Who guards the guards?

The Fire Protection Association (FPA) and RISCAuthority conducted a research project on the tests for fire safety of external cladding systems.

Cladding system fires have been increasing globally, with the UK reporting the third most casualties per population since 1990. Combustible cladding systems are approved on high-rise buildings in the UK by the performance-based route to compliance, using the BS 8414 test method and BR 135 assessment criteria. This test involves installing a representative cladding system on a 10m-high test rig and subjecting it to a large fire (representing a room in flashover), venting out of a window and attacking the front face of the cladding system. Serious questions have been raised, however, regarding the appropriateness of the fuel source, test construction, construction detailing, assessment criteria and availability of test results.

The FPA, through the RISCAuthority, conducted a research project with the aim of enhancing the safety of tall buildings by preventing fire spread through external cladding systems. The project culminated in the development of RISC 501, a fire safety assessment test for external cladding systems. It is a large-scale fire test and assessment method that is designed to be conducted alongside BS 8414 so that the results can confer compliance with both BR 135 and RISC 501.

First things first

A database of cladding fires was built and populated with data from various fire tests of external cladding systems conducted according to the BS 8414 test method and evaluated against the BR 135 performance criteria. The results were then compared to the fire performance of cladding systems in the real world. The primary objective was to generate data that would serve as a foundation for critically evaluating the BS 8414 test methodology by exploring the strategies and interpretations employed to meet the BR 135 criteria.

The analysis identified several irregularities with the construction detailing between real buildings and the representative test set-ups. Unsurprisingly, all of these irregularities resulted in the test set-up being better protected from fire ingress and fire spread than the real buildings. During fire tests, the edges of the test rig would be sealed, preventing the natural ventilation seen on real buildings. This reduces the oxygen available for combustion and can have a significant impact on the growth of the fire.

There is a burn chamber at the bottom of the test rig, where the fuel load for the fire test is placed. The burn chamber opening was often overprotected using materials and methods that were not repeated on real buildings. This overprotection prevents fire from entering any cavities behind the cladding system and attacking the combustible insulation in a way that fire normally would – for example, through weaknesses such as windows.

The most severe form of overprotection came from the overuse of cavity barriers. Most systems contain cavity barriers and fire breaks to prevent fire spread through cladding systems and comply with Approved Document B. Clever interpretation of the test rig allowed for the use of five horizontal cavity barriers per test, when only two or three would be used over the same space on a real building. This will undoubtedly have a significant impact on the performance of the cladding system during a fire test.

Performance criteria

The BR 135 performance criteria specifies a maximum temperature limit that must be met during the first 15 minutes of the fire test. The number of combustible insulation products that would survive this 15-minute period before rapidly developing into a fully involved cladding fire is astounding. This gives the impression that they had been chemically engineered to last 15 minutes and no longer.

The scientific justification for this 15-minute assessment period has never been fully revealed and neither has the temperature limit, which is just over 600°C. It has been argued that 500°C would be a more appropriate limit for judging the combustion of typical cladding materials.

Even if this assessment of a rate of fire spread of 15 minutes per floor had some relevance to life safety, it clearly has no relevance to property protection. There are many cladding systems on the market that can prevent fire spread to multiple floors. There is no reason why this shouldn’t be the minimum requirement.

Creating RISC 501

RISC 501 was developed, specifying a test method and assessment criteria for the fire performance of external cladding systems. The method was designed to be conducted concurrently alongside BS 8414. Some of the notable differences include:

• added design specifications impacting floor heights and the position of cavity barrier
• added design specifications influencing the detailing surrounding the edges of the burn chamber opening and the test rig edges

• different criteria for temperature sensors

• different placement for temperature sensors

• additional requirements for gas sampling

• additional requirements for mechanical performance; and

• prerequisite for material characterisation.

The test construction criteria now mandate that all systems must design against the same virtual building. The virtual building is not actually constructed on the test rig, but exists to specify locations for floor heights and window openings (see diagram above). This ensures that all cladding systems tested are designed against the same building. This means that results are comparable, as well as preventing systems from being overprotected from fire spread by designing against an unrealistic building.

An impact assessment was conducted to identify the impact of lowering the temperature criteria and extending the measurement period. It was noted that with the current temperature sensor locations, many non-combustible systems would fail to meet stricter requirements (due to flame spread in the cavity). This suggests that the test would be too onerous, with unrealistic performance requirements. This was partly due to the close proximity between the temperature sensors and the flames from the burn chamber. The temperature sensors were therefore moved further away from the fire, allowing the temperature criteria to be lowered and the measurement period to be extended to the full duration of the test.

The virtual building also specified a floor height between the peak flame temperature and the temperature sensor locations. This meant that a typical cladding system would place a cavity barrier between the test fire and the temperature sensors, allowing for an assessment of the cladding system’s ability to prevent fire spread to multiple floors.

Other criteria

During real cladding fires, it is regularly reported that fallen material impedes firefighting operations, which will clearly have an impact on both life safety and property protection. Mechanical performance criteria were therefore set based on the performance requirements for firefighting helmets.

Smoke toxicity is an emerging issue across the entire fire industry, so criteria were set for extracting and measuring gases from the rear of the cladding system. No performance criteria were set, as further research is required on the relationship between the concentration of toxic products produced by a cladding system and the potential concentration of toxic products in an adjacent room inside the building.

To address the potential issue of manufacturers using superior performing products in fire tests, compared to those actually installed on buildings, a method for material characterisation was developed. Products can differ between fire tests and real-world applications for a variety of reasons:

• continuous improvement means products continually undergo alterations to match market changes. Minor adjustments might not mandate re-evaluation of fire ratings. However, when these alterations accumulate, they could result in significant changes. While significant changes should prompt retesting, this step might not always be taken
• test specials, when manufacturers might employ chemically-enhanced products exclusively for fire tests, deceiving testing facilities. These specialised products might be economically unfeasible for mass production, leading to cheaper, inferior products being sold on the market; and

• lack of precise specifications in the testing standards might permit significant differences between a product used in a fire test and its real-world application.

A method was therefore developed to produce a representative chemical fingerprint of the products tested. It used small-scale tests designed to identify significant chemical variations between products submitted for testing and those supplied to market. This method employs microscale combustion calorimetry and infrared spectroscopy to establish a unique chemical fingerprint for each major component of a cladding system. This approach was found to be capable of identifying new chemical substances; detecting the absence of existing ones; assessing changes in combustion properties; and quantifying process reactions, such as the formation of Isocyanurate rings (see diagram above).

Validation tests

Validation tests were conducted against the proposed RISC 501 standard to compare the results of the predicted performance of cladding systems during the test to their actual performance, with the aim of identifying:

• any physical impossibilities with the implementation of the prescribed test set-up, including construction detailing and sensor locations
• the test’s ability to identify flame spread through known combustible cladding systems

• the test’s ability to identify a lack of flame spread through known non-combustible cladding systems

• the test’s ability to identify mechanical failures; and

• the test’s ability to measure a significant difference in the smoke toxicity between a combustible and non-combustible system.

Ultimately, RISC 501 was found to clearly differentiate between systems that promoted fire spread and systems that did not, while offering a much clearer distinction than the BR 135 assessments using stricter performance criteria.

In summary, RISCAuthority conducted a study to investigate critical issues related to cladding system fires and the fire testing of external cladding systems, particularly in the UK, where combustible cladding systems are approved for high-rise buildings. The findings from that research were used to develop the fire test and assessment method RISC 501 – Fire Safety Assessment Test for External Cladding systems.

This article only covers some of the main changes, but there are a significant number of other changes mandated based on the extensive research project. The research behind this work will be shared with the British Standards Institute and the European Commission. 

Download RISC 501 – Fire Safety Assessment Test for External Cladding systems at bit.ly/FPA_guidance

Watch the FPA RISC 501 webinar at bit.ly/FPA_webinar

Image credit | FPA