Elsevier

Fire Safety Journal

Volume 120, March 2021, 103139
Fire Safety Journal

PV modules on buildings – Outlines of PV roof samples fire rating assessment

https://doi.org/10.1016/j.firesaf.2020.103139Get rights and content

Highlights

  • Fire Safety of Photovoltaic System.

  • Review of PV module international test protocols.

  • A novel fire behavior test protocol for PV modules.

  • The assessment proposed test focuses on the fire behavior of the PV roof sample.

Abstract

The photovoltaic (PV) systems fire risk has grown up reaching a size that is no more negligible. PV fire events have happened mostly on systems installed on residential and industrial buildings and they have set in motion a process mainly aimed at developing product and installation regulations focused on the minimization of that risk. Such a process is growing up with the participation of different actors (authorities having jurisdiction, standardization bodies, PV modules manufacturers, installers, etc.). In this regard, in Italy a first result has been achieved by some new regulations issued by the Italian National Fire Rescue and Service – CNVVF – of the Ministry of Interior focusing on PV modules and roofs coupling, which contain procedures to categorize when a PV installation could be considered safe. To this end, such measures contain reminders to some national reaction to fire technical standards which, however, were developed in the past for products other than PV modules (e.g. wall panels, etc.). The research activity summarized in this paper was carried out in the last years and it was focused essentially on: analysis of Italian and European legal regulations and technical standards about the PV fire risk and identification of their shortcomings; development of new test protocols focused on PV roofs fire rating (based on some of the existing construction products harmonized standards). In particular, the test protocols are using PV roof samples with both PV modules and roof specimens.

Introduction

The rise in photovoltaic installations figures has increased over the last years in many countries (2015, global installed capacity 227,1 GWp, [1]), including Italy (e.g. end of 2015, Italy achieved a total installed capacity of 18.892 GWp, +0.298 GWp over the previous year, 688,398 systems in operation, [2]) and it has been accompanied by cases of accidental combustion whose number has been growing (2012, Italy: 479 interventions of Fire Departments [3] - Fig. 1; 2012, Germany: 390 fires of this type [4]).

Because of that, standard development processes started at International level (IEC TC 82 “Solar photovoltaic energy systems”), European level (CENELEC TC 82 “Solar photovoltaic energy systems” - WG01 “Wafers, cells and modules”) and National level (Italy: CEI TC 82 “Systems of photovoltaic conversion of solar energy” - WG11 “The fire risk in photovoltaic plants”, USA: UL) [5].

Furthermore, the result of a series of tests to investigate whether and how the presence of PV mounted array could affect the fire class of common roof material has been reported in Ref. [6]. The tests were conducted to examine combined effects of modules and roof coverings as a system when exposed to fire using the test rig configurations of UL 790 standard to evaluate different combinations of modules, stand-off heights, and roofing materials. The experimental study of the “Flammability Testing of Standard Roofing Products [6]” points out the increased temperature and heat flux due to “channeling effect” through which the PV module holds hot gases and flame closer to the roof surface; therefore the presence of either “Class C” or “Class A” modules mounted above “Class A” roof materials resulted in the roofing assemblies failing to meet the “Class A” spread of flame test requirements. In order to better understand the “channeling effect” of rack-mounted PV on the fire rating of roof assembly some experiments were conducted and reported in Ref. [7].

The tests in Ref. [7] have been focused on the determination of the minimum gap between the PV and the rooftop to maintain the roof covering original fire rating and to collect data on the effect of sheet metal interposed to block the passage of flames between the PV module and the roof assembly. Other experimental tests have been conducted to evaluate the PV module specific positions (low slope and steep slope) on roof decks during spread flame tests [8].

Meanwhile, several research programs have been formulated in order to support the standards development and the processes of manufacturing and installation. One of those is the program carried out by RSE S.p.A. as part of the research for the Italian Electrical System. Initially, in such research program, a brief analysis of the PV systems fire cases (about PV on buildings, but without BIPV systems) and a study of existing legislation and technical standards were performed.

The collaboration with standardization bodies such as CEI (Italian Electrotechnical Committee) and CENELEC (European Committee for Electrotechnical Standardization) was part of the activities during the program. In particular, about that, the CENELEC TR 50670: 2016 "External fire exposure to roofs in combination with photovoltaic (PV) arrays - Test methods" was published and it describes several tests focused on the PV module only and without any rating purposes. The activities subject of this test protocol were aimed to develop other test protocols, based on those contained in the CLC TR 50670 and with some fire rating criteria. Such protocols involve the use of PV module plus Roof samples installed within a testing room with some kind of "building roof type" geometries.

The testing room is already used in the context of construction products European legislation (Single Burning Item - SBI testing room; ref. CEN EN 13823:2010+A1:2014 “Reaction to fire tests for building products - Building products excluding floorings exposed to the thermal attack by a single burning item”). Furthermore, according to the collected test parameters, the fire behavior of the modules can be expressed using the same quantities applied for classifying reaction to the fire rating of conventional construction products. They allow the rating of module coupled with the coverage (roof) package, which, currently, do not exist in the legislation and standards published at European level, where tests are focused on fire spread feature. Fig. 2 illustrates some PV modules samples carrying out tests according to the following test protocols:

  • i)

    SBI (EN 13823, original version);

  • ii)

    new protocol with a flat roof version;

  • iii)

    new protocol with a sloped roof version.

Mainly, the new tests revealed the significant influence played by some features of the sample (PV plus roof) on the fire behavior characterization, in particular: sample inclination, initial/ignition flame (power, size, duration) and roof fire safety features. Moreover, these results are substantially in agreement with the results of the experimental campaign reported in Refs. [[6], [7], [8]].

Section snippets

Research activity

The research activity carried out was mainly based on an experimental program made-up by tests according to European Horizontal Specifications for building products (EN 13823:2010 "Reaction to fire tests for building products - Building products excluding floorings exposed to the thermal attack by a single burning item", i.e. “SBI”; EN ISO 11925-2:2010 + AC: 2011 "Reaction to fire tests - Ignitability of products Subjected to direct impingement of flame - Part 2: Single-flame source test", i.e.

Results

The PV samples considered during the experimental campaigns have been classified according to the Italian reaction to fire classification (class 0 to 5). These samples have been coupled with the European Regulation of Construction Products (CPR n. 305/2011) [11] classified substrates (Broof or Froof) acting as a roof in order to fulfill the new testing protocol configurations (see Fig. 3). Fig. 5 illustrates the test sequence of the fire test protocol SBI* b) configuration.

The sample technical

Conclusions

The experimental activity based on existing test protocols and on new tests (variants focused on PV features) allowed to verify the fire behavior of some PV modules in combination with different types of roof materials and it helped to identify some weaknesses of European and Italian testing protocols about fire behavior rating, mainly due to special design and outside installation features of PV modules (i.e. module inclination, ignition flame power and time duration, initial deterioration of

CRediT authorship contribution statement

Piergiacomo Cancelliere: Conceptualization, Data curation, Formal analysis, Funding acquisition, Validation, Visualization, Writing - original draft, Writing - review & editing. Giovanni Manzini: Conceptualization, Data curation, Formal analysis, Funding acquisition, Validation, Visualization, Methodology, Project administration, Resources. Giombattista Traina: Methodology, Project administration, Resources. Marco Gabriele Cavriani: Supervision.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The experimental program was carried out in collaboration with Politecnico di Milano (Dept. of Energy, FireLab, Milan, Italy) where some tests were performed; Brandoni Solare S.p.A. (Castelfidardo, Italy) and SUPSI-ISAAC (University of Applied Sciences and Arts of Southern Switzerland - Institute for Applied Sustainability to the Built Environment, Canobbio, Switzerland), which provided samples of PV modules; Istituto Giordano S.p.A. (Reaction to Fire and Electronics labs, Gatteo, Italy), which

References (11)

  • G. Manzini et al.

    The fire risk in photovoltaic installations - test protocols for fire behavior of PV modules

    (2015)
  • AA.VV.

    Snapshot of global photovoltaic market, Photovoltaic power systems programme

    (2015)
  • AA.VV.

    Rapporto Statistico 2015 – Solare Fotovoltaico

    (2016)
  • P. Cancelliere

    Ministero dell'Interno, Dipartimento dei vigili del fuoco, del Soccorso pubblico e della Difesa civile – Direzione centrale per la prevenzione e la sicurezza tecnica, Contenuti della “Guida per installazione impianti FV – Ed.

  • A. Kreutzmann et al.

    Passi avanti dei produttori nella prevenzione degli incendi – primi tra tutti gli Stati Uniti, Photon

    (June 2013)
There are more references available in the full text version of this article.

Cited by (8)

  • Fenestration integrated BIPV (FIPV): A review

    2022, Solar Energy
    Citation Excerpt :

    However, currently, most of the BIPV reports on power generation, heat loss and gain properties, semitransparency level. Noise protection and fire protection (Cancelliere et al., 2021) ability by BIPV is not a very popular research interest; nevertheless, they are also the most prominent sector where more exploration must be prepared. If BIPV windows are treated with replacement of traditional windows then snow (Borrebæk et al., 2020) and dust (Ghosh, 2020b) accumulation must not be ignored.

  • Failures of photo

    2022, Applied Energy
    Citation Excerpt :

    There are several fire resistant materials that can be used for module back sheet. For example, tedlar or polyester is more resistant to fire as compared to EVA encapsulant [63]. The EVA encapsulant is rapidly combustible and upon burning it emits gaseous fuels.

  • Solar Installations & Their Occupational Risks

    2021, Proceedings of the Human Factors and Ergonomics Society
View all citing articles on Scopus
View full text