Adiabatic Lecture.

Yup.

Normally we check the MCB elements of the RCBO give us compliance and if so then the RCD element of the RCBO is just a back up (supplementary protection). If not then we might be relying solely on the RCD element

And? The problem here is that whilst everyone wants RCDs, they are not prepared to trust them! It seems to me that the RCD is being "perverted" to additional protection only, except in TT installations. Why is this, it is certainly not a requirement of BS7671?

David Stone said:

The problem here is that whilst everyone wants RCDs, they are not prepared to trust them!

What is that based on?

RCDs can be used to provide the sole means of fault protection for ADS, at least since 17th Ed, BS 7671:2008. See Regulation 411.4.5

However, they cannot provide protection against overcurrent.

In terms of ensuring an RCCB isn't overstressed by earth fault currents, suitable overcurrent protection is required.


What are we looking at here? I have a 63 A RCCB. It is protected upstream by an overcurrent protective device, rated in accordance with the manufacturer's instructions, or a suitable device operating within the parameters of the largest device according to the manufacturer's rating.

So, for all earth faults up to 63 A, the RCCB can provide fault protection. For faults exceeding that, tbe RCCB is backed up by an overcurrent protective device which can protect the RCCB from harmful earth fault currents ... and also operates on Line-Line or Line-Neutral faults (as appropriate for the installation).


There is no quoted let-through energy for RCCB's, because they require a form of upstream overcurrent protection. To size conductors, you need to use the ratings of the overcurrent protective device.


So, you might ask, what about TT systems?

Well, if I were to ask a group of electricians how they measure the prospective earth fault current in a TT installation, they would say by taking an EFLI measurement to the earth electrode disconnected from the MET.
NOW, this is perfectly correct, for the worst-case conditions for ADS.
BUT it is NOT correct for the worst-case conditions in relation to protection against overcurrent (in this case, earth fault current).

Let me provide a simple example.

I have an earth electrode in a TT installation, R_A = 80 Ω, and a combined supply loop impedance (supply earth electrode R_B plus transformer plus line conductor) Z₀ = 5 Ω. So, Z_e = 85 Ω, and I think I have an I_ef = 230/100 = 2.7 A. (And of course I use this value for ADS calculations, meaning I have to use RCDs for ADS).

However, I connect extraneous-conductive-parts, which could be extensive steelwork, possibly LPS, or extensive buried metallic pipework, with equivalent combined earth electrode resistance of 1 Ω.

Now, my equivalent Z_e = 6 Ω, and I have a fault current at the origin of an I_ef = 230/21 = 38 A, and in this case, we might be looking at instantaneous tripping of mcb or RCBO for faults on some 6 A circuits, perhaps even some 10 A circuits.


In some cases, however, we might be talking about substantial earth fault currents when extraneous-conductive-parts are connected. For example, large steel-framed multi-tennant building where the first tennant gets a PME supply and the rest TT ... when extraneous-conductive-parts shared with the PME installation are connected, I_ef could actually be kA (the same as an L-N fault), simply because they have a path back to the supply neutral through the PME service MET.

Now, my fault current is

David Stone said:

The problem here is that whilst everyone wants RCDs, they are not prepared to trust them!

What is that based on?

RCDs can be used to provide the sole means of fault protection for ADS, at least since 17th Ed, BS 7671:2008. See Regulation 411.4.5

However, they cannot provide protection against overcurrent.

In terms of ensuring an RCCB isn't overstressed by earth fault currents, suitable overcurrent protection is required.


What are we looking at here? I have a 63 A RCCB. It is protected upstream by an overcurrent protective device, rated in accordance with the manufacturer's instructions, or a suitable device operating within the parameters of the largest device according to the manufacturer's rating.

So, for all earth faults up to 63 A, the RCCB can provide fault protection. For faults exceeding that, tbe RCCB is backed up by an overcurrent protective device which can protect the RCCB from harmful earth fault currents ... and also operates on Line-Line or Line-Neutral faults (as appropriate for the installation).


There is no quoted let-through energy for RCCB's, because they require a form of upstream overcurrent protection. To size conductors, you need to use the ratings of the overcurrent protective device.


So, you might ask, what about TT systems?

Well, if I were to ask a group of electricians how they measure the prospective earth fault current in a TT installation, they would say by taking an EFLI measurement to the earth electrode disconnected from the MET.
NOW, this is perfectly correct, for the worst-case conditions for ADS.
BUT it is NOT correct for the worst-case conditions in relation to protection against overcurrent (in this case, earth fault current).

Let me provide a simple example.

I have an earth electrode in a TT installation, R_A = 80 Ω, and a combined supply loop impedance (supply earth electrode R_B plus transformer plus line conductor) Z₀ = 5 Ω. So, Z_e = 85 Ω, and I think I have an I_ef = 230/100 = 2.7 A. (And of course I use this value for ADS calculations, meaning I have to use RCDs for ADS).

However, I connect extraneous-conductive-parts, which could be extensive steelwork, possibly LPS, or extensive buried metallic pipework, with equivalent combined earth electrode resistance of 1 Ω.

Now, my equivalent Z_e = 6 Ω, and I have a fault current at the origin of an I_ef = 230/21 = 38 A, and in this case, we might be looking at instantaneous tripping of mcb or RCBO for faults on some 6 A circuits, perhaps even some 10 A circuits.


In some cases, however, we might be talking about substantial earth fault currents when extraneous-conductive-parts are connected. For example, large steel-framed multi-tennant building where the first tennant gets a PME supply and the rest TT ... when extraneous-conductive-parts shared with the PME installation are connected, I_ef could actually be kA (the same as an L-N fault), simply because they have a path back to the supply neutral through the PME service MET.

Now, my fault current is

That's why I likes me earth fault loop tester. It gives a complete picture.

Z.