Testing times

Ever since the appalling loss of life in the Grenfell Tower disaster, the UK construction industry has been under intense scrutiny, from the types of building materials that we use, to the quality of installation, the way we design, and how buildings are signed off.

Each of these important aspects have received due consideration in the Independent Review of Building Regulations and Fire Safety by Dame Judith Hackitt. At the time of writing this article publication of the full Review is imminent; however, we have already had a clear indication of the direction of travel in the Interim report published in December 2017.

Three of the most hotly debated topics in the months following Grenfell have been: the use of combustible materials in high rise or occupancy sensitive applications; the part played by large scale system testing in identifying safe constructions; and the use of desktop studies.

Combustible or non-combustible?

To the uninitiated it appears perfectly logical that the best way to reduce the risk of fire in buildings is to simply use ‘non-combustible’ materials. But fire is a complex phenomenon and needs to be treated as such if we are to truly make our buildings safer.

Let’s start by looking at how the fire performance of materials is classified in England & Wales. We have a series of National Classes that are defined in Building Regulations: namely “non-combustible”, “limited combustibility”, “Class 0”, “Class 1”, “Class 2” etc.

There are two systems of testing that can be used to justify one of these National Classes – one based on testing in accordance with the BS 476 suite of tests, and one based on a suite of European Standard test methods. These European tests are used to assign a Euroclass rating, which is equated to the National Classification system as follows:

British Standard Euroclass

Non-Combustible A1

Limited Combustibility A2

Class 0 B

Class 1 C

Whilst individual product testing is important to provide a baseline for performance, it has severe limitations. For example, the classifications of ‘limited combustibility’ and ‘non-combustible’ are arrived at using bench-scale tests on small samples of individual products with no consideration of how those products will interact with other components of a complete building element, such as a rainscreen façade.

The small-scale nature of tests to classify Class 0 (and worse) products may also not produce results that truly reflect overall performance. For example, a product with a thin facing which burns off quickly producing an initial brief spike in heat output, could achieve a poor Euroclass rating, even if it does not then go on to propagate a fire and may, in fact, help to resist fire spread.

For rainscreen facades, the most simplistic route to compliance with the Building Regulations, given in Approved Documents B, relies on performance according to these National Classes. Insulation and fillers must be of “limited combustibility” or better and the outside surfaces of the cladding system must also be Class 0 or Class 1 for defined heights and locations.

The main test used to determine whether a product achieves Euroclass B and thus also Class 0 or 1, was adopted on the basis that it was seen as a good small-scale proxy for the ISO 9705 large scale test. ISO 9705 is a test for flashover in a room caused by “wall and ceiling linings”. A ventilated façade presents a very different environment for fire, and a lining test is therefore inadequate to assess risk for facades.

Finally, it is almost impossible to build a façade that is truly non-combustible. Even when the insulation and cladding materials are classed as ‘non-combustible’ or of ‘limited combustibility’, combustible materials such as thermal breaks, sealants, vapour barriers and tapes can make up a surprising portion of the overall construction. It is the interaction between all these different components and how they are spatially arranged that will determine how a façade system behaves when exposed to fire.

Full system testing is therefore the best way to assess how any given combination of products will perform, regardless of whether the insulation and cladding is classed as ‘non-combustible’, ‘limited combustibility’ or not.

Large Scale System Testing

BR 135 ‘Fire performance of external thermal insulation for walls of multi-storey buildings’, is directly referred to in both Approved Document B (ADB2) and Technical Handbook 2 (THB2), and forms one of the routes to compliance with the Building Regulations. BR 135 sets out the pass/fail performance criteria for testing to BS8414-1 or 2.

According to the ADB2 and THB2, and BCA Technical Guidance Note 18: Issue 1 Jun 2015, products successfully tested to these standards, as part of a complete external cladding system, can be deemed compliant for the applications in which they have been tested. The key here is that the complete façade build-up is tested, rather than relying on the stand-alone performance of individual components; an approach which does not account for the overall design of the facade.

The BS8414 test is designed to replicate a fire starting inside a room, breaking out through a window and exposing the external façade to fire. The complete façade system is erected on an 8-metre test wall and exposed to flames from a combustion chamber at its base for 30 minutes. This is not a test of individual products. Only the complete external façade system that was tested can be deemed compliant via this test. It designed to test the rate and extent of fire spread within a façade system providing a clear assessment of how a fully installed insulated façade system will perform in the event of a fire.

BS8414 is a rigorous large-scale system test. It was developed specifically to test non-loadbearing external cladding systems. It provides a high level of confidence that systems that have passed are safe to use in high-rise and high-risk buildings. Recent research conducted by independent fire safety engineering consultants, Tenos, examined the severity of the BS8414 test and concluded that it was more onerous than the other internationally used large scale test, NFPA285. As part of the research, it was also discovered that no cladding system, that has passed or would likely pass BS8414, has ever made a significant contribution to the spread of a façade fire, that has been publicly reported on and that Tenos could uncover.

Lightweight, high-performing rigid phenolic insulation has featured in a significant number of BRE 135 classified systems, having successfully passed BS8414.

The BRE holds a public register of BR 135 classified external cladding systems, which can be viewed at: www.bre.co.uk/regulatory-testing.

More testing is being carried out all the time, and new results will not yet appear on the register. Contact manufacturers for more information and to view the full reports.

Desktop Studies

Façade systems that have been shown to meet the criterial given in BR135, when tested to BS8414, are considered to be compliant. However, the test results only apply to that precise design and specification. Changing details such as the cavity width, or brand of product, would require another test. With the vast array of available products and potential small variations in design, it would be impossible to test everything at that scale – there simply is not the testing capacity.

This is why desktop studies were introduced as another acceptable route to compliance. The issue with this approach is that there are no common rules on how they should be conducted.

A good solution might be to require that these judgements be made based on data from a standard intermediate-scale system test.

To find out how closely intermediate scale system tests mirror the results of large-scale tests in assessing the fire performance of whole systems, a series of tests were carried out using the existing standard ISO13785-1: 2002 (Reaction-to-fire tests for façades – Part 1: Intermediate-scale test.).

The tests used the same design and overall specification as the full-scale BS8414 tests run by the Government following the Grenfell Tower disaster.

The correlation between the two sets of tests was good, which shows that testing to ISO13785-1 may have promise in supporting the extension of scope of BS135 classifications.

Requiring desktop studies to include test data at an intermediate-scale on the actual system being analysed could make this route to compliance far more robust, but, critically, it is also realistic to achieve.

For more information, go to www.kingspaninsulation.co.uk