9Kw pool heater, earth rod, seasonal drought + total tiling

[Ace1]

My neighbour had a 9Kw inline (in-pipe) water heater fitted - cables look well installed, with trip protection as requested by the heater installers.
On connection it pulled the house trip.
They suggested (rightly or wrongly) to install a separate supply before the house distribution panel - this was carried out by an electrician.
They returned when the owner was absent, re-installed the heater, and apparently the trip pulled, but it's not clear cos there were no witnesses - but they left with the heater.
1 year later they are to return (job has been paid for - ouch!)

Clearly the heater itself could have been duff.
However, I'd be interested to learn your thoughts on whether a 'bad earth' could be responsible, due to the fact that:
1. They only have one rod (and no transformer connection - TT)
2. It's fitted in-house (and tiled over ie. only entry cable is visible)
3. The entire garden/enclosed courtyard is tiled (water drained away)
4. We suffer seasonal drought

Having spent some time researching this (to get an overview), it appears(?) that a bad earth could cause tripping, but could also prevent tripping, depending on the protection used.

Time to get some firm advice from the community, in order to inform him of both the general risks/problems of a bad earth - the risk that a short could electrify the pool - and the possibility (or not) that the earth can't take the 9Kw connection, so causing the trips.

Note: there is no available surrounding land for multiple test stake testing.

 

[Deleted member 9648]

The quality of the earth should have been tested at the time of install,and can easily tested now. What the limits are in france I dont know but in the UK the maximum resistance on the rod should be 200 ohms...but if the system is protected by a 30ma rcd protection is assured at resistances up to 1667 ohms...(UK regs)
An IR test would show any faults on the system which may cause tripping......Proper testing is all that is required to identify the likely cause of your problem.

 

[topquark]

I thought that regs in France dictated that everything was run as radial direct to device (and rcd protection)?

A "bad" earth is more likely to cause it NOT to trip or trip too slowly than trip prematurely.

 

[Ace1]

Yes....this question is, to some extent, the key.
The research clearly indicates that, the rod is not ideal, due to the ground it is in, never getting any rainfall ie. it likely relies on flooding to lift the water table to ground level.
But there is a 'well' 5m from the rod, so moisture anyway drains away.
Another forum historic post suggested locating plants above a rod, with the watering of, ensuring moist earth.
This makes sense to me, so another earth rod seems a logical step to take.
But....
In the mean-time, post recent flood (ie. any test would not represent typical operational conditions), the question remains:

Would a 'bad earth' cause tripping (last summer when in drought)?
If it would definitely not, then we must assume that the original heater was truly duff.

That confidence would allow us to push the installer to try another heater.

Here are the trip details:
Merlin Gerin
Multi 9
ID CLIC
40A
30mA
23246

Thing is..... if the bad earth does cause tripping then, getting the installers to try again would likely be pointless (unless they did it immediately while the ground is still damp).
Then the next job would be a 2nd rod and test.

Bear in mind also your remark vis a vis French regs.
The regs can be one thing, but following them is another.

My first rented house here had cloth covered cables(!) something that I believe disappeared in the UK in the 60's - half the house was on a 2 wire sys, the other half on on +earth but using a water pipe connection that no longer exists due to pvc.
Some poor sod has just bought it so I think I need to warn him.

But thanks to wirepuller for the test data - 200 to 1667 ohms.

 

[Engineer54]

As far as i remember, ... French TT systems require a Ra of not more than 100 ohms and a RCD protection device of not more than 500mA Based on .....R=50 volts divided by 0.5A = 100 (ohms) R being the touch voltage requirement....

Oh, and a bad earth would Not cause the protection device to trip, rather the opposite!! ...lol!!!



Ω

 

[Ace1]

Just to be clear on this engineer54 - are you suggesting that:
Rather than 200 - 1664 ohms (inc. RCD)....... the actual test results for the earth should be no more than 100 ohms?

 

[Ace1]

Maybe we should just bang in another rod.
There is a well watered border of plants on the opposite side of the courtyard about 15m from the current earth rod.
Would it be correct to then simply connect this to the earth cable at the trip, midway between the two?

This would surely provide a more consistent earth than the one tiled over, so dealing with the risk of the RCD not tripping.

As for the heater tripping the RCD's in the first instant, it looks as though this has nothing to do with the earth, so indicating that the heater was probably duff after all.

However, on a positive note - that thinking has lead to an investigation into the earth, and the discovery that it is unlikely to be very good, and pretty much guaranteed not to hit the 100 ohm mark in summer.
Trevor Linsley (Advanced Electrical Installation Work)
Advanced electrical installation work - Google Books
....suggests that for countries experiencing drought, resistance could be thousands of ohms.

Hence my statement earlier, re having tough regulations, that are never followed.

 

[Engineer54]

Just to be clear on this engineer54 - are you suggesting that:
Rather than 200 - 1664 ohms (inc. RCD)....... the actual test results for the earth should be no more than 100 ohms?

No, not suggesting, just quoting the French regulations for a TT system...

Obviously making provision of a better Ra value is always going to be beneficial to the installation. I personally, do not like totally relying on an RCD device for earth fault protection. So if it was my installation that was subject to drought conditions, i would most definitly be either driving deep rods in, or borehole and using a rod enhancer like bentonite to get deep enough past any drought affected ground. If it needed more than one deep driven rod location to get a good solid Ra value, so be it, another would be installed...

 

[Engineer54]

Maybe we should just bang in another rod.
There is a well watered border of plants on the opposite side of the courtyard about 15m from the current earth rod.
Would it be correct to then simply connect this to the earth cable at the trip, midway between the two?

A very good idea, As suggested above!!!

If you drive another rod in, try and make it as deep as you can by using extension rods that screw/couple together. Really need to go down 3 meters minimum to get past the drought affected level and still have enough contact depth with the rod.
Connect both rods together using at least 4mm cable, ( but i always use a minimum of 6mm cable ) Then take the same sized cable from the connected rod, to your MET (main earth terminal)

 

[Ace1]

Thanks for confirming that, and the tip re bentonite.
I just researched what it was, learning that it is most readily available as 'cat litter'.

However, before winding up this thread, I think that some interesting grey areas still remain - in particular concerning the protective devices used (once you have established a 'good earth').

Residual-current device - Wikipedia, the free encyclopedia
1. RCDs (GFCI or GFI?) are designed to disconnect quickly enough to mitigate the harm caused by such (electric) shocks although they are not intended to provide protection against overload or short-circuit conditions.

From this - is it the case that; for a submerged pump system (eg. Desjoyaux) - If the shaft seal failed, and water flooded the motor; then it would be the mains entry 'dejuncter' that would trip (eg. 500mA DBD 11 45A). ie. that this would trip regardless of the prescence of an earth?
Note that I don't know what the specific UK term for a dejoncter is.

2. In the United States, the National Electrical Code requires RCD (GFCI) devices intended to protect people to interrupt the circuit if the leakage current exceeds a range of 4–6 mA of current (the trip setting is typically 5 mA) within 25 ms. A GFCI device which protects equipment (not people) is allowed to trip as high as 30 mA of current; this is known as an Equipment Protective Device (EPD).

Yet, when we examine the installed RCD data (earlier post), it is rated at 30mA ie. from the US regs, this is not the correct rcd - or - are the US regs over tough?

3. Total resistance of the whole earth loop must be low enough to permit the protective device to disconnect the supply to the circuit in 0.4 seconds for socket outlets, and 5 seconds for fixed appliances (regulations 413-02-09 and 13).
Advanced electrical installation work - Google Books

This I just don't understand because all other talk is of 25ms............. here the talk is of 400ms up to 5 seconds?

I think it is worth clearing these issues cos, when you are in a pool, and you get a belt you're stuck.......... your hand doesn't fly back as is usually the case LOL.

 

[Deleted member 9648]

Regarding point 3...this is to an obsolete version of bs 7671.....on a TT now disconnection should take place in 0.2s for both socket outlets and fixed equipment.

Regarding E54's post.
IMO if you are using earth electrodes as your means of earthing (TT system in the UK) you are ALWAYS reliant on an RCD for fault protection. Even if multiple electrodes achieve a low enough Ra to operate an overcurrent device at the time of installation, ground conditions vary,and there can be no guarantee that at other times the Ra will remain that low. Therefore IMO you have to rely on the RCD.
The point about the reliability of RCD's is however very important,but in my view the way to deal with that is to install more than one RCD...ie,a 100ma time delay at the origin and a 30ma on the final circuit. The chances of two RCD's failing to operate is miniscule......this makes the expensive and still unreliable option of multiple earth rods pointless.

 

[Engineer54]

Here we go again!!!!

I personally wouldn't rod any domestic installation unless i had no other recourse. If that is the case then i'd make dammed sure that, the roded installation is a good one and NOT just one of your 1.2m thin rods. A good deep driven or made roded system can remain effective and stable, for in excess of 30 years, in fact often betters with time.... I wonder how effective those RCDs of today will be in 10, 15, 20 years time, let alone 30 years time, and i do mean either or both of them... That ''Miniscule'', will get rather lager as those years pass by!!!

 

[Marvo]

Where I am it's not uncommon to need in excess of a dozen rods around 10 meters depth or more to achieve the required Ra. We wouldn't require by regs an external submersible pump (bore hole or well point) to be RCD protected. We do however install lightning protection on bore hole pump supplies as a matter of course.

 

[Engineer54]

Where I am it's not uncommon to need in excess of a dozen rods around 10 meters depth or more to achieve the required Ra. We wouldn't require by regs an external submersible pump (bore hole or well point) to be RCD protected. We do however install lightning protection on bore hole pump supplies as a matter of course.

What soil type and what minimum Ra are we talking about here Marvo?? A dozen or 10 metre Rods seems excessive to say the least!!! I've done several smaller TT field systems using say 10 locations each 3 X 3/4'' X 3.2m Rods (with rod enhancement) in the Middle East and walked away more than happy with sub 1 ohm values... That was in mostly compact sand and cracked sandstone ground, but with quite high salts, sulphate and chloride levels....

 

[Marvo]

Soil type is almost pure sand, I'm not sure about salt content but compaction is usually low due to high seasonal winds causing movement and we're aiming for <10ohms. We often use earth mats laid along the full length of foundations trenches before casting for new buildings and supplementary earth rods are usually still needed. Getting a decent Ra around here is a combination of science and art. A good dose of prayer or foul language has also proved invaluable on many occasions.

 

[Engineer54]

Haha!! ... I think i know what you mean, but having that problem ''every'' time you do a rod install must be a real downer for you sparks out there!!! Ever tried using a rod enhancers?? I think a Bentonite slurry might be your best choice in an non-compacted sand situation, maybe with some further enhancing chemicals if you don't have any water table catchment situations to worry about.... Just a thought!!!