Bonding

You can introduce a (small) risk of shock that could otherwise be avoided.

In general if someone for whatever reason ends up in contact with a hazardous live part (in theory shouldn't happen, but in the real world it can - e.g. from a damaged flex or faulty Class I appliance where ADS has failed for some reason (e.g. broken c.p.c. or failed RCD on a TT system)) then having an earthed part to hand means the circuit can be completed and they receive a shock - whereas if the part had been left floating, they'd feel nothing (like a bird perched on a bare overhead line).

For parts outside of the equipotential zone there's the additional risk of having two different "Earth" potentials to hand (especially nasty when PME earthing is used (during an open-PEN event), or a TT system fails to disconnect on an Earth fault when it should).

The risks are generally small, certainly smaller than not bonding when it is required, but if it's easy and cheap to avoid (and not bonding something is usually pretty cheap and easy) most people would welcome the additional safety, however small the benefit.

   - Andy.

May I add a supplementary question please?

If yer bonding is unnecessary because e.g. the water comes in a plastic pipe, but complete (including the kitchen sink and two pairs of taps), what are the risks of removing it.

My thinking is that it should be all (15th Ed) or nothing.

Chris Pearson said:

My thinking is that it should be all (15th Ed) or nothing.

This is a very interesting discussion.

A good example within buildings related to the 15th Ed thinking, is that of the metal window frame (or similar), which could be accessed from either within, or outside, the building. Within the building, perhaps no issue ... outside the building, window cleaner with ladder at Earth potential (with capital E ... i.e. the 'general mass of the earth') and window frame at fault-level touch voltage, the window cleaner is in real danger of falling off their metal ladder !

In 17th Ed, main bonding is only applied within buildings ... but even then, one person's building is another's mansion.

You're not telling me that the bonding in Heathrow Terminal 5 necessarily makes the potential at part of the installation near (not necessarily at) one of the top corners of the structure the same as that at the diametrically opposite corner in the basement ?

Conversely, in many domestic installations where protection by RCD or RCBO is provided for every circuit, the resistance between any two exposed-conductive-parts, via cpc's and MET, is likely to meet the requirements for additional protection by supplementary equipotential bonding, making arguments over disconnection times completely irrelevant regardless of whether the .earthing arrangement is TN-C-S, TN-S or TT.

gkenyon said:

Conversely, in many domestic installations where protection by RCD or RCBO is provided for every circuit, the resistance between any two exposed-conductive-parts, via cpc's and MET, is likely to meet the requirements for additional protection by supplementary equipotential bonding, making arguments over disconnection times completely irrelevant regardless of whether the .earthing arrangement is TN-C-S, TN-S or TT.

Graham, you have confused me. :-(

Suppose that you have a 15th Ed installation with main and supplementary bonding as it should have been and you change your DB for a no-expenses-spared 18th Ed one with all RCBOs, have you now got belt and braces?

My question is whether removing some of the bonding could be worse than removing all of it. Andy suggests that it might be.

So my example is a kitchen which was planted on the back of the house when it was extended 100 years ago. In the course of 21st century improvements, the obsolete supplementary bonding becomes no longer connected to the MET because it is in the way, but the cables between the double sink and all four of its taps, the gas cooker, and the gas boiler remain because they are not. The main bond to the metal gas supply remains.

Is the risk now that the sink at earth potential, as opposed to floating, is a greater risk in the event that e.g. the toaster develops a fault which is not cleared by ADS.

I might add that a lot of this seems very theoretical and the risks of being killed or maimed on the roads must still be much higher.

the obsolete supplementary bonding becomes no longer connected to the MET because it is in the way, but the cables between the double sink and all four of its taps, the gas cooker, and the gas boiler remain because they are not. The main bond to the metal gas supply remains.

My turn to be confused! If the remaining bonding includes the boiler (presumably class I) and the gas pipework (presumably continuous metallic within the building) then there's still a connection back to the MET?

To my mind:

Worst - metal part can introduce a potential and isn't connected to the building's Earthing system (unacceptable)

Better - such a part is bonded to  the building's Earthing system (acceptable)

Best - part can't introduce a potential - i.e. it has such a high resistance to anything else that touching it and a live part won't result in enough current to flow to cause a shock. (preferable - but only an option where the part isn't a true extraneous-conductive-part).

   - Andy.