Chassis or signal grounding why it works

[Middy]

I want to make my own Star grounded distribution strip.For my power cables.. But for a case ground..

 

Reading here and WBF I am still visually struggling to see in my mind how to 'ground' with a short path. Braided Earth strap for EMI case grounding.

Or is it more important to use a heavy gauge wire from the cases to the electrical outlet?

From seeing large building grounding schemes would your EMI ground scheme be Star wired to a central point from the cases,  then off to the outlet?

If so keep the star wires of a equal length?

@RogerD A few pictures detailing from the case or any other info would be gratefully received if you have time.

Or I have this wrong....

 

Good luck 

Dave 

 

[Speedskater]

If some of your components have 3 wire AC power cords, there is no need to connect to a wall outlet. As the AC cords will do the connecting.

If a component has a double insulated chassis (square within a square safety symbol, don't connect to that unit).

Even though in the past, some experts suggested equal length wires, it's not a good idea. Just make all the wires reasonably short. But don't sweat it.

[Middy]

Thanks Speedskater  @Speedskater

I have my phone as source with extra battery injection. Ups sine wave and Balanced isolation transformer driving an sbooster 13V into my Mytek Brooklyn. The source feeds an intona into the Iso regen with the LPS1.

So now the is no ground to earth unless the Neutral on the LPS Sbooster counts in this instance? As rodgerD sees his methodology??

 

Good luck my friend

Dave 

[Middy]

oh I signed up to this ages ago. I thought people here may find it useful.

https://interferencetechnology.com/category/standards/

[fas42]

4 hours ago, mmerrill99 said:

Yep & what you are achieving in your direct approach seems to have exactly the same effects as the various USB isolator products (lets not get into other grounding products)

 

So my belief is that the same underlying mechanism is at play - the prevention of noise from intermingling with signal pathways.

 

It would be interesting to see people who crave measurements to measure this - I bet they will need to be very creative & careful to find the differences on the analogue outputs.

 

Yes, this is at the heart of the matter - the goal is to reduce the impact of any interference or noise generating mechanism. There are myriads of ways of doing this - I've played with various schemes and techniques over the years; and just about anything out there will probably have some type of effect, good and bad.

 

At the end of the day, the smartest, most effective, cheapest way of achieving this is what should be aimed for - something to still resolve ...

[Speedskater]

'Dave' I'm not really familiar with any of your components. How many of the audio units have a 3 wire cords?

[Middy]

Sorry  @Speedskater

Uk mains Balanced isolation transformer and Sinewave Output UPS. 3 wire Live neutral earth.

But they feed a linear power supply 13v DC to the preamp Dac unit.

 

Feeder SMPS to Uptone LPS1. super cap flip flop supply

 

The power chain and uptone SMPS both sit on a single simple power strip.

Just 2,   3 pin UK plugs.

 

I thought of adding a drain to the case of the pre amp and LPS1 going to the case of the Balanced isolation transformer? 

 

Any thoughts always welcome and thank you for replying.

 

Good luck

Dave

 

 

 

 

 

 

 

[Le Concombre Masqué]

On 19/7/2017 at 2:31 AM, Speedskater said:

If some of your components have 3 wire AC power cords, there is no need to connect to a wall outlet. As the AC cords will do the connecting.

If a component has a double insulated chassis (square within a square safety symbol, don't connect to that unit).

Even though in the past, some experts suggested equal length wires, it's not a good idea. Just make all the wires reasonably short. But don't sweat it.

yes that what puzzles me : is not the chassis connected to earth via the power cable  ?

 

And since everything is balanced in my system, including the phono cartridge that feeds the phono preamp trough a xlr, I guess all my chassis are connected via all the xlr cables pin 1

 

And then I would assume at least one my components has its chassis connected to the ground that is to the rods beneath my house...

 

now experiment :

 

I took a cable sitting in my garage for a reason I ignore since I did not buy it : the color is yellow & green so here it means it's designed for grounding something, there's a female plug on one hand that loosely fits the ground prong of the wall plug and on the other hand a metal clip with a powerful spring for a good grip.

 

I pinched the ground connector on my phono preamp (a large screw where the ground part of the phono cable has to be fit to limit most of the nasty hum, for those too young to know...). It does not remove any further the phono noise. there's clearly a soft sound that modulates when I plug into the wall's electric outlet ground prong. Then it settles and I can not say that the noise out of my phono entry (volume maxed) has changed.

 

How about the sound ?

 

the cable on his own plugged to the phono pre but not earthed might well be a hasard, the sound thinner, did not spend real time on that

 

when it's all plugged, it seems that Savall's violas ensemble (Bach's Art of fugue, Astrée LPs) are woodier, the experience preferable

 

Might be all placebo effect ; I decided to keep the cable installed for a while and decide later...

 

1st : do I do the right thing to try out chassis grounding ?

2nd : should not the ground screw on the phono pre chassis be already connected to earth via the power cord ?

Thus am I just creating a change but a worsening (though it seems pleasant) not a betterment via a loop or antenna effect ?

[Cornan]

Correct me if I'm wrong, but is'nt a way to find out that you have a true star ground when you can use unshielded ICs without GND and can still achieve a DAC handshake?  This means that the DAC handshake will be performed via the safety ground connection between the devices instead of via ICs. So to be able to fully enjoy a star ground all ICs must be unshielded or atleast the shield must be lifted or isolated.

[Le Concombre Masqué]

15 hours ago, Le Concombre Masqué said:

yes that what puzzles me : is not the chassis connected to earth via the power cable  ?

 

And since everything is balanced in my system, including the phono cartridge that feeds the phono preamp trough a xlr, I guess all my chassis are connected via all the xlr cables pin 1

 

And then I would assume at least one my components has its chassis connected to the ground that is to the rods beneath my house...

 

now experiment :

 

I took a cable sitting in my garage for a reason I ignore since I did not buy it : the color is yellow & green so here it means it's designed for grounding something, there's a female plug on one hand that loosely fits the ground prong of the wall plug and on the other hand a metal clip with a powerful spring for a good grip.

 

I pinched the ground connector on my phono preamp (a large screw where the ground part of the phono cable has to be fit to limit most of the nasty hum, for those too young to know...). It does not remove any further the phono noise. there's clearly a soft sound that modulates when I plug into the wall's electric outlet ground prong. Then it settles and I can not say that the noise out of my phono entry (volume maxed) has changed.

 

How about the sound ?

 

the cable on his own plugged to the phono pre but not earthed might well be a hasard, the sound thinner, did not spend real time on that

 

when it's all plugged, it seems that Savall's violas ensemble (Bach's Art of fugue, Astrée LPs) are woodier, the experience preferable

 

Might be all placebo effect ; I decided to keep the cable installed for a while and decide later...

 

1st : do I do the right thing to try out chassis grounding ?

2nd : should not the ground screw on the phono pre chassis be already connected to earth via the power cord ?

Thus am I just creating a change but a worsening (though it seems pleasant) not a betterment via a loop or antenna effect ?

i would compare the soft modulated noise to the cartridge response to gentle brushing

I definitely think there's an improvement : makes my analog sounds more digital that is tighter more defined (on the other hand, I'm too tired to compare but the Wall sounds greater than it sounded for a while and last times were digital...)

[Speedskater]

6 hours ago, Cornan said:

Correct me if I'm wrong, but is'nt a way to find out that you have a true star ground when you can use unshielded ICs without GND and can still achieve a DAC handshake?  This means that the DAC handshake will be performed via the safety ground connection between the devices instead of via ICs. So to be able to fully enjoy a star ground all ICs must be unshielded or atleast the shield must be lifted or isolated.

Now I'm totally lost! Just what kind of interconnects are these?

a] RCA unbalanced analog

b] RCA digital SPDIF

c] XLR balanced analog

d] XLR digital AES/EBU

e] USB

f] other

[Cornan]

9 hours ago, Speedskater said:

Now I'm totally lost! Just what kind of interconnects are these?

a] RCA unbalanced analog

b] RCA digital SPDIF

c] XLR balanced analog

d] XLR digital AES/EBU

e] USB

f] other

 

If my assumption is correct any IC would work without a ground connected shield and GND with a proper star ground. I thought that that was the genral idea with a star ground that each device only have one connection to one star ground point (via ac mains connection) and have no shared ground paths inbetween devices?This would mean no GND or ground connected shields in ICs right? As I said in my previous post: correct me if I'm wrong.

[marce]

On 18/07/2017 at 6:08 AM, Ralf11 said:

is that a piezo-electric effect?

 

like with crystals?

No tribo

[Cornan]

Here is a picture (posted by Karin at Sbooster) from a previous thread that pretty well describes the optimal star ground scenario with USB:

 

[Le Concombre Masqué]

hard to believe it's more attracted by the DAC than by the main's earth 

[RogerD]

Recorded this last night on my iphone6

 

[Cornan]

22 minutes ago, Le Concombre Masqué said:

hard to believe it's more attracted by the DAC than by the main's earth 

 

It is not a matter of attraction. The thing is that the handshake is performed via the AC mains connection and no ground noises is shared between the devices. 

 

Ground currents are just lazy buggers choosing the quickest and easiest path to ground. They will go in any direction to reach that goal. If you setup a star ground path and eliminate all ground paths between the devices you will ensure optimal performance IMO. Make no misstake. This is not an easy task with a complicated setup.

[Speedskater]

32 minutes ago, Cornan said:

Ground currents are just lazy buggers choosing the quickest and easiest path to ground. They will go in any direction to reach that goal. If you setup a star ground path and eliminate all ground paths between the devices you will ensure optimal performance IMO. Make no misstake. This is not an easy task with a complicated setup.

Yep, all currents are lazy buggers. But ground is not their destination, it's just the  easiest path to their destination.

What is their destination, you might ask?

Well it's their voltage source which is the power company Neutral.

But a major goal of the grounding system is to hold all of your components chassis at about the same potential.

 

Ground can be a common reference point or a path to the current's voltage source, but never ever a destination.

[Cornan]

20 minutes ago, Speedskater said:

Yep, all currents are lazy buggers. But ground is not their destination, it's just the  easiest path to their destination.

What is their destination, you might ask?

Well it's their voltage source which is the power company Neutral.

But a major goal of the grounding system is to hold all of your components chassis at about the same potential.

 

Ground can be a common reference point or a path to the current's voltage source, but never ever a destination.

 

Great explanation Speedskater! ?

[Cornan]

To explain myself a little further the importance of using no ground connected shields on ICs between devices is because ground currents are travelling via cable shields and are creating multiple ground paths. This will interfear with the star ground path since you provide an alternative path for the ground currents to travel, ie you cannot really be sure they'll take the star ground path as you have set up and really want them to take. Same thing with GND paths. By limiting the possible ground paths you can "force" (not really, but limit their choises) them to choose a path selected by you, which ofcourse is the star ground path.

Now, there is several ways to ensure that the shields is not connected to ground. Either you simply use no shield on your ICs, active galvanic isolation (which creates possible additional ground paths if you do not use floating psu) or simply lift the shields somewhere down the line.

 

What you want to acheive using a star ground is a one way ticket for the ground currents to safety ground. This means creating a single ground point from all devices and elliminating all other possible ground paths inbetween devices. 

 

Now lets say that your only safety ground path is via a 1:1 isolation transformer (IT with a floating secondary) Where do the ground currents end up? ?

(Clue: read Speedskaters latest post)

[Cornan]

Here is the answer in green letters! ?

 

[Le Concombre Masqué]

5 hours ago, Cornan said:

 

It is not a matter of attraction. The thing is that the handshake is performed via the AC mains connection and no ground noises is shared between the devices. 

 

Ground currents are just lazy buggers choosing the quickest and easiest path to ground. They will go in any direction to reach that goal. If you setup a star ground path and eliminate all ground paths between the devices you will ensure optimal performance IMO. Make no misstake. This is not an easy task with a complicated setup.

What would be your recommandation : 

 

all balanced pre

balanced  connection to DAC

balanced  connection to phono pre

balanced  connection from cartridge to phono pre

 

I can't resurrect the Corning in my sig any longer and there's a Belkin usb entering first generation Regen (powered by the basic psu )hard connected to the DAC

and that's it

 

 

[Cornan]

28 minutes ago, Le Concombre Masqué said:

What would be your recommandation : 

 

all balanced pre

balanced  connection to DAC

balanced  connection to phono pre

balanced  connection from cartridge to phono pre

 

I can't resurrect the Corning in my sig any longer and there's a Belkin usb entering first generation Regen (powered by the basic psu )hard connected to the DAC

and that's it

 

 

 

Well my recommendation is to use a DC blocker trap filter (not recommended by JS) pre a 1:1 isolation transformer with floating secondary, use only floating PSUs, use both AC and DC starquad cables troughout, use star ground wired powerstrips without filters or overload circuits, use balanced ICs where ever possible and unshielded USB with GND-lift. Battery chargers and turntables should be connected to a powerstrip that by-passed the IT. YMMV ofcourse! ?

 

Furthermore I also recommend a wireless adapter into a grounded network switch using Cat 6a U/UTP 500mHz cables. If you're using cloud content it is very important to power the router with a floating PSU and galvanicly isolate the router input.

 

I hope that helps! ?

[Cornan]

Removed

[Middy]

@RogerD 

Could you do a few pictures please so I can see how you have grounded your set up please. At the cases through to the receptacle Ground. If that's ok?

Thanks