Driving a new ground spike

[JohnRoberts]

So earth is not connected to neutral in an RCBO breaker? That is odd. In that case fault current would go through the earth ground electrode, through soil to neutral at the utility pole. Probably not a good arrangement for a studio.

Update:

Actually it looks like any imbalance (Residual Current Breaker with Overload) will trip it. So as long as the earth ground electrode impedance isn't too high, EGC doesn't need to be connected to neutral for safe operation.

View attachment 92150

Thanks, we seem to be confusing multiple topics...

Indeed GFCI/RCD protection devices do not need to be ground bonded to work, these devices simply compare the current flowing in the two active conductors (typically Line and Neutral) and trip open circuit if they don't null out. The GFCI per outlet protection uses a trip threshold of only 6 mA imbalance, RCD branch circuit protection requires more like 30mA imbalance. The RCBO is an RCD + conventional overload breaker making these suitable for use in mains breaker panels to replace old school overload only breakers.

These devices do not care where the missing current goes, only that it isn't flowing where it should be....

[TMI] Back several years ago when I was messing around with designing a premium talent protector I combined a GFCI outlet with a 3 pole relay so in addition to opening the line conductor, I opened all three circuits. My logic was that live performance talent shock hazards include external energized voltage sources exposing the talent to hazard conducting through a still present safety ground. In my premium protector I sensed current flow through the safety ground path and opened all three conductors at 6mA or more current in the ground.

It is clearly debatable wether a robust safety ground "could" ever be a shock hazard. I didn't even bother trying to convince UL, they like their safety ground bonds. It is not unusual for talent to get shocks from grounded microphones that provide a return path for external energized sources. [/TMI]

JR

 

[ccaudle]

yes they are but only at the mains distribution panel.

That does not appear to be the case in the diagram which moamps was asking about.

Is there something about that diagram that would stop fault current from getting back to neutral?

There does not appear to be a reliable path from protective earth back to neutral. With North American style equipment that would not be a reliable trip if there was a line to case fault in the power wiring. The diagram shows an earth connection to the RCBO, but it is not obvious to me how much current you would be guaranteed to have flowing through PE. With a RCD breaker for an entire branch circuit the trip point has to be set higher than for the single outlet GFCI type devices used in North America, so you have to make sure that the minimum guaranteed current is still enough to trip the breaker.

The second set of devices which have a single large RCD feeding three breakers do not appear to have any path from protective earth to either neutral or to the RCD, so apparently if those devices have a line to chassis fault the chassis just sits at 230V until someone is unlucky enough to touch it and sink enough current to trip the large RCD.

From some info I found elsewhere, the transformer feeding the houses or buildings has neutral bonded to earth, but then the resistance between earth and neutral at the house would depend on distance, soil composition, moisture, etc. I'm not sure how that the protective earth resistance is guaranteed to be low enough to trip the protection device. Perhaps if the resistance is high enough that the RCD does not trip it is not considered to be a health hazard.

It also seems that the North American style neutral bonded to protective earth at the service entrance is now allowed in the UK, and referred to as "protective multiple-earth," the "multiple" there referring to neutral being bonded to earth at the power transformer and also at each house or building.

So as long as the earth ground electrode impedance isn't too high

I will trust that the European regulatory agencies have done their work to make sure the standard is safe, but as someone who is only familiar with North American wiring standards it looks like they are relying on literal dirt to make a protective connection.

 

[JohnRoberts]

That does not appear to be the case in the diagram which moamps was asking about.

that diagram stipulates UK... I have no wisdom about UK standards or veracity of the diagram.

There does not appear to be a reliable path from protective earth back to neutral. With North American style equipment that would not be a reliable trip if there was a line to case fault in the power wiring. The diagram shows an earth connection to the RCBO, but it is not obvious to me how much current you would be guaranteed to have flowing through PE. With a RCD breaker for an entire branch circuit the trip point has to be set higher than for the single outlet GFCI type devices used in North America, so you have to make sure that the minimum guaranteed current is still enough to trip the breaker.

30 mA is higher than US GFCI 6 mA, while drawing 30 mA from 230VAC only requires 7.6k of resistive path.

The second set of devices which have a single large RCD feeding three breakers do not appear to have any path from protective earth to either neutral or to the RCD, so apparently if those devices have a line to chassis fault the chassis just sits at 230V until someone is unlucky enough to touch it and sink enough current to trip the large RCD.

RCD and GFCI only compare current leaving and returning for balance over two wires. They don't much care where it strays off to.

From some info I found elsewhere, the transformer feeding the houses or buildings has neutral bonded to earth, but then the resistance between earth and neutral at the house would depend on distance, soil composition, moisture, etc. I'm not sure how that the protective earth resistance is guaranteed to be low enough to trip the protection device. Perhaps if the resistance is high enough that the RCD does not trip it is not considered to be a health hazard.

Earthing mains power distribution AFAIK is mainly about lightning mitigation, soil is not trustworthy enough low impedance to provide fault overload trip current path, likely enough conduction to trip RCD/GFCI. Lightning is high enough voltage that soil resistance is not a problem.

It also seems that the North American style neutral bonded to protective earth at the service entrance is now allowed in the UK, and referred to as "protective multiple-earth," the "multiple" there referring to neutral being bonded to earth at the power transformer and also at each house or building.

IDK

I will trust that the European regulatory agencies have done their work to make sure the standard is safe, but as someone who is only familiar with North American wiring standards it looks like they are relying on literal dirt to make a protective connection.

IDK

JR

 

[Rob Flinn]

if one looks at the page moamps diagram came from it gives examples for wiring European & North American consumer units. It also explains that the RCBO is a more expensive option since it provides both differential & over current protection but is best if you don't want one branch circuit taking out a load of other circuits. If you go cheap you use one RCD (which provides the differential protection) feeding an MCB (which provides the over current protection) for each branch. This is cheaper, but if one branch trips the RCD then all the power goes off in the other branches. Therefore in a studio it may be a better but more expensive option to use RCBO's since if one circuit trips it won't necessarily take out your pro tools rig mid session.

Also from looking at the moamps page it does appear that in the U.K it is not a requirement to bond Neutral & Earth in the consumer box, however I was taught many years ago that they are bonded at the power station.

 

[Bo Deadly]

Therefore in a studio it may be a better but more expensive option to use RCBO's since if one circuit trips it won't necessarily take out your pro tools rig mid session.

In a studio I would think noise performance of sensitive analog gear might benefit from EGC being connected to neutral at the panel. Otherwise, you could have common mode noise on the power lines relative to 0V and chassis.

If the studio is divided into studio and "house", then using RCBO for the house side with appliances and HVAC and sump pumps and so on might be a good arrangement. But for the studio side, my instinct would be to have EGC connected to neutral.

 

[nielsk]

"Relying on cable shields for safety grounds is dangerous "
When an audio system is interfaced with shielded audio cable that is terminated at both ends and none of the equipment is grounded, this is exactly what is happening.
I have personally seen power supply failures that resulted in a direct short between AC Line and Chassis, resulting is 120 VAC on the audio cable shield.
"it 9the Safety Ground0 does not go to the ground rod ! Never ever do this."
Why not? This system maintains 100% bond between Neutral and Technical Ground. This does not negate the existence of a Safety Ground system where the ground BUS (thank you, I never catch all the typos!) is bonded to the Neutral BUS in the panel, in fact it requires it.
Dirt may not be a good conductor, but water is. Referring to my first post, a good Technical Ground system requires the interconnected ground rods to penetrate the ground water table, and if this is not possible more exotic methods must be employed to ensure it's functional integrity.
As to the earth not being able to sink current, why it it that when a power line breaks and falls to the ground, it can create a 100 foot long 6 foot deep glass lined trench? (I have personally seen this).
If ground rods are only for lightning protection, what happens when the Neutral from the power pole is faulty or breaks? (I have personally seen this)
I don't intend to be argumentative, but the NEC (and , it would seem, other electric codes) are not written with recording studios in mind. Most pieces of professional recording equipment have the "ground" wire from the 3rd pin on the power connector bonded to the 0 volt reference in the power supply as well as the chassis, and Neutral only connects to the primary side of the P/T. Most other electrical appliances and equipment are not designed and built this way, the "ground" wire is ONLY attached to the metal housing.
My main point is that (here at least) it is rare to be the only customer on the secondary side of the last transformer from the grid. This often results in "noise" on the AC power lines, and there are well established techniques and materials that can be employed to minimize this type of trouble.

 

[JohnRoberts]

That is the obvious trade off between GFCI outlets that only take down one outlet, and RCDs that take down an entire branch, BUT if it is preventing a lethal shock hazard that is a small price to pay. GFCI outlets can be daisy chained in series creating two protected outlets but again if one trips they both go.

My GFCIs are pretty stable but I've had a few false trips, mainly not coming back on after a power interruption. I have no first hand experience with RCDs but have heard complaints about false triggers in damp weather from buildings with dodgy wiring.

YMMV when in doubt I would err on the side of human safety.
======
Nobody is suggesting that we use cable shields as an EGC, wouldn't pass the UL 50A ground bond test.

JR

 

[Rob Flinn]

In a studio I would think noise performance of sensitive analog gear might benefit from EGC being connected to neutral at the panel. Otherwise, you could have common mode noise on the power lines relative to 0V and chassis.

If the studio is divided into studio and "house", then using RCBO for the house side with appliances and HVAC and sump pumps and so on might be a good arrangement. But for the studio side, my instinct would be to have EGC connected to neutral.

For me I think it would the other way round. I don't care if the lights tripping takes the power out for the fridge, but I do care if the desk tripping switches off the PT rig losing clients work.

 

[JohnRoberts]

For me I think it would the other way round. I don't care if the lights tripping takes the power out for the fridge, but I do care if the desk tripping switches off the PT rig losing clients work.

a good application for a UPS....

JR

 

[Rob Flinn]

a good application for a UPS....

JR

That too ! Ahem, maybe not the best example, but I would still rather be using the RCBO's for the studio.

 

[leadbreath]

You guys realise the "neutral" on the supply (where usually 11000V is transformed down to 400V/3 phase) side is the centre tap of the star winding taken to earth with a ground spike or "matt"....?

 

[abbey road d enfer]

If you go cheap you use one RCD (which provides the differential protection) feeding an MCB (which provides the over current protection) for each branch. This is cheaper, but if one branch trips the RCD then all the power goes off in the other branches.

This is common practice in France. Regulations asks for at least two RCBO's, one of type A ( oven, range, washing machine...) and another of type AC (for wall outlets, heating, lighting...). No more than 8 breakers can follow an RCBO.
On my panel I have 40 breakers, I should have at least 5 RCBO's; I have only 4.

Also from looking at the moamps page it does appear that in the U.K it is not a requirement to bond Neutral & Earth in the consumer box

Neither in France.

, however I was taught many years ago that they are bonded at the power station.

I believe it's so in all parts of the world.

 

[leadbreath]

Once again please before doing any electrical modifications in your house based on instincts or hunches, i give you....

 

[leadbreath]

however I was taught many years ago that they are bonded at the power station.

I believe it's so in all parts of the world.

In UK/SA its not at the power station but at your local supply transformer....

 

[Matt Syson]

For what it is worth the French regulations demand that the breakers cut both live and Neutral and all circuits must be protected by 30milliamp 'differential' types (RCD). This neatly takes me on to the 'fact' that this means the mains cabling (to gear) must therefore be current balanced, and that as 'interference' is likely to be magnetic rather than capacitance coupled into/out of cabling it makes 'balancing' transformers unnecessary for the mains. Neutral is taken to ground at the local power transformer and I think other places as well. My neighbours think I am a bit weird, looking at the mains distribution in the neighbourhood but hey, at least they are friendly.

 

[MisterCMRR]

Sorry I'm late to this "party" but yes, I see lots wrong in that diagram - based on US electrical code:
1. You can put a switch or breaker in the neutral line (as shown at "Main Switch" in your diagram)
2. The "RCBO" wiring is unclear! I see N and P coming from the tops, but where does N enter? The RCD is wired correctly, just as a GFCI would be wired in the US. It detects difference in current between the two circuits and, at a threshold difference (4 to 6 mA here) opens the circuit. Neither uses or needs a "ground" connection to work properly. I have no idea what the function of the RCBO is.
3. As many others have pointed out, there is no connection I can discern that ties N and ground together - anywhere. Under fault conditions, current will flow from ground/earth to N but it will critically depend on soil resistance. In the US, authorities have, properly IMHO, decided that soil cannot be part of fault current protection. Only a tie ("bond" is the term in US code) between N and third/grounding terminals of outlets and any metallic housings of wiring will guarantee fault currents high enough to rapidly trip a circuit breaker. The earth ground rod ("electrode") is added to this bond point for the main purpose of lightning current diversion and a secondary purpose of offering a low level protection from wiring faults external to the building (such as shorts or breaks in utility wiring caused by auto accidents or, in California, earthquakes)

Some years ago, John Woodgate (a colleague and standards expert in the UK) showed me a number of utility power grounding/earthing schemes. I was horrified at a couple of them because they depend on soil resistance to carry fault currents. Even if I lived in a salt-water swamp, I wouldn't trust some of them with MY life! Not meaning to insult Europeans, some of these schemes are apparently based on the "equi-potential earth" theory, which is demonstrably a myth!

Flame suit fitted ...

 

[MisterCMRR]

So earth is not connected to neutral in an RCBO breaker? That is odd. In that case fault current would go through the earth ground electrode, through soil to neutral at the utility pole. Probably not a good arrangement for a studio.

Update:

Actually it looks like any imbalance (Residual Current Breaker with Overload) will trip it. So as long as the earth ground electrode impedance isn't too high, EGC doesn't need to be connected to neutral for safe operation.

View attachment 92150

The arrangement is bad for any use of electricity, not just a studio. I'm still at a loss as to the function of the connection marked "FE". But at least the breaker only interrupts L, and not N.

 

[MisterCMRR]

Thanks, we seem to be confusing multiple topics...

Indeed GFCI/RCD protection devices do not need to be ground bonded to work, these devices simply compare the current flowing in the two active conductors (typically Line and Neutral) and trip open circuit if they don't null out. The GFCI per outlet protection uses a trip threshold of only 6 mA imbalance, RCD branch circuit protection requires more like 30mA imbalance. The RCBO is an RCD + conventional overload breaker making these suitable for use in mains breaker panels to replace old school overload only breakers.

These devices do not care where the missing current goes, only that it isn't flowing where it should be....

[TMI] Back several years ago when I was messing around with designing a premium talent protector I combined a GFCI outlet with a 3 pole relay so in addition to opening the line conductor, I opened all three circuits. My logic was that live performance talent shock hazards include external energized voltage sources exposing the talent to hazard conducting through a still present safety ground. In my premium protector I sensed current flow through the safety ground path and opened all three conductors at 6mA or more current in the ground.

It is clearly debatable wether a robust safety ground "could" ever be a shock hazard. I didn't even bother trying to convince UL, they like their safety ground bonds. It is not unusual for talent to get shocks from grounded microphones that provide a return path for external energized sources. [/TMI]

JR

Regarding safety ground being a hazard: I was an expert called in to help in a case in a Texas church where a preacher was electrocuted (that's death, not just a shock as some folks think the word is defined) as he was handed a microphone. He was standing in a "hot tub" what used a small electric water heater to warm the water. The sound equipment, including microphone, was all properly grounded. But the amateur electrician who had replaced the water heater a few weeks earlier had neglected to fasten the green (grounding) wire to the housing of the water heater. The heating element had developed a corrosion pin-hole which then allowed current flow into the water. This current would have been harmlessly routed back to neutral/safety ground had the wire been in place. Obviously, the church congregation was horrified as the preacher collapsed into the water (the current through his arm exceeded the 10 to 15 mA "let go" threshold current, so he couldn't just let go of the microphone. Which makes me wonder, how did Europeans come up with 25 to 30 mA for RCDs?

 

[neil.johnson]

The page that moamps linked to is hosted in Serbia. For any safety-critical information I would consult a qualified and NICEIC registered electrician in your locality, not some random website written by goodness-knows who, or what. Here in the UK the relevant documents are the IET Wiring regulations that leadbreath pointed to. And note that we are now at the 19th edition as standards change/improve over time.

Here in the UK there are several different earthing systems - in my 16th edition guide book there are three different earth systems: TT, TN-S, and TN-C-S.

And there are specific rules for different environments, and they change. A while back the UK introduced stricter electrical rules under the "Part P" title, so for example only qualified electricians are allowed to do ANY electrical work in wet areas (kitchens, bathrooms, toilets, etc) precisely because of the kinds of faults that Bill described.

For more information than you might care to know about RCDs try this document from BEAMA (British trade association for electrical manufacturers):
http://www.beama.org.uk/asset/7EE14AAB-81DB-4870-A9ACFBC5C58A94A2/
Neil

 

[Rob Flinn]

This is common practice in France. Regulations asks for at least two RCBO's, one of type A ( oven, range, washing machine...) and another of type AC (for wall outlets, heating, lighting...). No more than 8 breakers can follow an RCBO.
On my panel I have 40 breakers, I should have at least 5 RCBO's; I have only 4.

Neither in France.

I believe it's so in all parts of the world.

Thanks, that's useful to know some of the regs in France, since I am in the process of building a studio at my house in Aude. I keep asking my French girlfriend to ask/translate questions to her electrician relatives, but I think there's a lot getting lost in the translation.....

Do you use an isolation transformer or filter for the studio gear ?