Speaker cables don't need shielding anyway
It's not the point. As described in the method, the shielding gets attached to the conductor at the end of the run, and thereby, becomes part of the cable. It's not a shield, at that point.
If you use two conductors inside a shield, that causes phase shift. It sounds wrong.
Using 1 conductor and a shield isn't really a problem, but makes less sense on amps that are balanced output. I cannot see any appreciable reason to do it.
I can. It's a lesson that my father taught me. Quite simply, it's "use what you have".
As I recall,the shield and core had distinctly different resistance. So when the cross occurs, the currents distribute core to shield differently. Because each coax no longer has a sum of zero cross sectional current, the inductance goes up.
That's the "what it does" EE version.
If your question is what does it do audio wise, your guess is as good as mine. Higher highs, lower lows...who knows...
Jn
Last edited:
A quote by the cable's designer, Jon Risch:
“Cross-connection is used to reduce the inductance to an absolute minimum. Merely paralleling the center wire and shield would create two separated different polarity composite conductors with an inductance much higher than the cross-connected pair.”
It seems an attempt to keep L,C and R at lowish levels Loudspeaker Cable Characteristic Impedance
Interconnect and speaker cable whitepaper
“Cross-connection is used to reduce the inductance to an absolute minimum. Merely paralleling the center wire and shield would create two separated different polarity composite conductors with an inductance much higher than the cross-connected pair.”
It seems an attempt to keep L,C and R at lowish levels Loudspeaker Cable Characteristic Impedance
Interconnect and speaker cable whitepaper
The funny thing is he increased the inductance, not lowered it by the cross connect.
Heck, if all you want to do is decrease inductance, just twist and parallel zip. One zip runs about 180 to 200 nH per foot, 4 in parallel runs 50 nH per foot, his cross connect runs 70 nH per foot.
Just use 14 AWG, resistance will be low enough for most apps.
He really didn't have a good handle on e/m theory, it spills into his writings.
Rod Elliot whom you linked to, is a fairly reasonable guy, writes well and is fairly technically accurate. While he also has some mistakes, I would always recommend his papers. Remember, I'm kind of "nitpicking". His mistakes are of little consequence to the writings overall accuracy.
Jn
Heck, if all you want to do is decrease inductance, just twist and parallel zip. One zip runs about 180 to 200 nH per foot, 4 in parallel runs 50 nH per foot, his cross connect runs 70 nH per foot.
Just use 14 AWG, resistance will be low enough for most apps.
He really didn't have a good handle on e/m theory, it spills into his writings.
Rod Elliot whom you linked to, is a fairly reasonable guy, writes well and is fairly technically accurate. While he also has some mistakes, I would always recommend his papers. Remember, I'm kind of "nitpicking". His mistakes are of little consequence to the writings overall accuracy.
Jn
Last edited:
Right, that's all cool, but we're talking about crossing the shield from one cable, to the conductor of another, and vice versa - so that the ends of 2 separate pieces of coax are permanently joined in a "Y" configuration. I can't really understand why you can't just join shield and conductor in the last few inches of the same cable. What's the difference? The cables are the same length...
In the end, you end up bundling the two together in a fishnet stocking, and wrapping the ends in gimp vinyl. (sans zippers) I'm not an EE, but intuitively, I see zero value in doing the cross. In fact, I see more of a potential to do harm (via a short) than anything.
I reserve the right to be wrong.
Mogami makes a dedicated coaxial speaker cable,with center conductor and shield being equal at approx. 14ga,W3082:
MOGAMI® - W3082 Superflexible Studio Speaker Cables
MOGAMI® - W3082 Superflexible Studio Speaker Cables
Similar to starquad? I was thinking earlier, zip with a spacer then twisted could do similar, perhaps like what Dave (Planet 10) mentioned, two wires wrapped around a tube and crossing at right angles?
Last edited:
Reading there marketing BS, I just lost all respect for Mogami cable. I've been putting it in studios, and home for 30 years. Now they seem to be making exotic snake oil cables and marketing them the same as any other snake oil.
Your intuition is correct.
If you say something wrong, I'll let you know..I an the consummate nit picker.
So far, you've been on the money.
Jn
Star quad lowers inductance by geometry. Two pairs is the max number that can be run untwisted and still remain orthogonal (no magnetic coupling). Beyond that, the only way to eliminate coupling is to twist at different pitches, just like cat5e cable did.
My recommendation to twist the zips assumes that the total number of twists on any zip will be random and therefore orthogonal to all other zips.
...orthogonal is just a really fancy way of saying different number of turns..but you have to admit, it sounds really smart yes??
Jn
I'm still a bit confused about exactly how this is wired (I need a diagram!) but I can see that destroying balance will increase inductance. Curious that someone thought that it might do the opposite.jneutron said:
Generally true of people who offer cable recipes. Some think they understand EM; others think they don't need to understand EM as conventional EM theory is all wrong anyway and doesn't apply to audio.
I can't think of any good reason why the signal and return conductors have to have the same thickness/resistance.audio king said:
Here you go. Cross-Connected Belden 89259 DIY Hi-Fi Speaker Cables
Perhaps the idea is twisting the coaxes reduces inductance
Ah, thanks for the link. It's been a long time. I forgot it wasn't split in the middle.
I think the idea was, instead of a twisted pair with it's inductance, cancel the fields like a coax does. The problem was, the shield and core have different resistances, so the current doesn't divide evenly. Had they been the same, the current would have balanced and the end result would have been two coaxes in parallel, with an inductance half of the coax's value.
Not applying ohms law was the problem.
Jn
Thanks for providing the link. It is quite funny that people all over the world will be carefully following these instructions to make an electrically weak speaker cable (but mechanically quite stiff) when all they need to do is simply parallel two coaxes - if for some reason they want a coax speaker cable. As it is they have basically a wide-spaced twisted pair, slightly modified by the inner current subtracting some magnetic field.
Sure it will make a good speaker cable with the mechanical problems that you noted. The subject sometimes came up in the news-groups.
Some old RG-8, RG-9 & Rg-11 all about 0.4 inch diameter might be available on the surplus market. Radio Shack sold RG-8 #278-1312.
