Skip to main content
Observer Ethical awards Winners 2011

YouGen Blog

Lightning strikes and solar PV

Posted by Cathy Debenham on 17 September 2012 at 9:36 am

YouGen reader Alan Drever lives on the south west coast of Isle of Skye. This part of Scotland has a higher than average lightning incidence, so when considering installation of a solar PV system it was something he wanted to take into consideration.

The risk of lightning strike is not just theoretical: he has had two fax machines knocked out since he moved there in 2003. Yet despite significant sleuthing before starting his solar PV installation he didn't find any information about potential lightning damage risk to PV. Nothing was mentioned by the Energy Saving Trust (EST) representative (who visited site) or his installer.

Alan had a 4kW solar PV system installed in May 2011. During following winter, an off-grid PV system in SW Skye suffered from lightning strike and the inverter was badly damaged. "This obviously sent alarm bells ringing!," says Alan. He takes up the story:

Risk and remedy

"I then embarked on what turned out to be an absolute saga trying to get to the bottom of risk level & safeguards!  

"First I contacted my installer, who offered to fit an earth spike from the array base into ground.  Christmas got in way of acting on this so I decided to consult others.  From discussions with EST, SMA (maker of Alan's Sunnyboy 4000TL inverter) BRE (Building Research Establishment), MCS (microgeneration certification scheme & CAT (the Centre for Alternative Technology)."

What follows is the advice he was given:

EST

EST based its advice on the Accredited Installers Handbook, Photovoltaics in Buildings: Guide to the Installation of PV Systems, which is described as the installers Bible. Apart from being long in tooth (2006 - a new edition is due this month - ed) requirements are not that clear! Installers are required to undertake site by site Risk Assessments (RA). Fitting of an earth spike (from the base of array into the ground) is the only measure specifically cited as common requirement.

SMA

SMA stated that surge protection should be installed on DC side (installer did only on AC side).  It indicated that varristors in the inverter are designed to protect it against surge damage from lightning strikes nearby, but not direct hits. The inverter warranty cover doesn't extend to any damage from this (bet house insurance wouldn't either methinks). So SMA recommends as standard that additional protection is fitted - regardless of the fact that this area experiences higher than UK average lightning frequency & house hillside location.  Measures involve surge arrestor fitted on DC side of AC Converter.  

BRE

Reluctant to engage unpaid.

MCS

Advised (via their PV Working Group) that an earth spike could heighten (my italics) lightning risk through potentially increasing strike risk. This has been refuted elsewhere on basis that it's not what lightning hits that matters, but where it ends up.  MCS Working Group were very defensive of the installer.

CAT

Advised that an earth spike from base of panel array is required; that it should not have more than 10 ohms resistance; and that it needs to be cross-bonded to mains electrical earth. The sequence of metal parts between panels base & ground (fitment) need to be in a particular order to minimise corrosion risk. Particular issue on west coast obviously.

Despite SMA's advice & warranty position, Alan's installer has been reluctant to install the necessary device to protect against DC surges.

In addition to lightning issues, he has experienced problems with system restart failure following power failure due to bad weather. 

"With 2 severe storms over Christmas, when I was away, a neighbour found that the PV system failed to restart on both occasions. The cut out device had to be manually switched back on.

"I have surmised (unsurprisingly) that this system should come back on automatically after mains power is restored. CAT had never heard of this happening & reckoned on cut off over-sensitivity.  SMA confirmed that system must be set too sensitively, suggesting 30 milli-amp breaker must be involved, when 100 milliamp is sufficient. My installer has been reluctant to substitute the circuit breaker."

Alan is keen to hear other people's experiences of solar PV and lightening, and also of any problems or solutions for inverters not coming back on after power failures. Please leave comments below.

Photo: Bo Insogna

By

If you have a question about anything in the above blog, please ask it in the comments section below.

Like this blog? Keep up to date with our free monthly newsletter

Comments

1 comments - read them below or add one

SunGift Energy

SunGift EnergyComment left on: 17 September 2012 at 1:15 pm

Hi Alan,

The amount of energy in a direct lightening strike is massive and unfortunately not something that can be readily protected against.

Luckily, if you can call it that, the majority of damage from lightening actually comes from induced voltages. In your case a nearby lightening strike could induce larger than normal voltages in the DC cables from your solar panels, and the AC wiring of your house or the power lines supplying it. In this case it would be advantageous to install surge arresters on both the AC and DC side of the inverter.

A surge arrester could also be specified to protect the whole of the property, if installed at the mains incoming position. This may protect you from further fax machine deaths!

Your PV array should be earth bonded unless the inverter installed provides separation of the AC and DC by means of an isolating transformer and class II insulated cables, connections, junction boxes and panels having been installed.

It is generally considered unsafe to take a PME earth outside. This is why a separate earth and earth spike/rod would be used. Therefore cross bonding to the mains electrical earth is not recommended with PME earth connections. It also would be another route for induced high voltages to be introduced into the household electrics.

Inverters by their nature do produce relatively large amounts of earth leakage, often protection by a 30mA RCBO or RCD may be too sensitive. This may not become apparent until later on in the inverters life when the inverters earth leakage may creep over the 30mA threshold. Alternatively certain environmental conditions may cause an increase which I feel is more likely the case in this instance. Changing the RCBO to a 100mA version should stop this nuisance tripping in the future.

Kevin

report abuse

Leave a comment

You must log in to make a comment. If you haven't already registered, please sign up as a company or an individual, then come back and have your say.