PV array earthed to MET

[Guy]

My partners mum has just had a pv install and it looks as if the array frame had been earthed to the MET. The fault current would be dc so is this ok?

 

[binky]

If you're mum has TN-c-s then this apparently needed. Though I never saw how it helps, but it's probably better than a live frame. Fault repairs and finding will be an interesting experience which I'm sure will escalate over the coming years

 

[Sidewinder]

Nice lightning magnet your mum now has on her roof!

 

[ivor dispatch]

Its in the DTI guide hereDepending on circumstances i personally disagree with it if the installation is designed and installed correctly with the correct kit

But thats for another debate, lets wait for Ivor :slap

Basically by doing so as SW states you now have a massive lightning attraction device on your roof....

Sorry pal, not enough info for me to start calling peeps names , I will leave it to you:consoling

 

[matt.leung]

A lot of solar frames do not need to be bonded unless you can touch them from the skylight window, plus most pannels now are also

class II so no earthing required...

this came from my MCS assessor

 

[Leeds Solar]

is it an extraneous conductive part? (test to verify)

if no, don't bond, if yes... is it in the equipotential zone?

if no, don't bond, if yes, bond to MET

so on houses it's virtually never needed, although DTI guide version 2 says different with TL inverters, version 3* says what I say above, and was written after TL inverters started using all pole RCD protection. The issue is related to capacitance between the panels and ground, but this is such a low level issue unless you're using certain types of thin film panels / metal backed panels that it's more dangerous to be bringing the equipotential zone to the roof than it is to risk a tiny tingle.

* version 3 is still in draft form, but the lead author has stated that it will not be changing and installers should work to it if their CP scheme allows it (which they all should if they have half a clue what they're doing).

Extraneous-conductive-part - A conductive part liable to introduce a potential, generally Earth potential, and not forming part of the electrical installation. .

On the d.c. side of the PV installation the designer will have usually already selected double or reinforced insulation as the protective measure

and therefore the component parts of the installation will be isolated from earth, the frame of the array has to be assessed as to whether it is

likely to introduce a potential into the installation. This aim of this assessment is to find out if the frame has any direct contact with ground that would make it introduce a potential.

The details on carrying out these tests are best given in the IET

 

[steptoe]

nice 1st post Leeds Solar,

---------- Post Auto-Merged at 19:41 ---------- Previous post was made at 19:40 ----------

and, NO, I wouldnt be happy with it bonded, unless it can be touched from INSIDE the EZ.

 

[Hefferl]

Sorry to drag this back up.

As far as I can see version 3 still remains unpublished.

Are others already following the advice in version 3 regarding bonding? i.e. if the frame is not extraneous (>22kΩ) then don't bond (even with TL inverters)?

 

[Hefferl]

See here for full list

http://www.microgenerationcertification.org/mcs-standards/installer-standards

This is the latest pv one

http://www.microgenerationcertification.org/images/PV%20Book%20ELECTRONIC.pdf

Its all in there

Thanks - that second document refers to latest "Guide to the Installation of Photovoltaic Systems" which I found a link to on the same site.

This latest version of the guide appears to use the same decision tree as the draft Version 3 quoted by Leeds Solar's earlier post which if I'm reading correctly confirms what I thought i.e. "if the frame is not extraneous (>22kΩ) then don't bond"