There's also abundant evidence the ~40dB CMRR provided by a basic balanced implementation is insufficient for inaudibility in mildly inclement conditions such as not the greatest grounding, mains circuit to mains circuit spans, or not so great audio component implementations. Hence the market for DI boxes and other fix its approaching 100dB CMRR. Usually that much is overkill but maintaining input signal integrity is one reason why the Mod and Parallel, as high end builds, use THAT 1200s.
twest820
Balanced cabling is probably one of the easiest ways to improve system CMRR while meticulous system grounding of pieced together components is probably one of the most difficult.
How to improve grounding related s/n with components that are physically large and maybe physically distant also? You are talking about CMRR approaching the s/n of the average 16 bit DAC here, being achieved by a system with huge antenna loops maybe even looped in the wall.
Best way seems balanced or digital, then electrical isolation via transformer/optical/wireless, then look into the grounding if there are still intractable issues.
I am just questioning what the purpose of the big heavy pretty box is. In terms of impedance it appears to serve no purpose and maybe even make things worse.
Amplifier grounding with Entreq.
Yesterday I had the pleasure of demoing a new application of Entreq grounding namely grounding the amplifier (in my case Vitus SIA 025).
---snip---
...Big improvement virtually immediately noticeable.Sounded like a new recording, Fresher, clearer, better instrumental separation, imaging, dynamics and sound staging. No ifs buts or downsides.
The language does not agree with the potential benefits of better s/n as I understand them. Only perceptible benefits would be from reduction of the noise itself plus reduction of any intermodulation due to the noise. Dynamics should be completely unaffected except in case where the noise is eating up considerable amp capacity.
So earth cables from each of the amplifier's negative speaker terminals, both channels, to two separate Silver Minimus.
Doesn't work using the input jacks of an amplifier...
This seems to make some sense. The speaker grounds are going to have low impedance path directly back to the internal amplifier ground and that internal ground should also have low impedance path to internal star connection between case/earth and signal grounds.
Bridging that speaker terminal to an external system star point via low impedance path could improve s/n, if such benefit were obtainable that way.
It seems that using a component with metal case installed in a metal rack with the other associated components is a better way to achieve same result, short of that a thick cable attached directly to the star point inside the component might help. Typical consumer probably has no better unambiguous ground reference than the speaker terminal(s).
Grounding to the line input side of an amp could inject common mode noise across higher impedance signal ground connection between line input stage and power output stage. The input stage may also have some cable ground isolation built into it that would be effectively short-circuited.
This new amplifier grounding should be done additionally and completely separately from any other Entreq grounding...Doesn't work...by using an existing ground box being used for other components or Entreq i/cs.
Now I am lost again. Why is another box required, if the important thing we are doing is improving ground impedance
between components? The fewer the boxes and the shorter the cabling, the better. More stuff just adds more impedance that interferes with the objective of shunting noise out of the interconnect signal ground.
Is there some ground domain isolation going on, with separate star connections that have different noise signals on them being added in some systematic fashion as 'earth grounds' in all these separate boxes?
If so, how does that improve CMRR? All of these somewhat isolated grounds with electrical potential between them are spanning both ends of the signal path and injecting that electrical potential directly into the signal ground path.
It is contradictory to both shunt ground noise away from signal interconnect through low impedance case/earth ground while electrically isolating that low impedance ground between the components that share it. Makes no conceptual sense. I would have expected all those speaker terminal grounds to terminate at a single star point on a single earth ground that all the system components share.
If the grounding box is actually floating the chassis grounds from the mains ground then I could see some potential benefit as long as they are also ensuring a solid star interconnect somewhere else in the system that is independent of the audio interconnects, but that approach presumes low impedance earth/case ground path
somewhere between individual components or the signal ground on the audio interconnect has to carry all the current picked up via antenna action.
If through the grounding box then why use separate boxes that increase the impedance of the star connection with physical separation between the earth/case grounds of the components? If not through the grounding box then where?
The contradiction leaves me confused and wondering if I fell down the rabbit hole somewhere.
Skeptical but willing to be convinced by measurements and theory. If a grounding box actually reduces noise, that effect should be measurable and its benefit explicable by physics.
