Don't skimp on interconnects, they are very important...

So when I was messing with home brewed interconnects, the HF consistently sounded brighter, but more detailed and smoother with silver wire than copper .. thin (22-24ga) solid core wires, laid out parallel on masking tape .. it was consistent effect, and I noticed a similar sound to silver when I compared manufactured interconnects ..

Why would that be likely ??
 
Well, you have to remember that essentially electrons don't flow- the energy is exchanged primarily through the propagation of fields at a significant fraction of the velocity of light. The "drift" velocity of electrons is quite low.
The question is actually pretty difficult to answer in any kind of definitive way.
The fields propagate within the body of the conductor and it's because of this that skin effect exists. So, near surface conductivity has little effect at LF and more effect as the frequency goes up. However, the skin depth is proportional to the square root of the resistivity (the fields penetrate further for a lower resistivity) so the skin depth for copper which has 6% higher resistivity than silver is about 3% greater. As a result the tiny difference in net conductivity for a copper wire coated in silver is at least somewhat offset by a better skin depth. The skin depth of copper at 10kHz is 652u, and for Ag it's 634u, and at 20kHz it's about 70% of this. So, if you keep the wire diameter or annulus depth substantially less than 1mm (0.04") skin effect is essentially irrelevant for audio.
18gauge solid wire has a diameter of c. 1mm, so if you really care you should make your conductors out of 18 gauge or smaller wire and bundle a bunch of them in parallel, and Cu would be better than Ag.
Of course, all you're really doing is changing the frequency dependence of the wire resistance, so the best answer is to make the wire as short as possible, and if you really care about the HF then depending on your speakers the speaker wire inductance might become important, so having the return conductor tightly coupled to the signal conductor becomes essential (it minimizes the "loop area" and hence the inductance , as does having multiple parallel paths.
And this doesn't even address the issue of capacitance.
If this sounds confusing- it is. There really is no perfect answer- but there also seems to be no reason to assume that Ag is superior to low defect Cu.
By the way, Ag also oxidizes more readily than Cu, however due to electrode potentials Ag coated Cu is more resistant to corrosion than Cu or Ag on their own, so military wires are often Teflon coated silver plated Cu. However as the environment that these wires are designed to resist is not found in the home, there really seems to be no need for it.
This is entirely consistent with my own subjective / uneducated perception of interconnects / sound.

I'll show you guys what I use, when I have more time.

It's mil spec 20 gauge stranded, OFC copper plated with Ag.

Effects the high end only.
 
Thank you Wynn ..

I have to reread that a few more times but I think that you just at least partially answered my silver question ..
 
I switched to 2497 from Blue Jeans LC-!, a big improvement for me as well. I wonder if the double shielding in the bjc's is the culprit. They sound rolled off in comparison.
interesting, the RCA interconnects that I am planning on making this weekends actually use double shield, these will be the longest cables that a have ever used ~ 10 feet so I thought id giv double shielding a try. I will compare them against a pair of awesome neotech cables to see whats what. I will use the KLE RCAs.

I will also compare them with a pair of pure silver RCAs that use radioshack brass connectors.
 
Yours definitively ...
Making the point that at some level the other components in the chain mask the subtle changes that cables make ..
Obviously a 4 watt transistor amp into 4" speakers isn't gonna benefit much, and I hesitate to start a flame war by mentioning receivers, but obviously under-powered receivers with veiled loudspeakers aren't going to point up much difference either ..(although I'd bet that a lot of them would be surprised at the difference that a good set of speaker cables could make)

I've just noticed that denials of cabling effects on the sound of the system rarely come from folks who are listening to high end, high resolution systems .. the whole mid-fi, hi-fi argument gets nowhere fast because often folks get defensive about whatever they're listening to .. when the system resolves those small details of the music that add realism, I call it high resolution ..

Yours is because a VPI is all about vibration control and therefore fine movements of the cartridge are resolved and your downstream chain isn't masking any detail either.. a tonearm with a penny on top can't..
The RME ADC/DAC is measurable as well as theoretically the highest resolution element in the system, followed by the Otari professional tape deck at 15IPS. I'm currently in the process of getting people to compare the digitized output from the turntable and tape deck versus the original.
Everyone is certain that they can do it, but as long as I keep the signal processing to a minimum, the level matching near perfect and the digitization at 24/96kHz no one has yet been able to actually do so in a way that I would consider to be statistically significant.
The RME is comparable in price to the refurbished Otari and about 1/5th the price of the VPI system, just to put it in perspective.
 
The RME ADC/DAC is measurable as well as theoretically the highest resolution element in the system, followed by the Otari professional tape deck at 15IPS. I'm currently in the process of getting people to compare the digitized output from the turntable and tape deck versus the original.
Everyone is certain that they can do it, but as long as I keep the signal processing to a minimum, the level matching near perfect and the digitization at 24/96kHz no one has yet been able to actually do so in a way that I would consider to be statistically significant.
The RME is comparable in price to the refurbished Otari and about 1/5th the price of the VPI system, just to put it in perspective.

Vinyl is expensive .. all that machinery has to be high tolerance or you limit resolution .. add in a preamp, cables, cartridges ..
 
Last edited:
I've been casually interested in scientific evidence related to interconnects and their impact on the quality of sound emanating from hi fi components ever since I saw a set of RCA cables going for $50 back in the 90's. There are all kinds of studies that can be found online; some are quite solid (theses, juried studies in journals, etc.). Others are independent small scale studies that individuals have conducted. The latter are typically flawed by researcher bias. I think this area is ripe for a large scale meta analyses that examines not only the body of quantitative research, but qualitative and mixed methods as well since that seems to be where the real controversy lies.

Forgive me if this has been done (I'm sure it has). I've been retired for five years, and I try to keep up with my guitar skills instead of scholarly pursuits these days!

After reading many different studies over a period of about 20 years this is what I have concluded for the time being:

Quantitative studies indicate that there is little difference between the same basic type of interconnects of the same length. The exception is total harmonic distortion where researchers consistently demonstrate measurable differences between cables. I couldn't determine whether these differences were statistically significant or audible.

Qualitatively, we quickly enter the area of psychoacoustics. If you perceive that expensive interconnects are improving the audio performance of your system, who's to say that this is not valid? If you are enjoying your Petula Clark records more after spending thousands on several pairs of RCA cables, I'm not going to tell you, "No you're not!" The fact is that you probably are. The fact also may be that your listening methods have changed based on expectations of those expensive wires, but the perceived quality of the cables is real.

And let's not forget that our entire economic structure is based on perceptions of how the economy is doing. Perceptions are powerful and they are connected to both psychological and physiological responses.

The one consistent claim that I have seen in the quantitative literature is that there is no significant correlation between a cable's measured performance and its price. You can find many instances where a $5 RCA cable measured lower THD than a $200 cable of the same length.
 
Last edited:
I've been casually interested in scientific evidence related to interconnects and their impact on the quality of sound emanating from hi fi components ever since I saw a set of RCA cables going for $50 back in the 90's. There are all kinds of studies that can be found online; some are quite solid (theses, juried studies in journals, etc.). Others are independent small scale studies that individuals have conducted. The latter are typically flawed by researcher bias. I think this area is ripe for a large scale meta analyses that examines not only the body of quantitative research, but qualitative and mixed methods as well since that seems to be where the real controversy lies.

Forgive me if this has been done (I'm sure it has). I've been retired for five years, and I try to keep up with my guitar skills instead of scholarly pursuits these days!

After reading many different studies over a period of about 20 years this is what I have concluded for the time being:

Quantitative studies indicate that there is little difference between the same basic type of interconnects of the same length. The exception is total harmonic distortion where researchers consistently demonstrate measurable differences between cables. I couldn't determine whether these differences were statistically significant or audible.

Qualitatively, we quickly enter the area of psychoacoustics. If you perceive that expensive interconnects are improving the audio performance of your system, who's to say that this is not valid? If you are enjoying your Petula Clark records more after spending thousands on several pairs of RCA cables, I'm not going to tell you, "No you're not!" The fact is that you probably are. The fact also may be that your listening methods have changed based on expectations of those expensive wires, but the perceived quality of the cables is real.

And let's not forget that our entire economic structure is based on perceptions of how the economy is doing. Perceptions are powerful and they are connected to both psychological and physiological responses.

The one consistent claim that I have seen in the quantitative literature is that there is no significant correlation between a cable's measured performance and its price. You can find many instances where a $5 RCA cable measured lower THD than a $200 cable of the same length.

As far as I remember the THD+N numbers were at the <-90dBv level for a standard 1dBv signal and the IMD was c. -100dBv- and the THD+N was essentially thermal noise not harmonic distortion.
The IMD was close to the measurement limit of the test gear. The evaluation was for interconnect- not speaker cable.
All of these numbers are, based on psychoacoustic evaluations, inaudible.
So essentially, cables look like lumped components with negligible non-linearity, the cost has no bearing on the "performance", and it all doesn't matter anyway...
In any case- these results are simulatable. Knowing the RLC lumped equivalent network of the interconnect, knowing the output impedance of the source and the load impedance, you can determine the change in frequency response and the decrease in SINAD of the system.
The other aspect is that changes in load CAN effect the performance of the source, and to a much lower extent, the performance of the load. Low impedance/capacitive loads can cause harmonic distortion in the source, high source impedance can cause distortion and as stated above, bandwidth reductions in the load- but these are actually design imperfections in the source and the load and not due to the cable per se...
This certainly applies to small signal interconnects. For speaker cables the situation is more complex as the load impedance is much smaller and more complex and there are issues such as amplifier stability that come into play...
 
I've been casually interested in scientific evidence related to interconnects and their impact on the quality of sound emanating from hi fi components ever since I saw a set of RCA cables going for $50 back in the 90's. ..And let's not forget that our entire economic structure is based on perceptions of how the economy is doing. Perceptions are powerful and they are connected to both psychological and physiological responses.

The one consistent claim that I have seen in the quantitative literature is that there is no significant correlation between a cable's measured performance and its price. You can find many instances where a $5 RCA cable measured lower THD than a $200 cable of the same length.

Agreed, but qualified by the fact that the measurements referred to are "electrical"..
Hearing is not electrical... It is, and will always be a "pychoaccoustic" phenomenon ..
A deaf man isn't going to "perceive" the differences .. without perception there is no noise ..

Essentially.. the lack of electrical differences doesn't presuppose a lack of audible differences ..
 
Last edited:
Agreed, but qualified by the fact that the measurements referred to are "electrical"..
Hearing is not electrical... It is, and will always be a "pychoaccoustic" phenomenon ..
A deaf man isn't going to "perceive" the differences .. without perception there is no noise ..

Essentially.. the lack of electrical differences doesn't presuppose a lack of audible differences ..
It does presuppose a lack of audible differences that can be determined in careful experiments, and everything else is just opinion.
 
So when I was messing with home brewed interconnects, the HF consistently sounded brighter, but more detailed and smoother with silver wire than copper .. thin (22-24ga) solid core wires, laid out parallel on masking tape .. it was consistent effect, and I noticed a similar sound to silver when I compared manufactured interconnects ..

Why would that be likely ??
That's not enough information to even hazard a guess. Number of wires, spacing, routing of signal and return, source impedance, load impedance, length, dielectrics, separation from ground, coupling to other signals, all make an electrical difference that may become an audible difference, and not just the gauge and material.
 
It does presuppose a lack of audible differences that can be determined in careful experiments, and everything else is just opinion.

Thus spoke Zarthustra ..
Are you stating that every acoustic phenomenon has an associated electrical measurement ??
or that Perception is electrical .. ???

In which case I'd like the measurements for the smells of skunk and rose ..
 
Thus spoke Zarthustra ..
Are you stating that every acoustic phenomenon has an associated electrical measurement ??
What I am saying, precisely, is that two items with identical characteristics- output impedance, frequency response, distortion of all kinds, impulse response, signal to noise, both ratio and spectral characteristics, etc., will not be distinguishably different using any objective metric, including controlled listening experiments.
Those that deviate by very small amounts will also be subjectively identical, under those same conditions, provided some subjective thresholds are not exceeded, and one of the purposes of psychoacoustics is to determine those thresholds.
Many of the supposed differences that people hear do not survive controlled ABX testing, reducing them to the classification of opinion.
 
Thus spoke Zarthustra ..
Are you stating that every acoustic phenomenon has an associated electrical measurement ??
or that Perception is electrical .. ???

In which case I'd like the measurements for the smells of skunk and rose ..
By the way, the chemical make up of skunk and rose can be precisely determined by spectrometry results, which are in actuality electrical (i.e. EM) measurements. So, yes, I could, potentially, give you the electrical meaurements of the smell of skunk and rose, and if I deviate slightly from that definition, you probably could not tell- but my dog could...
 
And the taste of good steak .. good wine , and fine women ..
All three of these items have objective differences that could be measured that may be sufficient for you to declare them better or worse- however the parameter space may be much greater than an audio electrical signal, so I wouldn't necessarily suggest such an approach for say, selecting a mate.
 
All three of these items have objective differences that could be measured that may be sufficient for you to declare them better or worse- however the parameter space may be much greater than an audio electrical signal, so I wouldn't necessarily suggest such an approach for say, selecting a mate.

Hopefully we can agree to disagree .. would that it were as simple as a few resistors out of range....
There would be a far healthier population on the planet if perception were so well understood ..
and it's that non-electrical better or worse judgement that I'm gettin at ..
 
Last edited:
Hopefully we can agree to disagree .. would that it were as simple as a few resistors out of range....
There would be a far healthier population on the planet if perception were so well understood ..
and it's that non-electrical better or worse judgement that I'm gettin at ..
Yes, I'm not fanatical about it- however it's always a fascinating conversation, particularly since, many years ago, I was in the other camp (but I got better :))
 
Back
Top Bottom