Cable Facts

SergeAuckland

Certified Measurist
Wammer
May 6, 2008
18,607
1,915
173
Bury St Edmunds, UK
AKA
Serge
HiFi Trade?
  1. No
Hi MickCable measurements can tell us more than just impedance, or DC resistance. As S2000db has said, the only noticeable audible difference in two cables with differing resistance could be a slight volume difference. I'm sure you've seen the articles on cable manufacturers demonstrating their cables against a "standard" cable where during a show, they'll craftily turn the volume down when demonstrating the standard one and then turn it up a tad demonstrating theirs so that the audience will "hear" more detail etc. Some cables are quite high in their capacitance and depending on the output and input impedances of your kit, a high capacitance cable can rob a system of some top end detail. For those with perfect hearing, they may be able to detect a few dB rolled off above 15KHz so that cable may be the one to "tame" bright systems. For most though, there's little evidence (that I can find anyway) that even a 1m pair with say 250pF capacitance makes much difference in most systems. The single greatest impact on how a cable sounds IMHO and experience is its S/N ratio. Two identical cables with the same resistance values to the signal core could sound very different if one is poorly screened and the other is well screened. There's a misconception that screening is only needed or of benefit in noisy environments but as (more) importantly, the lower the screen (or signal return) impedance, the better the S/N ratio and for a given capacitance, in theory the more dynamic range and detail you'll hear. Most differences in commercial cables are down to S/N ratio and perhaps that's what most people hear as differences. The minimum accepted hifi standard is an S/N ratio of 100. I've made cables ranging from S/N of 90 to 140 and there was an audible difference between them in my system. Hopefully this explains why calculations or measurements are still valid when comparing cables.
Paul,

Firstly, I am unaware of any standard for S/N ratio so why 100dB? Perhaps because that's better than CD provides, but why not 103dB or 110dB?

Secondly, the S/N ratio depends both on the N and the S, and as the N is fixed for any one cable or installation, it then depends on the S. A bigger S will provide a bigger S/N ratio. Unless you mean something different by N, like the thermal noise in the cable.

Noise pickup will depend on the field strength for the interfering noise, the impedances involved at both ends of the cable, and of course on the amount of screening. This is effectively per unit length, so a long cable will pick up more interference than a short cable, which is why long cables are preferably balanced.

There have been integrated amplifiers with very short unscreened links between pre and power sections, and yet the S/N ratio of the amplifier is unchanged when removing the unscreened links and replacing these with a length of screened cable. If screening affected the S/N ratio, I would have expected it to me worse with the unscreened links.

S.

 
P

Pac67

Guest
Paul,Firstly, I am unaware of any standard for S/N ratio so why 100dB? Perhaps because that's better than CD provides, but why not 103dB or 110dB?

Secondly, the S/N ratio depends both on the N and the S, and as the N is fixed for any one cable or installation, it then depends on the S. A bigger S will provide a bigger S/N ratio. Unless you mean something different by N, like the thermal noise in the cable.

Noise pickup will depend on the field strength for the interfering noise, the impedances involved at both ends of the cable, and of course on the amount of screening. This is effectively per unit length, so a long cable will pick up more interference than a short cable, which is why long cables are preferably balanced.

There have been integrated amplifiers with very short unscreened links between pre and power sections, and yet the S/N ratio of the amplifier is unchanged when removing the unscreened links and replacing these with a length of screened cable. If screening affected the S/N ratio, I would have expected it to me worse with the unscreened links.

S.
Hi Serge

indeed there is no standard for hifi S/N but there are commonly accepted standards. I for one wouldn't use any IC which performed worse than 100 as it defeats the object of benefitting from the low noise floor of modern digital sources and yes, it depends on the other kit, leakage voltages etc. Noise can be from two sources: RFI/EMF or direct from leakage voltages between interconnected kit. Sorry, I should have stipulated that this all refers to single ended RCA applications as things are slightly different with balanced connections.

However, the point is that it's good practice to aim for the lowest possible impedance in the shield, or of a shield isn't present, the ground/return wire. That's why my Litz braided designs are calculated to give a minimum of S/N = 100 for most applications based on averaged leakage voltages. This also refers to a generalised voltage figure for signal. These days, most pieces of kit have published signal voltages, so its fairly straight forward to do the maths.

However, rather than get too hung up on detail, the point hinges around low signal return conductor impedance as being a good thing.

 

SergeAuckland

Certified Measurist
Wammer
May 6, 2008
18,607
1,915
173
Bury St Edmunds, UK
AKA
Serge
HiFi Trade?
  1. No
Hi Sergeindeed there is no standard for hifi S/N but there are commonly accepted standards. I for one wouldn't use any IC which performed worse than 100 as it defeats the object of benefitting from the low noise floor of modern digital sources and yes, it depends on the other kit, leakage voltages etc. Noise can be from two sources: RFI/EMF or direct from leakage voltages between interconnected kit. Sorry, I should have stipulated that this all refers to single ended RCA applications as things are slightly different with balanced connections.

However, the point is that it's good practice to aim for the lowest possible impedance in the shield, or of a shield isn't present, the ground/return wire. That's why my Litz braided designs are calculated to give a minimum of S/N = 100 for most applications based on averaged leakage voltages. This also refers to a generalised voltage figure for signal. These days, most pieces of kit have published signal voltages, so its fairly straight forward to do the maths.

However, rather than get too hung up on detail, the point hinges around low signal return conductor impedance as being a good thing.
Amen to that!

By the way, how do you measure the S/N of the cable? What do you terminate it with? What bandwidth is the measurement made over? Sorry if this is a lot of detail, but S/N of cables isn't anything I've ever bothered with before as it's always been swamped by the noise from the equipment.

S.

 
P

Pac67

Guest
You can meaure it Serge but you don't need to. All you need to do is to input a typical leakage voltage figure for for the "noise" end of the equation, and then use the signal voltage value from your source and calculate the S/N ratio using:

20 Log(Vsig/Vnoise)

A conservative figure for noise is 25uV so you'd need some very sensitive test equipment.

Sounds inaudible but in fact a lot of people are using kit not to it's maximum potential in terms of lowest possible noise floor getting through to amplification and speaker sections because the leakage voltage between connected components isn't being silenced and this is then amplified. It doesn't matter that signal return wiring within amps isn't beefy because the mechanism for most of the noise is the connection between items of kit. You could try an experiment where you use unshielded long (say 3m) IC's using twin and earth then compare the noise floor with a good thin solid core coax signal cable with a good hefty shield. 9 times out of 10 you'll hear an audible difference which can, if needs be, be measured.

 

SergeAuckland

Certified Measurist
Wammer
May 6, 2008
18,607
1,915
173
Bury St Edmunds, UK
AKA
Serge
HiFi Trade?
  1. No
You can meaure it Serge but you don't need to. All you need to do is to input a typical leakage voltage figure for for the "noise" end of the equation, and then use the signal voltage value from your source and calculate the S/N ratio using:20 Log(Vsig/Vnoise)

A conservative figure for noise is 25uV so you'd need some very sensitive test equipment.

Sounds inaudible but in fact a lot of people are using kit not to it's maximum potential in terms of lowest possible noise floor getting through to amplification and speaker sections because the leakage voltage between connected components isn't being silenced and this is then amplified. It doesn't matter that signal return wiring within amps isn't beefy because the mechanism for most of the noise is the connection between items of kit. You could try an experiment where you use unshielded long (say 3m) IC's using twin and earth then compare the noise floor with a good thin solid core coax signal cable with a good hefty shield. 9 times out of 10 you'll hear an audible difference which can, if needs be, be measured.
Sorry, but I don't understand the term "leakage voltage". Is this the DC leakage from the source, or the noise voltage from the source or what?

My millivoltmeter measures down to 15uV full scale, so can easily read down to 2uV with a bandwidth of 100khz.

S.

 
P

Pac67

Guest
Hi Serge

that's the leakage voltage from the source which usually goes to ground and is picked up as an audible hum (although very low in some cases) as it creates a voltage differential between source and amp. The trick is to lower the value of the voltage differential which ideally would be zero. However, thats not possible, so the best we can do is to lower the impedance of the screen or return (ground) and hence less interference noise is picked up by the single ended signal, so producing a higher S/N ratio than say a much higher impedance ground.

 

lehoako

Wammer
Wammer
Mar 16, 2011
37
0
0
That may be a valid personal observation in the 2 Ch debate but as we're on the technical debate section, moreover, as we're discussing things (now only loosely related to) my earlier articles on cable facts, it bears no relationship with any factual basis. There's no evidence for copper conductor's being directional and I would like to take you to task on the statement in your other post that copper produces more distortion than silver.Copper (or any other conductor for that matter) does NOT produce distortion...in fact it produces nothing. Your amp or source components might though. IC's don't produce anything, they simply conduct an electrical signal in the audio bandwidth.

Please read the AUP. Your claims could mislead others and in this section you must either back up those claims with fact or take the argument to the 2 Ch debate. No-one argues with what you hear, but plenty will argue with what you state as fact. If you don't mind me saying so, you don't seem to have much of a grasp of electronics.
Apologies for the OFFTOPIC comments, thanks to the mods for trying to keep the topic from flooding :). Sorry again.

As you mention, I am a dumb for electronics, I agree. That is why I am experimenting, which will lead me to anywhere in this cable-jungle. I try to get a deeper insight in cables/measuring/facts universe, and I'd like to know your opinion about this:

http://www.psaudio.com/uploads/files/PerfectWave_AC_Series_white_paper.pdf

I found it quite informative, especially in my quest for optimal mains supply. I hope also the other fellow wammers would take a minute to read and comment it.

 

SergeAuckland

Certified Measurist
Wammer
May 6, 2008
18,607
1,915
173
Bury St Edmunds, UK
AKA
Serge
HiFi Trade?
  1. No
Hi Sergethat's the leakage voltage from the source which usually goes to ground and is picked up as an audible hum (although very low in some cases) as it creates a voltage differential between source and amp. The trick is to lower the value of the voltage differential which ideally would be zero. However, thats not possible, so the best we can do is to lower the impedance of the screen or return (ground) and hence less interference noise is picked up by the single ended signal, so producing a higher S/N ratio than say a much higher impedance ground.
Still don't understand where this "leakage voltage" comes from. Is it from the air, i.e. electromagnetic fields from all the various sources in a home, or is it from the equipment itself. If the former, then the output impedance of the source and the input impedance of the sink has a bearing on how big this induced voltage is, as well as the resistance of the screen. I agree that a zero screen impedance with perfect screening will reduce the differential voltage to zero, and so is desireable.

In the measurements I've just made taking a cheap white/red interconnect of 60cm length, and terminating one of the pair in 75 ohms, leaving the other end open circuit, into my millivoltmeter with a high impedance input (don;t know exactly how much, perhaps 1Mohm). Without the termination, i.e. open circuit, I get a voltage of 1.5mV, predominantly hum. The 75 ohm terminated side gives 4uV, also mostly hum, which shows to me that the source impedance is very important to the amount of hum induced.

S.

 

lehoako

Wammer
Wammer
Mar 16, 2011
37
0
0
I don't find measurements anywhere, not even on the pages of well established manufacturers/researchers.

However, this might be interesting for the topic too, a brief quote:

Dedicated and Isolated Power Lines

For those that are able to have dedicated lines installed, these are perhaps the greatest gift you can give to the DTCD performance of a recording or playback system. There is a lot of confusion surrounding dedicated lines and isolated outlets among audiophiles. Let's start with the definition of a dedicated line. Most power circuits in a home are daisy chained with multiple outlets on a single circuit. When you “flip” the circuit breaker to the outlet powering your audio system go check the other outlets in the room and adjacent rooms. You will find that several other outlets are being powered by a single circuit breaker. A "dedicated line" is a term used to describe a circuit breaker that is dedicated to a single outlet. It is not connected to any other electrical outlet or switch. The in-wall wires are all dedicated to that single outlet and cannot be shared with any of switch or outlet.

From here:

http://www.shunyata.com/Content/ac_guide.html

 
P

Pac67

Guest
Still don't understand where this "leakage voltage" comes from. Is it from the air, i.e. electromagnetic fields from all the various sources in a home, or is it from the equipment itself. If the former, then the output impedance of the source and the input impedance of the sink has a bearing on how big this induced voltage is, as well as the resistance of the screen. I agree that a zero screen impedance with perfect screening will reduce the differential voltage to zero, and so is desireable.In the measurements I've just made taking a cheap white/red interconnect of 60cm length, and terminating one of the pair in 75 ohms, leaving the other end open circuit, into my millivoltmeter with a high impedance input (don;t know exactly how much, perhaps 1Mohm). Without the termination, i.e. open circuit, I get a voltage of 1.5mV, predominantly hum. The 75 ohm terminated side gives 4uV, also mostly hum, which shows to me that the source impedance is very important to the amount of hum induced.

S.
As I understand it Serge, leakage current predominantly exists from current which is said to leak from other sources directly to ground via the chassis and earth shields. Examples include capacitively coupled current (direct to the ground on a shield) or AC line frequency hum caused by capacitance between AC primary and secondary windings in a components power supply meaning that a small element of AC line voltage will always be coupled directly to signal ground.

Add to that RF interference in AC mains supplies geeting through to the signal, and different components sharing their own safety grounds but coupled by ground shields, there's another element of induced voltage in the interconnect shield.

What I am trying to get people thinking about is that without minimising IC ground impedance, the value of the total "noise" induced into the signal will be (comparitively) high resulting in greater noise predominantly in the form of 50/60Hz hum.

I proved it to myself yesterday, after a hifi move in the listening room, I needed different length IC's from my previous ones, so I made up a new set with higher theoretical S/N ratio (120 v's 100) than my previous set and hey presto, achieved a significant reduction in overall humm from my highly sensitive speakers to the point where it's almost inaudible at the listening position. Another benefit was (with lower noise floor) better detail, and a better overall balance to the sound. I think when most people hear differences in audio cables, a good proportion of what they hear as beneficial is down to the S/N ratio being high.

 
P

Pac67

Guest
I don't find measurements anywhere, not even on the pages of well established manufacturers/researchers.However, this might be interesting for the topic too, a brief quote:

Dedicated and Isolated Power Lines

For those that are able to have dedicated lines installed, these are perhaps the greatest gift you can give to the DTCD performance of a recording or playback system. There is a lot of confusion surrounding dedicated lines and isolated outlets among audiophiles. Let's start with the definition of a dedicated line. Most power circuits in a home are daisy chained with multiple outlets on a single circuit. When you “flip” the circuit breaker to the outlet powering your audio system go check the other outlets in the room and adjacent rooms. You will find that several other outlets are being powered by a single circuit breaker. A "dedicated line" is a term used to describe a circuit breaker that is dedicated to a single outlet. It is not connected to any other electrical outlet or switch. The in-wall wires are all dedicated to that single outlet and cannot be shared with any of switch or outlet.

From here:

http://www.shunyata.com/Content/ac_guide.html
Indeed.

Dedicated power spurs are a good thing. I temnd to switch of my fish tank aeration pump when listening to music as that induces unwanted hum which even gets through my dedicated power supply for my hifi. The dishwasher noise (not the spray of water!) comes through too as the motor induces noise into the mains ring which is also picked up by the hifi, ditto the fridge motor. It's a compromise without a dedicated mains spur, but you can go a good way to improving matters with a heavily shielded audio IC cable.

 

SergeAuckland

Certified Measurist
Wammer
May 6, 2008
18,607
1,915
173
Bury St Edmunds, UK
AKA
Serge
HiFi Trade?
  1. No
As I understand it Serge, leakage current predominantly exists from current which is said to leak from other sources directly to ground via the chassis and earth shields. Examples include capacitively coupled current (direct to the ground on a shield) or AC line frequency hum caused by capacitance between AC primary and secondary windings in a components power supply meaning that a small element of AC line voltage will always be coupled directly to signal ground.Add to that RF interference in AC mains supplies geeting through to the signal, and different components sharing their own safety grounds but coupled by ground shields, there's another element of induced voltage in the interconnect shield.

What I am trying to get people thinking about is that without minimising IC ground impedance, the value of the total "noise" induced into the signal will be (comparitively) high resulting in greater noise predominantly in the form of 50/60Hz hum.

I proved it to myself yesterday, after a hifi move in the listening room, I needed different length IC's from my previous ones, so I made up a new set with higher theoretical S/N ratio (120 v's 100) than my previous set and hey presto, achieved a significant reduction in overall humm from my highly sensitive speakers to the point where it's almost inaudible at the listening position. Another benefit was (with lower noise floor) better detail, and a better overall balance to the sound. I think when most people hear differences in audio cables, a good proportion of what they hear as beneficial is down to the S/N ratio being high.
Indeed.Dedicated power spurs are a good thing. I temnd to switch of my fish tank aeration pump when listening to music as that induces unwanted hum which even gets through my dedicated power supply for my hifi. The dishwasher noise (not the spray of water!) comes through too as the motor induces noise into the mains ring which is also picked up by the hifi, ditto the fridge motor. It's a compromise without a dedicated mains spur, but you can go a good way to improving matters with a heavily shielded audio IC cable.
Thanks for the explanation, I understand your position better now.

I must be unusual, as I don't have any hum, even with my ears against the 'speakers, let alone at the listening position. On LP, there's a really small amount of hum audible only with ears against the 'speakers, but completely inaudible at the listening position. I live in an incredibly quiet rural location, so any noise would be very apparant.

As to machines creating noise (other than mechanical noise), again, I don't have any. If I did, I'd want to do something about it, but in my case, nothing. Nor has there ever been at my previous houses and with previous systems.

What am I doing wrong? ;-)

S.

 
P

Pac67

Guest
Thanks for the explanation, I understand your position better now.I must be unusual, as I don't have any hum, even with my ears against the 'speakers, let alone at the listening position. On LP, there's a really small amount of hum audible only with ears against the 'speakers, but completely inaudible at the listening position. I live in an incredibly quiet rural location, so any noise would be very apparant.

As to machines creating noise (other than mechanical noise), again, I don't have any. If I did, I'd want to do something about it, but in my case, nothing. Nor has there ever been at my previous houses and with previous systems.

What am I doing wrong? ;-)

S.
I guess it depends to some extent on the sensitivity of the speakers to very low level noise. As mine are around 100dB/1w I notice noise more than with my previous Lumley 300 AB's. I have an annoying ground loop due to the arrival of the Droplet CDp in the system but even that was minimised by the use of shielded phono plugs and a heavily shielded IC. Many valve amps produce noise (the worst I heard was from a set of new Quad II's) and with the SS standby amp in place, the system is pretty quiet. I don'thave a star earth on the preamp as it's an unpowered control unit, hence everything is start earthed at the power amp which may be another reason for some of the noise.

I'd say you're doing nothing wrong but obviously have house wiring and a system that majors on low noise! I'm happy that even though I can just hear a slight fuzzy hum (only just) once the music starts, it's totally unnoticeable even between passages so I'm happy with that.

 

Chumpy

Wammer
Wammer
Dec 3, 2005
14,040
112
0
Bristol UK
AKA
Charlie
HiFi Trade?
  1. No
I advise ensuring that your suppliers/manufacturers o.k. your choices re electrical compatibility etc - if all hunky-d., enjoying the subtle actual or imagined differences. And not overspending too much.

 

lehoako

Wammer
Wammer
Mar 16, 2011
37
0
0
I guess it depends to some extent on the sensitivity of the speakers to very low level noise. As mine are around 100dB/1w I notice noise more than with my previous Lumley 300 AB's. I have an annoying ground loop due to the arrival of the Droplet CDp in the system but even that was minimised by the use of shielded phono plugs and a heavily shielded IC. ...
OK, 2 short comments.

1.) My boxes have a sensitivity of 84 dB @ 4 Ohms, and I had the experience hearing these disturbances in the mains, caused by hoovers, washing machines, air condition, etc.

2.) [browellm: PLEASE DO NOT ADVOCATE DANGEROUS MAINS MODIFICATIONS]

Cheers, still running in the path, soundstage grew considerably. Shall I measure it in inches or centimeters? :geek: :geek: :geek:

 

lehoako

Wammer
Wammer
Mar 16, 2011
37
0
0
I can assure everyone that the modifications that have been made to my mains were supervised by adequate personnel and therefore no lives are in danger.

I came back with the info that the path seems to be ran in, but unfortunately, lacking the measurement technique/unit to measure how enjoyable the sound became, I cannot measure it in SI.

Where can be the missing link? What is that measurements do not take in consideration? / Except of course the enjoyable character of the sound.

 

lehoako

Wammer
Wammer
Mar 16, 2011
37
0
0
Dear Chumpy,

Against overspending I am definitely insured :-(((.

Thanks for the good wishes anyway, and wishing as many happy musical moments to everyone, as many I have now.

Cheers.

 

browellm

Wammer
Wammer
Sep 9, 2008
23,599
482
128
I can assure everyone that the modifications that have been made to my mains were supervised by adequate personnel and therefore no lives are in danger.I came back with the info that the path seems to be ran in, but unfortunately, lacking the measurement technique/unit to measure how enjoyable the sound became, I cannot measure it in SI.

Where can be the missing link? What is that measurements do not take in consideration? / Except of course the enjoyable character of the sound.
I don't know, but this is the technical section so you have to be able to support your subjective findings with some measurable parameters. If you want to spunk off in 2-ch about how good your mains is, be my guest.

 

Latest posts

Forum statistics

Threads
113,444
Messages
2,451,263
Members
70,783
Latest member
reg66

Latest Articles