. The post by Ralph/Atmasphere strikes me as particularly informative.

It is evident, though, that there is considerable diversity of opinion as to how well Maggies will perform with low or medium powered amps, even after allowing for the expectable differences in listening habits, the dynamic range of the music being listened to, etc.

From a technical standpoint, the reasoning you expressed is correct, IMO. The one additional point I would cite is that the impedance rise that occurs across a narrow range of frequencies around 1.7kHz will cause those frequencies to be emphasized a bit more with a SET or other tube amp, in comparison with the frequency response that would occur with a solid state amp. That may or may not be preferable in your particular setup, of course.

Best regards,
-- Al
Thanks to everyone for their opinions and help.

Al, I did look up that thread you linked regarding Ralph's post. I found Duke's post interesting too. :-)

Liz, re-braided one that didn't look very uniform, now it's sit, forget and enjoy for awhile.

Tgrisham, Jedinite24: No problem. I like seeing people like Al get the praise they deserve. They don't get paid for their time, just do it out of the goodness in their heart.