Understanding Grounding in Audio (Video)

[amirm]

Technical explanation of different grounds in your audio system and all the misconceptions around ground loops, buzz and hum.

Bill Whitlock Presentation: https://bennettprescott.com/downloads...
Entreg Olympus Minimus grounding box review: https://www.audiosciencereview.com/fo...

 

[jasonhanjk]

Bookmark.

 

[Duckeenie]

Presentations like this can quickly become overwhelming but you manage to do it in ways that even I can understand. Thanks Amir.

 

[JSmith]

Not all earthing systems are the same in each country;

https://www.studiecd.dk/cahiers_techniques/System_earthings_worldwide_and_evolutions.pdf

https://www.lsp-international.com/power-supply-system/

… none of this means putting a wire into a box of dirt behind your amp does anything though, just some further info for the thread.



JSmith

 

[Wombat]

Technical explanation of different grounds in your audio system and all the misconceptions around ground loops, buzz and hum.

Bill Whitlock Presentation: https://bennettprescott.com/downloads...
Entreg Olympus Minimus grounding box review: https://www.audiosciencereview.com/fo...

Add: https://www.ranecommercial.com/legacy/library.html

 

[daftcombo]

Very welcome video.

 

[prasanth.nath]

Thank you Amir. Some American university seems to have lost a very good teacher!

Could you do one on pin1 and signal-shields when you can, please?

I am curious about this one:

From https://www.diyclassd.com/documenten/download/860 page 17 (a note from Hypex NC 400 manual) ,
Signal shield is connected to chassis and not to pin 1 directly. pin 1 is connected to chassis using shortest possible path.

However, both signal shield and pin 1 are connected to chassis anyways, is that not equivalent to connecting signal shield to pin 1?
What is 1 gaining by following the recommendation of not connecting signal shield to pin 1?

@Buckeye Amps - I believe you had asked this question too. Wondering if you knew the answer.

Thanks.

 

[Wombat]

Pin 1 problem: https://www.ranecommercial.com/legacy/note165.html

 

[axbarker]

As an ex-power distribution engineer - Amir's explanation is spot on for the level of understanding required. I found it a great insight and has motivated me to go out and replace my RCAs with Balanced connectors - I have the option on my equipment to do that. No doubt I don't need to buy "expensive" balanced connectors - right Amir?

 

[PeteL]

Thank you Amir. Some American university seems to have lost a very good teacher!

Could you do one on pin1 and signal-shields when you can, please?

I am curious about this one:

From https://www.diyclassd.com/documenten/download/860 page 17 (a note from Hypex NC 400 manual) ,
Signal shield is connected to chassis and not to pin 1 directly. pin 1 is connected to chassis using shortest possible path.

However, both signal shield and pin 1 are connected to chassis anyways, is that not equivalent to connecting signal shield to pin 1?
What is 1 gaining by following the recommendation of not connecting signal shield to pin 1?

@Buckeye Amps - I believe you had asked this question too. Wondering if you knew the answer.

Thanks.

it’s explained here:
https://www.hypex.nl/img/upload/doc/an_wp/AN_Legacy_pin_1_problems.pdf

 

[pma]

Rane and Bill Whitlock documents are far better than that of hypexnl.

 

[diddley]

Thank you Amir, now i know the difference between un and balanced.
I,ve got everything unbalanced no noise, call myself lucky.
Except for the guitars there will be sometimes loud buzz.

 

[Lambda]

Alternative video Title: "Why balanced is better"

this reference should also be here!:
https://web.mit.edu/jhawk/tmp/p/EST016_Ground_Loops_handout.pdf

I missed the part about induction and actual loop size. why it is(can be) important to have cables tidy together to minimize loop area

https://www.angelfire.com/electronic/funwithtubes/Eliminating_Ground_Loops.html

Also most of this mains filter devices conduct a lot of noise to ground:

Since the ground impedance at higher frequency can be relatively high there can be noise on the ground and it can be influencing audio performance!

So a "non grounded" or some how ground lifted audio system can have noise benefits!
Depending on your local code this this might not be Legal but it helps to get rid of Low voltage low frequency noise:

This can be save but it also can be illegal depending on the implementation and your local code!
Some (processional) audio devices use this internally if the "ground lift" is enabled
It is Also used with Ships/bots called "galvanic isolator"


The Save way to get an extra ground is only with Isolation transformer and RCD

 

[Helicopter]

As an ex-power distribution engineer - Amir's explanation is spot on for the level of understanding required. I found it a great insight and has motivated me to go out and replace my RCAs with Balanced connectors - I have the option on my equipment to do that. No doubt I don't need to buy "expensive" balanced connectors - right Amir?

You don't need expensive cable assemblies. I like to get good connectors though, from the likes of Neutrik or Amphenol. Switchcraft are good too. This is because they break less. They don't sound different, and if they are going to have an easy life on the back of your gear for a long time, then super cheap will be fine.

 

[charleski]

I recently did a small survey of Amir's results from testing DACs with balanced and unbalanced connections. The difference in SINAD was around 2dB, and since they were all excellent DACs (except one) from Topping and other makers, all came in below the threshold of hearing using both types of connection. The one exception was a DAC which clearly had some sort of fault in its balanced outputs. The only example I could find in which an unbalanced connection produced noise that passed the threshold was in an old measurement by Archimago, in which he tested he tested dynamic range when sending an attenuated signal (as you might when using a DAC with a volume control), in which the difference between balanced and unbalanced went up to 7.5dB.

So if you have a ground loop problem, then using a balanced connection might fix it and is certainly worth trying, and it's certainly a good idea if you use long cables (i.e. over a few metres). If you don't, then don't get worked up about balanced connections, as there's no evidence you'll hear any difference.

For a rather more complex article on cabling and ground loops, I found an interesting article here, which covers some measurements made for pro-audio installations. In this paper source a) is a reasonable analog of an unbalanced output, and source d) corresponds to a typical differential balanced output. They passed 5A through the shields, and the cables were over 27m long, so they put them under a lot of stress. The difference in CMRR varied quite a lot depending on the cable construction, with one showing only a mild ~5dB improvement between a) and d) while another (an old cable with a high shield resistance) improved by almost 20dB. Their real purpose is to investigate the effects of shield-bonding, though, and they end up recommending that the cable shield is connected to the chassis at both ends for optimal results.

 

[Koeitje]

I really like these down to earth explanations.

 

[PeteL]

I liked the video, and when you listen to it it all makes sense what you mean, but my two cents, the first slide, on it’s own, may rise a few eyebrows and for some lead to dismiss it, I’d just be carefull with those generalities especially on the opening of the video. Star grounding, maybe not for the electrician but for the electronic designer, means a different thing that joining all your mains earth pin in a single outlet, proper star grounding in equipment solves hums, but I realise that you are talking to non engineers about hooking up their equipment, but it was not necessarily obvious from the very top. Dedicated ground for audio also have benefits, but yes as you rightfully said doesn’t mean disconnecting it from the neutral and planting an isolated pole from everything, As you showed, there is voltage differentials on ground outlets. Having a ground that is not the same as your inductive electrical motors help with that, but obviously not a must for typical household. Again, nothing to argue about what you say, just the first page, without context present “myths” that are debatable.

 

[Lambda]

Question, In Europe where all mains plugs have 3 pins, do you have to worry about grounding?

Yes. it looks like you don't read or understand the linked references?!
also not everything in Europe is using 3 pin mains connectors

Star grounding, maybe not for the electrician but for the electronic designer, means a different thing that joining all your mains earth pin in a single outlet, proper star grounding in equipment solves hums

Yes this is true.

 

[Lambda]

MYTH:Most Noise is Caused by“Improper” AC Power Wiring Small voltages between outlet safety grounds is NORMALin proper wiring

Chassis voltage can approach 120 volts but current is limited by parasitic capacitances0.75 mA maximum for consumer electronicsThis “LEAKAGE” current will flow in signal cables connected to other equipment

of Amir's results from testing DACs with balanced and unbalanced connections. The difference in SINAD was around 2dB,

You realize the scope oft this tests was the DACs output and only the DACs output performance under (near)ideal conditions.
What's interesting is the total system performance!
DAC, Preamp in and out , Poweramp...

So the Preamp can have leakage current of .75mA or a multiple of it depending how many devices are connected to the preamp.
DAC and Poweramp can have up to multiple Voltes of Ground potential difference

MYTH:Most Noise is Caused by“Improper” AC Power Wiring Small voltages between outlet safety grounds is NORMALin proper wiring

Even if everything is ideal and battery powered:

There is loop area between the audio connectors themself.

TL;DR
Real world performance of the system is way worse then the DAC output in an analyzer
I "proof" this to you by even performing a measurement
https://www.audiosciencereview.com/...o-does-it-make-sense.20978/page-6#post-704315
And so did KSTR.
But you seem to be ignorant or have lack of understanding of the fundamental problem.

 

[boselover61]

Professor Amir back at it again