Fixing TT earthing problems and gas bonding for an EV charger

[ericmark]

No bonding at all would be a C2

So any TT installation is a C2? I do remember my father-in-law getting a contractor to rip up and re-lay the floor in an operating theatre, as the wrong glue used, and the floor's bonding was compromised. Seems it needed a glue which could conduct to stop static build up.

Nearly every other electrical item used in one's garden is class II, it seems crazy why EV's are not class II, It seems all the rules are being rewritten in the push to have electric cars, anything else, caravans, boats etc, TN-C-S supplies are banned with good reason. And an AC supply has to be designed, so one can never get more than 50 volts to true ground. Instead of making them safe, they just change the rules.

 

[JohnW2]

So any TT installation is a C2?

It's fairly easy to argue that a TT installations does not need any main bonding, since it is almost always the case that the potential 'introduced' by an extraneous-c-p will be ('true') earth potential, which should be about the same potential as the earth electrode (hence MET) in a TT installation.

However, strictly speaking there is (per definition of extraneous-c-p), at least theoretically, a very slight chance that, in extremely unusual circumstances, an extraneous-c-p could introduce a potential other than (higher than) earth potential - so I suppose one can argue that main bonding is required?

And an AC supply has to be designed, so one can never get more than 50 volts to true ground. Instead of making them safe, they just change the rules.

One can't deny that a 50V maximum for 'touch voltage' is 'better than nothing' but, particularly for 'wet people', 50V can be plenty enough to result in a fatal shock.

 

[pete01]

So any TT installation is a C2?

I would bond extraneous parts on a TT system.

 

[JohnW2]

I would bond extraneous parts on a TT system.

I imagine that we all would (I certainly do in my TT installation) and it's certainly required for compliance with BS7671 (which makes no exception of TT installations).

 

[ericmark]

Sorry I intended to say IT. Yes need bonding with a TT.

 

[JohnW2]

Sorry I intended to say IT. Yes need bonding with a TT.

As I wrote, bonding is seemingly almost never going to be of any value in a TT installation, but it is 'required'.

 

[StuB]

In fairness very, very few existing DBs would be suitable, and they would also be unlikely to stock all different types/makes of protective device. Many brands don't have double pole switching RCBOs in a single module, which will rule many boards out.

As I said near the start of this thread, I do understand some of their reasons for trying to standardise their installs particularly regarding using an existing CU.

 

[plugwash]

As I wrote, bonding is seemingly almost never going to be of any value in a TT installation,

In a TT installation, bonding reduces the touch voltage in the time between when a fault occours and when a RCD trips (which for distribution circuits could be several seconds).

 

[plugwash]

Sorry I intended to say IT.

A problem with a plain IT installation with no earthing/bonding is that the "first fault is free" but it's also silent. The second fault can be shocking.

 

[JohnW2]

In a TT installation, bonding reduces the touch voltage in the time between when a fault occours and when a RCD trips (which for distribution circuits could be several seconds).

Do you mean "by reducing the Ze of the earth electrode" (by effectively adding a further path to earth in parallel with the actual earth electrode)?

If so, then I agree with what you say, but I would not call it "bonding" (since it is not there as "Protective equipotential bonding", and is not there to create an equipotential zone), since it really is simply an 'improvement' in the earth electrode.

 

[plugwash]

Do you mean "by reducing the Ze of the earth electrode" (by effectively adding a further path to earth in parallel with the actual earth electrode)?

No, I mean by reducing (ideally eliminating, but bonding conductors have resistance) the potential between different items within the "equipotential zone" during the fault.

 

[JohnW2]

No, I mean by reducing (ideally eliminating, but bonding conductors have resistance) the potential between different items within the "equipotential zone" during the fault.

I suppose so, but as far as I can see, only really significant IF a bonded extraneous-c-p had an impedance to earth which was very small in comparison with that of the TT electrode.

If I've got my nocturnal mental arithmetic right, if the extraneous-c-p had, say, a 1Ω path to true earth (and the TT electrode 'much greater' than that), then the pd across a 30m long 10mm² bonding conductor would be about 15V - so that would be the greatest possible 'potential difference between items' within the 'equipotential zone'

Edit: On reflection (when properly awake!) my comments above (stimulated by your comment about 'bonding conductors having resistance') relate to the situation in which there IS bonding present. As you've been saying, in the absence of bonding in a TT installation, the pd between two items (e.g. an exposed-c-p and something electrically connected to an extraneous-c-p) (i.e. the 'touch voltage') could be very high, possibly approaching the supply potential, for the duration of a fault (until cleared) -so, yes, bonding definitely is desirable (as well as 'required') in a TT installation.

 

[ebee]

As I wrote, bonding is seemingly almost never going to be of any value in a TT installation, but it is 'required'.

Well I can agree with "Almost Never" but I would never say "Never" - that chap who said to me "An Earth is an earth is an earth" and I replied no it is not, equalising earths is important and the more we equalise our earths in the location we are using then the "safer" we might be, whatever system our earth refence is, the ideal zero is impossible to achieve but the nearest to zero then all the better.

 

[JohnW2]

Well I can agree with "Almost Never" but I would never say "Never" ....

Nor would I - which is why, as you correctly quoted, I wrote "Almost Never"

One of the very first things that was drummed into me during my higher/'professional' education was that, in the real world, any statement including superlative words such as "always", "never", "all", "none", "everyone", "no-one" etc. etc. etc. were nearly always going to be incorrect!

- that chap who said to me "An Earth is an earth is an earth" and I replied no it is not, equalising earths is important and the more we equalise our earths in the location we are using then the "safer" we might be, whatever system our earth refence is, the ideal zero is impossible to achieve but the nearest to zero then all the better.

It's not really possible to argue with that, but one does have to (should) at least consider what is, and is not, 'significant' - as we often discuss, attempting to minimise 'risks' which are already incredibly small may not be 'sensible'.

However, just to be clear, I have in no way been suggesting that one should even think of omitting main bonding in a TT installation.

Kind Regards, John

 

[SUNRAY]

I suppose so, but as far as I can see, only really significant IF a bonded extraneous-c-p had an impedance to earth which was very small in comparison with that of the TT electrode.

If I've got my nocturnal mental arithmetic right, if the extraneous-c-p had, say, a 1Ω path to true earth (and the TT electrode 'much greater' than that), then the pd across a 30m long 10mm² bonding conductor would be about 15V - so that would be the greatest possible 'potential difference between items' within the 'equipotential zone'

Whoa there, if I have this right and eaually my nocturnal 3 r's may be letting me down

Assuming a TT earth of 15Ω and a similar extraneous 15Ω on something metal and while being in contact with both earths there is a direct 230V (protected by a 16A OCPD) earth fault:
.

What voltage appears across Sunray until the RCD does it's job?

EDIT: Reading this again now I realise your were referring to a the 2 earths being bonded.