cross bonding

[westie101]

Right or wrong

I would say that if there is the possibility that the pipe and what it connects to could be at earth potential, that in the event of an internal or external fault other metal work could become live at mains voltage and both sets of metal could be touched - then yes it should be done

 

[JohnW2]

Interesting. If there were multiple (electrically separate) bits of wholly-internal metal pipes in amongst the plastic ones (not unusual when people start gradually replacing copper with plastic), would you main bond each and every one of them?

Yes.

Again, interesting - but how often is that really done in practice? .....

....But.
What is the tipping point at which you do not need to bond?
For example I was doing some work in a house recently that had no main protective bonding to either gas or water. No problem bonding the gas but the incoming plastic water supply went to the brass stop cock and then to plastic inside the house. However there was a single piece of 10m of copper pipe that connected in between the plastic incoming pipes and the plastic pipes to the sinks, baths etc.
I bonded the 10m with 10mm earth of copper pipe. Right or wrong?

That's really a very simple example of the question I was asking, to which you have answered 'Yes' above. I've seen a good few houses in which the originally all-copper pipework is gradually being changed to plastic, as repairs/modifications/additions are effected and/or when there happens to be access to pipework. As a result, there are often many lengths of copper pipe, some of them very short, attached to plastic at both ends. One sometimes sees bonding put across short plastic interruptions in copper pipes, but I don't think I've ever seen each of the 'isolated' bits of copper 'main bonded' back to the MET with ≥10mm² cable. Is this something you often see, or even do?

Kind Regards, John.

 

[riveralt]

That's really a very simple example of the question I was asking, to which you have answered 'Yes' above. I've seen a good few houses in which the originally all-copper pipework is gradually being changed to plastic, as repairs/modifications/additions are effected and/or when there happens to be access to pipework. As a result, there are often many lengths of copper pipe, some of them very short, attached to plastic at both ends. One sometimes sees bonding put across short plastic interruptions in copper pipes, but I don't think I've ever seen each of the 'isolated' bits of copper 'main bonded' back to the MET with ≥10mm² cable. Is this something you often see, or even do?

Which is why I raised

Yes.
But.
What is the tipping point at which you do not need to bond?

I think westie101 hit the nail on the head in his description

I would say that if there is the possibility that the pipe and what it connects to could be at earth potential, that in the event of an internal or external fault other metal work could become live at mains voltage and both sets of metal could be touched - then yes it should be done

I haven't personally seen what you describe - But I have seen the token (but absolutely useless) 500mm of copper pipe at the water inlet on the customer side bonded to the MET - after which everything else is plastic.

 

[JohnW2]

The point of all this is to take a step back and understand the basic principles of why bonding is required (whatever name you wish to put to it), it's not about blindly following a book but working out the implications of fitting it or not.
With PME, as I keep commenting, it is all about the Faraday Cage.
Understand how that works and you will "get it"

I really think that I do understand the principles, and that is, indeed, one of the things which causes me some problems.

As you've probably seen, I am far keener on bonding exposed metalwork together than many of the people here (who will often say that it's not necessary). However,as far as I am concerned, what interests me is my 'local Faraday Cage' which essentially relates to two or more pieces of metalwork close enough for me to touch two or more of them simultaneously. I create that local cage by bonding every bit of metalwork together. What I struggle to understand is why the creation of that cage is in any way helped or enhanced by not only bonding the metalwork together, but also bonding it to the MET (i.e., with a TN supply, the DNO's provided 'earth').

I thought that, again with TN systems, the whole idea of BS7671-style 'main bonding' to incoming services was to ensure that the house could not simultaneously contain a true earth potential and the DNO's earth' (which, under certain fault conditions could be way above true earth). If no metal service pipes enter the building from outside, there does not seem to be such a risk to protect against.

Furthermore, as I frequently say, where I really do struggle is to understand why there is a requirement for 'main bonding' with a TT system (i.e. a piece of 10mm² G/Y joining together what are effectively two earth electrodes). The chances of a service pipe introducing a potential appreciably different from the potential of the earth electrode must be vanishingly small. Again, I would have thought that 'local Faraday Cages' were all one needs for personal safety.

Kind Regards, John.

 

[RF Lighting]

If you had a live to earth fault on a TT installation, all earthed metalwork will rise in potential until automatic disconnection happens.

If the water pipe is bonded to the MET, it will be at the same potential as the earth electrode, and there's your Faraday cage safety coming into play again.

If the water pipe isn't bonded to the MET, it may be at a different potential to the earthed metal work carrying the fault current.

You do not normally need to use 10.0mm² cable for bonding on a TT installation. 2.5mm² is usually fine if it is protected from mechanical damage.

 

[Adam_151]

The importance of equipptenial bonding on a TT system is greater than that on a TN system due to the higher touch volatages during faults.

A fault on a TN system might present a touch volatage between exposed metalwork and the ground of maybe 150V or so (half supply volatage, and then some, to take account of reduced CSA in 6242Y), with TT as the earth side is much higher in impedance than the line side, you might see almost fully supply volatage between earthed items and the ground outside of the resistace area of the electrode (roughly a circle of radius 2x rod length around it). Thefore its most critical to bond all services together.

To put it another way,yes both bits of metal are in the ground, but the resistance of an electrode comes from the soil immediatly around it rather than the planet has a whole, so the argument only holds if they are physically next to each other

A.

 

[westie101]

OK lets try this
my house is fairly modern, has a PME supply, the incoming gas pipe is plastic, the incoming water pipe is plastic so neither can be said to be at earth potential.
The DNO (my employer's) earth is connected to the internal gas piping which is metallic and to the internal water piping which is copper. Other bonding ensures that all internal pipework is connected to this "system"

Whilst I have not tested it (could be interesting) if that bonding was not in place most of it will be at earth potential as it is "embedded" within the building!
On the ground floor all the gas is buried in concrete, all the central heating, so hence the water, in the concrete so it must be bonded to the DNO earth as it is touchable and there is a risk of a potential difference.

Now to isolated bits of pipe, there can be no general rule, if the bit of pipe is inaccessible and does not connect (direct metal to metal) to anything that is there may be a case for not bonding to it.

Through all this we are not seeking to find an earth for the system which is an important distinction to make

 

[JohnW2]

I think westie101 hit the nail on the head in his description

I would say that if there is the possibility that the pipe and what it connects to could be at earth potential, that in the event of an internal or external fault other metal work could become live at mains voltage and both sets of metal could be touched - then yes it should be done

I essentially agree with what westie said, but:

• 1...as I've just written to him, that hazard can be addressed by local bonding ('the local Faraday Cage') and I don't see why it needs to be 'main bonded' back to the MET.
  2...We're talking about a water supply which enters the property in plastic, so I don't think westie's 'if' condition would be satisfied. There may, of course, be a metal gas supply pipe (which could introduce an earth potential and transfer that to the internal metal water pipework) - but that's a different matter, and a reason why main bonding of the gas (not water) pipe would then be required.

Kind Regards, John.

 

[RF Lighting]

Trouble is, most people dont understand what bonding is and why it's installed.

Like you've just mentioned, it's not just the incomming services that need bonding and thats it, sometimes the services may not need to be bonded, but other thinks will like your gas pipe buried in concrete will.

 

[EFLImpudence]

it is all about the Faraday Cage.
Understand how that works and you will "get it"

What would you do to a totally isolated piece of metal inside the Faraday Cage assuming you also were in there?

 

[RF Lighting]

SFA

 

[westie101]

I see the point as: -
We all agree the need to bond this bit of pipe and, say, a sink to the DNO earth; what seems to be the problem is what "type" of bond it is, to fit within the definitions in the regulations regulations.

That being the case I take the view that it is whatever type of bond the installation's designer decides it should be. If you are designing an installation it is your decision as to how you follow the regulations, not a books decision! As long as you can show how and why you comply there is not an issue.

 

[westie101]

What would you do to a totally isolated piece of metal inside the Faraday Cage assuming you also were in there?

Asses it, decide if there will be a risk of a dangerous potential difference between it and the cage and decide to connect it to the cage or not, to ensure safety.
I wouldn't start trying to look it up in a book!

 

[JohnW2]

The importance of equipptenial bonding on a TT system is greater than that on a TN system due to the higher touch volatages during faults.

Thankn you. That makes total sense, and I'm surprised that no-one has said it to me before - and now you and RF Lighting do so more-or-less simultaneously!

I need to think this through; there may be supplementary questions!

Kind Regards, John.

 

[JohnW2]

What would you do to a totally isolated piece of metal inside the Faraday Cage assuming you also were in there?

Asses it, decide if there will be a risk of a dangerous potential difference between it and the cage and decide to connect it to the cage or not, to ensure safety.

Is assessment actually necessary? If it's isolated and entirely within the cage, then (unless it contained a power source!) it is surely impossible that it could aquire a potential difference relative to the cage?

Kind Regards, John.