Leakage Currents in audio

[jkeny]

I see John Swenson has posted his technique for eliminating leakage currents & guess what - it's effectively the same method recommended by AP that I stated in this closed thread & as practised here by RogerD & others - grounding of chassis between devices. This approach was not plucked out of thin air - it was recommended by the Audio Precision company itself

It was quiet amusing when I read the objections to this grounding technique - one of which was that 2 pin SMPS have plastic cases & therefore can't be grounded. makes one query the level of technical know-how being demonstrated by such objections.

Swenson showed some measurements he made after he had built "a very high impedance (10 giga ohms) differential probe to properly measure leakage current." (necessary to avoid the measurement probe affecting what is being measured)

Measurement of leakage current on the output of SMPS


Measurement of leakage current with ground connection in place


An improvement of > 50dB reduction.

More measurements here when SMPS is powering network switch.

Some people need measurements before they can admit to what the rest of us can hear with our ears - this is for these poor lost souls!

 

[RogerD]

I see John Swenson has posted his technique for eliminating leakage currents & guess what - it's effectively the same method recommended by AP that I stated in this closed thread & as practised here by RogerD & others - grounding of chassis between devices. This approach was not plucked out of thin air - it was recommended by the Audio Precision company itself

It was quiet amusing when I read the objections to this grounding technique - one of which was that 2 pin SMPS have plastic cases & therefore can't be grounded. makes one query the level of technical know-how being demonstrated by such objections.

Swenson showed some measurements he made after he had built "a very high impedance (10 giga ohms) differential probe to properly measure leakage current." (necessary to avoid the measurement probe affecting what is being measured)

Measurement of leakage current on the output of SMPS
View attachment 35542

Measurement of leakage current with ground connection in place
View attachment 35543

An improvement of > 50dB reduction.

More measurements here when SMPS is powering network switch.

Some people need measurements before they can admit to what the rest of us can hear with our ears - this is for these poor lost souls!

John, The evidence keeps mounting everyday as many with high level systems and respect on WBF hear the difference . Any studies that provide empirical evidence is just icing on the cake.

 

[jkeny]

John, The evidence keeps mounting everyday as many with high level systems and respect on WBF hear the difference . Any studies that provide empirical evidence is just icing on the cake.

Yes, Roger & what I like about Swenson & your approach is that it resonates with my philosophy - no need for expensive grounding boxes, etc when a simple approach will suffice. Sometimes simple approaches are later validated by empirical evidence, as you state.

Roger, do you have any SMPS powering devices in your system? Not that it's only SMPSes that give rise to leakage currents.

 

[RogerD]

Yes, Roger & what I like about Swenson & your approach is that it resonates with my philosophy - no need for expensive grounding boxes, etc when a simple approach will suffice. Sometimes simple approaches are later validated by empirical evidence, as you state.

Roger, do you have any SMPS powering devices in your system? Not that it's only SMPSes that give rise to leakage currents.

Hi john,

No SMPS devices,but I use 5 SS amplifiers and my tube DAC has a toroidal transformer. Since the grounding scheme is cumulative I would contend that all components benefit that carry current.

 

[microstrip]

I see John Swenson has posted his technique for eliminating leakage currents & guess what - it's effectively the same method recommended by AP that I stated in this closed thread & as practised here by RogerD & others - grounding of chassis between devices.(...)

John,

I should note that this not the method used by most WBF members who use "grounding" - they use "bizarre" devices that are not formal "grounds" or "strange" grounding cables. Supreme heresy - some even claim that using more than one grounding device increases sound quality!

And I was astonished to read John Swenson comment that "You would think that the transformers would block the leakage current" concerning Ethernet - since long it was commented why the use of simple galvanic isolation did not eliminate leakage currents at high frequencies. Galvanic isolation is usually tested at DC or 50/60 Hz.

Unfortunately there is a lot we do not know about EMI in high-end audio systems.

 

[jkeny]

John,

I should note that this not the method used by most WBF members who use "grounding" - they use "bizarre" devices that are not formal "grounds" or "strange" grounding cables. Supreme heresy - some even claim that using more than one grounding device increases sound quality!

Yes, I noticed on WBF far more interest in esoteric, expensive 'ground solutions' - one of the reasons I raised/praised JS & Roger's approach.

And I was astonished to read John Swenson comment that "You would think that the transformers would block the leakage current" concerning Ethernet - since long it was commented why the use of simple galvanic isolation did not eliminate leakage currents at high frequencies. Galvanic isolation is usually tested at DC or 50/60 Hz.

I don't know if this is his opinion he is expressing or the 'received wisdom' of audiophiles?

Unfortunately there is a lot we do not know about EMI in high-end audio systems.

I'm not sure that there is "a lot we don't know"?

 

[adyc]

John,

For the benefit of some of the members in here, can you briefly share how Roger and you approach to the grounding?

 

[RogerD]

John,

For the benefit of some of the members in here, can you briefly share how Roger and you approach to the grounding?

Hi John,

Read from page 5 on.. http://www.whatsbestforum.com/showthread.php?1769-My-2-channel-experiment/page5
I ask you to do your own research. Here is a good start...http://www.rane.com/note151.html

Roger

 

[opus112]

I'm not sure that there is "a lot we don't know"?

There is a lot that we do know - like for example running balanced cables makes a system much less sensitive to leakage currents. Yet who runs balanced?

 

[microstrip]

Yes, I noticed on WBF far more interest in esoteric, expensive 'ground solutions' - one of the reasons I raised/praised JS & Roger's approach.

I have tried grounding everything to a central point - the so called star earth system. It did not improve sound quality in my system. I get much better sound quality using for example expensive power cables - something I can not understand.

IMHO people use the exoteric solutions because they result in better sound quality than the classical technical solutions.
I have not tried the "esoterics" yet, but they are in my wish list.

I'm not sure that there is "a lot we don't know"?

One example. When using the star ground I tried enveloping the speaker cables with aluminum film connected by a wire to the central ground. It killed my system - it become dull and lifeless. Do you have a possible reason for it?

How can we explain large differences in sound quality using for example shielded and non-shielded similar cables? IMHO it is an EMI effect, but I do not have an explanation for its interference mechanism.

 

[microstrip]

There is a lot that we do know - like for example running balanced cables makes a system much less sensitive to leakage currents. Yet who runs balanced?

Question of preference, IMHO. Balanced solutions have a typical sound quality - some people like it, others prefer the single-ended.

My preamplifier has balanced and non balanced outputs, as well as the amplifier. I prefer the single ended connection, although unhappily it is much noisier.

 

[jkeny]

John,

For the benefit of some of the members in here, can you briefly share how Roger and you approach to the grounding?

My setup is not comprised of commercial units so I may not be the best representative of the norm here.

My first inkling about all this was when I started to use battery power. I didn't fully realise, at the time, what were the many benefits of this approach - I only knew that it sounded excellent.
Initially, I put it down to these batteries being a very low noise power source. Since then I believe that being a very stable, unwavering PS is probably of more importance & this stability comes from a combination of there being no leakage currents & the batteries ability to deliver the needed current is not constricted in any way.

Leakage currents are one of the main sources of disturbance to this stability

 

[jkeny]

I have tried grounding everything to a central point - the so called star earth system. It did not improve sound quality in my system. I get much better sound quality using for example expensive power cables - something I can not understand.

IMHO people use the exoteric solutions because they result in better sound quality than the classical technical solutions.
I have not tried the "esoterics" yet, but they are in my wish list.



One example. When using the star ground I tried enveloping the speaker cables with aluminum film connected by a wire to the central ground. It killed my system - it become dull and lifeless. Do you have a possible reason for it?

How can we explain large differences in sound quality using for example shielded and non-shielded similar cables? IMHO it is an EMI effect, but I do not have an explanation for its interference mechanism.

I'm not even going to try to tease out what's going on with your configuration - in my experience trying to remotely diagnose something like this is fraught with difficult - it requires a hands-on approach.

There are also too many variables with mains connected equipment & why I simplify matters by going off grid, where possible

 

[microstrip]

My setup is not comprised of commercial units so I may not be the best representative of the norm here.

My first inkling about all this was when I started to use battery power. I didn't fully realise, at the time, what were the many benefits of this approach - I only knew that it sounded excellent.
Initially, I put it down to these batteries being a very low noise power source. Since then I believe that being a very stable, unwavering PS is probably of more importance & this stability comes from a combination of there being no leakage currents & the batteries ability to deliver the needed current is not constricted in any way.

Leakage currents are one of the main sources of disturbance to this stability

John,

IMHO you are just hitting the critical point - there are so many different solutions and implementations in high-end audio that there is not such think as the "best representative of the norm". What works in a system can be a complete failure in another.

 

[microstrip]

I'm not even going to try to tease out what's going on with your configuration - in my experience trying to remotely diagnose something like this is fraught with difficult - it requires a hands-on approach.
(...)

We fully agree!

 

[Folsom]

There is a lot that we do know - like for example running balanced cables makes a system much less sensitive to leakage currents. Yet who runs balanced?

I prefer single ended with less "leakage current" than balanced. Yup. That's what I run. There's ways to mitigate it without multiplying the problems you already have with double circuitry. I guess big ground straps is one possible way, but it depends on the range, too.

 

[opus112]

Double the circuitry isn't mandatory with balanced cabling. An input transformer to go bal-SE is sufficient.

 

[Folsom]

I have tried grounding everything to a central point - the so called star earth system. It did not improve sound quality in my system. I get much better sound quality using for example expensive power cables - something I can not understand.

Performance is really subjective, the sound becomes "better" but that doesn't necessarily mean any objective measurements will show that, if not the opposite.

Yes, Roger & what I like about Swenson & your approach is that it resonates with my philosophy - no need for expensive grounding boxes, etc when a simple approach will suffice. Sometimes simple approaches are later validated by empirical evidence, as you state.

Roger, do you have any SMPS powering devices in your system? Not that it's only SMPSes that give rise to leakage currents.

Grounding boxes do something 100% different than connecting your gear with large ground straps. Different people will prefer different results.

 

[jkeny]

Performance is really subjective, the sound becomes "better" but that doesn't necessarily mean any objective measurements will show that, if not the opposite.

I don't agree - if measurements don't show anomalies then we are back to our auditory perception to evaluate the sound.

Grounding boxes do something 100% different than connecting your gear with large ground straps. Different people will prefer different results.

Do you know what grounding boxes do with regard to leakage currents? Just trying to figure out how you know they do "something 100% different"?

 

[Speedskater]

I see low frequency, low impedance noise & interference current getting mixed with high frequency, high impedance noise & interference voltage. These are two different problems and require two different solutions.