gjsmo <
gabj...@gmail.com> writes:
snips
>> But if the problem is merely 50-60 dB down, we're talking millivolts
>> against volts and that's likely audible.
>>
>> Just because from a *linear* frame of reference we'd be inclined to
>> say that a 1/2 millivolt signal should be lost against the swamping
>> tide of a 10 volt signal, that's not necessarily true.
>But wouldn't all of this be influenced FAR more by the amp, monitors,
>ambient noise and room than by the actual WIRES? If you're that
Yes, all those items play a role; some larger than others. The original point
was that some small effects (some, but not all) can be heard alongside much louder
components, particularly in the ever-changing tapestry of a music signal.
>concerned about a millivolt or two of interference or crosstalk,
No, not crosstalk. Dioding and LCR affects within the wire that alter the way the
*entire* circuit performs, not just the wire as an isolated thing on a test bench.
>double-shield the send and return separately, and then triple shield
>the whole thing, then connect them all to a 10 foot copper pipe in the
>ground (on one side only, of course - no ground loops here), using
>solder with silver in it (mine has 2% silver). Still isn't going to
>get you anywhere near the price of the ridiculously expensive stuff
>sold to blind believers. And I would sooner spend money on a multi-
>thousand dollar amp with unmeasurable THD than those wires.
I'm one water mile away from a 50KW AM radio station; I know all about dealing with
RF. Clean wiring practices from the get-go will preclude most problems. Special
shielding was not required.
>Maybe you're not implying what I'm challenging, in which case I never
>said the above.
Not sure what you're challenging; it's possible you're looking past the abstract
point I've tried to make.
Right, but music isn't white noise. (Well, there might be some exceptions,
particularly if mishandled by a mastering engineer who wants to "make it reallly
loud," but I don't want to start a personal-taste-in-music holy war. <g>)
The concept of the test was to have two things of vastly different "sizes", and
show how the tiny one was still readily perceptable, even at a 1000x, 2000x or more
times smaller than the big one.
Now, if you have diode effects taking place, smears or overshoots from unwanted RCL
circuits that can form within the wire -- again, small in magnitude but damaging to
the small components we routinely hear (such as harmonics) -- you have potential
issues to address.
snips
>> The general point is that if an improbable observation comes your
>> way, eliminate what you can as the cause. If it persists, you might
>> indeed have found something that initially seemed unlikely.
snips
>> But back (finally!) to your statement:
>>
>> >> I'm still not buying any cable that costs much more than the
>> >> copper and rubber that's in it.
>>
>> Theoretically, yes, absolutely. I'm right there with you.
>>
>> But in the real world, things might not be quite so simple. We need
>> to specify:
>>
>> - how much impurity we'll allow in the copper, and the affiliated
>> parasitic dioding, if a particular impurity leads to that (mere
>> resistance or reactance we can likely tolerate -- up to a point). Our
>> answer depends on the application. And it's more granular than
>> "audio". A tracking studio might tolerate "little" failings in their
>> monitoring more than a mastering studio.
>I've never even seen purity marked on a roll of copper wire. Silver
>maybe, but then you're crazy.
Generally, you won't, but certain applications require a specified purity. A
reputable manufacturer can provide such information or guarantee a certain wire to
be made a certain way.
Poorly made copper wire might be called "copper" but have excessive iron or carbon
(or other elements) in its structure, which could be acceptable for many
applications, but might be iffy for audio. (I've seen what happens to high-iron
content copper wire after it's been out in the world for a number of years. It can
corrode, and then you have the potential for diode problems. It might power your
light bulb just fine, but play havoc with audio.)
>> - what price for the chosen level of purity?
>>
>> - what level of diminishing returns for that price?
>>
>> - what return on investment?
>>
>> (There's yet another discussion about the geometry of the wire,
>> starting with solid or stranded, and branching out from there. Again,
>> not as simple as we might like.)
>Well, theoretically stranded should carry more current due to
>increased surface area. Except that's irrelevant except possibly in
Again, depends on how the stranding is done and the purity level.
Doug Sax of the Mastering Lab, the crew at A&M Mastering, Bob Katz, et al, are all
rather picky about EVERY component in their mastering facility, wiring included.
As just one example, A&M rewired with solid core wire throughout their multi-room
plant. Line level was done with a solid core 22 gauge Alpha wire product; speakers
were done with a 12-2 solid-conductor twisted pair product from Belden. Their
contention was that over time and with any impurities or corrosion, RCL networks
would form throughout the stranding. So, preclude the problem out of the gate by
using solid core. (There were other advantages as well that slip my mind at the
moment. But, solid core makes a lousy stage mic cable! No such thing as a free
lunch.)
The Belden 12-2 is an interesting product: it's essentially house wiring (solid
copper), but in a twisted pair format with special insulation -- not for audio, but
to meet fire alarm system specs in extreme environmental conditions.
Turns out to be a coup for audio. The copper is high-purity for low corrosion issues
over a long number of years. The twisting helps here in preventing RF getting back
into the system at the amp output (yes, RF can do this). The insulation dialectic
is quite good and very stable.
>truly ENORMOUS speaker setups - that is, concerts. No studio or home
>sound system would require any more current than can be carried over
>average 14-2 house wiring, which will take a full 15 amps (and more -
>it's underspec'ed, like many things) before doing anything weird, like
>melting. And that's solid-core.
We're not talking running 15A into the speakers; the objective is to push any small
parasitics (some of which might be dynamic in nature) so far down that they're never
an issue. Oversized wire capacity is one way to help in this -- but again, rational
price/performance comes into play.
With speaker efficiency here at the listening levels employed, 22 gauge wire would
be fine for the speakers purely from a load capacity standpoint. But there would be
other potential problems, particularly with R. 4-0 wire (1/2" thick feeder cable)
would be ridiculous overkill, as would pure silver, though it might be a nice choice
if the price were similar to copper.
Instead, the idea with the Belden product (and the thinking of the A&M folks) was to
get an copper product optimized for one situation, one where life and property --
and lawsuits -- were at stake. But this meant that production runs meeting specs
were assured; and, that same product just happened to be ideal for another
application, in this case audio. (And believe me, the A&M folks did a lot of trials
on a lot of wire before settling on this one. And, when you're a six room facility
with top-flight engineers getting top money for their service, facility improve
prices were no object. They could have spent huge bucks for interconnects, but
didn't do so.)
>Incidentally, if you want an instance of a wire which COULDN'T handle
>the current... I once shorted a lawnmower battery with some 20 gauge
>wire. Or at least I thought I did - there wasn't anything left after
>the current vaporized the copper and insulation. The wrench I
>subsequently dropped across the terminals, however, was welded to the
>soft lead terminals - note to self, cover terminals or be more careful
>next time. That took a hammer to remove.
Yep. You've got the perfect welding configuration: high current, low voltage.