PE advice by Napit (hot tubs)

How does the 1 ohm that align with the NOTE in 702.410.3.4.3, which says 20 ohms, and only a "mat or electrode' - not a grid?

With difficulty ?

That is a very odd note. Its position seems to imply it only relates to zone 2 - even though ADS is permitted in zones 0 and 1 in some circumstances. 

In my defence, I'd point out it doesn't say 20 ohms, but “suitably low resistance” and gives an example of “20 Ohms or less". If the intention is to hold protective conductors at no more than (say) 50V above true Earth, then Ohm's Law suggests that the current flowing to the Electrode can't be more than 2.5A … for a domestic installation, let alone one that includes an electrically heated hot-tub - that doesn't seem to me to be “suitable”.

The suggestion that the electrode need only be connected to local supplementary bonding is also interesting from a diverted N currents point of view … with perhaps 1.0mm2 c.p.c.s being pressed into acting as main bonding conductors as it were.

   - Andy.

 I totally agree.  It is mildly amusing that the “how best to earth things” debate has been lurching back and forth in various incarnations  for about the last hundred years or more. It is far from straight forward, with several layers of ‘gotcha’ to trap the unwary, especially when earthing domains and methods are mixed. Hot tubs and cars are the latest pinch points, but by no means the only ones.

 I sometimes wonder if the creators of some ‘authoritative texts’ have it clear in their heads before setting pen to paper, they certainly do not always succeed in the stated aim of getting the concepts out of the author's head and into the reader's without error or misunderstanding.

Mike.

mapj1: 
 

Very low electrode resistances are not always good, nor is it always safe to omit the RCD and rely on a fuse or MCB  if you have them.

Mike

 

And vice-versa … in a TT system if you take the simple path for the Zs loop through the main consumer earth electrode resistance Ra (or in a TN system, the supplier's earthing terminal only) as the “worst case” for automatic disconnection, it's not always safe to assume it's the “worst case” for prospective earth fault current once supplementary electrodes are connected. Yet in TT systems this is often ignored “because the RCD does the job” in an earth fault …

Perhaps it is referring to grids, but then a figure of 20 ohms is a bit arbitrary and irrelevant, and for step voltages arguably a higher electrode resistance  is better as there is less gradient as there is less current flow. The floor above the grid is then closer to CPC voltage. It might be better if the intention was spelt out.

In extremis you could create a very good zero step voltage zone by laying down a sheet of polyethylene with aluminum foil over it, and connect that to the CPC. As an electrode connecting to terra-firma, it would be less use than the proverbial chocolate teapot, but in terms of keeping both feet at the same voltage as the supply CPC, about as good as it gets.  Until you step off it of course, for that you need to cut the plastic sheet bigger than the foil by a metre or two extra  all round.

There is another issue, not unique to PME if your local electrodes are too good, if the ones that the DNO use are comparable or higher resistance than yours. Unlikely in a city centre with many electrodes on the network, but a concern for those of us fed by pole transformer  and overhead singles to a small group of houses.

If a fault grounds the live conductor to terra-firma very well,  say with an electrode resistance of a few ohms, but the DNO neutral/cpc resistance to terra-firma earth is higher, maybe between 10 ohms and that upper limit 20 ohm figure, then 10-20A odd may flow from live to fault through earth, and back up the DNO electrodes to the transformer neutral. But this means the total 230V voltage drop is split across the two sets of electrode resistance - with as much or more across the DNOs than the users.  Now the DNO neutral and any CPC derived from it is at a dangerous voltage to terra-firma. If the fault to ground is  on a high current non RCD circuit maybe it would not be spotted for a while.

Very low electrode resistances are not always good, nor is it always safe to omit the RCD and rely on a fuse or MCB  if you have them.

Mike

mapj1: 
 

If one ohm is reasonable  rather depends on the other impedances in the rest of the PME system fed from the same substation. 20 ohms may be a book figure, but to expect that to do much equalization against all but the weediest PME supply is optimistic in the extreme.

I agree … but I think the trick is that G12/4 is talking about a grid under the “poolside areas” here, to control step potential, rather than controlling a touch voltage. In that case, the actual resistance to earth matters less … I'm guessing the 20 ohm figure in the note in Section 702 dropped out of what's considered reasonable for PME as an “end of line” electrode in line with current industry practices …

well I would of course, but then I am quite happy to justify my exceptions.

However, not being a member of any approved scheme, for my own domestic stuff I occasionally find myself trying to convince a third party or a building control chap that something I have done is kosher. Resistance to anything that is not like ‘normal’ is remarkable.

However, having worked for a little while in a country where 3 phases at 32A each was the norm for a house, and 16A/ phase for a small flat or bedsit, I must say I appreciate the advantages. water heaters in 3 cores of 2.5mm, thinner sub-mains lower PSSC, negligible neutral to terra-firma offset voltages, what is not to like. Oh yes, the bloody awful plugs, especially the older Russian ones. Still with lights and power on combined 16A radials the fuseboards are simple, and the no of fuse sizes or breaker ratings used is negligible, well one size really, 16A C type…

Certainly until recently plastic 3 phase consumer units existed. No use in BS7671 land, but the rest of the planet seems OK with the idea. Needs a 3 part bus bar under the breaker and separate neutral bar.

Mike.

mapj1: 
… but to suggest 3 phase in a normal UK house  that would bring out most domestic electricians and some DNO employees out in a nervous rash.

Why would you not put it in, especially with a new connection? The extra cost is very little.

But it's human nature to be afraid of, and to avoid things which you do not know about. I have failed to convince family members to get it.

The downside is the bulk of TP DBs and AFAIK, there are no “domestic” ones in the UK market. They are also much more expensive than SP ones. TP RCBOs are eye-wateringly expensive and there is no option of having a “split load" board.

If one ohm is reasonable  rather depends on the other impedances in the rest of the PME system fed from the same substation. 20 ohms may be a book figure, but to expect that to do much equalization against all but the weediest PME supply is optimistic in the extreme.

The supply to this  hot tub is not like a 5A cut out feeding one streetlamp. The load of the hot tub heaters will be a few thousand watts, so in the range of 15-30 ohms itself. If in lost neutral fault we found just the hot tub, and not even the rest of the house, trying to return to ground via a 20 ohm electrode, all the exposed CPCs for the house would be well over the 50V safe limit for dry conditions. A hot tub user is unlikely to be to dry. 

To be safe, rather than merely regs compliant, while wet and with one foot earthed ,you may prefer 25 volts at a likely diverted neutral current of  perhaps 15- 20A.  Not so far off the one ohm figure really.

It would be far safer if the supply to the hot tub was 3 phase of course, as then the neutral current would only be the imbalance of the phases, but to suggest 3 phase in a normal UK house  that would bring out most domestic electricians and some DNO employees out in a nervous rash.

Mike.

AJJewsbury: 
 

I might be talking out of my hat here, but why not combine both systems? Why not export PME via the swa and 3rd core, tie both together at both ends, the tie both to a rod at the destination end too? Belt & braces?

That would still suffer all the disadvantages of PME earthing. Unless the rod had a ridiculously low resistance to Earth (<< 1 Ohm) you'd still risk a hazardous potential on the earthing system during an open PEN event and the rise in potential of the soil around the rod would likely be far too steep to provide anything like equipotentiality in the vicinity of the tub. A buried grid under the entire area would probably be the minimum if you were going for that sort of approach.

   - Andy.

Andy,

How does the 1 ohm that align with the NOTE in 702.410.3.4.3, which says 20 ohms, and only a "mat or electrode' - not a grid? I do accept that there's a difference between BS 7671 and G12/4 in that the latter discussed a buried grid under zone 2 (implying exactly what you're saying), but that does not give earth electrode resistances … so would I still refer back to BS 7671 for that?

Would you be happy with this solution for a hot tub?