At Section 8.8 in the COP for Grid connected Solar PV systems it clearly indicates that the inverter should not be connected to the load side of any RCD that is shared with other circuits. The note explains the reasons without elaboration.
I am trying to understand how an inverter would continue to supply the fault after the RCD has operated. Is it not a separated circuit at the loss of supply?
Is it not a separated circuit at the loss of supply?
There's no intentional separation ... I suppose it could be IT, and there might already be a fault on the "disconnected" side of the RCD, making it the equivalent of TN one way or another.
With modern inverters, RoCoF (Rate of Change of Frequency) should cause almost instantaneous disconnection of the inverter, but inverters to older versions of G83/G59 might continue to provide power for some seconds.
There's another "BUT" to consider. Where RCDs are specified for additional protection they don't just protect against faults of negligible impedance to earth, BUT also against accidental contact with live parts. If the RCD operated because of that, and the power is still being supplied (say with an older inverter) then we still have a potential "additional protection" issue that we didn't disconnect. If someone is in contact with only one live conductor, and there's no fortuitous or impedance earthing (which could come from capacitances in the inverter itself) then perhaps this isn't so much an issue, provided neutral conductors are disconnected from the supply as well as line conductors ... BUT if the user of the installation is in contact with two live conductors, additional protection is certainly not provided for.
if the user of the installation is in contact with two live conductors,
Not sure you can do a lot for someone who presents as a valid L-N or L-L load ;-)
Thinking even how to detect, let alone any kind of ADS.
Mike.
Thinking even how to detect, let alone any kind of ADS.
RCD, except in IT or separated systems.
(In TN and TT systems, N is at 'earthy' potential, so current flows between L and feet as well as L and N ... the RCD might just detect enough residual current and operate in a time that might save you, but there's no guarantee of safety!)
so current flows between L and feet as well
Seems pretty unlikely unless the victim is barefoot on wet ground, or happens to be simultaneously in bare skin contact with an exposed- or extraneous-conductive-part. Body resistance normally limits shock current to a few hundred milliamps, and you'd need a sizeable proportion of that to run to Earth for the RCD to trip - i.e. that path would need a loop impedance not significantly higher than the direct skin contact L-N path - I'd suspect even a pair of sweaty socks would provide enough resistance for 90%+ of the current to stick with the easier L-N route and be missed by the RCD while the victim gasps their last.
Nothing's perfect of course - there are always risks especially where multiple simultaneous faults are possible. ADS can be defeated entirely by a simple pre-existing c.p.c. break. The long standing general principles only provide shock protection between two or move live parts by insulation, enclosures,etc and good maintenance; not ADS. It doesn't seem fair to make things more onerous for local generation.
- Andy.
Body resistance normally limits shock current to a few hundred milliamps,
10s of mA is enough to kill ... and only 30 mA needed to operate a 30 mA RCD in 0.3 seconds (admittedly, up to 250 mA might be needed to operate in 0.04 seconds if the manufacturer selected that option instead of 5 IΔn for their product as per BS EN 61008-1 or BS EN 61009-1)... as I said, protection not guaranteed, but you still have more chance than no RCD.
It doesn't seem fair to make things more onerous for local generation.
It doesn't seem fair to say that protective devices have to disconnect the source, and then circumventing that with a generating set supplied downstream of a protective device that's intended to provide protection of final circuits.
Is it actually more onerous, or just common-sense?
It doesn't seem fair to say that protective devices have to disconnect the source,
Not for L-N shocks though.
- Andy.
It doesn't seem fair to say that protective devices have to disconnect the source,
Not for L-N shocks though.
- Andy.
We're about to take you to the IET registration website. Don't worry though, you'll be sent straight back to the community after completing the registration.
Continue to the IET registration site
