Give it 10 years before it happens:-)
Or are we looking at the thin edge of the wedge where all 17th edition installations will have to be tested by law every ten years to check the RCDs work correctly? Possibly after someone has died due to a failed RCD?
I do a swimming pool EICR every year and have done so for the last 5 years.
Two RCBO's have gone down in that time (they were installed 10 years ago).
My house was wired (presumably not by a professional) with 10mm^2 green/yellow wire parallelling all the T&E circuits from the CU branching repeatedly with connection to all the pipework in every room. Since it is all surface wiring (for structural reasons) in boxes and trunking of various kinds I tend to strip out great reams of it every time I redecorate or alter anything. As all circuits (except the economy 7 one) are fast 30mA RCD protected it is hard to see what the objective was, but it seems to imply a confusion between bonding and circuit protection. Most of the rooms have no locally unearthed metalwork in them apart from the central heating, which is all copper anyway. For reasons which seem arbitrary, even earthed electrical fittings (such as outside lights) seem often to have a parallel 10mm^2 earth connection back to the CU.
I have another POV on this. I can't quote chapter and verse as I do not have my book to hand and my memory is s**te.
However, the 17th requires, IIRC, that any extraneous conductive parts aka water pipes entering a bathroom/etc to be bonded to the main earth as one of the conditions of not having supplementary bonding? g So assuming the bonding is good and equally so across all such pipes entering the bathroom, even without an RCD, the situation is hardly worse than if supplementary bonding were present.
As far as I can see, all supplementary bonding does is keep it all local, reducing the chance that it gets bypassed (eg plastic pipe sections) or a local fault exists in the CPC to say the shaver socket.
So SB makes for perhaps a more robust solution and one that *may* be easier to verify[1] but 17th style if done correctly ought to be as good, even if the RCD fails.
Thoughts?
[1] In reality, SB wires disappear under the floor and are often clamped in inaccessible places like under the bath so how easy is it to verify these, as opposed to just doing a loop test on all extraneous conductive parts relative to the MET?
Yup the main bonding must be in place - so incoming services, and extraneous metalwork that can distribute a potential around the building (typically CH pipework)
It keeps it local, and also electrically short - so more tightly controlled voltage differentials.
Although the loop test does not find the metal that does enter the EQ zone, and could be made live by a fault, but is not bonded at entry (say a partial metal pipe install where the incoming main and initial pipework is plastic, and hence not included in the main bonding)
I need to dig out the big red book (I know..) and find the exact wording.
But from memory, doesn't the 17th say: plumbing that is extraneous must have continuity back to the main bond, or something along those lines.
In other words, if you have pipes entering the bathroom which do not have continuity to the MET, then you cannot meet the conditions for not having SB.
The one thing I am not sure of is what "continuity" means here, and I really do need to check.
It could mean "sufficient to trip an RCD" or it could be "sufficient to disconnect a fuse/breaker in 0.4s".
I have a 17th design, but I've got hot, cold and eventually all CH main feeds cross bonded to each other in the hall cupboard as good measure - mostly to ensure hot and cold pipes and radiator (when fitted) have good continuity.
Sufficient to trip a 30mA RCD.
Which of course may might when then the electrician did the work but may fail when some pipework is altered (and that pipework need not be in the bathroom).
Local supplementary bonding removes that risk.
Perhaps someone read this and got confused.
Thanks Adam.
I agree that it's more foolproof in general (with the caveats that it's not always easy to inspect once things are panelled in).
I'm happy with my house being sans SB as it was designed with that in mind from the get go, but I do have cross bonding straps in the cupboard where the pipes disappear off to the bathroom - mostly to ensure continuity remains between hot, cold and heating pipes (when they exist) and are not reliant on the boiler manifold (when it exists) to achieve this.
I also test loop impedance is sufficient to trip the relevant MCBs in
0.4sec - I am surprised that was not written into the regs as it would have reduced the reliance on the RCD element.That's more identifying a likely scenario where things might get buggered up (eg boiler removed or replaced) and dealing with that rather than any book standard approach.
I was keen to avoid SB as I have all my pipes on show - but if that weren't a consideration, I'd have put it in conventionally.
In fact I will possibly build SB into the shower room as the pipes are mostly bunched in the underbasin cupboard.
That's a thought, but seeing it was an ordinary house and done about thirty years ago I suspect it was just a misunderstanding of supplementary bonding. It is a PME installation.
It's worth noting (from the Red Book, I've mislaid my green):
701.415.2" ... Supplementary equipotential bonding may be installed outside or inside a room containing a bath or a shower, preferably close to the point of entry of extraneous conductive parts into such rooms."
(I'd be grateful to know if that wording remains in the latest 17th?)
...So you don't have to make a mess of a tiled room and have clamps on show everywhere to meet SB.
I've been thinking about your argument Adam, and it is persuasive. And I would not be surprised either if they revise it - maybe in 10 years when a couple of people who never test their RCDs have got a belt.
I am going to build SB into my shower room (we're laying the floor screed for that in 2 weeks) - that's really trivial as the shaver socket and pipes all pass through to or back onto the lobby cupboard so the clamps can go there.
I am tempted, at some point to add SB back to the main bathroom at some point. The 17th exemption seemed like a good idea at the time, but...
Only problem there is the SB runs are tortuous. Pipes exit under the stairs 1m from bathroom - that's OK, convenient and fairly obvious clamping point. But to get to shaver socket and bit of pipe to high level loo cistern (if that's in the zones, have to check) require going up through the ceiling and some 5-7 metres around. It would work and I can see nothing in the regs that precludes this - it just won't be terribly obvious what's going on to anyone else.
What's your opinion on that (long SB wire routing)?
Cheers,
Tim
It's possible to work out a maximum touch voltage between two metallic objects (one at 230V and one at 0V) bonded by a 4 mm earth cable.
However I have been in the pub for 6 hours (mt apologies to Wakefield) and I'll pass on trying that calculation today.
Thanks Adam.
If that's all it comes down to, fair enough.
However, the potential difference would also depend on what's connecting the Ov and 230V sides. 4mm2 is not much compared to the area of copper in a 15mm water pipe which is about 32mm2 for 0.7mm walled pipe.
OTOH 4mm2 is vastly larger than the 1 - 1.5mm2 CPC in the lighting circuit which also has a long run to origin.
Less true with a shower though that might have a 4mm2 CPC and larger live conductors.
So how is the theory of supplementary bonding handled? Presumably some allowance must be made for ADS being operational?
You keep the touch voltages between two objects to less than 50V.
If the ADS is still working in 10 years time (ie the RCDs) then fine.
The maximum potential difference with EEBADs is down to the resistance of the 4mm cable between the two points connected by that cable.
And yes I would expect an MCB or fuse to trip before this sort of thing happened.
No matter what you design for some pillock will still manage to mess it up.
PSCC of domstic wiring is anything upto 6kA.
4mm2 voltage drop is 5.5mV/A/m or 5.5v/kA/m You can calculate from that, adding in the R of the supplying lighting circuit.NT
Mine is >12kA.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.