Chapter 60: The new smoke ventilation system

¶60.1 If a fire breaks out in a residential building, it is likely that some smoke will enter the corridors or lobbies, for example, through the door of the flat on fire when the occupants open it to escape. The purpose of a smoke ventilation system is to protect the escape route. In a high-rise building that may involve preventing smoke from entering the stairs and it may also involve at the same time removing it from the corridors and lobbies, which will thereby also obtain a measure of protection.

¶60.2 There are different ways of achieving that goal. Natural ventilation systems provide a means for hot smoke to escape through its own buoyancy, typically through a shaft that is open to the atmosphere at the top and can act as a chimney.²³³⁵²³³⁵ Lane, Phase 1 Report, Appendix J {BLAS0000031/12}. Mechanical ventilation systems use fans, ducts, vents, shafts and other features to draw smoke away from the stair and common corridors.²³³⁶²³³⁶ Lane, Phase 1 Report, Appendix J {BLAS0000031/12}. In some cases mechanical means are used for both the inflow and exhaust functions (sometimes called a "push-pull" system); in others they may be used for the exhaust function alone (sometimes called a "mechanical extraction, natural inlet" system).²³³⁷²³³⁷ Lane, Phase 2 Module 3 Report {BLARP20000035/69} Sections 1-9 and 11.

¶60.3 The type of system originally installed in Grenfell Tower was a system based on natural ventilation, but with the addition of fans that firefighters could operate to assist the movement of air into and out of the building. If the fans were not operated, the system functioned as a natural inlet and outlet system.²³³⁸²³³⁸ Lane, Phase 1 Report, Appendix J {BLAR00000025/14}.

The original system

¶60.4 The original natural smoke ventilation system installed at Grenfell Tower consisted of fresh air shafts and smoke extraction shafts serving the lift lobbies on each residential floor. Each lobby had a pair of fresh air inlets at low level on the south side connected to a pair of shafts drawing air from an inlet above the Walkway level, and a pair of exhaust vents at high level on the north side connected to a pair of shafts venting to an outlet on the roof. The pair of fresh air inlet shafts each had an opening area of 0.24m2, giving a total inlet area of 0.48m2. The pair of exhaust shafts were the same size, giving a total exhaust area of 0.48m2. Each vent had a set of motorised dampers that kept the vents closed. If smoke was detected, actuators opened the dampers on the relevant floor, while the dampers on the other floors remained closed. The mechanical ventilation "boost" function was operated by means of a switch on the ground floor.

214

¶60.5 The following diagram shows how the system was intended to work:²³³⁹²³³⁹ Lane, Phase 1 Report Appendix J {BLAS0000031/18} Figure J.5.

¶Figure 60.1 – Operation of the original smoke ventilation system on the fire floor

¶60.6 Accordingly, the system operated by ventilating the lobby area with the intention of protecting the escape route through the lobby.

¶60.7 The system did not follow the recommendations of British Standard Code of Practice CP3 1971 (the standard which appears to have been used when designing the building),²³⁴⁰²³⁴⁰ Phase 1 Report Volume I paragraph 5.1. section 3.4.3.1(4) of which required that for buildings with a single stairway, cross ventilation in common access areas should be provided by an opening with a free area of at least 1.5m2.²³⁴¹²³⁴¹ Lane, Phase 1 Report, Appendix J {BLAR00000025/6}. The available aggregate free area of 0.48m2 at Grenfell Tower was significantly less than that.

The decision to refurbish

¶60.8 As set out in Chapter 43 of this report, the original smoke ventilation system at the tower had fallen into disrepair and had failed to work in a previous fire in April 2010. In March 2014, following a visit to the tower, the LFB issued a deficiency notice to the TMO stating that about a quarter of the system's dampers were not working²³⁴²²³⁴² LFB Deficiency Notice for Grenfell Tower dated 24 March 2014 {LFB00032101/3}. and by October 2014 the TMO was aware that the system was beyond repair.²³⁴³²³⁴³ Email from Alex Bosman to Claire Williams, Janice Wray and Carl Stokes dated 9 October 2014 {CST00001244/1}.

215

¶60.9 During 2011, the TMO's Assets and Regeneration Health and Safety Committee repeatedly discussed carrying out a feasibility study for the installation of a replacement smoke ventilation system.²³⁴⁴²³⁴⁴ Minutes of the Asset Investment and Engineering Health & Safety Group dated 3 February 2011 {TMO10000823}; Minutes of the Asset Investment and Engineering Health & Safety Group dated 31 March 2011 {TMO10000842}; Minutes of the Asset Investment and Engineering Health & Safety Group dated 26 May 2011 {TMO10000868}. At a meeting on 15 September 2011, the Asset Investment and Engineering Health and Safety Group decided that an invitation to tender for the work should be published immediately with a view to beginning work in December 2011.²³⁴⁵²³⁴⁵ Minutes of the Asset Investment and Engineering Health & Safety Group dated 15 September 2011 {TMO10000920/1}.

¶60.10 In October 2011 AECOM, an engineering consultancy firm, prepared a draft tender for the refurbishment of the dampers and the conversion of the manually activated mechanical boost function to a system activated automatically by the presence of smoke.²³⁴⁶²³⁴⁶ AECOM tender report {SEA00000004}. The existing system did not comply with the statutory guidance current at the time of the refurbishment because of the size of the shafts but the AECOM proposal would have made the existing system more robust, thereby satisfying the requirement in regulation 4(3) of the Building Regulations 2010 not to make the building any more unsatisfactory in that respect than it had previously been.²³⁴⁷²³⁴⁷ McQuatt {MAX00017292/19} page 19, paragraph 82.

¶60.11 At a design team meeting on 18 July 2012, the TMO instructed Max Fordham to expand the scope of the tower's refurbishment to include improving the smoke ventilation system, as AECOM had recommended.²³⁴⁸²³⁴⁸ Studio E Notes from Meeting 5 dated 18 July 2012 {MAX00000147}. Even if the existing smoke ventilation system had been working at that time, it would have been necessary to extend it to the new lobbies that were being created on the lower floors.²³⁴⁹²³⁴⁹ McQuatt {MAX00017292/17-18} pages 17-18, paragraph 71.

Exova's consideration of the smoke ventilation system

¶60.12 In the draft Existing Fire Safety Strategy for Grenfell Tower dated 16 August 2012 Exova summarised the position on smoke ventilation, noting that the existing shafts did not provide sufficient capacity to meet modern standards and recommending that they be ventilated by mechanical means.²³⁵⁰²³⁵⁰ Existing Fire Safety Strategy for Grenfell Tower dated 16 August 2012 {EXO00001074/10-11} section 3.4. The existing system was described as "significantly undersized".²³⁵¹²³⁵¹ Existing Fire Safety Strategy for Grenfell Tower dated 16 August 2012 {EXO00001074/11} fourth paragraph.

¶60.13 Exova also said that unless the fans of any new system were capable of producing flow rates equivalent to a compliant modern system, the improved system would not meet current standards.²³⁵²²³⁵² Existing Fire Safety Strategy for Grenfell Tower dated 16 August 2012 {EXO00001074/11} fourth paragraph. It pointed out the particular importance of a smoke ventilation system in a building that had travel distances between flat entrance doors and the doors to the stairs of more than 7.5m. The maximum distance was said by Exova to be approximately 8.3m.²³⁵³²³⁵³ Existing Fire Safety Strategy for Grenfell Tower dated 16 August 2012 {EXO00001074/10} paragraph 3.3.6. Exova said that provision of ventilation was critical to the tower's "stay in place" evacuation principle and it therefore strongly recommended that the performance of both the existing and proposed smoke ventilation systems be assessed.²³⁵⁴²³⁵⁴ Existing Fire Safety Strategy for Grenfell Tower dated 16 August 2012 {EXO00001074/11} fourth paragraph.

216

¶60.14 Minutes of a project meeting on 6 September 2012 record that Exova needed to understand the existing situation, including whether the LFB tested the smoke ventilation system twice a year and what lay behind the proposal to improve it. The minutes recorded that action would be taken by Exova.²³⁵⁵²³⁵⁵ Studio E Notes from Meeting 8 dated 6 September 2012 {MAX00000311/2} section "Fire".

¶60.15 On 10 September 2012 Cate Cooney of Exova sent an email to Terence Ashton stating that the existing smoke ventilation system was "questionable" and that adding additional residential floors to the building would make the existing condition worse. She noted that it was proposed to improve the smoke ventilation system but that it was not known what standard the new system would achieve.²³⁵⁶²³⁵⁶ Email from Cate Cooney to Terry Ashton dated 10 September 2012 {EXO00000388}.

¶60.16 In Issue 3 of the Outline Fire Safety Strategy dated 7 November 2013 reference to the smoke ventilation system was limited to noting that it would be overhauled as part of the refurbishment of the building and would be covered in a separate report by Max Fordham.²³⁵⁷²³⁵⁷ Exova Grenfell Tower Outline Fire Safety Strategy Issue 3 dated 7 November 2013 {EXO00001107/7} section "Smoke Ventilation of Lobbies, Walkway +1 Level". There was no further consideration of the system as part of the fire safety strategy.

Max Fordham's initial proposals

¶60.17 Max Fordham's initial proposals for the refurbishment of the smoke ventilation system took a similar approach to AECOM's and involved an improved version of the existing system.²³⁵⁸²³⁵⁸ McQuatt {MAX00017292/20} page 20, paragraph 85. In November 2012 Mr McQuatt sought advice on developing the design from Atrium Airconditioning Ltd,²³⁵⁹²³⁵⁹ Email from Andrew McQuatt to Atrium Airconditioning dated 8 November 2012 {MAX00003356}. which passed his enquiry to PSB, a specialist smoke ventilation company.

¶60.18 On 9 November 2012 PSB responded to Mr McQuatt saying that it did not see anything wrong with the proposed design.²³⁶⁰²³⁶⁰ Email from Fergus MacGregor to Andrew McQuatt and others dated 9 November 2012 {PSB00000002/2}. Following the consultation with PSB the design intent for the improved system was revised to bring it as close to the current regulations as possible within the limitations imposed by the existing shafts.²³⁶¹²³⁶¹ Cross Smith {MAX00017304/10-11} pages 10-11, paragraph 28. In addition to providing a working smoke ventilation system, it was intended to use the system to provide ventilation to the lobbies to prevent overheating.

RBKC Building Control

¶60.19 During 2013 there were continuing discussions between designers and other contractors engaged in the refurbishment of the tower about the modernisation of the smoke ventilation system. They included Max Fordham, Studio E, and Exova. Max Fordham's proposals for the new system did not undergo any changes during that period, but it proved difficult to demonstrate to building control that the new system would be an improvement on the existing system or at least would not make it any worse.²³⁶²²³⁶² Cross Smith {MAX00017304/13} page 13, paragraph 36. The fact that the existing system was not in working order and the paucity of documentation surrounding its installation made it difficult for the designers to compare its performance with the expected performance of any proposed replacement.

217

¶60.20 On 25 October 2013, Max Fordham repeated its proposals for the new system, stating that, as it was not possible to adapt the existing system to meet current standards, its intention was to bring the existing system up to as high a standard as possible.²³⁶³²³⁶³ Cross Smith {MAX00017304/10-11} pages 10-11, paragraph 28. At that stage, therefore, the proposal was still to improve the existing system rather than replace it with an entirely new one. Max Fordham acknowledged that there were no directly applicable standards that could be used as a basis for the design of the improved system but considered that it was reasonable to aim at providing 15 air changes an hour.²³⁶⁴²³⁶⁴ Max Fordham Grenfell Tower Smoke Control Proposals Rev A, dated 25 October 2013 {MAX00000867}. That may have been based on conversations with Exova.²³⁶⁵²³⁶⁵ Cross Smith {Day157/174:16-20}.

¶60.21 On 25 October 2013 Bruce Sounes of Studio E submitted Max Fordham's proposals to RBKC building control, together with the draft Outline Fire Safety Strategy for the building prepared by Exova. In his covering email Mr Sounes said that the design team thought that agreement on the smoke ventilation system was the single biggest risk to the proposals, but that they did not think it was reasonable to leave the existing system in place.²³⁶⁶²³⁶⁶ Email from Bruce Sounes to John Allen and Paul Hanson dated 25 October 2013 {SEA00000121}.

¶60.22 On 11 November 2013 John Allen, the building control manager at RBKC, told Mr Sounes that the information submitted by Studio E was not adequate to enable an effective consultation to be held with the LFB.²³⁶⁷²³⁶⁷ Email from John Allen to Bruce Sounes dated 11 November 2013 {SEA00009805}. However, he said that if it could be shown that the new system was no worse than the existing system, the design would be acceptable. He suggested that if there were no data available on the existing system, a way forward might be to measure the flow rates of the existing system and provide information about the capacity of the proposed new system. For the purpose of consulting the LFB Mr Allen suggested that information about the existing system be obtained, including the method of activation, the size of the shafts, and the powered extraction rate, that could be compared to similar information about the proposed improved system.²³⁶⁸²³⁶⁸ Email from John Allen to Bruce Sounes dated 11 November 2013 {SEA00009805}. In the light of that response Mr Cross Smith decided to contact PSB to obtain further specialist advice on the design of the new system.²³⁶⁹²³⁶⁹ Cross Smith {MAX00017304/14} page 14, paragraph 38.

¶60.23 On 6 December 2013, in a memorandum to Mr Allen, Paul Hanson, then responsible for the Means of Escape Group within RBKC building control, provided further observations on the proposals.²³⁷⁰²³⁷⁰ Memorandum from Paul Hanson to John Allen dated 6 December 2013 {RBK00003014}. He noted that it was not proposed to redesign the system to bring it up to modern standards, as that was likely to require larger shafts, but that building control had indicated that, provided the performance of the system was not made worse, the Building Regulations did not require it to be replaced. He suggested that a comparison of the flow rates of the system in its current and improved forms could provide a way forward. He did not think it was acceptable to base the design on a particular number of air changes an hour, as Max Fordham had proposed, as it did not correspond to any known guidance for lobby ventilation systems, being based on a generic value used for car parks.²³⁷¹²³⁷¹ Hanson {Day154/77:1-9}.

¶60.24 In his memorandum Mr Hanson also said that, if the designers wished to design the system to achieve a particular performance, they would need to provide evidence to support the proposed rate of extraction, including performance modelling. He suggested that a reasonable approximation of ventilation rates could be obtained using computational fluid dynamics (CFD) or a simpler zone model.²³⁷²²³⁷² Memorandum from Paul Hanson to John Allen dated 6 December 2013 {RBK00003014}.

218

The approach to PSB

¶60.25 At around the end of April 2014, Mr Cross Smith asked PSB to assist with satisfying building control's request for a revised submission. That led to a proposal for the use of a pressure differential system as a possible alternative²³⁷³²³⁷³ Cross Smith {MAX00017304/15-16} pages 15-16, paragraph 44. which was described in a document written by Hugh Mahoney of PSB, entitled Smoke Ventilation Technical Proposal For Stair De‑pressurisation Systems at Grenfell Tower, Regeneration Project, dated 22 April 2014.²³⁷⁴²³⁷⁴ Smoke Ventilation Technical Proposal for Stair De-Pressurisation Systems at Grenfell Tower, Regeneration Project {PSB00001233}. PSB also suggested a target extraction rate of 5.0m3/s for the system, which Mr Cross Smith incorporated into Max Fordham's latest proposals in May 2014. Apparently, that was the rate PSB always used for those purposes.²³⁷⁵²³⁷⁵ Cross Smith {MAX00017304/16} page 16, paragraph 47.

¶60.26 In late April or early May 2014, PSB and Max Fordham discussed CFD modelling, but by then there was insufficient time to carry it out, so it was not pursued.²³⁷⁶²³⁷⁶ Mahoney {Day155/163:20}-{Day155/164:13}; {Day155/21:5-22}; {Day155/26:10-25}; {Day155/32:14-24}. Simon Lay, one of the experts from whom we heard evidence, was of the opinion that, since the new system was being designed specifically for Grenfell Tower, Exova, Max Fordham and PSB ought to have ensured that a CFD analysis was carried out.²³⁷⁷²³⁷⁷ Lay {Day286/169:15-23}. He considered that in failing to do so they had fallen below the standards reasonably to be expected of them.²³⁷⁸²³⁷⁸ Lay {Day286/170:5-17}. Dr Lane also considered that Exova and Max Fordham should have made sure that a CFD analysis was carried out.²³⁷⁹²³⁷⁹ Lane {Day287/136:14-17}. She said that she would have expected PSB to advise that that should be done or to do it themselves if they had the capability, and rejected the suggestion that shortage of time was a valid reason for not doing it.²³⁸⁰²³⁸⁰ Lane {Day287/137:3}-{Day287/138:1}.

¶60.27 It remains unclear to us why a CFD analysis of the proposed new system was not carried out once it had become apparent that it was not possible to obtain reliable information about the performance of the existing system. If such an analysis had been carried out, it would have ensured that the performance of the new system had been ascertained with reasonable confidence and suitably documented. We agree that it should have been considered further and that PSB should have suggested it to the design team. Pressure of time was not a reasonable justification for failing to carry out such an analysis. The existing system had not been in working order for a long time and a further delay of weeks or even months was not a sufficient reason not to analyse properly the performance of the proposed replacement.

¶60.28 On 24 November 2014 at a meeting between building control, Studio E, the TMO and Max Fordham to discuss the system, PSB and JS Wright (Rydon's electrical and mechanical sub-contractor) put forward an alternative system, the outline of which was contained in a technical submission from PSB dated 12 November 2014 produced by Mr Mahoney.²³⁸¹²³⁸¹ Smoke Ventilation Technical Submission, Revision 0 dated 12 November 2014 {PSB00000207}.

219

¶60.29 Mr Mahoney understood that the essential purpose of the proposed work on the smoke control system was simply to make it better than the existing system.²³⁸²²³⁸² Mahoney {Day155/25:1-7}. In his view, provided it could be shown to deliver an extraction rate of 3m3/s, it would be acceptable to building control on the basis that that would be significantly better than the 1.2m3/s which was the estimated capacity of the existing system.²³⁸³²³⁸³ Mahoney {Day155/110:20}-{Day155/111:2}.

¶60.30 Mr Mahoney regarded the PSB design as an alternative approach. It did not follow any published guidance but would in his view be effective to protect the stairs and ventilate the lobby. He considered it impossible to install a system that complied in all respects with current standards.²³⁸⁴²³⁸⁴ Mahoney {Day155/86:5-12}.

The design of the new system

¶60.31 The design contained in PSB's technical submission was subsequently developed, culminating in Revision 6 issued on 15 March 2016 which contained the design of the system that was ultimately installed at Grenfell Tower. All versions except the final one were produced by Mr Mahoney.

¶60.32 It is not necessary to describe every aspect of the system's design because we are concerned only with those that could have played a part in the fire. In summary, the following are worth considering: a. The new system extracted air from the relevant lobby through all four vents and

discharged it either at the top of the building through the north shafts or at the level of Floor 2 through the south shafts. Air entered the lobby around the door to the stairs in response to the change in pressure. Mr Mahoney saw the extraction of air from the lobby at the rate of 3m3/s as the system's primary purpose,²³⁸⁵²³⁸⁵ Mahoney {Day155/90:21-24}. thereby moving air from the stairwell into the lobby and protecting it from the ingress of smoke. In other words, the objective was to remove smoke from the lobby to prevent it entering the stairs.²³⁸⁶²³⁸⁶ Hanson {Day154/135:1-7}.

¶b. According to PSB's technical submission, the system was designed to provide an average air velocity of 2m/s across an open door between the lobby and the stairs. That velocity was derived from BS EN 12101:6, which contains standards for pressure differential systems, but the system was not intended to work as a pressure differential system and was not designed in full compliance with that standard.²³⁸⁷²³⁸⁷ Smoke Ventilation Technical Submission, Revision 6 dated 15 March 2016 {PSB00000214/3}.

¶c. The system was activated by smoke detectors in the lobbies. When smoke was detected on a particular floor, the dampers on that floor were designed to remain open and the dampers on all other floors to close, allowing the full capacity of the system to extract air from the lobby in which smoke had been detected.²³⁸⁸²³⁸⁸ Smoke Ventilation Technical Submission, Revision 6 dated 15 March 2016 {PSB00000214/3}.

¶d. A new pair of smoke extraction fans were installed at Level 2 and also an environmental extraction fan.²³⁸⁹²³⁸⁹ Smoke Ventilation Technical Submission, Revision 6 dated 15 March 2016 {PSB00000214/4}.

220

¶e. Pressure sensors were also installed on every floor to control the speed of the extraction fans in order to ensure that the reduction in the air pressure in the lobbies did not prevent the door to the stairs from being opened. According to PSB's technical submission, if the pressure differential fell below -25 pascals ("Pa") (for example, if the door to the stairs were opened), the speed of the fans would increase until the differential of -25Pa had been restored.²³⁹⁰²³⁹⁰ Smoke Ventilation Technical Submission, Revision 6 dated 15 March 2016 {PSB00000214/18}.

¶f. New dampers were fitted to the openings in the south and north shafts on every floor. In addition, the new system included dampers within the ductwork on Level 2 to bypass the environmental ventilation ducts when the system was in smoke extraction mode. That reflected the dual purpose of the system: smoke control when smoke was detected and environmental ventilation when the building became too warm.²³⁹¹²³⁹¹ Smoke Ventilation Technical Submission, Revision 6 dated 15 March 2016 {PSB00000214/4}. When the system was operating in environmental mode, the dampers would open on four floors at a time for 15 minutes in sequence to cool the building down to the activation temperature.²³⁹²²³⁹² Lane, Phase 1 Report, Appendix J {BLAS0000031/87}.

¶60.33 Below is an illustration of the way in which the new smoke ventilation system was intended to operate. The blue lines indicate fresh air from the roof vent at the top of the stairs entering the lobby through the door to the stairs; the red lines indicate the movement of smoke and hot gases out of the shafts on the north and south sides of the lobby.²³⁹³²³⁹³ Lane, Phase 2 Module 3 Report, Sections 1-9 and 11 {BLARP20000035/260-261}.

¶Figure 60.2 - Design intent of the smoke extract system in Grenfell Tower for Level 2 to Level 23

221

¶60.34 Mr Mahoney said that the advantage of the design was that by using four shafts rather than two to extract smoke from the lobbies, the velocity required to achieve the required rate of extraction was reduced. In his opinion, seeking to achieve a higher flow rate using only two shafts for extraction and two for the entry of air was not viable, because the higher air speed required through each shaft would have disrupted the flow. It would also have risked creating velocity jets in the lobbies where air entered the vents, which might have caused smoke to enter the stairs.²³⁹⁴²³⁹⁴ Mahoney {Day155/82:3-16}.

Dampers

¶60.35 Dampers are mechanical devices used to prevent air or gases from passing through ducts or from the openings of ducts into the space beyond. Typically they take the form of louvred plates that overlap slightly to provide a solid barrier when in the closed position. In the open position they allow the passage of air and other gases. They are often opened or closed by an electric motor known as an actuator. Different kinds of dampers with different performance characteristics are available. It may therefore be helpful to begin this section by describing the ways in which dampers are classified and the purposes they are designed to serve.

Fire dampers

¶60.36 Fire dampers sit within a duct or ventilation opening and are usually operated automatically. They are designed to prevent the passage of fire.²³⁹⁵²³⁹⁵ Approved Document B {CLG00000224/144}. A damper is classed as a fire damper if it achieves an integrity classification of at least "E" as defined in BS EN13501- 3:2005 when tested in accordance with BS EN1366-2:1999.

¶60.37 The integrity classification "E" reflects the damper's ability to prevent the transmission of fire as a result of the passage of significant quantities of flames or hot gases from the fire to the unexposed side, thereby causing ignition either of the surface not exposed to the fire or of any material adjacent to that surface.²³⁹⁶²³⁹⁶ BS EN 13501-3 (Fire classification of construction products and building elements – Part 3) {BSI00000810/9} paragraph 5.1.2. Its purpose is to prevent the ignition of whatever is on the other side of the damper rather than to restrict the amount of smoke adversely affecting the adjoining space. A fire damper, therefore, is expected to close in the event of a fire and remain closed during it;²³⁹⁷²³⁹⁷ Lane, Phase 2 Module 3 Report {BLARP20000035/86}. it is not intended to open and close repeatedly as part of a system's normal operation.

Fire and smoke dampers

¶60.38 Fire and smoke dampers are tested and classified by reference to their ability to limit the amount of smoke passing from one side to the other.²³⁹⁸²³⁹⁸ Approved Document B {CLG00000224/144}. Such dampers must have been tested in accordance with BS EN 1366-2:1999 and meet both the integrity and smoke leakage criteria in EN 13501-3:2005. They are then classified "ES". The function of a fire and smoke damper is also to close and remain closed during a fire,²³⁹⁹²³⁹⁹ Lane, Phase 2 Module 3 Report {BLARP20000035/88} Sections 1-9 and 11. but it must have a lower leakage rate than a fire damper.

222

¶60.39 The ability of a damper to minimise the leakage of smoke was directly relevant to its suitability for use in the smoke ventilation system of the tower, because the vertical shafts used to extract smoke passed through the lobbies, which were part of the escape route for the occupants of the flats on those floors. A damper that prevented the passage of fire alone would not have been suitable.

¶60.40 The performance necessary to achieve the relevant classifications are set out as follows in EN 13501-3:2005:²⁴⁰⁰²⁴⁰⁰ BS EN 13501-3 (Fire classification of construction products and building elements – Part 3) {BSI00000810/16}. a. For the classification "E" the leakage limit is 360 m³/h∙m²; b. For the classification "ES" the leakage limit is 200m³/h∙m² (corrected to 20°C). (The limit is the same at ambient temperature and during the fire test.)

Smoke control dampers

¶60.41 Smoke control dampers are described in BS EN 1366-10:2011²⁴⁰¹²⁴⁰¹ BS EN 1366-10:2011 (Fire resistance tests for service installations – Part 10: Smoke control dampers) {BSI00001777/12} paragraph 3.27. ("Fire resistance tests for service installations, Part 10 Smoke Control Dampers") rather than in Approved Document B. A smoke control damper is a similar device that can be automatically or manually opened or closed in its operational position to control the flow of smoke and hot gases into, from or within a duct.²⁴⁰²²⁴⁰² Lane, Phase 2 Module 3 Report {BLARP20000035/90} Sections 1-9 and 11.

¶60.42 A smoke control damper is subject to the same performance requirements as a fire and smoke damper, but differs from the previous two types of damper in that it is capable of both opening and closing during a fire.²⁴⁰³²⁴⁰³ BS EN 12101-8:2011 (Smoke and heat control systems – Part 8: Smoke control dampers) {BSI00000048/7}.

The design of the system

¶60.43 The shafts and ductwork for the smoke ventilation system were required to be in a protected area, as described in Approved Document B, in order to restrict the ability of fire and smoke to spread between compartments.²⁴⁰⁴²⁴⁰⁴ Approved Document B {CLG00000224/73} paragraph 8.9. The effectiveness of the dampers in preventing excessive smoke leakage from the shafts into lobbies on floors other than the one affected by a fire was therefore an important feature of the system. Leakage through the dampers also adversely affects the ability of the fans to extract smoke and thus the performance of the system as a whole.²⁴⁰⁵²⁴⁰⁵ Lane, Phase 2 Module 7 Supplementary Report {BLARP20000043/16}.

¶60.44 The design of the new smoke ventilation system envisaged that dampers might be changed from the closed to the open position during a fire, because the manual control function allowed firefighters to change the floor from which smoke was being extracted as required by the location of the fire. Accordingly, a damper that had previously shut on one floor might later be opened manually if the fire brigade required the system to extract smoke from another floor.

¶60.45 Only a smoke control damper is designed and tested to perform in that way. The key differences between the testing of smoke control dampers and the testing of the other two kinds of damper relate to:²⁴⁰⁶²⁴⁰⁶ Lane, Phase 2 Module 3 Report {BLARP20000035/92-93} Sections 1-9 and 11.

223

¶a. the durability of operational reliability, which is tested by carrying out a minimum number of opening and closing cycles before fire testing – 10,200 for smoke control dampers but only 50 for fire dampers and fire and smoke dampers; and

¶b. the ability to change from open to closed position and vice versa at elevated temperatures, which is not a requirement for fire dampers or fire and smoke dampers.

¶60.46 In this case durability of operational reliability was important because the dampers were part of a system that also provided environmental ventilation. They could therefore reasonably be expected to open and close many times each week. As a result of having been tested through over 10,000 cycles, smoke control dampers can be expected to be effective over a much longer operational lifetime.

¶60.47 The ability to move between an open and closed position at elevated temperatures was also of importance in this case because firefighters might need to control the system manually if the temperature on one floor rose considerably due to the fire.

¶60.48 Having regard to the properties of the three kinds of damper we have described, it is clear that only a smoke control damper could be expected to meet the requirements of the new system.

The Gilberts Series 54 damper

¶60.49 The Gilberts Series 54 damper is shown in the open and closed positions in the following photographs taken after the fire.²⁴⁰⁷²⁴⁰⁷ Lane, Phase 2 Module 3 Report {BLARP20000035/296}.

¶Figure 60.3 – Examples of dampers in an open and closed position

¶60.50 It is opened and closed by an actuator (a small electric motor)²⁴⁰⁸²⁴⁰⁸ Lane, Phase 2 Module 3 Report {BLARP20000035/297} Sections 1-9 and 11. that moves the blades. When smoke is being drawn up the shaft the damper prevents it from entering the lobbies. It therefore plays a vital role in protecting the atmosphere in the lobbies not directly affected by the fire.

224

¶60.51 The damper was described in a brochure published by the manufacturer in October 2011²⁴⁰⁹²⁴⁰⁹ Gilberts, Series 54 Smoke Evacuation Damper brochure {PSB00000201}. as a "smoke evacuation damper", a description not found in any published guidance. The brochure also stated that it had been fully tested to the requirements of EN1366 Part 2 for one hour.²⁴¹⁰²⁴¹⁰ Gilberts, Series 54 Smoke Evacuation Damper brochure {PSB00000201/2}. However, BS EN 1366-2 contains the standard for testing fire dampers and fire and smoke dampers; it does not contain the standard for testing smoke control dampers.²⁴¹¹²⁴¹¹ Lane, Phase 2 Module 3 Report {BLARP20000035/93} Sections 1-9 and 11. Moreover, the damper had been tested from the closed, rather than the open, position, as the notes attached to the original quotation confirmed. The statement in the brochure was therefore inaccurate and misleading, because the test method to which it referred requires testing from the open position. There was no suggestion by Gilberts in the brochure or otherwise that the damper was a smoke control damper in the recognised sense.

The choice of the Gilberts Series 54 damper

¶60.52 In January 2015 Hugh Mahoney asked Gilberts to quote for the supply of Series 54 dampers for use in a smoke ventilation system. At that stage he did not provide any information about the system in which they were to be used.

¶60.53 The quotation that Gilberts sent to Mr Mahoney on 22 January 2015 contained a note stating that the damper had undergone an EN 1366-2 test starting from the closed position which had lasted over 60 minutes for both fire integrity and smoke leakage (ES60) but that it had no formal certification.²⁴¹²²⁴¹² Quotation provided by Gilberts to PSB for Series 54 Dampers dated 22 January 2015 {GBL00000006/3}. (As we have said, the claim in respect of smoke leakage was wrong because the test had been started from the closed position.) The information that the test had been started from the closed position would have disclosed to a reasonably knowledgeable reader that the test had not been carried out entirely in accordance with the published standard, a conclusion reinforced by the reference to the absence of formal certification.

¶60.54 Another note recommended that the purchaser discuss the specific technical requirements of the damper with the relevant authority to ensure that it was acceptable but there is no evidence that any of the contractors involved in designing or installing the smoke control system discussed the matter with RBKC building control.

¶60.55 After receiving the original quotation from Gilberts, Mr Mahoney asked for further information about the Series 54 dampers, including test results, which were sent to him by email.²⁴¹³²⁴¹³ Email from Mark Griffiths to Hugh Mahoney {GBL00000005/2}. One of the documents attached to the email was a report dated 6 October 2011 of a fire resistance test on a Series 54 damper carried out by Exova on 23 August 2011 in accordance with BS EN 1366-2.²⁴¹⁴²⁴¹⁴ Email from Mark Griffiths to Hugh Mahoney {GBL00000005/2-3}. The report showed that the damper had been subjected to an integrity test with a threshold of 360m3/h∙m2 for over 60 minutes under a pressure of about 300 Pa.²⁴¹⁵²⁴¹⁵ Email from Mark Griffiths to Hugh Mahoney {GBL00000005/42}. The damper satisfied that test, with satisfactory performance for 74 minutes, but it failed to satisfy the requirement in relation to the leakage of smoke, for which the performance criterion was not to exceed 200m3/h∙m2 under a pressure of about 300 Pa.

225

¶60.56 In the Conclusions section of the report the test result for smoke leakage was recorded as "0 minutes", which represented an immediate failure to satisfy the performance requirement.²⁴¹⁶²⁴¹⁶ Email from Mark Griffiths to Hugh Mahoney {GBL00000005/43}. In the Test Procedure section,²⁴¹⁷²⁴¹⁷ Email from Mark Griffiths to Hugh Mahoney {GBL00000005/8}. Exova recorded that the specimen had been assessed on its ability to comply with the performance criteria for integrity, insulation and smoke leakage as required by BS EN 1366-2:1999 but that at Gilberts' request the damper had been in the closed position at the commencement of the test and that the test had therefore not been conducted fully in accordance with the standard. As a result, it was not possible for the damper to obtain any kind of certification.

¶60.57 In spite of the incorrect and misleading statements about certification and testing in Gilberts' 2011 brochure and its first quotation, any recipient of Exova's report who carried out even a cursory review of those results would have understood that the Series 54 damper held no certification whatsoever and had failed the smoke leakage test immediately. The recipients of that report included Mr Mahoney and, in due course, RBKC building control.

¶60.58 Mr Mahoney appears to have thought that a certified smoke control damper of a size and configuration suitable for use in the walls at Grenfell Tower was not available.²⁴¹⁸²⁴¹⁸ Mahoney {Day155/203:20}-{Day151/205:2}. He thought that the Gilbert's dampers were the best that could be found to fit within the wall of the tower.²⁴¹⁹²⁴¹⁹ Mahoney {Day155/213:20}-{Day155/214:1}. In any event, PSB chose to specify the use of the Gilberts Series 54 damper, despite the fact that it had not been classified in accordance with any of the standards mentioned earlier and had not been shown to meet the alternative specifications for dampers to be used in smoke control systems contained in the guides published by the Smoke Control Association. He could instead have specified performance criteria or a certification standard, such as ES60, against which a search of available products could have been made. Having carried out a limited internet archive search (including of the archived version of the LPCB²⁴²⁰²⁴²⁰ Loss Prevention Certification Board. Redbook dated 23 August 2014) for dampers available on the market at the time, Dr Lane identified two smoke control dampers that might have been suitable for use in the system,²⁴²¹²⁴²¹ Lane, Addenda and Errata Report {BLARP20000044/4-5}. although she acknowledged that she did not have access to enough information to tell whether either of them could in fact have been used.

¶60.59 Mr Mahoney was aware when he designed the system that the shafts had to be protected²⁴²²²⁴²² Mahoney {Day155/198:14-23}. and he discussed with Mr Cross Smith the need to have fire-rated ductwork,²⁴²³²⁴²³ Mahoney {Day155/200:2-7}. but he did not do enough to ensure that the dampers he selected to separate the protected shafts from the lobbies were fit for the purpose of protecting escape routes. Apart from anything else, in the absence of test data evidencing their durability there could be no confidence that the dampers would perform properly in a fire after many months of use in a combined environmental and smoke ventilation system.

226

¶60.60 In due course, as a result of changes to the sizes of some of the dampers required, J S Wright asked Gilberts for another quotation.²⁴²⁴²⁴²⁴ Email from David Bradbury to Mark Griffiths {JSW00003607}. Following the receipt of that quotation, J S Wright placed an order for the products which Gilberts supplied.²⁴²⁵²⁴²⁵ Jones {GBL00000010/3} page 3, paragraph 10. Gilberts did not provide advice on the suitability of the Series 54 dampers for the proposed system, about which it had little or no information.²⁴²⁶²⁴²⁶ Jones {GBL00000010/3} page 3, paragraphs 11-17; {GBL00000010/4} page 4, paragraph 21.

¶60.61 Mr Hanson said that he had checked the specification of the Gilberts Series 54 damper²⁴²⁷²⁴²⁷ Hanson {Day154/204:5-8}. and had concluded on the basis of the product literature that it was regarded as being of an acceptable standard by the Smoke Control Association Guide 2012.²⁴²⁸²⁴²⁸ Hanson {Day154/208:12-15}. He said he could see from the drawings that the ducts were adequately fire-resistant.²⁴²⁹²⁴²⁹ Hanson {Day154/210:1-2}. He accepted that the dampers were of a lower standard than smoke control dampers but, in his opinion, they were still of an acceptable standard.²⁴³⁰²⁴³⁰ Hanson {Day154/214:10-15}.

¶60.62 The Gilbert Series 54 damper had not been tested in a way that demonstrated it was capable of acting as anything more than a ventilator, which required the lowest standard of performance in the event of a fire.²⁴³¹²⁴³¹ Lane, Phase 2 Module 7 Supplementary Report {BLARP20000043/264}. It was far removed from a smoke control damper, which would have been the appropriate product to install in a complex and multi-functional system of the kind designed by PSB. Although it received a copy of the Exova report, PSB does not appear to have reviewed the specification of the dampers or to have discussed it with building control.²⁴³²²⁴³² Lay, Smoke Control System Report {LAY00000001/201}. It should have done so, but we have seen nothing to suggest that the implications of Exova's report were identified or acted upon. PSB should have realised that the Gilberts Series 54 damper was unsuitable for this application.

¶60.63 We do not know whether in 2015 fire and smoke dampers were available that would have been suitable for use in the system, but we are surprised that PSB does not appear to have tried to find appropriately certified dampers. At the very least, it should have told J S Wright about the limitations of the Series 54 damper so that the risks of using it could be properly assessed.

¶60.64 Mr Lay was of the opinion that, although the Series 54 dampers were not certified and did not meet the alternative Smoke Control Association recommendations, their performance characteristics could allow them to perform adequately as part of the system.²⁴³³²⁴³³ Lay, Smoke Control System Report {LAY00000001/203}. However, we do not agree. The Series 54 dampers did not meet the requirements for smoke leakage, which was important for protecting routes of escape, and had not been shown to meet the durability requirements. In the absence of reliable test results (i.e. results of tests conducted in accordance with recognised standards) from which to draw inferences, we cannot safely make any findings about how a component might have performed. The point of testing standards is to provide an assurance that a product meets the specified performance criteria and no such assurance was provided in relation to the Gilberts Series 54 dampers.

227

Effect on the system's performance

¶60.65 It is possible that the use of the Gilberts Series 54 dampers contributed to the breach of compartmentation by permitting more smoke to pass from the shafts into some of the lobbies than would have been the case if appropriately certified dampers had been used, but it is impossible now to tell whether that was in fact the case. However, circumstances surrounding the choice of dampers demonstrates insufficient concern for fire safety to which we think it important to draw attention as part of the events which led to the tragedy.

Other shortcomings in the design process

The lack of design records

¶60.66 A recurring theme of our investigation into the design of the smoke ventilation system was a failure to create a clear record of its objectives and how its performance characteristics related to the fire strategy for the building. The importance of creating such a record was made clear in the 2012 version of the Smoke Control Association's guide, which strongly recommended that, except perhaps in the simplest cases, the objectives of the system, the circumstances to be calculated or modelled, the modelling criteria, the expected reporting and the criteria for success should all be agreed and recorded before the commencement of design.²⁴³⁴²⁴³⁴ Smoke Control Association Guide 2012 {LFB00059241/5} Introduction and fifth paragraph.

¶60.67 The Smoke Control Association Guide 2012 advised that the records should include at least: i. A description of the residential area and the proposed ventilation system. ii. The design criteria and performance objectives. iii. The range of circumstance in which it was intended to operate. iv. Details of the techniques used and related information. v. The results of the analysis. vi. A statement whether the design criteria and objectives have been met.²⁴³⁵²⁴³⁵ Smoke Control Association Guide 2012 {LFB00059241/14-15} paragraph 5.4.

¶60.68 That information was required in this case to enable the wider design team, and ultimately the TMO as the responsible person, to understand the design of the system and its limitations but not all of it had been recorded.

¶60.69 Information relating to the design of the system should also have been incorporated in the fire safety strategy but that was not done either. Instead, Issue 3 of the Draft Outline Fire Safety Strategy discussed the smoke control system in very general terms and left it to Max Fordham to take the matter forward. We consider that Exova and the TMO, as the client, should have ensured that the Outline Fire Safety Strategy specifically described the final design of the smoke control system and explained how it supported the fire safety strategy for the building. A fire safety strategy necessarily relies on certain assumptions, which should be made explicit.²⁴³⁶²⁴³⁶ Torero {Day289/7:5-25}. That was particularly important in this case, since the new system did not correspond directly to any design described in the available guidance.

228

Extended travel distances

¶60.70 We heard evidence about the distances between the entrance doors to flats and the door to the stairs. The guidance in Approved Document B is that the distance in an unventilated part of an escape route should not exceed 7.5 metres.²⁴³⁷²⁴³⁷ Approved Document B {CLG00000224/28-30} Diagram 7, flats served by one common stair and Table 1. BS9991 contains similar guidance.²⁴³⁸²⁴³⁸ BS 9991:2011 (Fire safety in the design, management and use of residential buildings – Code of practice) {BSI00000621/36}.

¶60.71 Measurements were made of the maximum distance between the front doors of the flats and the doors to the stairs on Floors 4 to 23. They ranged from 9.3 metres (BRE²⁴³⁹²⁴³⁹ BRE Grenfell Tower Fire Investigation Report {MET00039807/24} paragraph 32.) to 10.3 metres (Dr Lane).²⁴⁴⁰²⁴⁴⁰ Lane, Phase 1 Report, Appendix J {BLAS0000037/25}. We do not consider the difference between 9.3 metres and 10.3 metres to be material. What is significant is that, even at 9.3 metres, the maximum distances exceeded that indicated by Approved Document B.

¶60.72 Mr Hanson accepted that if the travel distance exceeded 7.5 metres, the Smoke Control Association Guide indicated that the primary objective of the system should be the protection of both the lobby and the staircase.²⁴⁴¹²⁴⁴¹ Hanson {Day154/167:16-22}. His view, however, was that for an existing building it would be unreasonable to apply the standards applicable to a new building.²⁴⁴²²⁴⁴² Hanson {Day154/169:5-9}.

¶60.73 Mr Lay also thought that if the travel distances in a building were excessive, it would not be appropriate to concentrate solely on protecting the stairs and that protection would also need to be provided to the lobby.²⁴⁴³²⁴⁴³ Lay {Day286/52:14}-{Day286/53:11}. He referred to the LGA Guide as indicating that extended travel distances of less than 10 metres in existing residential buildings would not be considered excessive and did not require mitigation. We consider, however, that some caution should be exercised in using the LGA Guide in this context, since it is intended to provide guidance on compliance with the Fire Safety Order and the Housing Act 2004, rather than the functional requirements of the Building Regulations and is directed at housing providers and enforcing authorities,²⁴⁴⁴²⁴⁴⁴ LGA Guide on Fire Safety in Purpose Built Blocks of Flats {CST00014237/13-14} paragraph 4.1-4.2. rather than designers of smoke control systems. Although that does not make it irrelevant, it is significant that the context is assessing existing fire safety arrangements, rather than controlled works under the Building Regulations. Furthermore, we note that Approved Document B stated at paragraph 0.21, that guidance documents intended specifically for assessing fire safety in existing buildings will often include provisions which are less onerous than those set out in Approved Document B and are therefore unlikely to be appropriate for use in relation to work controlled by the Building Regulations.²⁴⁴⁵²⁴⁴⁵ Approved Document B {CLG00000224/13} section 0.21.

¶60.74 The LGA Guide itself states that the primary guidance is that maximum travel distances should not exceed 7.5 metres;²⁴⁴⁶²⁴⁴⁶ LGA Guide on Fire Safety in Purpose Built Blocks of Flats {CST00014237/87} paragraph 58.24. goes on to recognise that some existing blocks of flats do not comply with that guidance,²⁴⁴⁷²⁴⁴⁷ LGA Guide on Fire Safety in Purpose Built Blocks of Flats {CST00014237/94} paragraph 62.1. giving rise to a need to consider the overall risk.²⁴⁴⁸²⁴⁴⁸ LGA Guide on Fire Safety in Purpose Built Blocks of Flats {CST00014237/94} paragraph 62.2. The LGA Guide also refers to compensatory measures, such as automatic fire detection and automatic fire suppression systems, which were not present in Grenfell Tower. It suggests that the acceptance of standards that differ from current benchmarks should be subject to careful scrutiny, which may require assessment by a suitable specialist,²⁴⁴⁹²⁴⁴⁹ LGA Guide on Fire Safety in Purpose Built Blocks of Flats {CST00014237/94} paragraphs 62.3-62.4. in this case a fire engineer.²⁴⁵⁰²⁴⁵⁰ Lay {Day286/63:9-25}. The LGA Guide goes on to say that in ventilated lobbies and corridors, increases from 7.5m up to 10m are likely to be acceptable in most situations with no additional measures.²⁴⁵¹²⁴⁵¹ LGA Guide on Fire Safety in Purpose Built Blocks of Flats {CST00014237/95-96} paragraph 62.7. However, we are not persuaded that any of the parties involved in the design of the new smoke ventilation system actually relied on the LGA Guide to support its design. It was not referred to in any of the design documents, the witness statements or the oral evidence of those who were involved at the time. Nonetheless, the travel distances at the tower exceeded 7.5 metres and some thought should therefore have been given to protecting the escape route within the lobbies.

229

¶60.75 Mr Hanson said that he had thought about extended travel distances at the time but had not written anything down, although he agreed that such information would have been helpful.²⁴⁵²²⁴⁵² Hanson {Day154/154:8}-{Day154/156:5}. Mr Cross Smith said that he did not recall any discussion about travel distances.²⁴⁵³²⁴⁵³ Cross Smith {Day157/138:12-21}. Mr Whyte said that there might have been some discussion of travel distances, but he could not recall when and was not himself aware that the travel distances were more than the guidance in Approved Document B.²⁴⁵⁴²⁴⁵⁴ Whyte {Day158/52:12}-{Day158/53:9}.

¶60.76 In the absence of any contemporaneous evidence we are not able to accept that there was any discussion with building control about the travel distances in the tower or the possible need for mitigation measures. We would have expected to have seen at least some record of such a discussion if it had occurred. Rather, the evidence suggests that the travel distances were overlooked or that it was assumed that they did not need to be considered because the increase in distance was only around 2 metres. That was not an acceptable approach. Proper consideration should have been given to whether the extended distance required mitigation. That is particularly so because the system was being changed and the focus of protection was being shifted from the lobby to the stairs. If that had been done, the design might have been considered acceptable despite the extended travel distance, taking in account the limited additional distance and the mitigation measures in place.²⁴⁵⁵²⁴⁵⁵ Menzies {Day169/32:22}-{Day169/33:10}.

¶60.77 Considering the system as part of the wider fire safety strategy was not the responsibility of PSB but of those who were responsible for drafting that strategy, principally Exova.²⁴⁵⁶²⁴⁵⁶ PSB Module 7 written closing submissions {PSB00001380/17} paragraph 82. When it was drafting the fire safety strategy Exova should have identified the extended travel distances in the lobbies and should have recorded its view of the need for mitigating measures to be provided. When the design of the system was changed, Exova should have been asked to consider the fire strategy again to ensure that it reflected developments in the design.

Leakage

¶60.78 Little consideration appears to have been given to how the new smoke ventilation system would operate under different conditions of air leakage that were reasonably foreseeable. In order to assess the likely performance of the system there would need to be at least some consideration of the circumstances in which it was expected to function. Leakage paths were relevant, because the ability of the system to protect the stairs would be affected by the source of air entering the lobby. For example, the system could extract air from the lobby at a rate of 5.0m3/s but might not be able to maintain an air flow velocity of 2.0m/s across the door to the stairs if air was able to enter the lobby from other sources.

230

¶60.79 Dr Lane and Mr Lay both carried out calculations to demonstrate the effect of different combinations of door and window openings on the velocity of air through the door to the stairs. Although their analyses differed, each demonstrated that leakage paths were potentially significant. The calculations involved retrospective attempts to quantify the performance of the system; there was no evidence that those who designed it had carried out any similar analyses at the time.

¶60.80 Mr Mahoney said that the figure of 50% that he had used in his initial proposals to allow for unforeseen losses from all sources when calculating the appropriate rate of extraction was a notional number taken from the British Standard.²⁴⁵⁷²⁴⁵⁷ Mahoney {Day155/98:13-15}. Mr Lay did not think that that was the right way to design a system of that kind, but acknowledged that it had become normal in the industry to make assumptions about such things.²⁴⁵⁸²⁴⁵⁸ Lay {Day286/151:22}-{Day286/152:6}. He said that he would not have approved PSB's specification without having calculations or a computer model indicating the leakage in the system.²⁴⁵⁹²⁴⁵⁹ Lay {Day286/152:7-14}.

¶60.81 Dr Lane's view was that relying on experience to justify a suitable rate of extraction without explaining what assumptions about leakage had been made or without demonstrating how the rate had been calculated was not a reasonable professional approach. She said that it was not clear how Mr Mahoney had quantified the different elements that would cause resistance in the proposed system, such as leakage through the builders' work shafts and blemishes in the ductwork. Dr Lane's view was that it would be normal to identify them.²⁴⁶⁰²⁴⁶⁰ Lane {Day287/130:21}-{Day287/132:5}.

¶60.82 In its closing statement PSB accepted that those were matters which, with the benefit of hindsight, could, and should, have been spelled out in more detail in its technical submission.²⁴⁶¹²⁴⁶¹ PSB Module 7 written closing submissions {PSB00001380/22} paragraph 98. However, although PSB obviously needed to rely on its previous experience when designing the system, there still needed to be some consideration of the building itself and the circumstances in which the system would be expected to operate. That required at least some consideration of the potential leakage from the building's structure and routes of escape that would counteract the rate of extraction, rather than relying on a generic figure. The 5.0m3/s rate of extraction was a key part of the proposal and the primary basis on which the new system was said to be superior to the old one. We do not understand how that figure could reasonably be put forward as appropriate (if, for example, it failed to achieve an air flow of 2m/s across the door under certain conditions) without there being at least some consideration of how the particular nature of the building might affect leakage in a fire. The absence of any such analysis represents a defect in the design process.

Air movement into the lobby

¶60.83 Dr Lane pointed out that if a door to one of the flats was left open the system might continue to operate at maximum capacity, even if the door to the stairs were closed, because the pressure difference between the lobby and the stairwell would have been lost, causing smoke to be drawn from the flat into the lobby.²⁴⁶²²⁴⁶² Lane {Day287/141:18}-{Day287/143:5}. She was of the view that if the designer considered that that risk was mitigated by the presence of self-closing devices on the doors of the flats, that ought to have been made clear.²⁴⁶³²⁴⁶³ Lane, Module 7 Supplementary Report {BLARP20000043/94}. It would have ensured that their importance was understood by those responsible for the operation and maintenance of the building.

231

¶60.84 We do not think it is possible to know how the system would have responded if a flat door had been left open. Apart from anything else, the precise circumstances could have varied widely and could have included a tendency for the reduction in the air pressure in the lobby to draw the door closed. The possibility was not fanciful, however, and it was important for the designers of the system to identify the circumstances under which it would no longer be able to perform in accordance with its design. The design criteria should have been taken into account in the overall fire safety strategy for the building.

Commissioning the system

¶60.85 Proper commissioning of a new smoke ventilation system is of critical importance in ensuring that it is capable in practice of performing the function required of it. It therefore needs to be carried out with great care and attention to detail.

¶60.86 Granville Partlow, PSB's Group Service and Engineering Manager,²⁴⁶⁴²⁴⁶⁴ Partlow {PSB00001309/3} page 3, paragraph 11. carried out the final commissioning of the new system between 26 and 28 April 2016.²⁴⁶⁵²⁴⁶⁵ Partlow {Day156/19:14-18}. He accepted that there had been errors in the commissioning method statement, which he attributed to his having adapted the document rather than preparing it from scratch.²⁴⁶⁶²⁴⁶⁶ Partlow {Day156/32:10-14}. He had not noticed the mistakes in the document, including in the description of the system, until he had read it again for the purposes of making a statement for the Inquiry.

¶60.87 Mr Partlow told us that during the commissioning process he had checked that all the fireman's override switches were working, but certain matters escaped him.²⁴⁶⁷²⁴⁶⁷ Partlow {Day156/94:21-23}. PSB's technical submission said the fireman's override switches were in the stairs, when in fact they were in the lobbies, but Mr Partlow did not notice the discrepancy. It did not occur to him, therefore, that in the event of a fire they might be affected by rising temperatures in the lobbies, potentially leading to unintended activation.²⁴⁶⁸²⁴⁶⁸ Partlow {Day156/177:18-24}.

¶60.88 There was also some confusion about measuring flow rates. Mr Partlow did not understand why building control was asking for readings of extraction flow rates in cubic metres per second, which suggests a failure of communication between building control and those designing the system.²⁴⁶⁹²⁴⁶⁹ Partlow {Day156/166:9-11}. PSB did not measure the flow in that way but Mr Partlow asked Mr Whyte to measure the area of one of the doors which he used together with the average velocity across the open door to calculate the flow in cubic metres per second.²⁴⁷⁰²⁴⁷⁰ Partlow {Day156/167:13-24}; Partlow {Day156/167:13-24}; Whyte {Day158/80:6-25}.

¶60.89 When building control approved the proposed system on 24 June 2015 it gave no reasons for its decision but did say that it should conform to the Smoke Control Association Guide 2012.²⁴⁷¹²⁴⁷¹ Hanson {Day154/143:11-23}. That was repeated in a memorandum sent by Mr Hanson to Mr Hoban on 26 January 2016,²⁴⁷²²⁴⁷² Hanson {Day154/144:8-16}. in which he said that by saying the system was acceptable he meant that it was compliant with functional requirements B1 and B5 of the Building Regulations.²⁴⁷³²⁴⁷³ Hanson {Day154/145:5-9}.

232

¶60.90 The need for a cold smoke test of the system was also debated. Ms Menzies said that such a test should have been carried out as part of the commissioning process in order to demonstrate to building control that the system operated effectively. In her opinion that was an important part of the commissioning and acceptance process, since without one it would not be possible to know whether the system worked under different door-opening conditions. In her view, the failure to require a cold smoke test fell below the standard to be expected of a reasonable building control body. Mr Hanson did not think that carrying out a cold smoke test was an important part of commissioning any system protecting escape routes with extended travel distances, although he accepted it could be a useful guide.²⁴⁷⁴²⁴⁷⁴ Hanson {Day154/151:22}-{Day154/152:6}. Mr Lay did not expect cold smoke testing to be part of commissioning; he would expect to take flow measurements.²⁴⁷⁵²⁴⁷⁵ Lay {Day286/160:24}-{Day286/161:3}.

¶60.91 Mr Partlow said that a cold smoke test had not been considered in this case because the building was occupied and it would have been very frightening for residents to walk out of a flat and be confronted by a lobby full of smoke.²⁴⁷⁶²⁴⁷⁶ Partlow {Day156/191:6-8}. However, Ms Menzies, who has many years' experience as a building control officer, said that she had never heard it suggested that a cold smoke test might be a problem in an occupied building.²⁴⁷⁷²⁴⁷⁷ Menzies {Day169/61:4-9}.

¶60.92 We think that unless detailed flow measurements under different conditions of door opening had been made, a cold smoke test should have been performed as part of the commissioning process. The practical challenges identified by Mr Partlow were far from insurmountable. The inconvenience to residents and the additional cost was justified by the benefit of assessing the system's operation in circumstances approaching those in which it would be expected to function and was to be preferred to measuring flow rates.

¶60.93 There were other respects in which the commissioning process had not captured all the information required to demonstrate that the system performed as intended.²⁴⁷⁸²⁴⁷⁸ Lane {Day287/213:7}-{Day287/214:1}. They included a failure to record the performance of the fans, the commissioning of all dampers, door opening forces, or pressure measurements.²⁴⁷⁹²⁴⁷⁹ Lane, Phase 2 Module 3 Report {BLARP20000035/409} Sections 1-9 and 11.

¶60.94 Mr Hanson accepted that there should have been a record of door opening forces, at least for a representative number of doors. He said that his ability to scrutinise the commissioning report was limited because the RBKC building control team no longer had an engineer. It therefore had to rely on the professionalism of the installer and designer of the system.²⁴⁸⁰²⁴⁸⁰ Hanson {Day154/201:19-24}. Ms Menzies did not think that was a reasonable approach for building control to take if it had not seen some evidence that the door opening forces had in fact been tested.²⁴⁸¹²⁴⁸¹ Menzies {Day169/48:14-21}. She also considered that building control could not have considered whether the system ensured an adequate means of escape without some form of testing which demonstrated that a flow rate of 2m/s through the door to the stairs could be achieved in a variety of different conditions.²⁴⁸²²⁴⁸² Menzies {Day169/47:15}-{Day169/48:8}.

233

¶60.95 There was another reason why, in the opinion of Ms Menzies, building control should not have accepted the commissioning report. An additional vent had been added to the system at its suggestion after commissioning had been undertaken. In her view, the entire system needed to be commissioned after all the work on it had been completed.²⁴⁸³²⁴⁸³ Menzies {Day169/62:5}-{Day169/64:12}.

¶60.96 Overall, we consider that the commissioning process was not sufficient for building control to be satisfied that the system would perform as intended. That was in part because it was not provided with a set of data that enabled it to be confident that the system operated correctly, but also because it did not require other forms of testing, such as a cold smoke test, to make up for that omission. More generally, the commissioning documents did not describe the performance expected of the system, details of the situations for which it was designed or how it related to the building's fire safety strategy.

Maintenance of the system

¶60.97 By the time of the refurbishment the original smoke ventilation system had fallen into disrepair, partly as a result of inadequate maintenance,²⁴⁸⁴²⁴⁸⁴ See Chapter 43. and the evidence suggests that maintenance remained inadequate after the system had been replaced.

¶60.98 The TMO had no formal procedures of its own for the maintenance of the smoke ventilation system, having entered into a contract with Allied Protection Ltd for the inspection and servicing of fire safety equipment, including the smoke ventilation system, which included the following: a. the annual servicing and testing of hardwired standalone and linked smoke detection equipment in communal areas in accordance with British Standards; and

¶b. the six-monthly servicing of the smoke ventilation system in accordance with the manufacturer's instructions, including checking interfaces with the fire alarm or control system panel and the checking and servicing of all damper activators and associated mechanisms.²⁴⁸⁵²⁴⁸⁵ Packer {LAK00000525/3} page 3, paragraph 7.

¶60.99 Although Allied Protection was required to service the system at 6-monthly intervals in accordance with manufacturer's instructions, the TMO did not provide those instructions until 14th June 2017.²⁴⁸⁶²⁴⁸⁶ Packer {LAK00000525/3} page 3, paragraph 8.

¶60.100 Generally accepted standards for the maintenance of smoke control systems published by the British Standards Institution²⁴⁸⁷²⁴⁸⁷ BS EN 12101-6:2005, BS 5839-1:2013 (withdrawn 31.08.2017 and replaced by BS 5839-1:2017) and BS 9999:2008 (withdrawn 31.01.2017 and replaced by BS 9999:2017). included the need to test each zone of a smoke ventilation system separately, ensuring that any fans and powered exhaust ventilators operate correctly and that dampers close or, where relevant, open.²⁴⁸⁸²⁴⁸⁸ Lane, Phase 2 Module 3 Report, Sections 1-9 and 11 {BLARP20000035/498-499}.

234

¶60.101 We were provided with copies of the inspection and servicing certificates produced by Allied Protection following maintenance visits to Grenfell Tower on 17 January 2017²⁴⁸⁹²⁴⁸⁹ {LAK00000009}. and 15 May 2017.²⁴⁹⁰²⁴⁹⁰ {LAK00000011}. The visit on 17 January 2017 was carried out by Karl Russell. The certificate records that the smoke detectors were not tested, although the inspection should have included an operational test of the 26 smoke detectors between the ground floor and Floor 23. Moreover, there was no record in the call history of the auto-dialler that the smoke detectors had been activated, although they should have operated if the tests had been carried out properly.²⁴⁹¹²⁴⁹¹ {THL00000019}. The evidence suggests, therefore, that either the smoke detectors did not work or that an operational test of the system was not carried out on that occasion. The latter is the more likely explanation, because if the system had been tested and had failed to operate, that would have been recorded.

¶60.102 The inspection on 15 May 2017 was carried out by Barry McAuliffe. There were 20 logged activations of smoke detectors,²⁴⁹²²⁴⁹² {THL00000019}. but there should have been 26 if all the smoke detectors had been tested properly. It was not possible to identify from the auto-dialler log which smoke detectors had been activated and which had not, but Mr McAuliffe said that he had not tested the ground floor lift lobby detectors because he had been called away to carry out an urgent repair.²⁴⁹³²⁴⁹³ McAuliffe {LAK00000522/4} page 4, paragraph 26.

¶60.103 Mr McAuliffe recorded his arrival time on 15 May 2017 as 12:15 and his departure time as 14.15, a period of two hours.²⁴⁹⁴²⁴⁹⁴ {LAK00000011}. On each operational test of the system the maintenance engineer was required, as a minimum, to carry out the following steps to ensure that the system and all its components were operating correctly:

¶a. Activate the system using a smoke detector.

¶b. Inspect the position of the dampers in the lobbies on all floors.

¶c. Inspect the position of the by-pass dampers at Level 2.

¶d. Inspect the operation of the smoke extraction fans at roof level.

¶e. Inspect the operation of the smoke extraction fans at Level 2.

¶f. Inspect the control panel to check that it correctly indicated the system operation, floor activation and status of all dampers.

¶g. Reset the control panel on the ground floor once all inspections of dampers and fans had been completed.²⁴⁹⁵²⁴⁹⁵ Lane, Phase 2 Module 3 Report {BLARP20000035/516} Sections 1-9 and 11.

¶60.104 Mr McAuliffe described testing one floor at a time by setting off each smoke detector head using his own artificial smoke and checking that the smoke extraction system started. He then proceeded to the control panel in the ground floor lobby and reset the system. He said he had done that for each floor, using the emergency stairs for access.²⁴⁹⁶²⁴⁹⁶ McAuliffe {LAK00000522/4} page 4, paragraph 25.

235

¶60.105 The period from the start to the end of the 20 activations on that day recorded in the auto-dialler log was about 1 hour 15 minutes.²⁴⁹⁷²⁴⁹⁷ {THL00000019}. That left Mr McAuliffe about 4 minutes a floor to carry out the required tests, which we do not think was long enough to carry out the tests he described.²⁴⁹⁸²⁴⁹⁸ Lane, Phase 2 Module 3 Report {BLARP20000035/518} Sections 1-9 and 11.

¶60.106 When he was sent to inspect the system at Grenfell Tower Mr McAuliffe had had previous experience of only more basic smoke ventilation systems. He had not been provided with information about the system before his visit and called his manager to ask if he could decline the job. However, he was told to go ahead, despite making clear his lack of expertise.²⁴⁹⁹²⁴⁹⁹ McAuliffe {LAK00000524/1-2} pages 1-2, paragraph 5.

¶60.107 Mr McAuliffe accepted that he had not opened the grilles on each floor to check that the dampers had activated properly during the tests²⁵⁰⁰²⁵⁰⁰ McAuliffe {LAK00000524/2} page 2, paragraph 9. and if the grilles had not been opened, the dampers cannot have been inspected in sufficient detail to ensure that they sealed properly when closed.²⁵⁰¹²⁵⁰¹ Lay, Smoke Control System Report {LAY00000001/315}. It was not possible for all 91 dampers to be serviced and their operation checked during the relatively short visits made by Allied Protection, so the maintenance records did not provide any assurance that the system would operate correctly if a fire were to occur.²⁵⁰²²⁵⁰² Lane, Phase 2 Module 3 Report {BLARP20000035/523} Sections 1-9 and 11.

¶60.108 Overall, the maintenance documentation provided for the smoke ventilation system was poor and did not satisfy the guidance in BS 9999: 2008.²⁵⁰³²⁵⁰³ The Code of practice for fire safety in the design, management and use of Buildings. Neither the limited weekly test carried out by Mr Steadman (a TMO estate services assistant) or Allied Protection's checks provided clear evidence of rigorous maintenance and inspection consistent with what was recommended by the guidance.²⁵⁰⁴²⁵⁰⁴ Lay, Smoke Control System Report {LAY00000001/316-317}.

¶60.109 The smoke ventilation system was complex and contained many separate mechanical components. Dampers were subject to repeated wear through regular opening and closing while operating in environmental mode and therefore required regular inspection and maintenance. However, Allied Protection carried out inspection and maintenance visits without sight of the manufacturer's instructions and sent an engineer to carry out the inspection on 15 May 2017 who lacked the skill to do the job properly.

¶60.110 There is little evidence that the fans on Level 2 had been properly inspected or maintained before the fire.²⁵⁰⁵²⁵⁰⁵ Lane, Phase 2 Module 3 Report {BLARP20000035/516} Sections 1-9 and 11. Mr McAuliffe said that he probably had inspected the operation of the smoke extraction fans at Level 2 but could not be certain.²⁵⁰⁶²⁵⁰⁶ McAuliffe {LAK00000524/6} page 6, paragraph 37. We are doubtful that he had, however, given that he was not aware of all the testing requirements,²⁵⁰⁷²⁵⁰⁷ McAuliffe {LAK00000524/6} page 6, paragraph 35. had little experience of complex systems of this type and had not been provided with a description of the system.²⁵⁰⁸²⁵⁰⁸ McAuliffe {LAK00000524/1-2} pages 1-2, paragraph 5. He did not test the by-pass damper at Level 2 and could not recall having tested the smoke extraction fans at roof level.²⁵⁰⁹²⁵⁰⁹ McAuliffe {LAK00000524/6} page 6, paragraph 37. Moreover, it would not have been possible for him to check the operation of the fans at Level 2 given the length of his visit

236

and the extent of the other work he had to carry out.²⁵¹⁰²⁵¹⁰ Lane, Phase 2 Module 3 Report {BLARP20000035/497} Sections 1-9 and 11. There is little evidence, therefore, that the smoke extraction fan at Level 2 was inspected and maintained in a way that would give one confidence that it would operate effectively if there were a fire.

Operation of the system on the night of the fire

¶60.111 The smoke ventilation system was designed to respond to a fire on one floor of the building only. It follows that the fire that occurred at Grenfell Tower on 14 June 2017 far exceeded anything that it was or could reasonably have been designed to deal with effectively. The system itself was complex and was significantly damaged by the fire. We have carefully considered the available evidence that might shed light on its performance on the night of the fire but given the nature and scale of the disaster that evidence, which is mainly derived from the condition of the dampers and fans after the fire, is inevitably of limited value. Unfortunately, it does not enable us to draw any reliable conclusions about how the system performed during the fire or how it would have performed under the circumstances for which it had been designed.

Conclusions

¶60.112 The design of a new smoke ventilation system for Grenfell Tower provides a good illustration of the difficulties that can be faced by those seeking to refurbish an old building. Time and standards move on and it may be impossible because of the constraints imposed by the existing structure to provide systems that comply in all respects with current standards. In those circumstances the choice is between doing the best you can and doing nothing at all.

¶60.113 The Building Regulations do not require a building that undergoes a material alteration to comply with the current functional requirements on completion of the work, provided that it is no more unsatisfactory in relation to any particular requirement than it was before the work was carried out (the "no-worsening" principle). There may be good reasons for retaining that principle, but it may be due for reappraisal because one consequence is that many older buildings, including some that have been refurbished, do not meet current standards in relation to one or more of the functional requirements, including those relating to fire safety. The solution adopted by PSB in the present case was to use the existing structure in a different way that enabled it to double the volume of the shafts available for the extraction of smoke and to provide protection for the stairs that represented the main escape route as well as the means of access for the fire and rescue service.

¶60.114 The criticisms that were made of the design were directed in the main to PSB's failure to record the basis for its design decisions and to ensure that certain processes and calculations, in particular a CFD analysis, were carried out, either by itself or others, that would have enabled a more rigorous assessment of the proposed system to be made. As a result, although it had good grounds for feeling confident that the new system would be considerably better than the original (and therefore satisfy the "no worsening" principle), it was difficult for PSB to provide evidence to that effect and difficult also for PSB to rebut the suggestion that it had not carried out the kind of rigorous analysis of the system that was to be expected. Its difficulty in demonstrating both to building control and to us that the new system was better than the old one was made all the more difficult by the fact that the original system was inoperable and its performance therefore incapable of being measured.

237

¶60.115 The failure to record the basis for design decisions is not in itself a criticism of the system that was produced, but it may indicate a failure to take the process as seriously as required. However, designers need to bear in mind that there are others who need to understand the basis of their decisions, not least any fire engineer charged with producing a fire safety strategy for the building and the person responsible for the operation and maintenance of the building when in occupation. Proper recording of the calculations and assumptions that underpin the design of a complex system of that kind is therefore essential for the safety of the occupants in the longer term. Those are matters of which building control must be aware and in relation to which it must insist on being provided with cogent evidence.

¶60.116 The need to choose equipment and materials that have been properly tested and shown to conform to the appropriate requirements (such as the dampers for the system) should not be a contentious matter, but again it requires the careful attention of the system designer, the fire engineer and building control to ensure that appropriate decisions and choices have been made.

239