Posts: 6659
Joined: April 4, 2000
Contributor Since: March 1, 1999
|
I have posted several things that might help you understand what the deal is on this type of construction.
See: About cable "ground": http://www.AudioAsylum.com/audio/cables/messages/503.html
DIY Cables and RFI/IGP http://www.audioasylum.com/audio/cables/messages/80108.html
Cable RF Termination and Cable Loading http://www.audioasylum.com/audio/cables/messages/80110.html
After reading those, come back here to finish.
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
Now, perhaps at this point, you can see that a regular coaxial cable used as an audio interconnect has the signal traveling on the shield braid just as much as it is traveling on the center wire.
In order for a shield to work, it must be conductive or a magnetically active material. A shield works by intercepting the interfering signal, and in doing so, currents flow within the shield. A coaxial cable shield is carrying the signal, but now, if an interfering signal is encountered, it induces currents in the shield, and if the shield is connected to "ground" at both ends, those currents can flow into one ground or the other, and corrupt the audio signal by being picked up and/or amplified by the audio gear being connected.
That's just the shielding aspect being involved, if you read the post on "DIY Cables and RFI/IGP", then you would also be aware that different pieces of audio equipment can have their "grounds" at different voltage potentials. This can lead to the classic "ground loop" problem, or the more insidious Interchassis Ground Potential effects, which can involve RFI injection, etc.
Note that when a shield is carrying current, it's effectiveness as a shield is reduced, thus a coaxial cable shield braid has multiple reasons it is less than ideal as a shield and a signal return path.
Now the geometry you show in your post (which more commonly has the outer shield grounded at the source end, not the load), does not have the audio signal currents traveling in the shield, they are in the 'ground wire' of the twisted pair. The currents in the twisted pair are equal but opposite, and thus tend to cancel out the inductance of the wires, (SEE: Cable inductance, and how it is controlled http://www.audioasylum.com/audio/cables/messages/17.html) as well as reject outside interference that might get through the shield, due to the twisted pair having the conductors located in nearly the same space, and thus, nearly the same electrical and magnetic fields.
This rejection is not as powerful and deliberate as it is with a true balanced signal set-up and balanced ins and outs, but it still helps nonetheless.
The fact that the "ground" wire is carrying the audio signal just as much as the "hot" wire, means that the nearby insulation will also effect the "ground" wire, as will the conductor purity, etc. So the geometry you show has the advantage of a purer and less contaminated signal path for the signal return wire compared to a coaxial cable shield.
So by separating the work done by the shield and the work done by the signal return path, both duties are handled just a little better than they would be otherwise.
Finally, many commercial and even "high end" twisted pair cables have the shield too close to the twisted pair, this means that the shield is close enough to partially "short out" the electrical field, and curtail and distort the magnetic fields, thus causing the well known sonic problems with a shielded twisted pair compared to the unshielded twisted pair.
Some of my DIY cable designs take this into account, and space the shield out away from the twisted pair, and even take the trouble to keep the shielding material out of the twisted pairs 'crease', so as to minimize the field distorting aspect of the shield, while retaining the benefits of it's interference protection.
I usually strongly recommend to beginner DIYer's that they build a shielded version of an interconnect first, to make sure they are not suffering from any sort of insidious or low-level RFI/EMI interference, and then build a unshielded twisted pair if they are so inclined. At least now they will truly know that the sound of the interconnect is better or worse compared to the shielded version instead of only building an unshielded version and declaring the DIY recipe to "suck".
I have had this happen several times, where the DIYer first built an unshielded interconnect, only to find they had residual or low level interference that make it sound poor; once they added a correctly grounded shield, the sound got better, and usually much better than the cheapo cables they were using before.
I hope this explanation helps in your understanding of the geometry of the shielded twisted pair used unbalanced.
Jon Risch
|