RCD location for TTd outbuilding

[JohnW2]

I recently heard this story from a friend. His house has a PME supply. He has a supply to (just) a couple of sockets (for very occasional, low power, use) in a metal-framed greenhouse. The supply comes via SWA from an RCBO in the house and the greenhouse is isolated from the PME earth and separately TTd. This was installed several years ago by a professional electrician, and visually looks to have been done competently.

An electrician recently on-site for something totally different apparently commented that reliance on the in-house RCBO for protection of the greenhouse was 'unsafe and totally unacceptable', seemingly suggesting that there ought to be an RCD within the greenhouse 'for the locally TTd installation'. Whilst I agree that it's probably not the most convenient of situations to have to go to the house (once in a blue moon) to reset (or attempt to reset) the RCD, as far as I can see the electrician was essentially talking total nonsense, and exhibiting a lack of understanding of basic electrical principles. Does anyone disagree with me?

Kind Regards, John

 

[ricicle]

The electrician was right. Once you create a new equipotential zone with a rod then another RCD will be required to disconnect any fault relating to that zone.

 

[JohnW2]

The electrician was right. Once you create a new equipotential zone with a rod then another RCD will be required to disconnect any fault relating to that zone.

Why? An RCD, wherever it is, knows nothing about 'earths' or equipotential zones - all it is aware of is an L-N imbalance, due to some current having found a route to some earth (no matter what/where). The L and N currents passing through an RCD in the greenhouse would be identical to the L and N currents passing through the present one in the house. I don't understand.

Kind Regards, John

 

[riveralt]

This was discussed in part here!

electrical.theiet.org/wiring-matters/16/elect-inst-outdoors.cfm

you may have to cut and paste the link.

 

[plugwash]

What is more concerning is that RCBOs are typically only single pole so they will not disconnct a neutral-earth fault.

 

[ericmark]

I think it all comes from the original ELCB-v where clearly they would not work if the earth system had some connection other than through the ELCB-v. However with an ELCB-c it would not matter one little bit where it was situated there would still be an unbalanced supply so it would still trip.

However this assumes there is enough voltage for the electronics within the RCD to work. Where with a major fault the loop impedance could allow the voltage to drop too low then either the RCD would need to be placed closer to the supply point or an active type used.

So really safety wise there is a very good reason to fit the RCD in the house rather than the green house as the loop impedance will be better and it is unlikely to get a volt drop so high as to prevent it working.

 

[JohnW2]

This was discussed in part here! electrical.theiet.org/wiring-matters/16/elect-inst-outdoors.cfm
you may have to cut and paste the link.

I'm familiar with that article, but (unless I'm missing something) I don't think it really addresses this issue.

I'm having great difficulty in understanding the disagreements here. How on earth can it make any difference whether an RCD is connected in the path of L and N conductors in one place or in the path of exactly the same L and N conductors a few metres upstream? As I said, a belief that there is a difference seems to me to indicate a lack of understanding of one of the most basic principles of electricity!

Kind Regards, John

 

[JohnW2]

What is more concerning is that RCBOs are typically only single pole so they will not disconnct a neutral-earth fault.

Agreed, but is that really a problem? A single-pole RCBO will operate in the case of a N-E fault. As you say, it won't 'disconnect' the N-E (the 'E' being the TT earth) fault from the main installation's neutral, but I don't really see what harm that does. Even if one postulates a co-incidental (vanishingly improbable?) Bernard-like high neutral potential in the house, the amount of current that could flow from that N into the TT rod would surely be too low to represent any hazard.

Kind Regards, John

 

[ericmark]

What is more concerning is that RCBOs are typically only single pole so they will not disconnct a neutral-earth fault.

Good point, but I have used double pole RCBO's in a small box many times to supply sockets for radio chargers and the like.

But I have not seen many active RCBO's most are passive and in the green house it would likely need to be active type due to volt drop under fault conditions.

I had not considered the voltage required until I came across some RCD's fitted to a split phase 110 volt supply and found when one leg was lost there was not enough voltage for the RCD to trip. The problem is we test with a tester also designed for 230 volt. So although we should fit and test RCD's on 110 volt in practice most of us have no way to test them.

I had to use some resistors to make a tester as the test button would not work with a phase missing. It resulted in quite a problem as we realised we had fitted a batch of 230v RCD's in error the box was marked at 110 volt.

But this did alert me to the volt drop problem which I had just not considered before. All the 110 volt RCD's were active so failed safe.

 

[bernardgreen]

I think it all comes from the original ELCB-v where clearly they would not work if the earth system had some connection other than through the ELCB-v.

They would work in that the touch voltage ( the voltage on the CPC relative to ground ) would not go above 50 volts. The current to ground via the parallel paths could be very high and the fuse in the Live would probably blow.

If it was a Neutral to CPC fault and supply neutral was not at ground potential then high currents could flow from neutral into parallel path. In those days the parallel path could have been the metallic water main all the way back to the sub-station. The impedance of the water main could be lower than the network neutral so the fault current could be very high, maybe the reason for fuses in the neutral in those days.

 

[JohnW2]

However with an ELCB-c it would not matter one little bit where it was situated there would still be an unbalanced supply so it would still trip.

Exactly my point, and I really wasn't expecting to see people here disagreeing!

Kind Regards, John

 

[Taylortwocities]

Just having a head scratch here.

Its an RCBO, correct.?
The RCBO has a functional earth and this should be connected to the earth of the circuit(s) that are being protected. That is the earth spike.

The greenhouse spike is probably in a different voltage place to the (PME) earth in the house.

 

[bernardgreen]

Bernard-like high neutral potential in the house,

I do not claim a monopoly on bouncing neutrals.
I agree the current from neutral into a ground rod from a Neutral - Earth fault will be at most an amp or two, but if the "rod" is very low impedance it is effectively a path in parallel with the network neutral to the sub-station.

I would opt for a two pole RCD in the TT zone for that reason.

 

[ericmark]

A double pole RCBO does not normally have a functional earth connected.

We know that using a single pole RCBO does not really comply even if it is what is used here so not really worth talking about them.

The only reason I can see where the RCBO needs fitting in the out building is to comply with 314.1 so that lights would not be lost with a fault with power but this is hardly going to be a problem with a green house.

Do remember the Autumn Wiring Matters was written in 2005 when all sockets were not RCD protected and we did not have 17th Edition consumer units where all circuits were RCD protected.

At that time the RCD FCU was a rare beast today they are quite common.

 

[JohnW2]

Just having a head scratch here. Its an RCBO, correct.?

That's what I was told - but to be honest I didn't look. I will make a phone call and interrogate him!

The RCBO has a functional earth and this should be connected to the earth of the circuit(s) that are being protected. That is the earth spike. The greenhouse spike is probably in a different voltage place to the (PME) earth in the house.

As I understand it (as always, possibly incorrectly!), the only reason for the functional earth is to ensure that there is a power supply for the RCBO's electronics in the case of a totally lost neutral. That being the case, any connection to earth ought to suffice. Indeed, there seems to be debate out there as to whether the function earth is 'essential', and I think that there are even some RCBOs without functional earths.

Kind Regards, John