Where the faults lie in fire safety process

At the RIBA regulations and standards group and the Construction Industry Coun­cil we’ve been wanting to tighten up fire safety regulation for the past two years; CIBSE and RICS have been discussing it too. The tragedy has pulled it all sharply into focus, with everything pointing to endemic, systemic problems in the industry. 

Part of that is due to the sheer volume of codes and guidance on fire issues. I’m on the Fire Sector Federation steering group with the DCLG representation advising on regulation but even I have trouble navigating Approved Document B (ADB). While there’s a lot of good technical guidance in its two volumes it’s a complex and confusing document; there’s a general feeling among architects that it’s not set out in a manner that allies with the chronology of the design process.

And Part B is only the start. BS9999 (Code of practice for fire safety in the design, management and use of buildings) is massive and largely impenetrable to the general architect. It seems it was written by experts to encourage the use of fire consultants rather than to help architects  design buildings. Currently routes to compliance are by meeting ADB, a testing route, or by fire engineering, usually a desktop study – this last being the most interpretive of them all. But the sheer volume of codes and the fact that they need to be cross-referenced is asking for trouble.

While fire engineering using BS9999 is  complicated, using  BS7974 (Application of fire safety engineering principles to the design of buildings) goes back to first principles of calculating escape distances, fire sizes, smoke development and ventilation, door widths etc, necessitating fire consultants. Cladding testing regimes are in BS8414-1 (Test methods for non-loadbearing external cladding systems applied to the face of a building).  Then there’s BR135 (Fire performance of external thermal insulation for walls of multi storey buildings). Some of this guidance, such as fire brigade access, needs to be referenced at masterplanning stage or at the outset of planning applications, not when you’re already committed to a design.

The ambiguity of ADB and dependent codes and the level to which they can be interpreted is deeply problematic and this has been emphasised by the Grenfell fire. The issue with its overcladding is the apparent ambiguity  in Section 12 of the ADB guidance on ‘limited combustibility’ and ‘non-combustibility’ of the insulation material for buildings over 18m. Even though the external rainscreen cladding could have had a combustibility factor, it could still have been code compliant within the whole wall build-up. The crucial point in guidance is, how limited is ‘limited’? That will be a significant focus of the criminal investigation and public inquiry. 

Class 0 fire resistance for external cladding means meeting BS476 Part 6. Tests have to prove that the material is of limited combustibility – but at higher temperatures most things burn, so that is where misinterpretation occurs. It’s clear the government initially misunderstood these complicated test criteria. If 243 tests have since been carried out on towers and they’ve all failed, it suggests to me that it’s not about incorrect specification per se but that they’re setting fire to cladding systems that are of ‘limited’ – and allowable – combustibility under the current code. Full scale, properly set up fire tests are needed.

Guidance on construction of non-loadbearing external walls is more about fire spread to adjoining buildings than over the surface of the building itself, which is a further misunderstanding. No  adjoining buildings caught fire at Grenfell, so in that regard it worked due to separation distance. The cladding is non-structural, so it’s not about structural integrity in fire, which again has not been compromised. This isn’t trying to make excuses for industry, it’s trying to show that guidance is easily misinterpreted.

The DCLG seems to have sat on its hands after the Lakanal House fire in 2009, not wanting to accept that ADB was not fit for purpose. Working backwards, it’s a good document for capturing perceived current technical guidance, highlighting life safety issues and prosecuting people, but not actually for designing buildings. Yet 80% of its users of are design professionals trying to understand and apply it. In that regard I champion deregulation by government, not a means of getting rid of good legislation but to pare it down to make it simpler and more accessible.

Flawed procurement

Problems are as manifest in procurement as in guidance. Architects increasingly rely on industry players to say they meet certain criteria. As performance specifiers we can only demand that it complies with regulation – it’s for the facade engineers, approved inspectors and the manufacturing industry to look into the specificities. 

You’re in a procurement chain of trust that assumes others are taking the responsibility of ensuring the right product or suite of products gets specified and constructed. This can, accidentally or due to financial pressure, be reduced to a base denominator of minimal acceptability at best.

Under design & build it is difficult to ensure compliance. Clients, funding bodies and banks may prefer it for its budgetary surety and contractor control but it has undermined the fire and safety compliance process because there is no clear chain of responsibility. Architects are not involved enough in final specification, and with the demise of district surveyors and clerks of works, sign-off of passive installations is left to third party inspectors paid by contractors or self-certifying approved installers. You might have begun with a fire consultant specifying but you can end up with a fire stopping package sub-let to approved installers and PU foam used to seal penetrations in compartments. This cannot continue. A non-partisan and informed role, such as that of the London district surveyors or local authority building control inspectors, must be reintroduced into the process, with a greater degree of independent checking by owner or commissioning client bodies.

This is the thinking that led to our Plan of Work proposal being worked up with the Association of Specialist Fire Protection, and industry supplier Hilti, aiming to pull together all parties in the fire engineering process and to assign roles and responsibilities for decision-making, specifying or signing off every stage of the RIBA Plan of Work. 

And that means everyone. Not just the main consultants and the  fire and facade engineers but the client’s agent, insurers, even sub-contractors.

It’s a holistic, chronological approach to procurement which acknowledges that commissioning clients and architects are generally the only ones who  hold a thread of continuity throughout the project. It means more methodical approaches to the design. That entails getting sign-off from the fire brigade for access and firefighting cores and deciding on the use of sprinklers or early warning systems at the outset. If insurers want asset safety beyond building regulation life safety criteria, they need to state it clearly and early. These aspects will then become non-negotiable as they affect occupation and escape distances. It will, further down the line, involve local authority as well as third party accreditation and sign-off for design and materials installation. With input and accountability from the whole team, it will finally end the vagaries that riddle the procurement process. 

Embedding the process

The Fire Plan of Work process needs to be embedded in the ADB and ideally needs to dictate the way the document is ordered to reflect the chronological process of design. It should start with fire brigade access, entrance points and stair core positions – but none of that appears until Section 16. By contrast, compartmentation, cavity barriers and fire stopping, often barely mooted before building control stages, is in Section 3. A colleague on the RIBA Expert Fire Panel feels Part B needs completely rewriting, but I think it just needs to be made more intuitive and comprehensible. That means refining technically, improving clarity and co-ordinating the sections better with the RIBA Plan of Work stages. The veracity of desktop fire assessments should be seriously addressed and it should mandate full-scale testing of cladding systems as they will be built.

Building Regulation 38, ensuring information critical to the life safety of people in and around the building is communicated to the owner or occupier so that buildings can be operated and managed correctly, needs to be enforced. The fact we’re going to have to design and audit the construction of every compartment service penetration in a building will horrify a lot of people but it just takes a change of mindset. BIM has to be the vehicle; it’s the only way to accurately change-register designed and completed buildings – but it will cost. 

Our industry has evolved to where it is today through a combination of societal expectations and clients’ and lawyers’ contractual and economic perspectives. British Standards help you achieve Part B but they are no roadmap to compliance. We need to re-evaluate that. That’s why I’ve also been looking at reworking the CDM Regulations. 

It’s clear that cladding products on tall buildings will probably have to be ‘non-combustible’ but concentrating on this distracts us from a necessary root and branch reappraisal of fire guidance. In future, we must obviate the need to answer crucial questions raised by the Grenfell Tower disaster. 

• Why was the fire not confined to the flat of origin? 
• Why didn’t residents get out in time as a two-hour rated fire stair should have allowed? 
• Why were residents told to stay inside a burning building when evacuation was imperative in a single-staircase building? 
• Was the escape stair maintained and how was whole building compartmentation compromised? 

The Fire Plan of Work would frame those questions and demand answers from every participant. Pre-empting his­tory, it would help us guard against repeating it.