• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Interconnect vs Speaker cable Res/Cap/Induct

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
Oh yes, I do see that the cables with complex geometry are doing fairly well. Reactance[Cap/Induct] is the factors that play a lot of this low pass high pass filters.

However, to my knowledge I would have to say an RCA or XLR cable for non professional use will not have this issue. Most cables are suffering with greater lengths. At typical 3 ft RCA or XLR should we any competent build suffer from this issue.

My curiosity just stems from why the difference, and so far we see input and output impedance. The cables I am looking at are not cheap almost 50x more expensive than your average rca cable, and the measurements seem to be riding in the 14 gauge zip cord level which also probably costs somewhere in the 50x lower price.

So just using simple logic, why would you buy an RCA cable. Buy a speaker cable with a drain line and make it RCA. If it doesn't come with it you can simply make your own and save loads of money. However as we see the kicker right now is we do not know what difference in ohms is doing. We can say for certain less watts are going through with higher ohm, like with headphones running 300 ohms, but I am not sure how wattage plays on interconnects which runs on volts I am sure the volts can probably turned into watts and vice-versa.

Was hoping someone that is with an electrical background could chime in with definite results or answers and think we may have it with @DonH56
A number of posters have electrical background, probably higher than mine (BSEE, BSCompE, MSEE, ~40 years inexperience in analog IC design and test, mostly RF stuff).

My work week is insane lately so I am mostly a "fly by" poster; I did not read the whole thread. It came up when I clicked on "new threads". Sorry if my answer did not cover your original questions.

Regarding this statement: "So just using simple logic, why would you buy an RCA cable. Buy a speaker cable with a drain line and make it RCA."

The main reason is that RCA cables are usually coax and that offers much better shielding than speaker cable. That is more important for the small signal levels and high impedances found with low-level components (DACs, preamps, etc.) The outer shield protects the inner signal conductor, assuming the ground is good. The other reasons are mainly practical: coax is much easier to handle than big speaker wires, and fits nicely into RCA connectors, whereas you'll have to futz with speaker cables since they are generally much larger to handle the higher current and power levels speakers require.

At 4 Vrms into 10 k-ohms, an interconnect is handling about 0.0016 W. A speaker cable is handling power on the order of watts, about a thousand times more power. And because impedances are so low and signal levels so high, the chances of picking up noise are much, much lower for a speaker cable, so shielding is rarely an issue for a consumer installation. Maybe for a 100'+ run in a channel with other power and signal lines in a concert hall, but that is rarely the case in a home.

HTH - Don
 
Last edited:

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
Yes, I am fully aware that to avoid issues with frequency response you want the outputting device to be ~10x less than input device, which is why most amps are 10,000 and as you stated an output device would be 1000 at the higher points.
Well, true for audio, where matching is not required. For RF systems, optimal power transfer and signal integrity happens when the source, cable, and load are all the same impedance. For example, your antenna or cable box, cable, and TV receiver are all ideally 75-ohm characteristic impedance. That is a high-frequency impedance, not audio. For analog audio signals, while the cable still has 75-ohm impedance (*), it just does not matter as you are not working in a matched environment and the frequencies are so low as to be DC to the cable.

"These cables are designed to carry GHz RF signals" so is that what you are saying is coming out of the dac? the dac has no limit to frequency output? I think from amirs video's, which I did not really comprehend is higher 192khz is actually playing out to 192khz, I did not actually correspond those sample rate to what the dac is putting out. Didn't tie that together until it was pointed out. I do not want to talk about this subject, but at 192 khz you are able to hear about half of that. 44.1khz you get about 22k out which covers the entire band. Lets not talk about this though, this is just a light bulb moment.
No, the cable itself is usually an RF cable, because it is cheap and widely available. It is often the same cable (impedance, though larger in diameter) used by cable companies to send the signal many miles, and used for both video and some digital signals that are much higher bandwidth than analog audio. The cable will support signals to about a GHz. The DAC is putting out analog audio, with a bandwidth usually about half the sampling rate (look up Nyquist). So a 44.1 kS/s DAC has a top analog output around 22 kHz, and 192 kS/s tops out around 96 kHz. Coax is used because it is cheap, shield against noise, and easy to work with.

Yes, logically I did tie the conductor size to different electrical characteristics. With the tests on audioholics regarding speaker cables that's the pattern I saw. Even with two cables of equal gauge the proper geometry cable was having far superior performance in some aspects, but reactance is acting up. I mean on longer speaker cables this maybe something to consider, not a huge deal on interconnects especially for at home use.
Speaker cables handle much higher power and work at much lower impedances. Audio is still essentially "DC" so the only thing that really matters the vast majority of the time is the DC resistance, not the characteristic impedance. Coax is usually specified in terms of characteristic impedance for RF signals, but specs listed usually include R, L, and C per unit length. Unless you have hundreds of feet or more of cable the only time any of that matters for an interconnect is for the cable connection from a phono cartridge to the phono preamp, since most (MM) cartridges drive a very high-impedance load (e.g. 47 k-ohms) and are sensitive to the load capacitance. Coax cables run around 30 pF/foot, so you may have to change the preamp's load setting if you have a long cable.

I think probably to sum up everything, is that even the most expensive interconnect cable with greatest measurements is not going to provide better sound but rather not create any limitations or degradation to the frequency response.
IME the better interconnect cables use better connectors and have overall better construction, like welded connections and 100% shield vs. the 90% or so of standard standard coax (which is still plenty good -- that is what I have in my system). As for measurements, I would bet the vast majority measure about the same independent of price. It is hard to imagine a case wherein the cable would limit the frequency response in a home installation.

On a sidenote

I am very new to audio, just started in september of 2020, so I am just over a year and half in. Sometimes I get these random thoughts that I would like to learn about or better understand. Even after watching and investigating so much, and now even further justified by Amir and Audioholics. Just a very general knowledge I have in physics has told me that I am not going to invest a foolish amount of money into cables, it doesn't really make sense at all.
Learning is good, common sense is even better. Actually hard to have one without the other.

I just use some Monolith RCA cables and some mogami XLR's which measurement wise do have some good noise rejection with testing to prove it. Honestly see no issues there and compared to my old cables I do see that it[RCAs] does "sound" better which maybe just placebo. The only other thing that makes me wonder, and I do not wish to discuss it as we have no way of proving it, is if different metals have any sound characteristics as most people say. Measurement wise I do not think there will be a way to tell, and I am not sure if different metals will have different electrical characteristics which is what people are hearing. Even then that one is strange to me because you have a cable lets say is all silver, and then you put it on connectors which are silver plated and brass under or copper, and you would have to assume that's almost bottlenecking any performance you would actually gain.
I don't know anything about Monolith RCAs. Mogami is good cable, along with Canare and some other brands I have used over the years in both professional installations and at home. I think different metals having different sound is perception bias. Yes, different metals have different characteristics, easily measured in a lab, but for audio those differences are irrelevant assuming common materials like copper, silver, gold, nickel, etc. I have noticed a consistent trend for silver to sound "bright" and copper "warm" but personally think people are hearing with their eyes.

The only final factor which comes into play crystal structure and the crap they have in marketing. I honestly cant even truly understand that and my current research which started this topic shows that a cable with 99.99999 cable is actually measuring worse than it's 99.99 counter-part and costs double. So not seeing any real influence there. Also considered conductor size on this matter, and the better measuring cable does have twice as much conductor, however the rest of the performance is almost 3 times as bad so the purity of the conductor plays no role.
Crystal structure is a big deal for some cases but not audio. A cable for a 100+ GHz test system is critical and has special needs. A cable for a 20 kHz (or 100 kHz) audio system, not so much. Ditto the dielectric used (the material between center conductor and outer shield); it matters very much in some applications requiring very wide bandwidth or ultra-low noise, for applications many orders of magnitude grater than required for audio. Ironically, some of the dielectrics highly-regarded by audio marketeers, are worse for noise than cheaper, more common dielectrics. It does not matter because the noise is many orders of magnitude below what we can hear (and is difficult to measure even with fancy test equipment).

There have been endless marketing adverts over the years showing minute differences blown up on a graph to indicate vast superiority that is just not audible. A certain cable manufacturer showed a plot with their cable having much greater bandwidth and much lower loss than a competing "no-name" cable. The catch was the plot scales were not shown, and later found to be fractions of a dB and MHz in frequency. So the differences were real, and measurable, but ridiculously beyond audibility.

And so it goes - Don

(*) Characteristic impedance actually gets a little fuzzy at very low (and very high) frequencies, but for our purposes it does not matter.

Edit: Many years ago there was a big debate about how RF effects in speaker cables mattered, and I wrote a little blurb about it. I think it was transported to ASR, check the articles linked in my signature. The bottom line is that you can do the analysis, but for homes the effects die out in ns or so IIRC, far, far, far! faster than any speaker can respond or we can hear them.
 
Last edited:
OP
MacCali

MacCali

Major Contributor
Joined
Dec 21, 2020
Messages
1,122
Likes
539
The variations in metals that are used will create higher or lower resistance in the cable. When dealing with speaker cable the load resistance/speaker is say 4 or 8 ohms and the source impedance/amplifier is down in the fractions of a ohm or better. So the voltage drop across the speaker wire will not change things very much unless it is high enough to cause a decrease in sound output from the speaker.

A conductor has a certain range of electrons in the outer valence ring and that makes it a conductor or a semi-conductor. Normally, a conductor has three or less valence electrons, an insulator has five or more valence electrons, and semiconductors usually have four valence electrons. All the elements of which matter is made may be placed into one of three categories: conductors, insulators, and semiconductors.

(Simplified) So by changing the metals used in the speaker wire the resistance will change due to the quantity of valence ring electrons in the specific metal. The frequency response does not change due to the metal type used. The speaker wire is represented in a model circuit as a resister. A model resister does not affect frequency; it only affects the current through it and the voltage drop across it.

So the long and short of speaker cable is use the best conductor for the job although don't buy snake oil. 12 gauge is the best value for the money and does the job very well.
Indeed, I do know that 12-13 awg is the sweet spot. Just having fun and learning lol. There’s a whole bunch of crap to these damn cables and dielectrics etc.

I do not want to get into that but feel like those probably play some type of roll.

Also on another topic I’m not sure what signal propagation is but they talk about that bs too
 
OP
MacCali

MacCali

Major Contributor
Joined
Dec 21, 2020
Messages
1,122
Likes
539
Well, true for audio, where matching is not required. For RF systems, optimal power transfer and signal integrity happens when the source, cable, and load are all the same impedance. For example, your antenna or cable box, cable, and TV receiver are all ideally 75-ohm characteristic impedance. That is a high-frequency impedance, not audio. For analog audio signals, while the cable still has 75-ohm impedance (*), it just does not matter as you are not working in a matched environment and the frequencies are so low as to be DC to the cable.


No, the cable itself is usually an RF cable, because it is cheap and widely available. It is often the same cable (impedance, though larger in diameter) used by cable companies to send the signal many miles, and used for both video and some digital signals that are much higher bandwidth than analog audio. The cable will support signals to about a GHz. The DAC is putting out analog audio, with a bandwidth usually about half the sampling rate (look up Nyquist). So a 44.1 kS/s DAC has a top analog output around 22 kHz, and 192 kS/s tops out around 96 kHz. Coax is used because it is cheap, shield against noise, and easy to work with.


Speaker cables handle much higher power and work at much lower impedances. Audio is still essentially "DC" so the only thing that really matters the vast majority of the time is the DC resistance, not the characteristic impedance. Coax is usually specified in terms of characteristic impedance for RF signals, but specs listed usually include R, L, and C per unit length. Unless you have hundreds of feet or more of cable the only time any of that matters for an interconnect is for the cable connection from a phono cartridge to the phono preamp, since most (MM) cartridges drive a very high-impedance load (e.g. 47 k-ohms) and are sensitive to the load capacitance. Coax cables run around 30 pF/foot, so you may have to change the preamp's load setting if you have a long cable.


IME the better interconnect cables use better connectors and have overall better construction, like welded connections and 100% shield vs. the 90% or so of standard standard coax (which is still plenty good -- that is what I have in my system). As for measurements, I would bet the vast majority measure about the same independent of price. It is hard to imagine a case wherein the cable would limit the frequency response in a home installation.


Learning is good, common sense is even better. Actually hard to have one without the other.


I don't know anything about Monolith RCAs. Mogami is good cable, along with Canare and some other brands I have used over the years in both professional installations and at home. I think different metals having different sound is perception bias. Yes, different metals have different characteristics, easily measured in a lab, but for audio those differences are irrelevant assuming common materials like copper, silver, gold, nickel, etc. I have noticed a consistent trend for silver to sound "bright" and copper "warm" but personally think people are hearing with their eyes.


Crystal structure is a big deal for some cases but not audio. A cable for a 100+ GHz test system is critical and has special needs. A cable for a 20 kHz (or 100 kHz) audio system, not so much. Ditto the dielectric used (the material between center conductor and outer shield); it matters very much in some applications requiring very wide bandwidth or ultra-low noise, for applications many orders of magnitude grater than required for audio. Ironically, some of the dielectrics highly-regarded by audio marketeers, are worse for noise than cheaper, more common dielectrics. It does not matter because the noise is many orders of magnitude below what we can hear (and is difficult to measure even with fancy test equipment).

There have been endless marketing adverts over the years showing minute differences blown up on a graph to indicate vast superiority that is just not audible. A certain cable manufacturer showed a plot with their cable having much greater bandwidth and much lower loss than a competing "no-name" cable. The catch was the plot scales were not shown, and later found to be fractions of a dB and MHz in frequency. So the differences were real, and measurable, but ridiculously beyond audibility.

And so it goes - Don

(*) Characteristic impedance actually gets a little fuzzy at very low (and very high) frequencies, but for our purposes it does not matter.

Edit: Many years ago there was a big debate about how RF effects in speaker cables mattered, and I wrote a little blurb about it. I think it was transported to ASR, check the articles linked in my signature. The bottom line is that you can do the analysis, but for homes the effects die out in ns or so IIRC, far, far, far! faster than any speaker can respond or we can hear them.
Very well put and understandable. Appreciate your time, some of the things you mention in your last two comments are things I have heard in the past.

But ties it all up for a better more thorough understanding and better comprehension.

I think this about wraps it up. And want to thank everyone for any input provided.
 

Audiofire

Addicted to Fun and Learning
Forum Donor
Joined
Feb 8, 2022
Messages
625
Likes
347
Location
Denmark
So we have a general idea of what makes a speaker cable good when it comes to electrical properties like resistance, capacitance, and inductance. How come a lot of interconnect cables have so much higher readings?

Is it strictly based on conductor size? that's the only thing I can assume. The cables inside an interconnect dont tend to be 12 awg. Does it simply just come down to this principle?

Is it bad because it's much higher? I do not understand the precise difference between a signal coming from a dac into an amp vs exiting an amplifier going to a speaker. Frankly it seems like the same concept.
Resistance is determined by the gauge of the wire, but not capacitance and inductance. Those two are instead determined by distance to the neutral conductor. All three parameters are transmission line effects for normal/cheap audio cables. That means you would need an extreme length for your hi-fi cables, such as a mile/kilometer long.

Shielded or twisted pair or quad core. No straight wire.
I would be interested to know your reasoning. Shielded? Yes, but not for speaker wire that would possibly have an increased capacitance from shielding without any benefit. Twisted pair? Not for unbalanced audio cables. Quad core? Maybe for microphone cables. Only straight wire for unbalanced audio cables!

In either case twisting wires (reduces inductance, increases capacitance) is beneficial since it exposes both wires evenly to external electric or magnetic fields.
Does that really matter for unbalanced audio cables? My QED Profile Audio cables are unbalanced twisted pairs with RCA connectors. Customer support also told me about such suppositions as you. I then found out that I regretted buying the cables and will sell them, because QED is way too much involved in woo-woo reasoning.

For MIT Cable, they explain that cable can exist like low pass, band pass or high pass. This you have to read their white paper in their website. Generally, people treat cable as low pass but the frequency roll off is away above audible frequency. Many just ignore till it is able to damage on very niche cases like amplifier oscillation. Also, you will heard not to coil up your wire, it will cause inductance and deliver as low pass filter. I dunno what the value of this act that affect the audible frequency.

For information about shielding, twisting for two bare wire, quad core twisting. I think the website have enough explanation on it.
So I took a look at that and MIT means Music Interface Technologies. That company has some of the most laughable audiophoolery I have ever seen. Maybe, you can try to explain the information on their website...

Interconnects are typically coax like RG-59, RG-6 or similar cables for RCA
It is often the same cable (impedance, though larger in diameter) used by cable companies to send the signal many miles, and used for both video and some digital signals that are much higher bandwidth than analog audio. The cable will support signals to about a GHz.
Isn't it a little misleading to reference the RG cables? Those are 75 ohm characteristic impedance cables that have almost no measurable attenuation at around 1 MHz. The characteristic impedance of RCA connectors doesn't matter for audio. There was visible attenuation at around 200 kHz on the RCA cables here:

Audio is still essentially "DC" so the only thing that really matters the vast majority of the time is the DC resistance, not the characteristic impedance.
Isn't it a little misleading again to call audio frequencies DC when they are absolutely AC, while DC has no frequency at all?
 
Last edited:

fpitas

Master Contributor
Forum Donor
Joined
Jul 7, 2022
Messages
9,885
Likes
14,191
Location
Northern Virginia, USA
Terminating speaker cables is the best overall solution. Then you needn't worry about the L and C components very much. Here's an article discussing that:

 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
Isn't it a little misleading to reference the RG cables? Those are 75 ohm characteristic impedance cables that have almost no measurable attenuation at around 1 MHz. The characteristic impedance of RCA connectors doesn't matter for audio. There was visible attenuation at around 200 kHz on the RCA cables here:
Read again all of my post wherein I explain why it does not matter for audio and why "RF" cables are used at audio. The comparison was meant to show that a cable capable of carrying RF signals for thousands of feet or more will have no problem carrying audio signals in your house.

The bandwidth depends on many factors, all of which have been hashed out many times before, and I am disinclined to rehash it here. You can also take a look at https://www.audiosciencereview.com/forum/index.php?threads/interconnect-bandwidth.25441/ for an earlier article on interconnect bandwidth.

Isn't it a little misleading again to call audio frequencies DC when they are absolutely AC, while DC has no frequency at all?
Notice "DC" is in quotes. Compared to GHz (or even MHz) RF signals, audio is essentially "DC". The radar front-end guys I used to work with considered the 1 GHz receiver IF output "DC" in casual conversation when compared to the 40~100+ GHz input signal at the antenna.

Perhaps I have become too old, jaded, disingenuous, and misleading for ASR... I am missing the "fun" part of the forum lately.

Gotta' get back to work - Don
 

Chrispy

Master Contributor
Forum Donor
Joined
Feb 7, 2020
Messages
7,751
Likes
5,910
Location
PNW
Resistance is determined by the gauge of the wire, but not capacitance and inductance. Those two are instead determined by distance to the neutral conductor. All three parameters are transmission line effects for normal/cheap audio cables. That means you would need an extreme length for your hi-fi cables, such as a mile/kilometer long.


I would be interested to know your reasoning. Shielded? Yes, but not for speaker wire that would possibly have an increased capacitance from shielding without any benefit. Twisted pair? Not for unbalanced audio cables. Quad core? Maybe for microphone cables. Only straight wire for unbalanced audio cables!


Does that really matter for unbalanced audio cables? My QED Profile Audio cables are unbalanced twisted pairs with RCA connectors. Customer support also told me about such suppositions as you. I then found out that I regretted buying the cables and will sell them, because QED is way too much involved in woo-woo reasoning.


So I took a look at that and MIT means Music Interface Technologies. That company has some of the most laughable audiophoolery I have ever seen. Maybe, you can try to explain the information on their website...



Isn't it a little misleading to reference the RG cables? Those are 75 ohm characteristic impedance cables that have almost no measurable attenuation at around 1 MHz. The characteristic impedance of RCA connectors doesn't matter for audio. There was visible attenuation at around 200 kHz on the RCA cables here:


Isn't it a little misleading again to call audio frequencies DC when they are absolutely AC, while DC has no frequency at all?
Hope you don't expect a rational answer from kongwee....he's way "out there"
 
Top Bottom