Hot tub installation

I prefer to use a TT earthed supply with a separate earth rod positioned well away from the influence of any buried pipes etc. for outdoor hot tubs. The rod  positioned away from the hot tub area and covered by a plastic earth box. Buried pipes may be bonded to the house P.M.E. earth terminal, so the rod needs to be well away from them.  You may need a 30mA Type “A” R.C.D. if the hot tub has any electronics that may blind an ordinary Type AC R.C.D. 

Z.

The concern folk have is the possibility of an offset voltage between the terra-firma earth (wet feet) and the PME earth (the suppliers) . This is similar to the car charger problem, with the additional complexity of a wet and lightly clad victim,  and as per the previous reply, the ideal solution is the same, a TT island well away from the rest of the electrical system. 

How serious this really is depends on a number of things, not least if the user can credibly make good contact to both at once.  There will be situations, arguably most,  (hot tub on raised decking or other insulating surface come to mind) where the PME earth would be fine - and remember there are loads of PME earthed lamp-posts, bus stops etc that folk touch with impunity, as shoes and paint conspire to ensure any contact is not a dangerously low resistance.

A badly done TT where the zones are not properly separated or the ADS is not right, (pre RCD live in a metal container relying on the TT earth for example) is probably worse than using the  PME, if it actually increases the risk of shock in fault or of somebody bridging two earthing zones by bringing them close  TT earthing of a metal box on the wall beside some outside light or other thing supplied by the PME earth for example.

I suspect that a great many domestic hot tubs do in fact use the house earth, without incident.

Mike.

PS I also disagree with the advice in the PE article above that implies an earth rod of 20ohms or so in parallel the suppliers earth  is going to have any significant effect on all but the weediest of PME supplies in the event of a lost PEN.

More likely the rod, and a carrot shape of earth around it gets pulled up to the fault voltage increasing the step-voltage shock risk if the top of the rod is accessible.

Isn’t there a value of 1667ohms if it’s protected by a 30ma RCD

That's for ADS (the figure actually comes from ensuring that exposed-conductive-parts can't exceed 50V above true earth when the earth fault current/earth leakage is just too low to ensure the RCD will trip). But that's then subject to making sure that the TT electrode has sufficient contact with the lower layers of earth to ensure it's reliable over the seasons - which is often deemed to be reflected in a figure of less than 200 Ohms (or some prefer 100 Ohms).

But none of that makes a blind bit of difference if your means of earthing itself is at a raised voltage (either due to voltage drop along the PEN conductor in the case of ‘perceived electric shock’, or a close to full mains voltage in the case of a broken PEN) - since the RCD won't disconnect the c.p.c. even if it did trip.

I agree with Mike that 20 Ohms is a bit of a nonsense. I guess it's been copied from the practice of street lighting where max 20 Ohm electrodes are used to protect (to some extent) from broken PEN conductors - but that only work due to the very limited load currents (<2.5A to keep things below 50V) - which is fine for a street lamp or two, but hardly realistic for an entire domestic installation, especially one including an electrically heated hot-tub.

A buried grid - to force the ground in the vicinity to the same voltage as the local earthing system makes a lot more sense (in that case the resistance to the general mass of the earth is of much less significance), but it would have to go under the entire area. And there are still risks around the edges of the area protected in that way.

Life might have been so much simpler if the DNOs had stuck with TN-S.

   - Andy.

AJJewsbury said:

Life might have been so much simpler if the DNOs had stuck with TN-S.

Each of the earthing arrangements has its issues.

If we have a TN-S installation with no extraneous-conductive-parts (or extraneous-conductive-parts that don't have an extremely low fortuitous earth electrode resistance), and the supply protective conductor becomes damaged, one can still be left with hazardous touch voltage/touch current at exposed-conductive-parts due to the summation of protective conductor currents from appliance noise filters.

Yes, with respect to wet barefooted persons, diverted neutral currents in PME systems can cause "tingles" in normal operation, and I agree this isn't ideal ... but TN-S isn't without its problems (nor is TT).

If we have a TN-S installation with no extraneous-conductive-parts (or extraneous-conductive-parts that don't have an extremely low fortuitous earth electrode resistance), and the supply protective conductor becomes damaged, one can still be left with hazardous touch voltage/touch current at exposed-conductive-parts due to the summation of protective conductor currents from appliance noise filters.

Agreed - but dealing with a few tens or even a few hundred milliamps of leakage current seems relatively practical with say an additional electrode - much more so than the (dropped) proposal to do similar for PME systems.

Quite agree about TT systems too - the Earthing system could be by at anything up to 50V above true earth and still be compliant with BS 7671 requirements - yet it's often suggested as a means of avoiding perceived electric shocks of 10V or 20V from a PME system.

   - Andy.

Earthing system could be by at anything up to 50V above true earth and still be compliant with BS 7671 requirements

But in a system with a 30mA RCD, as required for safety of life, and almost any electrode system that is much more than the end of a wire in a flower pot, will do better than the 1500 ohms or so that would give a 50V offset, probably by a factor of ten or better.

So regs wise I agree, the voltage could be 49.99V, but is quite likely to be less than 5V, even with a fault in progress at a level that is certain to trip the RCD in a fraction of a second. Assuming the RCD is not to false trip, then a leakage more like 10mA should be seen as an upper limit if all is operating exactly as it should, so perhaps 1 to 2V when  in normal operation ?

On real TT systems, the offsets are often very low indeed - often non 50Hz components from elsewhere are dominant.

Mike.

But in a system with a 30mA RCD, as required for safety of life, and almost any electrode system that is much more than the end of a wire in a flower pot, will do better than the 1500 ohms or so that would give a 50V offset, probably by a factor of ten or better.

A single RCD yes, but what about multiple RCDs - e.g. a CU with a dozen or more 30mA RCBOs as tends to be the fashion these days?

Hopefully normal protective conductor currents will be small, but all it needs it a bit of wet getting in somewhere (e.g. an outside light or socket) and such assumptions can count for naught.

   - Andy.