Chassis or signal grounding why it works

[Ralf11]

1 hour ago, RogerD said:

Stupid......

 

you mean the signal is stupid, but the noise is smart???

[Cornan]

That's the tricky thing these days: being able to surprise people. "Joseph Kosinski"

[Cornan]

I would never try to convince people that my opinion is better than others. I can just tell you what I know by real life experiments in my own setup. In my setup unshielded twisted pair USB cables sounds best. In my setup a 5v lift sounds best. In my setup a GND lift sounds best. In my setup a star ground connected to a floating IT followed by a AC mains DC Blocker trap filter sounds best. I have evaluated and re-evaluated the other options many times.

[jabbr]

5 hours ago, marce said:

Return Path...  doesn't have to be ground...

My comments were again to try and dispel this belief that GROUND is a magic sink for noise...

Does noise just get trapped in electronic circuits in space?

 

Speak about floating -- literally  

[fas42]

It seems pretty simple to me ... an audio rig, no matter how complex or messily set up, is just one big electronics circuit - you can simulate the whole shebang in a Spice program, with perfect connections everywhere. Which means that something marked 0V is always that, no matter what's going on, irrespective of the frequency talked about, stretching up to 100's of GHz.

 

Now, in the real world all we have to do is ensure what actually happens matches our lovely simulation, at all times - that is, all 0V nodes or points are really that voltage, irrespective. But IME that is a bit tricky to get happening - a lump of wire from the scrap box is not a solution.

[marce]

12 hours ago, RogerD said:

Stupid......

Is less stupid than your comment, that implies noise goes to ground...

[marce]

13 hours ago, Cornan said:

I would never try to convince people that my opinion is better than others. I can just tell you what I know by real life experiments in my own setup. In my setup unshielded twisted pair USB cables sounds best. In my setup a 5v lift sounds best. In my setup a GND lift sounds best. In my setup a star ground connected to a floating IT followed by a AC mains DC Blocker trap filter sounds best. I have evaluated and re-evaluated the other options many times.

A lot of what some put up, myself especially is not opinions, its from experience in real world electronics, facts that I can and do back up with relevant information. If you want to look at the information and learn that is your prerogative, same as if you want to play with your system, its your choice. So a lot of people are NOT trying to convince you of their opinions, just presenting facts...

You are posting your thoughts and views on an open forum, so you are going to get some discussions regarding what you are doing, some supporting you some not...

If you are happy with what you are doing then there is no problem, just don't expect 100% support especially if what you are doing goes against a lot of EMC engineering stuff that is well documented, tested and has empirical data to back it up.

So it sounds best to you, then its a happy world.

[One and a half]

In trying to find an analogy for USB compared with an industrial practice of shielding and bonding, Profibus is a serial packetised protocol, that maxes out at 12Mb/s, a little short of USB's 480Mb/s, but digital is digital and Profibus suffers the same poor installtion practices as our USB cables. Profibus is used in industrial situations were signals MUST be correct otherwise too much or too little is added in process manufacturing creating problems. So noise mitigation is a big issue and this article is a very good start to show where noise comes from and how to reduce EMI that affects serial communications.

So in viewing what standard techniques work for Profibus, we might use a little of this knowledge and put it into practice for audio use.

 

When it comes to bonding enclosures together, no old school, but things that work, depends on the frequency that you want to control, or avoid being collected by the cables. For different frequencies, different bonding techniques are used.

So for low frequencies a cable is best, as the frequencies of noise are higher different cables, even braids or solid metal is used.

 

For audio, which one to use? The most difficult is to where to make the bond on audio equipment, for size 2 or 4 cable, you need a large rust free area to bond with, which means drilling holes into equipment which I don't fancy at all. At some point paint needs to be removed to bare metal, then anti rust primer added, plus a conductive surface that doesn't corrode, like nickel.... how do you apply this?

 

Are these higher frequencies for an amplifier important to remove? If the bandwidth of the amp peters out at 50kHz, is there any need for a braid or a bond at all? If the signal is USB, then all the more to maintain the bonds in the correct manner with large metal plates. How to fit this to a microrendu or an ISOREGEN? Good luck. We rely too heavily on the shielding of USB cables into the shells , there needs to be something else. 

 

When a Profibus cable enters a cabinet, the shield is hard terminated to a functional earth (clean ground). See picture. Now the other end of the cable is bolted to a functional earth bar, then a large cable can be used between the two functional earths to keep the potentials of the shields as close as we can get the same.

 

Having opposing cables on an Intona, or an ISOREGEN is problematic. In order to bond the shields effectively and keep the bonding short, The input and output USB cable should really be side by side.

 

[Jud]

18 hours ago, RogerD said:

Stupid......

 

Roger, it’s all electrical current.  So how does the “noisy” part of the electricity wind up taking a different circuit path than the “signal” part of the electricity?  There are ways to do this (for example, if most of the noise tends to be at different frequencies than the signal), but just having a return or ground path in the circuit alone isn’t sufficient.  *Every* circuit has a return path, and plenty of those circuits are noisy.

[jabbr]

36 minutes ago, Jud said:

*Every* circuit has a return path, and plenty of those circuits are noisy.

 

Consider an audio system in the space shuttle as I was alluding to above, Suppose there is a "ground" let's say a black box -- if someone suggests that current flows into the box , where does it go from there? Do you think the box "fills up" and needs to be literally returned to earth when the shuttle lands? That would be pure fantasy, yet the system plays music.

[marce]

Planes are an interesting case, you have a chassis connection, often connected to the plane body (or you are just unlucky and your electronics are hanging off a wing). The problem is lightening, it likes planes, so your whole chassis can become quite interesting for a while, a slight increase in potential... the problem is making sure this surge does not raise any system ground potentials and the fun that could cause. Like all system grounding its system dependant and takes careful consideration to get right.

My personal view is with today's miniaturisation of circuitry is have the DAC, amp etc in one box, the music storage in a server, NAS etc. connected wirelessly to keep the PC noise isolated, the whole lot remotely controlled with the shortest wires possible to the motors (speakers). You minimise all the ground loops different components create, in fact make it double insulated, with a remote main power supply and point of load supplies on board with filtering and you can remove the PE connection from the box. All shielded, minimal connections, noise controlled. I would use a flexi-rigid PCB with each section tailored to its function, DAC, power amp etc. You could then control the chain, its "grounding" etc. with finesse.

[fas42]

marce, I agree with your concept of how to do audio systems - 10 years ago I was getting a bit serious about creating something like that ... but it never happened, - instead, life happened ...

[marce]

We agree!!!

 

LOL

On a serious not, using a flexi rigid would get rid of a lot of connectors, just up your street.

[jabbr]

20 hours ago, marce said:

My personal view is with today's miniaturisation of circuitry is have the DAC, amp etc in one box, the music storage in a server, NAS etc. connected wirelessly to keep the PC noise isolated, the whole lot remotely controlled with the shortest wires possible to the motors (speakers)

 

Yes, although fiberoptic Ethernet is quite good and provides dedicated low latency bandwidth. My approach is to isolate digital and analog systems (floating power) using differential/balanced signaling between. "ground" is a common (0V) reference point (aside from safety).

[Jud]

Dumb question: “PE” connection?  Power entry?

[Ralf11]

if any metal items can act as inductors or caps, then HF (noise) might take a different path to wherever than LF's

 

actually that would make a good title for a jazz or fusion album: a different path to wherever

[mmerrill99]

10 minutes ago, Jud said:

Dumb question: “PE” connection?  Power entry?

Power earth

[One and a half]

1 hour ago, mmerrill99 said:

Power earth

Protective Earth 

[marce]

13 hours ago, Jud said:

Dumb question: “PE” connection?  Power entry?

Protective earth

What Keith Armstrong of EMC fame, calls the noise distribution network.

[Abtr]

To eliminate leakage currents I use an audio-grade isolation transformer (https://www.toroidal-transformer.com/shop/high-end-audio/audio-safety-transformer/709/audio-scheidingstrafo-900va.html) connected in balanced (or common) mode, with the center tap of the secondary winding (as well as the shield between primary and secondary windings) connected to (primary) Earth. In my system this alone makes a HUGE sound quality improvement. There appears to be truth in the idea that such an AC power arrangement causes balanced reactive currents and leakage currents from connected power supplies which cancel at the center tap (null/ground). (Note that all audio components including computer source are plugged into a single low impedance power strip connected to the iso transformer.)

 

From http://www.equitech.com/articles/enigma.html :
 

Quote

 

A fourth variation is Common Mode. The voltage potential traversing the circuit is referenced to its own zero crossing point. (Fig. 4) The power resembles a balanced audio circuit or an XLR input from an unbalanced to balanced audio transformer. In this configuration, a center tap (ground reference) is added to the transformer output winding which divides the output voltage into two 60-volt-to-ground potentials. The potential across the system's two main current-carrying wires remains the usual 120 Volts. From the center tap reference point to the transformer output terminals, the measured voltages are identical but inversely phased. The power output is symmetrical -- each side of the system is a mirror image of the other.

 

Figure 7 and Figure 8 are illustrations of a balanced power application. In both cases, inversely phased power elements meet at the common ground. The effect on the grounding system is also observable in the power grid with a near-total absence of locally generated noise under the balanced power/load condition. Note that when balanced power is applied, peak and inverse peak voltages are equally present with respect to ground. This means that reactive currents generated by the impedance load are also perfectly balanced. (In the electrical industry, the term "Counter EMF" is sometimes used to describe these power artifacts.) Counter power elements which are balanced will null (sum to zero) in a balanced system at the ground or center tap, an obvious benefit in sensitive audio and video production facilities where overall noise floor and dynamic headroom are directly affected by grounding noise levels.

 

 

[Abtr]

By the way, my recent experience with balanced power indicates that power amps (and active subs for that matter) should be disconnected from the center tap null/Earth. Power amps appear  to be quite sensitive to whatever residual noise current is still lurking there. "As a rule, high gain and high impedance equipment is where most noise is introduced." (http://www.equitech.com/articles/bpng.html)

[RogerD]

I thought I would post a final audio of my extreme grounding experiment. Now moving on to my HT horn system,which I'll be sure to use the ExGr method.

 

 

[fas42]

Very nice one, Roger - big thumbs up!

[RogerD]

5 minutes ago, fas42 said:

Very nice one, Roger - big thumbs up!

Hi Frank......thanks!