Grounding and Noise Floor

[Bill Hart]

People think of electrical current flow like water that pours from a hose into the ground and disappears. It's really a loop, with ground being the return path. You do not want the return path to differ from the signal (delivery) path or things get out of synch. As Mark notes, ensuring every signal line has a matching ground is one way to help. Normally signal ground flows through the shield of your interconnect, but that shield is often coupled (directly, through a small R, or capacitively) to the chassis or earth (outlet) ground. An extra ground connection creates another, potentially harmful, loop. The idea behind star grounding is that every component has just one path to the common ground point, it is the same path as the signal return current path, and that one ground point then provides the final earth ground.

Note the ground wire in your outlet is considered a safety ground and should not normally carry any current anyway.

Edit: I was writing while valkyrie was posting, he was faster. Our posts are in agreement.

That's what i didn't understand about the Granite Audio device. How could it work if you were still plugging everything into the wall with 3 pins?
Thanks Don. I digested Val's post whole, i'm surprised those snakes don't get indigestion.

 

[microstrip]

I(...) Take my Lamm ML2. For example.
Ever use it?

Sorry, no experience with Lamm. It is not distributed in my country. We will have the Audio Show this weekend, but no Lamm ...

 

[Robh3606]

I'm not raising this b/c I currently (no pun) have a problem but it's always an issue with this horn system it seems.

Hello whart

What horns are you running?? I have run a direct radiator horn combo so I feel your pain about noise. Not as bad as an all horn set-up but woofers at 98 mids at 101 and compression drivers at 110, no attenuation just HF compensation. It's a hodge podge of tranformer isolated and balanced Pro gear and unballanced consumer gear. A real PITA to get the gain structure right and keep the noise down.

Rob

 

[Bill Hart]

Hello whart

What horns are you running?? I have run a direct radiator horn combo so I feel your pain about noise. Not as bad as an all horn set-up but woofers at 98 mids at 101 and compression drivers at 110, no attenuation just HF compensation. It's a hodge podge of tranformer isolated and balanced Pro gear and unballanced consumer gear. A real PITA to get the gain structure right and keep the noise down.

Rob

Avantgarde Duos. 104db efficiency. Phono only, all tube electronics: tube phono stage, tube line stage, tube amps. The woofers, as you may know, are self-powered dynamics. There is no crossover between the main power amp and the mid horn, which is where the amp first connects. Then jumpers to the tweeter horn and the woofer cabinets (getting a feed from the main amps, rather than line level input).

 

[TBG]

Bill, I have always had the same question as you about having three grounds on every unit-one through the AC ground, one through your interconnects, and then having a star ground through some sort of a grounding unit. I must say straight away that I tried the Granite Audio unit and heard no benefit. This tainted my expectation when I first tried the Tripoint Troy long ago. It had a big impact or so I thought at that time. Before I went to the Troy, I had an H-Cat preamp where the manufacturer strongly recommended only having it grounded through the AC. I tried it using cheaters that had been treated on a Tesla Coil which was far superior to the untreated ones. I must say the sound was superior, with only the preamp on an AC ground. Of course the ICs were still grounded, but there was no star grounding.

After I bought the Tripoint Troy Signature, I again lifted the grounds on everything except the H-Cat. I heard little if any improvement and suspected that the Tripoint was keeping the star ground from having any negative impact, but still I had grounding through the ICs. Frankly, I cannot imagine how I might be able to lift their grounding.

So where I am right now is to assume that the signal ground through ICs, is not critical, at least not with my High Fidelity Cables.

I should also say that I have two dedicated ac lines. The one on the right side of the room feeds everything except the turntable. AC conditioning is largely responsible for this as I could not afford to use all that I have on the right side also on the left side.

Oh, and one final comment. I allow none of my ICs, PCs, or speaker wires to have any contact with the floor. They all have porcelain isolators under them.

 

[RogerD]

The only thing that matters and will really makes a difference in overall SQ is the "physical size" of the pathway cable. Too many units use a very small size cable and it is in effect pissing in a ocean of noise. That's why users keep adding units and cables....a very inefficient method. Adequate large pathway chassis ground will increase SQ and since most signal grounds are tied to chassis grounds,that's all that's needed.

 

[TBG]

The only thing that matters and will really makes a difference in overall SQ is the "physical size" of the pathway cable. Too many units use a very small size cable and it is in effect pissing in a ocean of noise. That's why users keep adding units and cables....a very inefficient method. Adequate large pathway chassis ground will increase SQ and since most signal grounds are tied to chassis grounds,that's all that's needed.

I know, however, that shielding grounding wires makes a great difference and that filtering in a star grounding scheme matters. I also suspect that the quality of the ground to Earth matters.

 

[TBG]

Yes, I have a friend who has repeatedly proven to me that internal grounding improvements rather than just using the chassis path to ground, make a big difference.

 

[Speedskater]

I know, however, that shielding grounding wires makes a great difference and that filtering in a star grounding scheme matters. I also suspect that the quality of the ground to Earth matters.

For general reference:
When running the wire to the ground rod through a metal conduit (ridge or flexible) the conduit needs to be connected to the wire at both ends, otherwise to conduit acts as a choke/inductor, exactly what you don't want.

 

[Speedskater]

Yes, I have a friend who has repeatedly proven to me that internal grounding improvements rather than just using the chassis path to ground, make a big difference.

Yep, Jim Brown pro-audio EMI/RFI expert and Ham radio operator wrote this yesterday:

An earth ground is vital for lightning safety, but does not matter for noise, nor does it make antennas work better (except for RX antennas that interact with soil to work). What DOES matter (a LOT) is proper and complete BONDING. Bonding is the interconnection of the chassis of every piece of equipment in your station, and the interconnection of that to every ground in your home/building. That bonding is also critical for lightning protection.

 

[microstrip]

Yep, Jim Brown pro-audio EMI/RFI expert and Ham radio operator wrote this yesterday:

An earth ground is vital for lightning safety, but does not matter for noise, nor does it make antennas work better (except for RX antennas that interact with soil to work). What DOES matter (a LOT) is proper and complete BONDING. Bonding is the interconnection of the chassis of every piece of equipment in your station, and the interconnection of that to every ground in your home/building. That bonding is also critical for lightning protection.

Could you post a link so that we know in what context this statement was written?

 

[RogerD]

Yep, Jim Brown pro-audio EMI/RFI expert and Ham radio operator wrote this yesterday:

An earth ground is vital for lightning safety, but does not matter for noise, nor does it make antennas work better (except for RX antennas that interact with soil to work). What DOES matter (a LOT) is proper and complete BONDING. Bonding is the interconnection of the chassis of every piece of equipment in your station, and the interconnection of that to every ground in your home/building. That bonding is also critical for lightning protection.

Yes Kevin in my experience bonding makes the difference. In Pro studios bonding is standard practice.

 

[Speedskater]

Could you post a link so that we know in what context this statement was written?

Sorry, it an E-mail group letter. It's about EMI/RFI getting into amateur Ham radio stations (shacks).

He was responding to:

My earth ground here is poor, but I intend to remedy that asap. 4 ea 8 foot copper-clad steel
ground rods, separated by 8 foot, driven to within an inch or two of the surface of the ground,
connected together by size 00 welding cable, then into the shack. about 15 feet, with 00 welding cable.

But here are about 50 Jim Brown papers and Power Points. Many on EMI/RFI topics in both Ham radio and hi-fi & pro audio:

http://www.audiosystemsgroup.com/publish.htm

 

[GaryProtein]

Yep, Jim Brown pro-audio EMI/RFI expert and Ham radio operator wrote this yesterday:

An earth ground is vital for lightning safety, but does not matter for noise, nor does it make antennas work better (except for RX antennas that interact with soil to work). What DOES matter (a LOT) is proper and complete BONDING. Bonding is the interconnection of the chassis of every piece of equipment in your station, and the interconnection of that to every ground in your home/building. That bonding is also critical for lightning protection.

Absolutely 100% true.

 

[TBG]

Yep, Jim Brown pro-audio EMI/RFI expert and Ham radio operator wrote this yesterday:

An earth ground is vital for lightning safety, but does not matter for noise, nor does it make antennas work better (except for RX antennas that interact with soil to work). What DOES matter (a LOT) is proper and complete BONDING. Bonding is the interconnection of the chassis of every piece of equipment in your station, and the interconnection of that to every ground in your home/building. That bonding is also critical for lightning protection.

I deal with lightening safety by detaching the power and grounding systems from the wall.

I have found that shielded grounding in a star ground connection with more and more shielding greatly improves the realism. I suspect this is all EMI and RFI interferance not getting to the audio signal but vibration also seems to be a factor, not surprisingly.

 

[Speedskater]

I deal with lightening safety by detaching the power and grounding systems from the wall.

Remember to disconnect the internet, telephone and cable TV wires also.
Living 400 feet from a big lake, I would spend a lot of time disconnecting wires. Yesterday we were getting the sailboat ready to launch. A sunny, windy day then a thundershower popped up out of nowhere, it hardly rain and disappeared quickly.

I have found that shielded grounding in a star ground connection with more and more shielding greatly improves the realism. I suspect this is all EMI and RFI interferance not getting to the audio signal

Some advanced grounding systems:

Grounding Systems

SRPP :: System Reference Potential Plane
STGP :: Signal Transport Ground Plane
ZSRG :: Zero Signal Reference Grid
ZSRG :: Zero Signal Reference Conductors
ZSRP :: Zero Signal Reference Potential
ZSRP :: Zero Signal Reference Plane
MESH-CBN :: Meshed Common Bonding Network
MESH-IBN :: Meshed Isolated Bonding Network
PEC :: Paralleled Earth Conductors
PBC :: Paralleled Bonding Conductors

Conductive Structure

but vibration also seems to be a factor, not surprisingly.

Unless it's turntables and vacuum tubes, that's a big surprise!

 

[TBG]

We have twice been hit by lightning that has taken out intercom systems, telephones, and blown a hole in a gas pipe to the water heater. I am really only worried about the expensive audio gear.

 

[amirm]

We have twice been hit by lightning that has taken out intercom systems, telephones, and blown a hole in a gas pipe to the water heater. I am really only worried about the expensive audio gear.

Best defense is an SPD (Surge Protection Device) right at service entrance where the best (lowest impedance) ground is available. Get an industrial one that is fused as otherwise, these units themselves can self destruct and catch on fire!

And get good insurance as nothing stops a real lightning bolt if it comes close to you.

When we lived in Florida we had one hit. My wife got shocked while talking on the (wired) phone. We actually had Romex wire behind the walls powerlight lights melt! This was early 1980s and I had a Carver receiver and it got blown to pieces. Carver repaired it (under warranty) and sent me a big bag of bad parts! The real deal has incredible amount of power and nothing stops it really.

 

[Brucemck2]

Best defense is an SPD (Surge Protection Device) right at service entrance where the best (lowest impedance) ground is available. Get an industrial one that is fused as otherwise, these units themselves can self destruct and catch on fire!

And get good insurance as nothing stops a real lightning bolt if it comes close to you.

Pay a LOT of attention to other methods of entry too. I had lightning hit my backyard. Everything going through the electrical panel was fine. But, a current surge went through some door contacts on the alarm system, and from there to the rs232 devices connected to the alarm system, which led to the rs232 connections on two AV receivers and two HVAC controllers. Based on that I unplug ALL external wiring from expensive AV gear and don't count on having it being unplugged from the wall as offering 100% protection.

 

[GaryProtein]

. . . . Based on that I unplug ALL external wiring from expensive AV gear and don't count on having it being unplugged from the wall as offering 100% protection.

That is a very scary thought