Driving a new ground spike

[JohnRoberts]

I recall the UL being picky about things labelled as "ground".... patchbays, not so much.

JR

 

[Newmarket]

Yes, gas and water lines must also be connected to building ground.

I have only worked on systems within a single building, but as Bill Whitlock (Mister CMRR) has pointed out to me in the past, because anything metal is connected to the building grounding conductor, if you have something like a patch bay in a commercial building, the exposed metal has to be connected to building ground. That is why the XLR cabling standard still has pin 1 for shield connection which is not connected to the connector shell in cables, so that when the cable shell is grounded by a patch panel connected to building metal, you do not get multiple connections of a cable shield to building ground on a long connection going between buildings, or between different electrical service entrances in a really large building.

I need a drawing to visualise what is being said here (= 1000 words etc)

 

[Speedskater]

METAL PARTS:
bonding required: for the NEC electrical items junction boxes, outlet boxes, panels, receptacle frames, conduit.

accessible large objects like interior pipes and building framing should be but not always required to be bonded.

gas pipes have their own rules depending on type

 

[nielsk]

Relying on cable shields for safety grounds is dangerous (= fuse to ground), as is not having all ground systems inside the same facility being properly bonded together.
Every piece of equipment in a studio system (that is not some POS with a 2 prong power cord) should have it's power cord complete with a ground pin connected to a properly grounded outlet. Power cord ground lifts are NEVER the correct way to deal with a ground problem such as a ground loop.
There are cases where the power cord ground is not bonded to the internal P/S ground (as in a power supply to a console where the P/S commons tie to ground at the console, and this P/S common is properly terminated to Technical Ground using the console's ground lug). In the "good ole days" many manufacturer's included a ground bond strap on the back of equipment for this purpose... but inside individual pieces of equipment this is a poor solution to a shielding / physical mounting fault.
BTW, "Everything goes to the console, and it goes to the ground rod" IS a Star Ground system. Of course it is only safe if that ground rod is bonded to all other ground systems in the facility.
The reasons why moving the bonding point for the main ground buss to the audio system from being tied to the Neutral at the panel to being bonded to the end of a series of ground rods that the panel (& conduit & Neutral bond) is bonded to often vastly lowers the noise floor of an studio system may be debatable, but the results are not. This is standard practice in "isolated ground" systems such as Hospitals and large computer facilities, where the conduit (system safety) ground and the 3rd pin ground are kept separate and are only bonded at this point.
Even better, a "2 hots & a ground" system (balanced AC power or 2 phase) solves the voltage on the Neutral problem nicely.
Of course, any piece of audio gear will turn on and pass signal without being grounded, but that does not mean it is safe or predictable, and as stated elsewhere, shielding and CMMR performance are improved with proper grounding.
I am certainly not working on a super-collider in my basement, but I hope all of us here understand enough basic physics to be able to safely implement audio systems and get great results from them!

 

[ccaudle]

I fail to see the logic of earthing a patchbay for safety or regulatory reasons

I think I phrased that incorrectly. I believe that the regulation is not actually that a patch bay would have to be grounded per regulations, but that building metal framework does have to be grounded, so in the case the patchbay is mounted to building framework for convenience, you get the patchbay grounded as a side effect.

 

[JohnRoberts]

Relying on cable shields for safety grounds is dangerous (= fuse to ground), as is not having all ground systems inside the same facility being properly bonded together.

cable shields are not safety grounds... UL ground bonding tests drive on the order of 50A though a "ground" path and measure the voltage rise... many shields will fuse open at 50A.

Every piece of equipment in a studio system (that is not some POS with a 2 prong power cord) should have it's power cord complete with a ground pin connected to a properly grounded outlet.

yup

Power cord ground lifts are NEVER the correct way to deal with a ground problem such as a ground loop.

yup

There are cases where the power cord ground is not bonded to the internal P/S ground (as in a power supply to a console where the P/S commons tie to ground at the console, and this P/S common is properly terminated to Technical Ground using the console's ground lug). In the "good ole days" many manufacturer's included a ground bond strap on the back of equipment for this purpose... but inside individual pieces of equipment this is a poor solution to a shielding / physical mounting fault.
BTW, "Everything goes to the console, and it goes to the ground rod" IS a Star Ground system. Of course it is only safe if that ground rod is bonded to all other ground systems in the facility.

can't quite follow what you said..

The reasons why moving the bonding point for the main ground buss

bus

to the audio system from being tied to the Neutral at the panel to being bonded to the end of a series of ground rods that the panel (& conduit & Neutral bond) is bonded to often vastly lowers the noise floor of an studio system may be debatable, but the results are not. This is standard practice in "isolated ground" systems such as Hospitals and large computer facilities, where the conduit (system safety) ground and the 3rd pin ground are kept separate and are only bonded at this point.
Even better, a "2 hots & a ground" system (balanced AC power or 2 phase) solves the voltage on the Neutral problem nicely.
Of course, any piece of audio gear will turn on and pass signal without being grounded, but that does not mean it is safe or predictable, and as stated elsewhere, shielding and CMMR performance are improved with proper grounding.
I am certainly not working on a super-collider in my basement, but I hope all of us here understand enough basic physics to be able to safely implement audio systems and get great results from them!

ground is a concept... audio signals are differential audio + wrt audio -....

The whole pin 1 "problem" drama explains that you can bond all shields and whatever to chassis ground and still pass clean audio with proper differential signal discipline.

JR

 

[abbey road d enfer]

Relying on cable shields for safety grounds is dangerous

Did anyone suggest that?

 

[Speedskater]

BTW, "Everything goes to the console, and it goes to the ground rod" IS a Star Ground system. Of course it is only safe if that ground rod is bonded to all other ground systems in the facility.

and you were doing so well.
But and it's a big but. it 9the Safety Ground0 does not go to the ground rod !
it goes to the power company Neutral in the main breaker box.

The reasons why moving the bonding point for the main ground buss to the audio system from being tied to the Neutral at the panel to being bonded to the end of a series of ground rods

Never ever do this.

 

[abbey road d enfer]

BTW, "Everything goes to the console, and it goes to the ground rod" IS a Star Ground system.

Of course it is only safe if that ground rod is bonded to all other ground systems in the facility.

Here, the subject is making a system as clean as possible.
Making it safe does not guarantee it.
Most of noise issues come from HOW the various "grounds, earths, shields, references..." are interconnected.

 

[dbonin]

He explains sooo much in that clip. Worth watching if you are working with mains electricity.

 

[MisterCMRR]

Go to https://centralindianaaes.files.wordpress.com/2012/09/indy-aes-2012-seminar-w-notes-v1-0.pdf
Read carefully starting at page 17. See especially pages 27 and 29. Earth ground rods are ONLY a measure against lightning. Earth grounds do NOT protect people from being electrocuted when equipment fails (as it often does) in a way to put full AC line voltage on its touchable parts. Safety ground provides a low-impedance (low enough to cause a high fault current to trip a breaker) path back to NEUTRAL, not earth ground. I had a guy who was preaching this gospel of "use your own ground rods instead of neutral" (as page 29) kicked out of a trade show - I reminded the show staff of their legal liability letting this moron promote this deadly practice!

Incidentally, UL allows for gear to have a 2-prong power connection = but, to get UL listing, it has to survive torture tests to prove that, under no circumstance does leakage current to the chassis or any touchable part exceed 1 mA (a tingle if it flows in your body). They usually accomplish this with a double-redundant insulation system and/or non-resettable thermal cutoff switches embedded in parts that can overheat (like wall-wart transformers).

 

[Tubetec]

Thanks Bill ,

 

[Bo Deadly]

Safety ground provides a low-impedance (low enough to cause a high fault current to trip a breaker) path back to NEUTRAL, not earth ground.

Just to be clear if people like me had trouble parsing this - if the load touches the chassis the fault current will flow through earth ground of the mains cable back to the panel, into the neutral bus and not to the ground electrode.

 

[abbey road d enfer]

Just to be clear if people like me had trouble parsing this - if the load touches the chassis the fault current will flow through earth ground of the mains cable back to the panel, into the neutral bus and not to the ground electrode.

Bo, you should proofread this sentence.
What "load" touches the chassis? Why this chassis would be live?
"Earth ground of the mains cable" does not make much sense.

 

[Bo Deadly]

Bo, you should proofread this sentence.
What "load" touches the chassis? Why this chassis would be live?
"Earth ground of the mains cable" does not make much sense.

Presumably we're just mixing up terms. Mains wiring is most commonly referenced as:

Live / Hot and frequently has a letter of 'L' which I might be confusing with "load" instead of "live" perhaps
Neutral which has letter 'N'
Ground / Earth / Protective Earth most commonly referred to as ground with letter 'E' (or the symbol "⏚").

So restated, if the live conductor touches the chassis the fault current will flow through the ground conductor of the mains cable back to the panel, into the neutral bus and not to the ground electrode.

Still a problem?

 

[abbey road d enfer]

Ground / Earth / Protective Earth most commonly referred to as ground with letter 'G'

In my area, it's "E". Never G.

 

[Bo Deadly]

In my area, it's "E". Never G.

Actually it looks like "E" is more common. The most common is the earth ground symbol "⏚" but "PE" is also used.

 

[JohnRoberts]

Presumably we're just mixing up terms. Mains wiring is most commonly referenced as:

not to be more pedantic than usual

Live / Hot and frequently has a letter of 'L' which I might be confusing with "load" instead of "live" perhaps

L typically refers to "line"

Neutral which has letter 'N'

Correct but UL tends to also refers to neutral as a line conductor.

Ground / Earth / Protective Earth most commonly referred to as ground with letter 'G'

EGC (equipment grounding conductor) aka safety ground. The EGC is bonded to neutral at the panel to provide an adequate current path to trip breakers from voltage faults. UL calls it a grounding pin (grounding conductor) as compared to the line blade(s) in a plug.

So restated, if the live conductor touches the chassis the fault current will flow through the ground conductor of the mains cable back to the panel, into the neutral bus and not to the ground electrode.

correctamundo...

Still a problem?

I just opened up a UL specification (S-1436) that I paid hundreds of dollars for when designing my outlet tester.

UL refers to both line and neutral as "line conductors", they also refer to "ungrounded" and "grounding" conductors. Presumably the "ungrounded" are referring to hot or energized conductors (line, neutral). In a schematic inside the spec they label the 3 terminals of a plug with ; H, N, and G.

I apologize if this UL word salad is confusing, I paid good money to get this confused by their spec.

JR

 

[scott2000]

So restated, if the live conductor touches the chassis the fault current will flow through the ground conductor of the mains cable back to the panel, into the neutral bus and not to the ground electrode.

EGC? Effective Ground-Fault Current Path? 2 birds?

 

[MisterCMRR]

To add to the terminology confusion, NEC (National Electrical Code or just "Code" in the US) refers to the 3 conductors as the "ungrounded" or "line," the "grounded" or "neutral," and the "grounding" or EGC (equipment grounding conductor). One has to read the language of the Code carefully to avoid confusing grounded with grounding.

But, to the point about fault current, it's important to remember that ALL current flow returns to the source of voltage that is causing the current flow. In the case of fault (or normal load) current, this is the transformer serving your building - and it returns via the neutral conductor from your breaker panel to the transformer. The normal load current will flow to the neutral via the neutral conductor in the branch circuit wiring. But, if an equipment fault occurs, the fault current flows to the breaker panel (and utility neutral0 via the EGC or "safety ground" conductor. Notice that earth ground is not involved in ANY of this. The situation would be identical in an airliner and the wiring is the same. Earth ground is not a universal "sink" for utility power currents. However, earth ground is exactly where lightning currents need to flow to in completing the circuit (for cloud-to-ground strokes - there are cloud-to-cloud strokes that do not involve earth).