Should/shouldn't an inverter be connected via an RCD?

User avatar
Stinsy
Posts: 2740
Joined: Wed Jun 02, 2021 1:09 pm

Re: Should/shouldn't an inverter be connected via an RCD?

#11

Post by Stinsy »

Obviously it is a "good idea" to provide RCD protection to all circuits these days. Cost used to be a driver, but £3 for a MCB vs £30 for an RCBO shouldn't make a huge difference. The location of the RCD in a very large premises, particularly one consisting of several buildings, should be carefully considered.

However to be nit-picky there is nothing in the regs to say that devices such as inverters MUST be RCD-protected in all situations. Also "Type-AC" RCDs can be blinded by the inverter. So if several circuits share the same RCD and you add the inverter to that RCD, you could prevent it working and create a dangerous situation.

It is interesting that the MIs require a DP RCD. A few years ago these didn't exist in the UK market. While they are now available they are not standard and generally only used to protect equipment used outside the equipotential zone.
Last edited by Stinsy on Sat Oct 01, 2022 1:31 pm, edited 1 time in total.
12x 340W JA Solar panels (4.08kWp)
3x 380W JA Solar panels (1.14kWp)
5x 2.4kWh Pylontech batteries (12kWh)
LuxPower inverter/charger

(Artist formally known as ******, well it should be obvious enough to those for whom such things are important.)
ChrisJEvans
Posts: 34
Joined: Mon Sep 12, 2022 12:33 pm
Location: Sussex coast

Re: Should/shouldn't an inverter be connected via an RCD?

#12

Post by ChrisJEvans »

I don't know if it makes any difference but the supply is overhead. There is an earthing stake but we had a problem with a fridge that didn't seem to like the 70 odd ohm earth-neutral resistance, IIRC the electrician said up to 180ohm was with in regs. I consulted UK Power Networks who came and fitted PME earthing.

Sounds like the only downside is a small cost, and I'm not worried about that I just wanted to do the right thing.
15 x JA SOLAR 385W JAM60S20MR (9 facing East 6 facing South)
Solis S5-EH1P6K-L Hybrid Inverter.
4 x Pylontech US3000
Oldgreybeard
Posts: 1873
Joined: Thu Sep 09, 2021 3:42 pm
Location: North East Dorset

Re: Should/shouldn't an inverter be connected via an RCD?

#13

Post by Oldgreybeard »

Stinsy wrote: Sat Oct 01, 2022 1:27 pm Obviously it is a "good idea" to provide RCD protection to all circuits these days. Cost used to be a driver, but £3 for a MCB vs £30 for an RCBO shouldn't make a huge difference. The location of the RCD in a very large premises, particularly one consisting of several buildings, should be carefully considered.


RCBOs are a bit cheaper than £30, now. Bought some double pole compact Type A ones a few weeks ago, when our second emergency supply consumer unit went in. IIRC they were around £16 each. No real excuse not to fit them now, given the price has come down even further, IMHO.

Stinsy wrote: Sat Oct 01, 2022 1:27 pmIt is interesting that the MIs require a DP RCD. A few years ago these didn't exist in the UK market. While they are now available they are not standard and generally only used to protect equipment used outside the equipotential zone.
Our old house was rewired some time around 2000 and that had a double pole RCD protecting half the circuits in the consumer unit. I believe double pole RCDs have been around for a long time before that, too. The new thing seems to be compact double pole RCBOs. Surprised me how small and cheap they've got recently.
25 off 250W Perlight solar panels, installed 2014, with a 6kW PowerOne inverter, about 6,000kWh/year generated
6 off Pylontech US3000C batteries, with a Sofar ME3000SP inverter
sharpener
Posts: 419
Joined: Fri May 20, 2022 10:42 am

Re: Should/shouldn't an inverter be connected via an RCD?

#14

Post by sharpener »

ChrisJEvans wrote: Sat Oct 01, 2022 1:30 pm I don't know if it makes any difference but the supply is overhead. There is an earthing stake but we had a problem with a fridge that didn't seem to like the 70 odd ohm earth-neutral resistance, IIRC the electrician said up to 180ohm was with in regs. I consulted UK Power Networks who came and fitted PME earthing.
How did UKPN go about fitting PME to an overhead supply?

Unfortunately the RCD is nothing to do with considerations of protecting the cable or following the general provisions of Part P.

A dedicated RCD for the inverter is an express requirement of the IET CoP for EESS which is in turn incorporated into the MCS standard 3012 (and all MCS member firms are required to have a copy).

For a fuller discussion of the earth spike issue see this thread viewtopic.php?f=14&t=1290&p=18777#p18777, so your conversion to PME was unfortunately in vain.
16 x 230W Upsolar panels S Devon, 4kW Steca, 3.9 MWh FITs/yr
8 x 405W Longi panels, 250/60 MPPT, 3.3 MWh/yr
Victron MultiPlus II-GX 48/5000/70-50
10.65 kWh Pylontec Force-L2
zappi 7kW EVCS
Villavent whole-house MVHR
5000l rainwater system
Vaillant 12kW HP
Oldgreybeard
Posts: 1873
Joined: Thu Sep 09, 2021 3:42 pm
Location: North East Dorset

Re: Should/shouldn't an inverter be connected via an RCD?

#15

Post by Oldgreybeard »

sharpener wrote: Sat Oct 01, 2022 3:47 pm How did UKPN go about fitting PME to an overhead supply?
Just by using ABC: https://en.wikipedia.org/wiki/Aerial_bundled_cable They do the same with ABC as they used to do with singles, run a copper earth tape down one pole every now and again to an earth for the intermediates, maintaining the multiple neutral to earth bonds. Last time I spoke with a crew replacing poles they had some poles on their truck that already had earth electrodes on the base of the pole. Not seen them before, but it looked to be a good idea.

Before that they ran singles, but I think they stopped running singles a long time ago. ABC has the advantage of being very strong, stronger than the poles sometimes. A mixed blessing, as I drove along the A30 towards Salisbury a few years ago, after a fairly strong storm the day before, and there were a row of poles broken, but with the ABC still intact and apparently supplying power, even though it was part on the ground, part on the hedge, part supported by broken poles.

We have 95mm² ABC coming across overhead to us, then underground as 95mm² wavecon where it comes up to our meter kiosk, and could almost have been PME, had it not been for the cable run being just a bit too long, which pushed the loop impedance up above the allowable limit of 0.35 ohms. I tried chucking a double length earth electrode in in parallel, but Zs was still a bit too high. Does mean we have a nice low Ra value for our TT installation, though!
25 off 250W Perlight solar panels, installed 2014, with a 6kW PowerOne inverter, about 6,000kWh/year generated
6 off Pylontech US3000C batteries, with a Sofar ME3000SP inverter
ChrisJEvans
Posts: 34
Joined: Mon Sep 12, 2022 12:33 pm
Location: Sussex coast

Re: Should/shouldn't an inverter be connected via an RCD?

#16

Post by ChrisJEvans »

sharpener wrote: Sat Oct 01, 2022 3:47 pm
ChrisJEvans wrote: Sat Oct 01, 2022 1:30 pm I don't know if it makes any difference but the supply is overhead. There is an earthing stake but we had a problem with a fridge that didn't seem to like the 70 odd ohm earth-neutral resistance, IIRC the electrician said up to 180ohm was with in regs. I consulted UK Power Networks who came and fitted PME earthing.
How did UKPN go about fitting PME to an overhead supply?
as here:
Image
15 x JA SOLAR 385W JAM60S20MR (9 facing East 6 facing South)
Solis S5-EH1P6K-L Hybrid Inverter.
4 x Pylontech US3000
sharpener
Posts: 419
Joined: Fri May 20, 2022 10:42 am

Re: Should/shouldn't an inverter be connected via an RCD?

#17

Post by sharpener »

Did they already have an earth conductor on the overhead (as OGB implies, with ABC) - in which case it would be trivial - or did they have to add one or install a new overhead cable which is what I first imagined?

Not sure I understand what might have been wrong with the fridge in the first place, presumably you will have to fix it or get a new one now!

At 60 ohms my earth spike is similar to your 70 so I would say that is not something to worry about, I would start at the fridge end. BTW the limit is 200 ohms not 180.
16 x 230W Upsolar panels S Devon, 4kW Steca, 3.9 MWh FITs/yr
8 x 405W Longi panels, 250/60 MPPT, 3.3 MWh/yr
Victron MultiPlus II-GX 48/5000/70-50
10.65 kWh Pylontec Force-L2
zappi 7kW EVCS
Villavent whole-house MVHR
5000l rainwater system
Vaillant 12kW HP
Oldgreybeard
Posts: 1873
Joined: Thu Sep 09, 2021 3:42 pm
Location: North East Dorset

Re: Should/shouldn't an inverter be connected via an RCD?

#18

Post by Oldgreybeard »

sharpener wrote: Sun Oct 02, 2022 11:27 am Did they already have an earth conductor on the overhead (as OGB implies, with ABC) - in which case it would be trivial - or did they have to add one or install a new overhead cable which is what I first imagined?

Not sure I understand what might have been wrong with the fridge in the first place, presumably you will have to fix it or get a new one now!
No, there is no earth conductor with PME/TN-C-S. The only low voltage distribution system that use a separate earth conductor is TN-S, and that isn't that common now.

PME has no earth conductor, just a neutral and three phases, but that neutral is connected to earth at the transformer (via a ground mat electrode) and is usually also connected to intermediate earth electrodes all along the cable run (hence the "M" in PME).

ABC has a neutral, L1, L2 and L3 usually, with the neutral being a protective earth and neutral combined (PEN) for the PME installations it supplies. No compulsion to use the incoming PEN as the earth conductor, though. Our's is almost usable, IIRC I think the earth loop impedance came out at around 0.4 ohms, only a little bit over the maximum allowed for PME of 0.35 ohms. Just down to the length of the cable run and the fact that the combined live and neutral impedance was just a bit too high.
25 off 250W Perlight solar panels, installed 2014, with a 6kW PowerOne inverter, about 6,000kWh/year generated
6 off Pylontech US3000C batteries, with a Sofar ME3000SP inverter
sharpener
Posts: 419
Joined: Fri May 20, 2022 10:42 am

Re: Should/shouldn't an inverter be connected via an RCD?

#19

Post by sharpener »

My confusion, I had assumed somehow that the OP didn't have ABC because then it would have probably been TN-C-S all along. So I had supposed that if it was TT then it was overhead singles across the fields - like my grandparents' house 50 yrs ago where their RCD was always tripping when it was windy.

Maybe it had been TT originally because of the loop Z as in your case and they upgraded it somehow to the point they could make it TN-C-S, perhaps @ChrisJEvans will tell us.

When I worked in Germany I was surprised to see the houses in villages were daisy-chained with singles going from rooftop to rooftop rather than coming off poles in the street, looked very messy (and hazardous for roof repairs).
16 x 230W Upsolar panels S Devon, 4kW Steca, 3.9 MWh FITs/yr
8 x 405W Longi panels, 250/60 MPPT, 3.3 MWh/yr
Victron MultiPlus II-GX 48/5000/70-50
10.65 kWh Pylontec Force-L2
zappi 7kW EVCS
Villavent whole-house MVHR
5000l rainwater system
Vaillant 12kW HP
AGT
Posts: 949
Joined: Sat Jan 29, 2022 11:26 am

Re: Should/shouldn't an inverter be connected via an RCD?

#20

Post by AGT »

Do remember if it’s an existing consumer unit, you should fit the same manufacturers equipment inside not other makes.
Post Reply