Outbuilding Earthing

[i_am_fubar]

Arnour terminated at the source end. Terminated into a non conductive transition box (*) at the shed end, conductors and inner sheath passing through the box uncut.

As an electronics engineer, this is what I would expect... What's with the * though?

Why do you need a 60A supply? Are you starting a waccy baccy farm?

Yea, I wish, would pay for a garden make over Alas, just some workshop equipment, welder, lathe, mill... etc. 60A due to high in-rush currents of some potential future things, as well as the possibility of putting in a 3ph inverter. Plan on having at least 1 32A socket, couple of 16A and a whole crud load of 13A and fused outlets for equipment. Most of the cost / effort is in digging trench, making off... etc. So may as well guarantee I never have to do it again

 

[EFLImpudence]

 

[bernardgreen]

What's with the * though?

I deleted the alternative of

(*) or glanded into a metal box that is isolated from the local ( shed ) Earth and protected from being touched.

I forgot to remove the (*)

workshop equipment, welder

You will need to check with the DNO as to whether the local network can cope with the in-rush current of a welder.

 

[i_am_fubar]

It's not a crazy powerful system. So I doubt inrush will get close to being a problem, but as I'll have computers running in the workshop too, I want a nice low impedance link to my supply.

How would the DNO spec that? if there are 10 houses on a phase down my road, each with a 100A incomer fuse, surely it's impossible to say how much headroom there is for my feed if the phase fuse is less than 1kA. I had, foolishly perhaps, thought that if I had a 100A fuse coming into my house... I could pull 100A without an issue. Or is this another case of system loads increasing due to housing developments without the infrastructure being upgraded?

On the note of the metal box, if the Armour was terminated outside the workshop (i.e. box bolted to the wall), that could be metal surely, then take the unarmored cores through the wall into the CU.

 

[JohnW2]

.... Alas, just some workshop equipment, welder, lathe, mill... etc. 60A due to high in-rush currents of some potential future things, as well as the possibility of putting in a 3ph inverter. Plan on having at least 1 32A socket, couple of 16A and a whole crud load of 13A and fused outlets for equipment.

Have any of the electricians you consulted discussed the adequacy (or otherwise) of your house's supply to service the anticipated loads (as well as the existing ones)? There is probably also, strictly speaking, a requirement to get your DNO's permission to install equipment such as you mention.

As for the 'earth issue', as you have discovered, opinions vary, which is a sure sign that there is no one 'correct' answer, and it's all about 'swings and roundabouts' judgements - particularly in relation to which of various (all very low probability) fault scenarios one considers to be the more likely. Just a couple of thoughts ....

One consideration is that the high inrush currents of some of your equipment may well require use to use Type C (or maybe even Type D) RCBOs, and if you were utilising the exported TN-C-S earth that could cape an impact on the required size of the CPS (core, armour or both) in order to get the required disconnection times (i.e. adequately low Earth Fault Loop Impedance) in the event of L-E faults. If the outbuilding were isolated from the house's TNCS earth and reliant on it's local earth electrode (i.e. TTd), then there would be no expectation of achieving the required disconnection times (in response to L-E faults) by means of the over-current function of the RCBO, it being accepted that protection against such faults would then be reliant on the residual current function of the RCBO.

In terms of swings and roundabouts, I am rather surprised that bernard has not mentioned one issue which makes some people disinclined to export TN-C-S earths. As you will be aware, certain (rare) supply-side faults can lead to a TN-C-S 'earth' (actually the incoming neutral) rising to a dangerous potential above true earth potential. If one exports that earth to an outbuilding, one can ensure that the inside of that building is all equipotential (particularly easy when there are no extraneous-conductive parts), so that no dangerous pds shoudl exist within the building, regardless of the actual potential of the touchable exposed-conductive parts. However, if it is possible for someone to touch one of those exposed conductive parts (e.g. a metalclad light switch just inside the door) whilst standing on ground outside of the building (hence outside of the equipotential zone), then there is a theoretical possibility of a dangerous of existing between one's hands and feet - and a problem which would/could not be cleared by any protective device. However, we're talking a bout a very rare aupply fault scenario, so one has to decide whether it is worth worrying about.

BAS's initial suggestion of a 'belt and braces' approach of using the exported TN-C-S 'earth' (via core, armour or both) AND a local TT electrode will to some extent reduce that risk, since it will shift the potential of the ground in the vicinity of the earth rod to closer to the (elevated) potential of the TN-C-S 'earth' under supply fault conditions.

As an aside, even if 10mm² SWA (however many cores) could be shown to be theoretically adequate, I would personally be inclined to go a fair bit larger than that (perhaps 25mm²), not to least to avoid the need to dig up the garden again should the future bring different requirements!

Just a few thoughts!

Kind Regards, John

 

[JohnW2]

How would the DNO spec that? if there are 10 houses on a phase down my road, each with a 100A incomer fuse, surely it's impossible to say how much headroom there is for my feed if the phase fuse is less than 1kA. I had, foolishly perhaps, thought that if I had a 100A fuse coming into my house... I could pull 100A without an issue.

DNOs seem to work on the assumption of a average load of about 2kW (around 9A) per household, with loads considerably greater than that being unusual, usually short-lived, and virtually never applicable to multiple houses simultaneously - in other words, 'diversity'.

The problem the DNO want to avoid is their getting complaints from other households because their lights dim significantly (and/or other equipment malfunctions) due to the inrush when you power up one of your machines. Their judgement about that will depend on several factors, not the least being how close you are to their transformer.

Kind Regards, John

 

[i_am_fubar]

In terms of swings and roundabouts, I am rather surprised that bernard has not mentioned one issue which makes some people disinclined to export TN-C-S earths. As you will be aware, certain (rare) supply-side faults can lead to a TN-C-S 'earth' (actually the incoming neutral) rising to a dangerous potential above true earth potential.

Wouldn't this be an issue with non exported earth, say a metal light switch just inside a utility backdoor of the main dwelling (Assuming no incoming metal and bonded services)?

average load of about 2kW (around 9A) per household

Wow... 9A... that's a reasonable daytime average, but around dinner time, I can see that being much higher on an hourly average. What's the rating on a phase fuse (With regard to blow time vs % over rate)?

Electric cars will be a problem when they all need charging. Makes me really appreciate the local energy storage concepts.

For the record though, all my current tools run of 13A, and only one at a time. It's only really the future I want to plan for where I may get something bigger. I could probably run what I have on 6mm (4 at a push) and down rate the fuse, so 10mm gives me the headroom I need if I have say, 2 people using manual machines + a small CNC and compressor running.

 

[JohnW2]

Wouldn't this be an issue with non exported earth, say a metal light switch just inside a utility backdoor of the main dwelling (Assuming no incoming metal and bonded services)?

Indeed - but it's less likely that people will be standing on 'bare ground' (and maybe with bare feet) when reaching in and touching that switch. The more common (and worrying to some) issue is that of an 'outside tap' which is plumbed in metal, and hence at the potential of the TN-C-S earth. In that case, the bare feet may be standing on wet bare ground! However, as I said, you may well feel that those are such low risks as to not warrant consideration.

Wow... 9A... that's a reasonable daytime average, but around dinner time, I can see that being much higher on an hourly average. What's the rating on a phase fuse (With regard to blow time vs % over rate)?

I think the substation fuses are typically 400-600A, and probably would require a load of about double their ratings to be sustained for an appreciable amount of time before they blew.

However, bear in mind that, even at "dinner time", a cooker (oven+hobs) will, due to its thermostatic control, represent a far lower average current draw (over time) than you might think.

Electric cars will be a problem when they all need charging. Makes me really appreciate the local energy storage concepts.

I somehow doubt that the present supply network could cope with a very widespread change to electric cars - which is why I think that, no matter what governments may think, it is likley to be many decades before such a widespread change would be achievable. Just a few years back, before people started thinking/talking about EV charging, the hot topic for discussion were the serious concerns that the UK generating/distribution system was not going to supply our 'present' needs for much longer. Add widespread EV charging vto that load and, as I said, I think one would probably be talking about 'many decades'.

For the record though, all my current tools run of 13A, and only one at a time. It's only really the future I want to plan for where I may get something bigger. I could probably run what I have on 6mm (4 at a push) and down rate the fuse, so 10mm gives me the headroom I need if I have say, 2 people using manual machines + a small CNC and compressor running.

Fair enough. However, as I said, the DNOs are not likley to be so concerned with their ability to supply the amount of energy you need - they are probably much more worried about the potential transient voltage drops to your equipment's inrush currents, and the possible impact of those voltage drops on your neighbours.

Kind Regards, John

 

[ban-all-sheds]

I somehow doubt that the present supply network could cope with a very widespread change to electric cars - which is why I think that, no matter what governments may think, it is likley to be many decades before such a widespread change would be achievable. Just a few years back, before people started thinking/talking about EV charging, the hot topic for discussion were the serious concerns that the UK generating/distribution system was not going to supply our 'present' needs for much longer. Add widespread EV charging vto that load and, as I said, I think one would probably be talking about 'many decades'.

Smart metering will help to manage the load.

One thing to bear in mind is that most people use their cars for short journeys, so overnight charging at home doesn't need to do a full charge - probably 10-20%.

 

[ericmark]

Wiring matters Outdoors: A supply to a detached outbuilding by John Ware Autumn 2005 is still considered to be about the best when deciding how to earth think this is the link

It was written in the days of 16th Edition, but still in the main valid points, the main consideration is the voltage gradient. If you have a housing estate with the houses arranged in a square back gardens being over looked by neighbours back gardens with road around the out side, even if all those gardens cover an acre then the supply cable with its multiple earths will surround the estate so using the TN-C-S supply is not a problem.

However same size garden with just fields behind then likely needs to be TT.

There is no right or wrong, it is down to the guy who signs to say I have designed the system to decide.

 

[SimonH2]

Im thinking otherwise the earth has 2 paths, or does that not matter.

Not only does it not matter, it's actually a benefit - you have two conductors in parallel so the resistance of the combination is lower than that of either conductor on it's own. Though as pointed out previously, it's possible for the wire armour to rust and negate the benefits - but if that did happen (most likely on a localised basis due to sheath damage) then earthing it at both ends still means you get the benefit of earthed armour over almost the full length of the cable.

 

[ban-all-sheds]

However same size garden with just fields behind then likely needs to be TT.

I don't think you'd need to have what would reasonably be thought a mansion, or palatial dwelling, to have a kitchen, or a conservatory, at the back of the house, 20m from where the supply enters at the front. A large house, maybe, but not a "stap me vitals, this is some place you've got" large.

So should the whole house be TT? Where should the rod be? Should the kitchen or conservatory be carved off as a separate TT sub-installation?

 

[ban-all-sheds]

would pay for a garden make over

How long do you have?

Could you persuade Mrs Fubar to go away somewhere without telling anybody? You then act a bit evasive when friends ask where she is, and you do a bit of night-time digging here and there and burn some of her old clothes.

Sooner or later some nice men in disposable white coveralls will come and dig all of your garden over for you.

 

[ericmark]

The belt and braces idea of having an exposed or extraneous conductive part bonded to the DNO earth means one has to carefully measure the resistance to ground of those parts and work out current flow with worst case scenario. So the part in the ground has a resistance of say 8Ω so 230/8 = 29 amp using 10 mm² cable so that's OK. However the rod will over time build up a better and better connection, so in 5 years time it may now measure 4Ω now looking at 58 amp, and remember there is no fuse, or even a way to disconnect it, you can pull that DNO fuse, it will not make the slightest difference still 58 amp flowing, what will make a difference is what other people in the street draw, it can vary between zero and 400 volt when the combined earth/neutral is severed. So could be 100 amp passing to that metal part.

What we hope is everyone in the street has exposed or extraneous conductive part bonded to the DNO earth so the load is shared between them, so for example if they all have metal water pipes then each one may carry 10 amps so no one pipe is overloaded, where the problem lies is where a fuel pipe is only used with some properties in the street so the load exceeds what the pipe can carry so they rupture of course with a spark so the fuel is ignited.

One method to stop this is to use an insulating bit of pipe, so excessive current can't pass through the pipe. Without this insulator or if some one bonds to wrong side of insulator should anything cut the earth neutral it can and has caused fires.

So what does the belt and braces do in real terms? It reduces the chance of a gradient in ground voltage as found in electric storms causing sparks to ground around the installation, this is important with a gas installation, but just in case the DNO loses its earth you also need a large resistor to limit the current, so put a 20Ω resistance between the two earths then you have limited the worst case scenario current to 20 amp but that resistor needs a rating of 8 kW the one on the gas terminal was about a 2 meter square block to be able to keep cool enough should the DNO lose its earth.

So one rod at 60Ω not really a problem, but is it really any advantage either? OK if every house in the street has a 60Ω earth rod then this would form the multiple earthing of the PME system, but then no one person has control, better the DNO sinks those earth rods then they can test them at regular intervals to ensure all are still connected.

So for a garden shed what does the belt and braces do in real terms, answer is nothing, it does not ensure you have an earth all it does is cause a point where your pet can get electrocuted if the DNO lose their earth. So either TT or TN not both.

I have seen once where the DNO lost their earth and there was a garden shed (shack) with a very good earth fitted because used to transmit from the shed (radio shack) and the 4 mm earth wire which had been tie wrapped to the supply cable to shack was reduced to copper balls on the floor.

Yes you can have an earth mat, but looking at cow shed design not a garden shed. Why have earth mat, because the cows heart lies between their legs, so even 25 volts can kill. We may have sensitive parts between our legs, but not an organ which will cause death like with animals with four feet.

 

[ban-all-sheds]

So one rod at 60Ω not really a problem, but is it really any advantage either? OK if every house in the street has a 60Ω earth rod then this would form the multiple earthing of the PME system, but then no one person has control, better the DNO sinks those earth rods then they can test them at regular intervals to ensure all are still connected.

So for a garden shed what does the belt and braces do in real terms, answer is nothing, it does not ensure you have an earth all it does is cause a point where your pet can get electrocuted if the DNO lose their earth. So either TT or TN not both.

AIUI, "both" is what the 18th will require?