Low level interconnect (RCA) measurements

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
IMO, at least at this second in time, none of those effects should be happening at audio frequencies

True

the cables are not being excited at frequencies where we have to abandon lumped time constants or worry about transmission lines

Cellphones, wifi, radio, switching power supplies, cable box, TV, PC, etc

Got any of those in the vicinity ?

The usual practice of using unshielded boxes, isolating RCAs from the back panel etc, means noise pickup is "optimized" to be the maximum possible :D

I'd bet that RF effects in cables explain lots of stuff...
 
udok, thanks so much for doing that. I've never really understood how to choose the loss components for a cap. Did you trial and error it, or is there some guideline as to proportioning them? It would appear you're clairvoyant- you modeled the inputs of the 1A7A plug-in correctly- you might own one or could have looked it up, but how did you know that the value of my Pomona input cable was 100 pF? I just measured it- 100 pF exactly!

When I have time I want to create a couple comparison wav files. What bothers me here is the errors don't fit my beliefs on what should be audible, yet they clearly are, unless I'm fooling myself. Surely that's never happened!

peufeu, I don't own a cell phone and I live in a Faraday cage. Well, OK, at least the first is true, but my rural RF setting is probably very different than if I lived in town or in a city. I've seen trouble with internally generated RF from supplies and agree that RF is often overlooked. The trouble is it just has to be taken on a case by case basis- in my experience only the most obvious problems have consistent symptoms, the rest can be subtle.
 
Last edited:
The resistances in the paralleled legs that create the DA equivalence has the effect of isolating the instantaneous pulse from the scope.
The scope initially sees only the effect of the pure capacitance and gives a vertical step. after a short delay the resistance fed microcapaitance is progressively seen and the scope reading rises progressively to a final value that represents all the micro capacitances summed to the pure capacitance.

This is much the same as a normal discharging.
We see the initial drop to near zero volts and on removing the load, the micro capacitances release their charge through their resistances, to recharge the pure capacitance.

Udok's sim shows this beautifully.
 
Last edited:
That's a good way to proceed. We can then try them out with the Foobar ABX ///

No harm in trying, but listening to a recording of a cable using many other cables is an intuitive needle in a haystack.

I think a blind-folded test subject listening to the weakest interconnect versus the Teflon interconnect, inserted by someone else via rolling a dice, should work nicely in theory, just "which cable is it now?", a single sound in present tense, identify X, instead of three sounds, A... B... X..., then you have to think "in past tense" as well. :)

If you see any errors in my blind test model please let me know.
 
Last edited:
Conrad Hoffman said:
Curious what people think. Misguided bunk or a clue?
Looks like a filter to me i.e. no distortion, just a filter. As it appears to be merely a tiny LF boost DA must be suspected.

It is easy to produce a frightening (time domain) graph, but more difficult to relate this to sound degradation. Cable vendors do this all the time, knowing that their customers are easily frightened and can't distinguish mild filtering from distortion.
 
No harm in trying, but listening to a recording of a cable using many other cables is an intuitive needle in a haystack.
...

The rest of the system, cables included, shouldn't matter. All I'm looking for is a change in some aspect of the sound that says, at some level, cable differences can be audible. Or, that even with the worst of the worst, it doesn't make any difference.
 
Of course cables can make a difference, but only if the cables or the equipment are unusually badly designed.

A really bad cable will allow in interference or even hum. This is how some audiophile cables work (that is, the ones which actually make a difference).

A really poor source will have high output impedance, and this will react with the cable capacitance to produce an HF rolloff - 'smooth' sound. If the cable has very high capacitance and the source has very high output impedance then there is a small chance that dielectric non-linearity could introduce a little HF distortion (especially if the cable is deliberately 'poled' with a battery) or dielectric DA could add a tiny bass or subsonic boost.

This is why competent cables attached to competent equipment all sound the same i.e. no sound at all.
 
Now look, we know that's completely ridiculous because of so much anecdotal evidence to the contrary.

I am, however, always willing to change my beliefs if the evidence is there. Thus, I did some more testing, constructing a small switch board so I could instantly and silently connect and disconnect the "bad" cable. I listened with music, tones and noise signals. I listened with speakers and headphones. Unless I did something really wrong, always possible, the results were crystal clear:

There was no detectable difference between any cables I compared. None, zero, zip and nada.

The brain and listening biases are an evil thing, but I've changed my page to reflect the new results.

Alas, this is the same thing that's happened when I looked into the differences between different metallic conductors and several other things. The closer one looks into the supposed differences, the further they recede, until all that remains is the grin, which slowly fades from view as well.

In general, we can measure far better than we can hear.
 
Conrad Hoffman said:
There was no detectable difference between any cables I compared. None, zero, zip and nada.
Alas, this is the same thing that's happened when I looked into the differences between different metallic conductors and several other things. The closer one looks into the supposed differences, the further they recede, until all that remains is the grin, which slowly fades from view as well.
Surely you are not expecting us to believe that naive circuit theory and materials science, as taught to millions of EEs down the years, is fully applicable to audio connections?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.