The IET is carrying out some important updates between 17-30 April and all of our websites will be view only. For more information, read this Announcement

This discussion is locked.
You cannot post a reply to this discussion. If you have a question start a new discussion

Solar supply on test reports

How do I document and test existing solar secondary supply when adding or modifying an installation that includes solar but not working on the solar system.

I have worked on a few installations now that have main supply plus solar, I have no training on solar installations.

I look to make sure cabling looks professional and that there are isolators on DC input to inverter and AC output from inverter, use them to isolate the supply while I work.

On the test report I record that there is a second supply and that it's solar but I don't do any testing or open anything up.

Is this enough or do I need to start learning more about solar and doing more?

I haven't worked on an installation with battery but same question.

Also if I was asked to do an EICR do I just agree with the owner that the EICR doesn't include the solar installation or do I need to decline the job until I get more knowledge on solar?

  • Actually "Island" modes, like generators and UPS's feeding final circuits as opposed to just one device,  all have the general problem that they may well be, and are intended to be, live when the mains supply is missing . At that point you have to assume that the supplier's earth is also missing - as it may be if the power is off due to an errant digger or something.  So you always need your own earth electrodes in parallel with the suppliers.

    Then what to do about N-E bonding is more complex than relying on the supplier to do it.

    If you add a bond at the inverter or generator, as you would in a stand-alone case. you may end up with two - one at your genset, and one from the supplier, or if you do not, then you may end up with none, if it happens the supply side fault also cuts the neutral off between you and the supplier's NE bond. In most single phase designs, simply operating the mains switch does this.

    Not having an earthed neutral may not matter for one item of equipment, it works OK for shaver sockets after all, but supplying into consumer units with an uncertain  mix of class1 and class II loads, and varying ADS it is not so simple. All but the largest solar installation is going to struggle to 'fire' large MCBs promptly on over-currents, and so the fault protection really relies on RCD when islanded, and that really does need an earthed neutral to work as intended.

    It is not safe to assume that all solar and similar installations have been done correctly with these factors in mind.

    Mike.

  • Agreed Mike.

    For clarity, all these points are covered in the IET CoP.

  • It'll need it's own local electrode, but the resulting system will almost certainly be TN-S (in island mode) rather than TT. It's the equivalent of source electrode that's being added rather than the consumer's one.

    Ideally TN-S, but there may still be requirements for certain parts of the installation which must remain on TT, as the installation will remain connected to PME - so PME conditions still apply (e.g. Section 722). Most importantly, in island mode, some types of open-PEN protection will cease operating effectively, yet there could still be a voltage on the PME earthing terminal.

  • The install standard on quite a few domestic PV installs is poor, and battery systems with the added complexity is even worse. When i was still installing solar I was asked to check some installs that were not working properly and found various issues,  AC connected as a spur of the ring, connected to an immersion and or lighting circuit at the fuse/mcb, T&E tails straight into the utility electric meter before going into a sub board and a couple where the PV was connected the wrong side of the utility meter so the user was being charged for their own generation. I did try to report 1 install to NAPIT as the test cert showed a sub board with RCD and B16 MCB had been fitted but in reality the AC circuit had been stabbed into the 30A BS3036 for the sockets, at the time they said the owner had to report it. The inverters weren't always set to the right countries either so not G83 compliant.