Grounding of system

[DaveC]

Dave, it's almost 2016, these days it's not difficult or expensive to build a balanced interconnect with 3 zeros between the decimal point and the first significant figure in the THD measurement. At that level it doesn't matter what harmonic the distortion is.
To use unbalanced interconnects in a high-end system is an oxymoron.

That's simply not the case, and your first sentence makes no sense to me, I already said it's not about the cable, it about balanced systems as a whole. Lots of high end systems and components are single ended only and many experienced audio designers prefer single ended home audio systems. In a home environment the common mode noise rejection of balanced systems simply isn't necessary to eliminate interference and provide a low noise floor. Balanced components need two of everything, it's more complicated and expensive for questionable gains. The 2nd harmonic distortion cancellation inherent to balanced systems (that's SYSTEM, not cable!) often makes the sound more mechanical, dry and less musical. I can provide tons of examples of extremely high quality single ended components, like every SET amp in existence for one, tube preamps without a single XLR jack in them, etc...

 

[Folsom]

It doesn't matter what process they use, it still doesn't affect noise currents.

Not true, as that opposes the fundamental difference in Ott's notes on ground/shield affect on attenuation. It also isn't consistent with recording studio practices.

 

[Barry]

The ground connection in balanced cables is quite audible. The advantages of balanced gear in home audio is hugely overstated and detrimental to musicality as it cancels 2nd order distortion while leaving 3rd untouched.

Dave, as of course you know, your view on balanced versus single ended circuits and 2nd and 3rd harmonic distortion is far from universally shared in the industry. For example, while obviously not the final word on this, here's Nelsen Pass' view on 2nd and 3rd harmonic distortion:

"I have built many examples of simple 2nd and 3rd harmonic “types” of amplifiers over the last 35 years. Anecdotally, it appears that preferences break out roughly into a third of customers liking 2nd harmonic types, a third liking 3rd harmonic, and the remainder liking neither or both. Customers have also been known to change their mind over a period of time. However the issue is partially obscured by the fact that the 3rd harmonic type amplifiers usually have lower total distortion. 2nd order type amplifiers seem to do particularly well with simple musical material, and 3rd order types generally seem to be better at more complex music. Nevertheless, whether you prefer 2nd or 3rd order type amplifiers, let's agree that we wish to minimize the total amount of distortion. And assuming that we have to put up with some distortion, let's also agree that we prefer 2nd and 3rd harmonic components over 4th, 5th, 6th, 7th."

In the "tube world", Ralph Karsten (Atmasphere) certainly has views on this as well. Oh yes, I do like the sound of single ended triodes...!

 

[DaveC]

Dave, as of course you know, your view on balanced versus single ended circuits and 2nd and 3rd harmonic distortion is far from universally shared in the industry. For example, while obviously not the final word on this, here's Nelsen Pass' view on 2nd and 3rd harmonic distortion:

"I have built many examples of simple 2nd and 3rd harmonic “types” of amplifiers over the last 35 years. Anecdotally, it appears that preferences break out roughly into a third of customers liking 2nd harmonic types, a third liking 3rd harmonic, and the remainder liking neither or both. Customers have also been known to change their mind over a period of time. However the issue is partially obscured by the fact that the 3rd harmonic type amplifiers usually have lower total distortion. 2nd order type amplifiers seem to do particularly well with simple musical material, and 3rd order types generally seem to be better at more complex music. Nevertheless, whether you prefer 2nd or 3rd order type amplifiers, let's agree that we wish to minimize the total amount of distortion. And assuming that we have to put up with some distortion, let's also agree that we prefer 2nd and 3rd harmonic components over 4th, 5th, 6th, 7th."

In the "tube world", Ralph Karsten (Atmasphere) certainly has views on this as well. Oh yes, I do like the sound of single ended triodes...!

Of course, everyone has their preferences... I've read that quote before and I agree with it entirely. I'm just disputing the statement that single ended systems can't be high end.

 

[Atmasphere]

Except that the ground you are talking about is chassis/earth grounding. Entreq grounds the signal. Not the same thing and is beneficial on its own.

I did a survey last year and found that in all cases (where people responded) where the Entreq and similar had a benefit, that the components in the system otherwise had bugs in the grounding technique used. It even says in the setup guide of some of these grounding systems that if the equipment used employs proper grounding technique in their design there will be no benefit. So my survey seemed to support that.

Some components do not isolate signal ground at all. If there is ONE component in your system that does not and everything else does than it "ruins" the isolation through the interconnect cables ground wires. There are no agreed on standards for ground isolation making the distinction impossible unless you know if and how each one of your components grounds are isolated, or not as the case may be. This being the case I'd expect the results of using aftermarket grounding to depend on the deficiencies in the audio systems existing grounding system.

+1

The best solution to any grounding issue is to use balanced interconnects. It almost completely sidesteps the effects of earth ground on signal being transmitted.

If it were only so! - but most high end audio companies do not support the balanced line standard (AES file 48). In fact we had to modify our grounding scheme on account of so many poorly set up components in the industry so even we (and we were the first to offer balanced line products to high end audio) don't support file 48 out of the box, although it is restored by tying any pin 1 of any XLR connector on our equipment to chassis.

The ground connection in balanced cables is quite audible. The advantages of balanced gear in home audio is hugely overstated and detrimental to musicality as it cancels 2nd order distortion while leaving 3rd untouched.

This statement is false. The ground connection is inaudible (if not then the equipment is really not operating the way it should...). The advantages for home use are the same as in the industry: immunity to cable artifact and reduction of noise and distortion, even if the connection is only 1 meter. What you are not mentioning here is that the 3rd is considered musical to the human ear (and is the only odd harmonics wherein this is the case), and that it will be at the same level or slightly lower that it would be in a single-ended system. Further, in a fully balanced system, distortions do not compound from one stage to the next due to cancellation and reduced IM. There is quite a benefit to musicality!

Once you hear a proper balanced system its hard to go back to single-ended. The problem we have in high end audio is that most high end audio manufacturers have no idea that there is even a standard for balanced operation. So if you hear a poorly designed system, that is not a problem with balanced any more than when you discover that square tires don't work so well on a car (you would not blame all tires as being thus defective, right?) One time we serviced out an Audio Note DAC that employed output transformers for its balanced outputs. It would have worked fine (and did after we fixed it) if the ground connections had not be set to pin 2 of the XLRs!

 

[Folsom]

Atmasphere, technically Dave isn't wrong. Fluctuating impedance on ground will have affects on the noise even in a balanced system. As far as how much it might be entirely negligible because you're obviously correct in that balanced (true) cabling has very good rejection of noise. If the grounds on your appliances had different potentials and exposure to noise from other things in a home, it could be apparent. But what's important to note is that artifacts aren't the only concern. In pro-audio they're paramount, but for audiophile it goes far beyond that and into a realm of how good it sounds. In that realm very small things have real consequences. Noise is real, it's a problem, and the most common mistake is to assume it'll represent itself as an artifact. The exceptional articulation found in audiophile stereos makes it worth ones wild to at least test with one's ears on things that seem negligible from an engineers perspective.

Personally I like balanced amplifiers, but haven't hard balanced cable based systems that were note worthy. They often seem to be a bit shy on finesse, from what I've heard.

 

[Atmasphere]

Atmasphere, technically Dave isn't wrong. Fluctuating impedance on ground will have affects on the noise even in a balanced system. As far as how much it might be entirely negligible because you're obviously correct in that balanced (true) cabling has very good rejection of noise. If the grounds on your appliances had different potentials and exposure to noise from other things in a home, it could be apparent. But what's important to note is that artifacts aren't the only concern. In pro-audio they're paramount, but for audiophile it goes far beyond that and into a realm of how good it sounds. In that realm very small things have real consequences. Noise is real, it's a problem, and the most common mistake is to assume it'll represent itself as an artifact. The exceptional articulation found in audiophile stereos makes it worth ones wild to at least test with one's ears on things that seem negligible from an engineers perspective.

Personally I like balanced amplifiers, but haven't hard balanced cable based systems that were note worthy. They often seem to be a bit shy on finesse, from what I've heard.

FWIW, in a balanced line system, there should be no (zero) signal current in the shield. All the signal occurs between pins 2 and 3. IOW, you won't have a fluctuating impedance.

Now in high end audio (and regrettably, in a lot of pro audio systems too) the balanced output occurs between pin 2 and ground (pin 1) and for the inverting output, between pin 3 and ground. Now, even though you are putting opposing currents through the shield, the fact is you are putting current through the shield and so now the construction and geometry of the cable becomes more important.

That is not how the balanced line system is supposed to work! Its supposed to work such that you really can't hear differences between cables, because if you rely on the ground connection for circuit completion, you're going to hear artifacts from the cable. Another way of look at this is that if you need the shield to make it work, you are doing something wrong. A twisted pair should work just about as well if the shield is there or not.

This fact is poorly understood in the high end audio world, and only fairly understood in the pro audio world. Rane has an excellent break-down of AES file 48 on their site: http://www.rane.com/note110.html

 

[Folsom]

Because all cables inherently have mutual inductance, the idea here is the shield should have a low enough impedance/resistance to drain most of the noise as the two hots are higher impedance so the division of noise to them is ideally minuscule. Their phase cancellation should, that's part of their differential operation, reduces noise greatly as well. But neither of these are absolutes. Improvements in the shield that's directly connected to chassis/enclosure that's typically safety grounded will equal overall improvements, but I can't tell you if you can hear it.

 

[rockitman]

I did a survey last year and found that in all cases (where people responded) where the Entreq and similar had a benefit, that the components in the system otherwise had bugs in the grounding technique used. It even says in the setup guide of some of these grounding systems that if the equipment used employs proper grounding technique in their design there will be no benefit. So my survey seemed to support that.

There in lies the problem....unless you are electronically as adept as you are Ralph, it's difficult to understand the grounding topology for a given component with a given company, so the Entreq stuff is a YMMV situation. The effect is beneficial albeit subtle at best. It made a difference in my system....in particular the phono stage and main preamp areas.

 

[Speedskater]

Because all cables inherently have mutual inductance, the idea here is the shield should have a low enough impedance/resistance to drain most of the noise as the two hots are higher impedance so the division of noise to them is ideally minuscule. Their phase cancellation should, that's part of their differential operation, reduces noise greatly as well. But neither of these are absolutes. Improvements in the shield that's directly connected to chassis/enclosure that's typically safety grounded will equal overall improvements, but I can't tell you if you can hear it.

I have been doing electronic (and electrical) engineering for over one/half century now. While your words are all common engineering words, your ideas are unfamiliar and strange. Could you be writing about Shield Current Induced Noise?

 

[Folsom]

Yes.

 

[Speedskater]

We should remember that a balanced interconnect system does not have a ground connection or ground conductor. It does have a shield conductor which may connect to the chassis at one or both ends. True it's probably does have a galvanic connect to other things that may be called ground, But that's not important to the operation of the interconnect system.

More than one/half century ago (when I was starting out). Then only broadcasting and high end pro audio (there was no budget pro audio) components had balanced interconnects and they were all transformer coupled. They were also all vacuum tubes, 600 Ohm impedance matched and no XLR connectors (they all had terminal strip connections. Then along came transistors and the shielding/grounding problems started. Thirty years later (1994/1995) Neil Muncy (RIP) wrote a paper on the proper way to wire interconnects (both balanced & unbalanced) and shields. A small part of this paper became ASE48 standard. It took about 10 years for the pro audio community to buy into the correct way to wire a interconnect system. Now yet another 10 years later the audiophile manufactures still don't get it. See Atmasphere's post #49

 

[Speedskater]

Shield-Current-Induced Noise - Part 1
Current flowing on the shield of balanced audio cables will be converted to differential mode voltage on the signal pair by imperfections in cable construction.
http://www.audiosystemsgroup.com/Shield_Current_Induced_Noise.pdf

Shield-Current-Induced Noise - Part 2
This is Part Two of a two-part article published in the SynAudCon Newsletter. It also includes recommendations for the termination of audio cables.
http://www.audiosystemsgroup.com/SCIN-2.pdf

About 50 more good Jim Brown papers:
http://www.audiosystemsgroup.com/publish.htm

 

[Speedskater]

What breakers sense Safety Ground/Protective Earth current?

The breakers have a host of names, but they're like the receptacle you have in kitchens that's a GFCI. The Wiki has several photos and such.

See 'amirm's' post in another thead:
http://www.whatsbestforum.com/showt...e-of-grounding&p=303370&viewfull=1#post303370

 

[Folsom]

What breakers sense Safety Ground/Protective Earth current?

See 'amirm's' post in another thead:
http://www.whatsbestforum.com/showt...e-of-grounding&p=303370&viewfull=1#post303370

Yes, it appears I've been duped by misinformation. I knew that GFCI receptacles could be used without ground, but have read incorrect bit on the breakers.

 

[Speedskater]

You are not alone. Countless people (some call themselves experts) have gotten confused about an GFCI works and then the publish papers & blogs with the wrong information.

 

[DaveC]

This statement is false. The ground connection is inaudible (if not then the equipment is really not operating the way it should...). The advantages for home use are the same as in the industry: immunity to cable artifact and reduction of noise and distortion, even if the connection is only 1 meter. What you are not mentioning here is that the 3rd is considered musical to the human ear (and is the only odd harmonics wherein this is the case), and that it will be at the same level or slightly lower that it would be in a single-ended system. Further, in a fully balanced system, distortions do not compound from one stage to the next due to cancellation and reduced IM. There is quite a benefit to musicality!

Once you hear a proper balanced system its hard to go back to single-ended. The problem we have in high end audio is that most high end audio manufacturers have no idea that there is even a standard for balanced operation. So if you hear a poorly designed system, that is not a problem with balanced any more than when you discover that square tires don't work so well on a car (you would not blame all tires as being thus defective, right?) One time we serviced out an Audio Note DAC that employed output transformers for its balanced outputs. It would have worked fine (and did after we fixed it) if the ground connections had not be set to pin 2 of the XLRs!

In a perfect world, I'd agree with you... not to the point that cables will be inaudible as the conductor material and insulation do make a big difference, but I do agree the ground connection shouldn't matter in a balanced system but it does ime, and in a less than optimal system it matters even more. Part of grounding is dealing with the fact the ground plane isn't a perfect, never-changing voltage reference.

 

[Ron Resnick]

There in lies the problem....unless you are electronically as adept as you are Ralph, it's difficult to understand the grounding topology for a given component with a given company, so the Entreq stuff is a YMMV situation. The effect is beneficial albeit subtle at best. It made a difference in my system....in particular the phono stage and main preamp areas.

Very interesting, Christian. Do you yourself have any special ground set-up?

 

[Atmasphere]

There in lies the problem....unless you are electronically as adept as you are Ralph, it's difficult to understand the grounding topology for a given component with a given company, so the Entreq stuff is a YMMV situation. The effect is beneficial albeit subtle at best. It made a difference in my system....in particular the phono stage and main preamp areas.

You should not have to be 'electronically adept'!

Quite literally, if one of these systems makes an improvement, it can also be seen that the equipment in use (amp, preamp, whatever), really should be sent back to the manufacturer to correct the grounding fault/design error. Stuff like this should not be going on at all!! IOW the fact that it does simply points to a lot of 'by gosh and by golly' going on with hobbyists turned 'high end audio designers' that lack basic grounding knowledge that should come with proper engineering.

 

[DaveC]

You should not have to be 'electronically adept'!

Quite literally, if one of these systems makes an improvement, it can also be seen that the equipment in use (amp, preamp, whatever), really should be sent back to the manufacturer to correct the grounding fault/design error. Stuff like this should not be going on at all!! IOW the fact that it does simply points to a lot of 'by gosh and by golly' going on with hobbyists turned 'high end audio designers' that lack basic grounding knowledge that should come with proper engineering.

I heard a quote a while ago, something like "nobody in high end audio is an expert." This seems more evident every day...