Does an RCD still function when Solar PV is generating?

[spud1]

Hi everyone,

I wondered if anyone could help me with something that has been niggling me please?!

When a Solar PV feed in circuit is installed and added onto a spare way of the RCD protected side of a split load dual RCD consumer unit (i.e. other final circuits are protected by that same RCD) and the inverter is generating power, can the RCD in the consumer unit still detect an earth fault that may occur on said other final circuits, if they are only drawing current from the generated solar?

In my minds eye the current flows backwards into the top of the feed in circuits MCB, across the busbar and back up the current using final circuits MCB into the final circuit. This bypasses the RCD completely doesn't it? Or am I missing something?

 

[pc1966]

The RCD can still detect an imbalance so it could trip - but the PV inverter could well keep powering the now-disconnected set of MCB for a lot longer than the allowed disconnection time before it realises that it is no longer synchronised to the grid and shuts down.

So while you might get some protection it is DEFINITELY not acceptable to have a source of power on the load-side of a common RCD.

 

[pc1966]

the inverter is generating power, can the RCD in the consumer unit still detect an earth fault that may occur on said other final circuits, if they are only drawing current from the generated solar?

Remember the neutral of the PV is (normally) referenced to earth on the supply-side of the RCD, so even though the flow of power is reversed, if you get a load-side path Line to Earth that still causes an imbalance in the RCD.

 

[westward10]

PV should not be connected through a shared RCD.

 

[spud1]

The RCD can still detect an imbalance so it could trip - but the PV inverter could well keep powering the now-disconnected set of MCB for a lot longer than the allowed disconnection time before it realises that it is no longer synchronised to the grid and shuts down.

So while you might get some protection it is DEFINITELY not acceptable to have a source of power on the load-side of a common RCD.

Thankyou for your reply. That does make sense. But I did just wonder, in the event of an earth fault in my scenario with the shared RCD, after it trips (within the prescribed time), in that short period after, where the faulty circuit is still being indirectly energised by the inverter, is the earth fault still dangerous? Because again in my minds eye, it is being supplied by inverter winding which I assumed (maybe rightly or wrongly) would isolated from and as such have no reference to earth ?

 

[westward10]

This is why PV should not be on a shared RCD. There could be a momentary hazard whereby the PV can still be generating line to earth under fault conditions until it collapses due to AC failure.

 

[timhoward]

I found a completely unlabelled PV setup last week on a shared RCD.
First I knew about it was an 750ms RCD test result. Even I could tell the RCD wasn’t taking 3/4 of a second to trip on the test button.

after it trips (within the prescribed time), in that short period after, where the faulty circuit is still being indirectly energised by the inverter, is the earth fault still dangerous?

Yes it is, especially if you are the earth. There’s no point in the RCD trying to save your life if the PV continues to pump current through you!

 

[littlespark]

If you came across PV on a shared RCD while doing an EICR, would that now be a C1 or C2?

My own house PV was put in 9 years ago now and is on a shared rcd. Might be time to upgrade.

 

[pc1966]

If you came across PV on a shared RCD while doing an EICR, would that now be a C1 or C2?

I would say C2 as its similar to a non-functioning RCD where one is needed (sockets that may be used outdoors, bathroom circuits)

My own house PV was put in 9 years ago now and is on a shared rcd. Might be time to upgrade.

Definitely!

Along with @James there seems to be a cobbler's convention looking at their own shoes

 

[spud1]

This is why PV should not be on a shared RCD. There could be a momentary hazard whereby the PV can still be generating line to earth under fault conditions until it collapses due to AC failure.

What you say makes absolute sense but then why, in Electrical Safety Firsts Best Practice guide 3 (See viii in BPG 3), do they say it is permissable to connect the dedicated feed in circuit to an RCD supplying other final circuits, if the RCD is double pole??

 

[littlespark]

I would say C2 as its similar to a non-functioning RCD where one is needed (sockets that may be used outdoors, bathroom circuits)

Definitely!

Along with @James there seems to be a cobbler's convention looking at their own shoes

The house itself is only 16 years old, so the lack of rcd on lights is a C3. I might just move the PV onto a dedicated RCBO within the board.

Some of the houses built on the later phase of building have rcd on multiple socket circuits like mine, but individual RCBO on other circuits…. Like it was an afterthought.

 

[pc1966]

What you say makes absolute sense but then why, in Electrical Safety Firsts Best Practice guide 3 (See viii in BPG 3), do they say it is permissible to connect the dedicated feed in circuit to an RCD supplying other final circuits, if the RCD is double pole??

Good point!

In terms of RCD being DP then it does make a big difference, but what you end up with immediately after disconnection is (hopefully) a single-faulted IT supply for the time it takes for the PV inverter to shut down. If you are luck enough to only have a single fault and no noticeable N-E path on the now-isolated side then you are protected. If there is anything on the N-E side below about 8k ohms then you are still at risk of a potentially fatal shock.

I don't have the 18th AM2 regs with me (yes, I'm such a travel bore!) but I was sure there is something now about this sort of situation beyond the previous regs not about RCDs having to still work as intended.

More generally, I would also worry that folks don't grasp the very serious nature of employing a SP RCBO in a similar situation (final circuit fed from the PV/energy source side) and it really not being protected during inverter shut-down.

 

[westward10]

The point is even with a double pole RCD it can feed back to the other circuits through it's protective device via the busbar.

 

[spud1]

Good point!

In terms of RCD being DP then it does make a big difference, but what you end up with immediately after disconnection is (hopefully) a single-faulted IT supply for the time it takes for the PV inverter to shut down. If you are luck enough to only have a single fault and no noticeable N-E path on the now-isolated side then you are protected. If there is anything on the N-E side below about 8k ohms then you are still at risk of a potentially fatal shock.

I don't have the 18th AM2 regs with me (yes, I'm such a travel bore!) but I was sure there is something now about this sort of situation beyond the previous regs not about RCDs having to still work as intended.

More generally, I would also worry that folks don't grasp the very serious nature of employing a SP RCBO in a similar situation (final circuit fed from the PV/energy source side) and it really not being protected during inverter shut-down.

Thanks pc great advice, really appreciate it

 

[spud1]

The point is even with a double pole RCD it can feed back to the other circuits through it's protective device via the busbar.

Thanks westward great advice as always