As the global community increasingly embraces sustainable practices and endeavors to achieve net zero emissions, the utilization of photovoltaic (PV) systems and their associated infrastructure has become commonplace. Some reflections on this was given in an opinion piece a few years ago.

However, a challenge has emerged alongside this proliferation, namely to ensure that the fire risk associated with rooftop PV installations. In this edition of Burning Matters, the subject of PV-related fire safety will be discussed, drawing upon three significant incidents as catalysts for analysis. Furthermore, insights and perspectives shared by followers and colleagues on LinkedIn are included, thereby enriching the discussion.

Important points from this fire that ended up damaging 5-600 panels:

1. The fire was quite big before it was reported. This warrants considerations for earlier detection, for example by monitoring the system performance or by using more conventional fire detection methods.

2. The fire spread across the membrane, which again confirms that membranes that pass the required fire tests for roofs will result in spread when covered with PV panels. This has been shown in laboratory tests for all types of membranes and for membranes with all Broof classifications.

3. The aisle separation distance was too small to prevent spread between aisles - both laterally and longitudinally. It would therefore be useful to carry out tests that can establish ‘fire breaks’ that work. This can be approached in different ways, and does not necessarily involve increasing the distances between aisles.

4. This was a new building, so there was no EPS on the roof, which reduced the fuel load. However, the mounting system had polypropylene footings, which contributed a significant amount of fuel to the fire. It is important to consider how all materials below the panels will influence the fire development.

5. The fire was contained to the roof and they could resume business after 3 or 4 days. Details of the in-depth damage is not available at this point, but it will be interesting to learn the construction and fire development details that created this outcome. In the case of retrofit with EPS present, this is more challenging and non-combustible mitigation layers on top of the EPS (if it cannot be removed) are highly recommended for a robust solution.

Based on research findings in literature and by the FRISSBE team, it is essential to test all planned solutions, and the tests should be on the system level (i.e., it should include panels and the entire planned roof composition), preferably with several panels. The recently published guideline has more details.

The reader is also encouraged to listen to the Fire Science Show episode with Jens Steemann Kristensen and Wojciech Węgrzyński.

The fire in Heerenveen, Netherlands on June 9, 2022, involved about 2000 PV panels and small fires continued the following day due to electricity activity associated with the panels.

1. Environmental concerns have been expressed for other PV fires - here is an article in PV magazine Global

2. I have been told (in the comments to the post) that the building was a total loss. Given the similarity to the fire in Peterborough, it would be interesting to compare the complete roof buildup and PV system in order to identify the key differences, and use these for improved constructions in the future.

3. Many comments related to maintenance being essential, which is very true. This adheres to the panels, the cables and accumulation of debris (e.g., from birds), among others.

4. The influence of the wind is often questioned. In strong winds, ignition will be more difficult, but a stronger wind will aid the fire spread later on in the fire. However, this is not something that can or should be the focus of the design. Ignition is more likely on a day with less wind, but once the fire spreads, the fire itself will result in air movement. Studies into how this air movement can be reduced could be worthwhile.

5. The comments also had the classic pointing to bitumen membranes being the problem, along with the type of panel. Research has shown that all types of membranes will result in spread below classic panel configurations, for all types of panels (and even for a metal plate substituted for panels).

The reader can see results with different panel geometries and the importance of a mitigation layer in a recent article by the FRISSBE team.

In April 2022, a solar panel damaged by birds triggered a fault in the electrical system, which then resulted in a fire.

1. The fire damage was contained to the roof and solar panels, but it took thousands of liters of hose water to extinguish the fire, which caused significant water damage.

2. The fire and the water damage has caused the building to be closed for more than 2 years - it is scheduled to open in the summer.

3. This fire highlights the need for firefighting considerations and the use of extinguishing medium. With a proper roof buildup, the amount of water can be minimized.

4. PV on all buildings means on all buildings - museums, schools, hospitals, ... - this means that each installation must consider what is below and how a fire and the ensuing firefighting will affect the building and the business.

5. Fire proofing of penetrations and cable systems are essential aspects for limiting damage from such fires. After this fire, six kilometers of wire had to be replaced.

Read more about the damages on this article from the museum.

In summary, three fires with very different outcomes - an in-depth comparison of the buildups could provide important details for improved designs in the future. It would be great if the fire investigation reports were made available.

Do you have any burning matters to discuss with Grunde? Reach out through the Burning Matters feedback form.

Interested in finding out more about the author? Check out this article 🙌🏻

Original LinkedIn Posts: