earthing Array and TL inverters

DTI guide says if TNCS fit a spike, if TNS just bond to the MET (when there is a TL inverter) what is your opinion of this? The spike for TNCS is to prevent anyone off a ladder getting a shock or if the neutral is lost etc but we could maybe insist on the system being shut down for cleaning? Its because I have seen a lot of these installs with no earth at all

Also has anyone had issues with DC isolators not really being dc isolators and what should we be looking for to ensure that they are dc should we ask the installer for evidence?

This has been a topic of some debate in recent times, especially when a transformerless (TL) inverter is being used. In my travels as an MCS assessor, I have observed installers using this type of inverter more and more, because they are cheaper and more efficient that their Isolation Transformer counterparts. I have also conducted voltage measurements using a digital voltmeter from the array frame to building earth and have measured between 20 - 80 volts ac on the framework with reference to the building MET. Apparently, this ac voltage is due to an AC capacitive coupling within the inverter itself.

The question is does this charge on the framework represent a risk to someone working at roof height possibly off an aluminium ladder? I have raised the matter with the ESC and suggested that a body like ERA Technologies is engaged to conduct experiments to establish exactly the risks to personnel should someone touch the frame working off a ladder. Then the NICEIC could provide definitive guidance on the topic and at the same time encourage the IET to include regulations covering this matter in Section 712. This would be infinitely better than what is happening now with so much mis-information being circulated in the industry on the subject.

At the present time there are no regulatory requirements to enforce an installer to ground/earth the array framework because of this functional issue and on top of that the modules are invaribly Class II. However, in my opinion, this does not negate the risk that is present and I would consider it good practice to ground the framework where a transformerless inverter is being used. In actual fact, I advise installers to speak to the manufacturers' of both the modules and inverters and then based on engineering judgement whether or not to ground the framework. Incidently, I am using the term ground in preference to earth in order to differentiate between protective earthing and functional grounding, which may lead to a debate about the colour of such a conductor.

I have spoken to the technical department at Fronious Inverters and they recommend that the array frame is grounded where a transformerless inverter is used as do Sanyo - Solar PV module manufacturer's. European countries and others like Australia and the US require the framework of arrays to be grounded for the same reasons. In fact the original DTI Guide on the subject suggests on Page 23 that the array should be grounded where a transformerless inverter is used, although the new draft DTI guide has ommitted this from the document and suggests instead that the installer asks the Solar PV equipment manufacturer for guidance.

Regarding the array framework being grounded to a separate earth electrode, I personally think this is the best solution, as it would remove the ac capacitively coupled charge and also any dc staic charges that may be present. It would also remove the posibility of a charge being transported from the grounding conductor via the MET if the earthing system is TN-C-S or should the PEN condutor of the supply cable break.

Returning to Fronious inverters I recently spoke to a gentleman called Joseph Clark who is the most knowledgable person I have ever spoken to on Solar PV matters. If anyone would like to speak to him at Fronious they are on 01908 512316. Interestingly, Joseph said he would be prepared to speak at one of our AE forums on the subject, which I for one would welcome.

I Have come across installations which have used a spike, my question to the contractor was how had he tested the spike.
What do we think regarding the testing of the spike?

When installing a Solar PV system and the installer wishes to ground/earth the array to an independent earth electrode it would be sensible to run the grounding conductor to a test terminal (poss insulated Henly block suitably labelled) local to the origin of the installation. An earth loop impedance test could then be conducted between a non RCD protected line conductor and the disconnected earth electrode grounding lead.

In many cases the inverter supply will be fed from a dedicated DB which is not RCD protected. Following safe isolation (NICEIC Pocket Guide 5) a conventional earth loop impedance test could be done with the DB cover removed not forgetting to observe safe testing procedures ie, GS 38 lead set and suitable eye protection when conducting the test.

Of course there is another way to do the test, as described by Guidance Note 3 using an earth electrode resistance tester - Test Method 1 under Section 2.7.12.