the signal must pass around a circuit. That circuit can begin at the source Hot terminal and MUST return to that same starting point.
The cable that connects the source to the receiver must have two paths for current flow. The Hot sends the current Flow and the partner returns the signal.
A coax or a twisted pair or a star quad can all do this Flow and Return thing.
The difference between all of these is what prevents interference entering the Flow and Return circuit.
An unbalanced coax that uses shield as both interference attenuator and as a return current route does not work as expected by the layman. Look at the different currents that can circulate through the shield to see what can happen.
An added shield that is connected ONLY at one end, cannot be part of a circuit. There is no route for the current to pass through. It acts solely as an interference attenuator.
The cable that connects the source to the receiver must have two paths for current flow. The Hot sends the current Flow and the partner returns the signal.
A coax or a twisted pair or a star quad can all do this Flow and Return thing.
The difference between all of these is what prevents interference entering the Flow and Return circuit.
An unbalanced coax that uses shield as both interference attenuator and as a return current route does not work as expected by the layman. Look at the different currents that can circulate through the shield to see what can happen.
An added shield that is connected ONLY at one end, cannot be part of a circuit. There is no route for the current to pass through. It acts solely as an interference attenuator.
Qusp, start here with this note, then have a good look around his site.
http://www.hottconsultants.com/pdf_files/Audio Interconnections.pdf
Shields connected at one end are not much use. Its a dipole effectively.
Some more interesting stuff
Technical Articles/Talks
And his presentation to the AES, just the puplic domain stills from his presentation:
http://www.hottconsultants.com/pdf_files/aes-2007.pdf
The I would sugest bying or lending his book:
Electromagnetic Compatability Engineering.
The chapters on cables and shielding will help dispell the myths that surround all this...
http://www.hottconsultants.com/pdf_files/Audio Interconnections.pdf
Shields connected at one end are not much use. Its a dipole effectively.
Some more interesting stuff
Technical Articles/Talks
And his presentation to the AES, just the puplic domain stills from his presentation:
http://www.hottconsultants.com/pdf_files/aes-2007.pdf
The I would sugest bying or lending his book:
Electromagnetic Compatability Engineering.
The chapters on cables and shielding will help dispell the myths that surround all this...
Dirkwrigt, The Canare Star Quad wire looks realy nice, thanks for the link. I'd like someone to explain the cross linking irridation technique in simpler terms! It seems to be a common process, but could not find a realy simple explanations.
It will be interesting to debate the shielding at one end versus two😀 in todays high frequency noisy world, I go more for the complete shielding with seperate twisted cores for signal and return, low resistance briaded shield connected at both ends for RF shielding. But every system and set up has its own problems and no one solution is never perfect.
It will be interesting to debate the shielding at one end versus two😀 in todays high frequency noisy world, I go more for the complete shielding with seperate twisted cores for signal and return, low resistance briaded shield connected at both ends for RF shielding. But every system and set up has its own problems and no one solution is never perfect.
Power cable (mains shielding) did a great sonic improvement in my setup, starting at the power amps and also the preamp in a less immediate effectiveness but by a level of magnitude. At least, some unshielded cable designs just around began to clearly show a great sonic potential 🙂
Best,
nAr
- sorry for being off topic -
Best,
nAr
- sorry for being off topic -
As for why a heavy return conductor is important in an unbalanced interconnect cable, Read Section 2.3 of the Bill Whitlock seminar paper.
http://www.jensen-transformers.com/an/generic seminar.pdf
On a long (25 foot) interconnect, a heavy return conductor can reduce noise by 24dB.
http://www.jensen-transformers.com/an/generic seminar.pdf
On a long (25 foot) interconnect, a heavy return conductor can reduce noise by 24dB.
While Canare Star Quad is a great balanced interconnect cable, it's far from my first choice for an unbalanced interconnect.
I'll look for Jim Brown's explanation in one of his many papers.
Audio Systems Group, Inc. Publications
Cheers Sy, I got some very deep explanations last night with , dont know why Wikki didn't come up. Lots of Far East sites selling cable with this as the insulation though. Although a few drinks may not have helped my searching abilities.
Qusp, being thinking further about what you said, if there is noise then the point of shielding it and tying the shield to ground is that it is where the noise will do the least damage.
Allowing for a heavy guauge ground wire is common practice on a lot of systems, PCB,s etc. You can follow all the standard practices, and still have noise. A nice thick cable between two grounding points soon sorts it out.
Plent of info regarding Audio shielding and cabling on the EMC club site (Keith Armstrong and Tony Waldron have excellent articles.)
Qusp, being thinking further about what you said, if there is noise then the point of shielding it and tying the shield to ground is that it is where the noise will do the least damage.
Allowing for a heavy guauge ground wire is common practice on a lot of systems, PCB,s etc. You can follow all the standard practices, and still have noise. A nice thick cable between two grounding points soon sorts it out.
Plent of info regarding Audio shielding and cabling on the EMC club site (Keith Armstrong and Tony Waldron have excellent articles.)
Last edited:
Guys, I don't see exactly the point to reduce cable sounding qualities to the degree of shielding they have. Sometimes i's irrelevant, e.g. not linked. Some Goto systems had terrible hum background noise audible even some feets away from speakers and sounded great. I'm personnally more interested in what makes a cable sounding better than another, even if their shielding or background noise are almost the same.
Someone ... The Essex Echo ? 😎
Best,
nAr
Someone ... The Essex Echo ? 😎
Best,
nAr
I could say the same..funnily enough you didnt provide any insight as to what possible use connecting one end of a shield on an RCA cable might have; preference for one that has a positive outcome
i prefer the mogami for quad, preferred to the canare. one end is pointless, its still part of the circuit that is connected at both ends, i truly dont understand how so many people, including so-called informational resources mention it as an option. with XLR you have the opportunity to have dedicated terminal for shield and i will sometimes use it then, but not on pin 1 of course, on the shield terminal.
I also prefer just a chunky return for single ended cable, larger than the signal carrier, preferably at least double.
sometimes i will use something like belden 1794A coax and connect at both ends (of course) it can be quite effective, but given proper layout and RF filters on input stages i cant say i ever notice a difference if shielded twisted pair is used. but then i dont have bat ears. i changed to balanced on everything so long ago now its only a very vague memory.
thanks for the links though, I can imagine it useful for RF technique
yeah i get that, but with RCA the shield HAS to actually BE PART of the return circuit, so it doesnt stop the noise entering the return path at all, just more efficiently gathers it and shunts it to the return. not arguing with your above quote btw, just replying to it.
for clarification, to further the above. do you mean to say that it shouldnt be part of the circuit in order to act as an attenuator, or that because its only connected at one end, it isnt part of the circuit? i dont understand how as in the example of coax that it isnt part of the circuit just because its connected at one end only? it still part of the return path so any noise that hits it will be collected and go to signal ground, it may attenuate the noise that may have directly contaminated the signal line, but it simply gathers more noise that might have just passed by and sends it to the signal return.
Nar,
if the shielding is done properly it is part of the circuit and can be part of the return path, just as the earth cable on the mains interface can be part of the return current for the lower frequecies.
if the shielding is done properly it is part of the circuit and can be part of the return path, just as the earth cable on the mains interface can be part of the return current for the lower frequecies.
Interesting 😉 I thought in alternative 230V/50 Hz or "mixed pseudo sinus/transcients" we use here as models in music, there is no "go path" and "return path" ??? 😉 The way I see things is the 50 Hz 230V or audio signal is an alternative potential, using the 2 wires only together as guiding rails, going the same direction (source to receiver) and those potential changing alternatively would travel (alternatively charge) the dielectric surrounding them ... 😎
The use of a shield on mains cable seems clever, as it seems to permit mains transmission potential / current without the eletromagnetic perturbation close to the amp. As a result, the LFE signals in music ( but not only ... ) can have less perturbation whilst beeing carried to the speakers. That what I heard, at least, and that is also how I try to explain it 🙂
Sorry for being once again off topic - except for the shielding question
Best,
nAr
You can build true balanced interconnects with only 2 wires/polarities in it 😉
This will work only with true differential gain stages
In that case one of the XLR's pin is of no use 😎
Best,
nAr
Someone trying to wire a mains or audio circuit without go and return paths will quickly find that such low frequency signals like to have wire to travel along. This roughly corresponds to TEM mode propagation.nar said:
You can get single-wire propagation, but at RF and it is difficult to control so rarely used. For microwaves you can have no-wire propagation - a waveguide.