Its one of those weeks. A simple job but..
We are installing a small G98 string inverter 3.6kw without any batteries its fed from 2 strings of 4 panels 450w @ 36.6v ea max 13.6 amps. 1 string is high up on a wall the other is at ground level. Th consumer unit is fully populated with type A rcbo's of varying sizes.
Earthing the inverter - it asks for a 6mm earth to the chassis direct from the main earth terminal.. so fine that's what the manufacturer has stated so thats what it gets.
Its also earthed back to the consumer unit internally with its 6mm Twin and earth ( all three conductors same size) on its feed in / out cable. This is connected to a Bi- directional 20amp RCBO, again what the manufacturer has ask for they get.
The supply earth purports to be a TN-S but is it really ? could be bonded out there on the way in.. we get a Ze of 0.34 from a 6mm earth cable clamped to the sheath.. so it could be. or there could be a big DNO stake out on the path... I have no idea.
So far so good... then its the panels..
As 1 string are most certainly at touch height there could be some capacitance voltage if the panels start to degrade and start leaking, which would be good to bleed off to ground.
Also consider if the transformer less inverter goes mad and shoves 240 out to the array. Whilst it says in the manual its 0v ac on the dc side... hmmmmm
So I have a bonding / earthing choice... there is an earth rod at the ground level , installed years ago for a long gone hot tub but its pretty sound giving 87 Ohms in the dry.
So do I;
ground the panels at the rod to dissipated any capacitance
or
Run the earth back to the MET - I know exporting the zone thus giving the inverter sight of any faults - if it needs that ?
Or
connect the rod and MET and panels together.
Couple of flow charts out there that say rod, met if not PME.. which would be handy testing the Insulation resistance periodical all done inside at night out of the wet.
Any ideas or thoughts.
We are installing a small G98 string inverter 3.6kw without any batteries its fed from 2 strings of 4 panels 450w @ 36.6v ea max 13.6 amps. 1 string is high up on a wall the other is at ground level. Th consumer unit is fully populated with type A rcbo's of varying sizes.
Earthing the inverter - it asks for a 6mm earth to the chassis direct from the main earth terminal.. so fine that's what the manufacturer has stated so thats what it gets.
Its also earthed back to the consumer unit internally with its 6mm Twin and earth ( all three conductors same size) on its feed in / out cable. This is connected to a Bi- directional 20amp RCBO, again what the manufacturer has ask for they get.
The supply earth purports to be a TN-S but is it really ? could be bonded out there on the way in.. we get a Ze of 0.34 from a 6mm earth cable clamped to the sheath.. so it could be. or there could be a big DNO stake out on the path... I have no idea.
So far so good... then its the panels..
As 1 string are most certainly at touch height there could be some capacitance voltage if the panels start to degrade and start leaking, which would be good to bleed off to ground.
Also consider if the transformer less inverter goes mad and shoves 240 out to the array. Whilst it says in the manual its 0v ac on the dc side... hmmmmm
So I have a bonding / earthing choice... there is an earth rod at the ground level , installed years ago for a long gone hot tub but its pretty sound giving 87 Ohms in the dry.
So do I;
ground the panels at the rod to dissipated any capacitance
or
Run the earth back to the MET - I know exporting the zone thus giving the inverter sight of any faults - if it needs that ?
Or
connect the rod and MET and panels together.
Couple of flow charts out there that say rod, met if not PME.. which would be handy testing the Insulation resistance periodical all done inside at night out of the wet.
Any ideas or thoughts.

