Chapter 22: Kingspan K15 insulation

The use of K15 insulation on Grenfell Tower

¶22.1 Most of the insulation used on Grenfell Tower during the refurbishment was Celotex RS5000. A small amount, probably about 5% of the total, was K15 Kooltherm manufactured by Kingspan Insulation Limited and marketed for use in rainscreen cladding systems.⁴¹⁷⁴¹⁷ Marketing literature 2001 {KIN00020720}; 2002 {KIN00009173} and {KIN00008018}; 2003 {KIN00002579}; 2006 {KIN00005071}; 2007 {KIN00002580}; 2008 {KIN00009703}; 2011{KIN00003545}; 2015 {KIN00000086}; 2016 {KIN00000070}.

¶22.2 K15 was a rigid phenolic foam board with a foil facer bonded to both faces. Phenolic foam is combustible, with a short time to ignition.⁴¹⁸⁴¹⁸ See Phase 1 Report Volume I, paragraph 6.20 for a description of the fire properties of phenolic foam. How K15 came to be supplied in May⁴¹⁹⁴¹⁹ {SIG00000012}. and September⁴²⁰⁴²⁰ {CCF00000019}. 2015 for use in the refurbishment of the tower has been described in Chapter 56 of the report. Kingspan was not aware until after the fire that K15 had been used on the tower.⁴²¹⁴²¹ Burnley {KIN00000554/5} page 5, paragraph 4.2.

K15 – the product

¶22.3 K15 was first manufactured at Kingspan's Pembridge site in Herefordshire in November 2002.⁴²²⁴²² {KIN00022307} Tab 2, Row 2, Column C. At that time, the foam was produced with a pentane blowing agent.⁴²³⁴²³ Rochefort {KIN00008838/7} page 7, paragraph 3.10; Rochefort {Day80/50:21-22}; {KIN00022307} Tab 2, Row 2, Column C. The foil facers of the product were unperforated, with a thickness of 7.6 microns.⁴²⁴⁴²⁴ Meredith {Day75/35:16-18}; Heath {Day79/13:8-10}; Rochefort {Day80/54:11}; {KIN00022307} Tab 2, Row 2, Column C.

¶22.4 Following its acquisition of a Dutch company in 2003,⁴²⁵⁴²⁵ Rochefort {Day80/46:15}; Rochefort {KIN00008838/6} page 6, paragraph 3.6; Heath {Day79/11:13}; Heath {KIN00020709/9} page 9, paragraph 3.4. Kingspan began to introduce a number of changes to its phenolic foam range. They included physical changes to the structure of the foam and the introduction of different chemical ingredients and manufacturing processes to achieve better thermal performance and rates of productivity.⁴²⁶⁴²⁶ Rochefort {KIN00008838/6-7} pages 6-7, paragraphs 3.6 and paragraph 3.8. In particular, Kingspan introduced the use of a mixture of pentane and isopropyl chloride as a blowing agent⁴²⁷⁴²⁷ Rochefort {Day80/50:25}. and perforations were introduced to the aluminium foil facers.⁴²⁸⁴²⁸ Heath {Day79/13:4-7}; Meredith {Day75/35:16-18}; Rochefort {Day80/54:6-8}. The new version of K15 came to be known within Kingspan as "new technology" or "Kesteren technology"⁴²⁹⁴²⁹ Kesteren is the name of the location in the Netherlands where Marec manufactured phenolic foam. K15 and the previous version as "old technology" K15.⁴³⁰⁴³⁰ Rochefort {KIN00008838/6} page 6, paragraph 3.6. Dr Rochefort said that by September 2006 the K15 available to the market was the "new technology" version of the product.⁴³¹⁴³¹ Rochefort {Day80/34:23-24}; {Day80/48:22-23}; Rochefort {KIN00008838/9} page 9, paragraph 3.17. Ivor Meredith, who had been a Project Leader in the Technical Projects team,⁴³²⁴³² Meredith {KIN00022312/1} page 1, paragraph 1. and his line manager, Philip Heath, Technical Manager until 2010,⁴³³⁴³³ Heath {KIN00020709/4-6} pages 4 and 6, paragraphs 2.8 and 2.18. both agreed, although they thought that production of "old technology" K15 had ceased in about August 2007.⁴³⁴⁴³⁴ Heath {KIN00020709/13} page 13, paragraph 3.18; Meredith {Day75/124:24}-{Day75/125:3}.

¶22.5 Phenolic foam is an organic polymer, so whatever blowing agents or methods are used to produce it, it is not a material of limited combustibility as defined in Approved Document B.⁴³⁵⁴³⁵ See Chapter 5; Table A7 in all relevant editions – 2000 {INQ00014107/120}; 2002 {CLG10000740/124}; 2006 {CLG10000007/132}; 2010 {CLG00000173/132}; 2013 {CLG00000224/132}. With the exception of Dr Rochefort, who professed not to have been aware at the time of the fire performance of K15 or the existence of a definition of limited combustibility in Approved Document B,⁴³⁶⁴³⁶ Rochefort {KIN00008838/23} page 23, paragraph 5.8; Rochefort {Day80/27:12-15}; {Day80/28:25}-{Day80/29:4}. all the Kingspan witnesses from whom we heard confirmed that they had been aware throughout their employment at Kingspan that K15 was not, and never could be, a material of limited combustibility.⁴³⁷⁴³⁷ Meredith {Day75/25:17-21}; Millichap {Day81/25:4-10}; Pack {Day86/22:1-5}; Burnley {Day85/134:1-3}; Pargeter {KIN00000494/6} page 6, paragraph 3.4; Heath {Day78/181:7–12}; Mills {Day77/15:3-6}.

Early marketing

¶22.6 From a very early stage, even before the first test in accordance with BS 8414 in 2005, Kingspan knew that, applying the guidance in Approved Document B, K15 was not suitable for use on buildings over 18 metres in height.⁴³⁸⁴³⁸ Heath {Day78/198:2}; {KIN00005054/1} where on 26 January 2004 Mr Meredith reported to Mr Heath on the work then being carried out to try to develop a K15 product "that will achieve acceptability for above 18m in facade construction by Building Regulations". However, it chose to disregard that fact when selling K15 and advising on its use. Kingspan's principal marketing literature from 2001 onwards described K15 as suitable for use in rainscreen cladding systems generally⁴³⁹⁴³⁹ {KIN00020720/1}. and brochures published between 2002 and 2006 described it as a product designed for use behind rainscreen cladding systems without any qualification.⁴⁴⁰⁴⁴⁰ January 2002 {KIN00009173}; November 2002 {KIN00008018}; March 2003 {KIN00002579}; June 2006 {KIN00005371}. None of those documents referred to the warning in Approved Document B about the use of combustible materials in tall buildings or of the guidance that insulation used on buildings over 18 metres in height should be of "limited combustibility". That was the case even in those sections of the documents that depicted typical design details, where no distinction was drawn between use of the product on buildings of over and under 18 metres in height.⁴⁴¹⁴⁴¹ See for example {KIN00005371/2}.

¶22.7 In 2004 Ivor Meredith was a Technical Advisor in Kingspan's Technical Projects Team; from 2005 he was a Project Leader.⁴⁴²⁴⁴² Meredith {KIN00022312/1} page 1, paragraph 1. On 19 March 2004 he wrote to contacts at Tower Hamlets Borough Council and LABC inviting comments. His email was headed "Insulation above 18m". In it he acknowledged that K15 did not meet the criteria for limited combustibility, but he expressed confidence in its fire performance and said that it had been accepted regularly for use on buildings over 18 metres in height in ventilated facade systems. He also said that K15 had been "pigeonholed" with combustible insulation, even though the char that he said would form when it was exposed to flame limited the combustibility of the product.⁴⁴³⁴⁴³ {KIN00003685}.

¶22.8 The language of that email was liable to mislead, because as Mr Meredith well understood,⁴⁴⁴⁴⁴⁴ Meredith {Day75/26:14-19}. insulation used on buildings of 18 metres or more in height had to be of limited combustibility if it was intended to follow what was known as the "linear route to compliance" in Approved Document B and K15 did not satisfy that requirement. He said that he had borrowed the language from a set of written standard answers, possibly produced by the European Phenolic Foam Association, then in circulation within Kingspan.⁴⁴⁵⁴⁴⁵ Meredith {Day75/27:13-21}. Whether that is right or not, however, the message provides some insight into the methods that Kingspan was using to sell K15 at the time and how it might have come about that, as Mr Meredith told us, K15 was being accepted for use on buildings over 18 metres in height in the period before any large-scale system test in accordance with BS 8414 had been carried out.⁴⁴⁶⁴⁴⁶ Meredith {Day75/28:13}-{Day75/29:3}.

Large scale testing: May 2005

¶22.9 Before the publication of the BS 8414-1 test standard in December 2002⁴⁴⁷⁴⁴⁷ {CEL00001205}. Kingspan had begun to explore the possibility that it could use successful testing to BS 8414 to promote and sell K15 for use on buildings over 18 metres in height.⁴⁴⁸⁴⁴⁸ K5, an identical phenolic foam to K15 with different foil facers, was tested to a draft version of BS 8414-1 as early as March 2002: Meredith {Day75/32:8-20}; {KIN00003685}. On the publication of the 2006 edition of Approved Document B in April 2007⁴⁴⁹⁴⁴⁹ Reference to the use of BS 8414 test data classified in accordance with the criteria in BR 135 as an alternative to the use of insulation material of limited combustibility on buildings over 18 metres in height was first included in the 2006 edition of Approved Document B published on 6 April 2007 {CLG10000007/95}. it encouraged and celebrated the adoption of the BS 8414 test,⁴⁵⁰⁴⁵⁰ Kingspan's AD B consultation response {CLG00002607/24}; {KIN00005292/9} second paragraph. which it viewed as an opportunity to create a new category of insulation products for use on buildings over 18 metres in height, namely, products that were not materials of limited combustibility but were nonetheless suitable for use above that height.

¶22.10 That was the foundation of the fundamental falsehood at the heart of Kingspan's marketing strategy in the years that followed. There was no such thing as an insulation product that was combustible and yet could be used generally on buildings above 18 metres in height while following the guidance in Approved Document B. The use of a combustible product on such a building in accordance with Approved Document B was possible only if it were incorporated in a system which had been tested in accordance with BS 8414 and had met the performance criteria in BR 135 (2003). There was therefore no "over 18 metre market" for combustible insulation products as such. Insulation materials that were not of limited combustibility could be used on buildings over 18 metres in height in accordance with the guidance in Approved Document B only on the basis of a test of the particular system it was intended to install. Nonetheless, once BS 8414 and BR 135 had become available, Kingspan treated them as a generally applicable "route to compliance" for K15 as a product, even though it could constitute only part of any such system.⁴⁵¹⁴⁵¹ Meredith {Day75/29:21-22}.

¶22.11 Following preliminary testing,⁴⁵²⁴⁵² March 2002 test on K5 and Permarock {KIN00003685}; November or December 2004 "naked" test without cladding {KIN00022357} Row 1. See also for example {KIN00005048}; {BRE00003278}. in 2004 Kingspan began preparing for its first full test in accordance with BS 8414-1 of a system incorporating K15. The test was arranged by Ivor Meredith,⁴⁵³⁴⁵³ Heath {Day78/173:8-19}; Meredith {Day75/4:19}-{Day75/5:15}. under the supervision of Philip Heath.⁴⁵⁴⁴⁵⁴ Technical Manager from 2001 to 2010. See Heath {KIN00020709/4-6} pages 4 and 6, paragraphs 2.8 and 2.18. Mr Meredith was responsible within Kingspan for the planning and design of the test rig and for the sourcing, purchase and delivery of the necessary materials, as well as for liaising with BRE during preparations for the test and in connection with its design.⁴⁵⁵⁴⁵⁵ Meredith {Day75/42:21}-{Day75/43:16}. Both he and Philip Heath told us that the decision to test in accordance with BS 8414-1 at that time had been taken by a wider group of managers and directors at Kingspan, as had been the choice of the components of the cladding system to be tested.⁴⁵⁶⁴⁵⁶ Heath {Day78/201:14-17}; {Day78/204:19-22}; Meredith {Day75/58:9-16}. They also agreed that the purpose of the test had been to use a successful result to promote K15 as a product for use on buildings over 18 metres in height.⁴⁵⁷⁴⁵⁷ Heath {Day78/202:14-15}; Meredith {Day75/58:19-20}.

¶22.12 The test was carried out in May 2005. Kingspan decided to test a system using "old technology" K15⁴⁵⁸⁴⁵⁸ {KIN00024104/2} paragraph 4; Meredith {Day75/88:19-24}; Pargeter {KIN00022610/52} page 52, paragraph 7.1. with a non-combustible cement particle or fibre cement board as the rainscreen. There was some disagreement between the witnesses about the precise nature of the rainscreen used. Philip Heath was unable to remember what it had been,⁴⁵⁹⁴⁵⁹ Heath {KIN00020709/79} page 79, footnote 48. whereas Ivor Meredith said that it had been a 6mm non-combustible cement particle board manufactured by a company called UAC,⁴⁶⁰⁴⁶⁰ Meredith {Day75/67:21}-{Day75/68:12}. which is the information recorded in the test report issued by BRE in December 2005.⁴⁶¹⁴⁶¹ BRE Report (Test 220876) {BRE00002511/6} third paragraph. Mr Meredith went on to tell us that the rainscreen had been a fibre cement board marketed in the UK as 'Supalux'.⁴⁶²⁴⁶² Meredith {Day76/207:18-24}. Other Kingspan witnesses, notably Adrian Pargeter, the Head of Technical and Marketing,⁴⁶³⁴⁶³ Pargeter Kingspan's Head of Marketing from November 2014 and Head of Marketing and Technical from June 2015 {KIN00000494/4-5} pages 4-5, paragraphs 2.7 and 2.8. Adrian Brazier⁴⁶⁴⁴⁶⁴ Adrian Brazier replaced Ivor Meredith as Technical Projects Manager in September 2015, Brazier {KIN00008828/4} page 4, paragraph 2.7. and Adam Heath,⁴⁶⁵⁴⁶⁵ Adam Heath was Kingspan's Technical Projects Leader from May 2014, with responsibility from August 2015 for large-scale fire testing. In April 2019, he moved to the role of Regulatory Affairs Manager, Heath {KIN00008834/3-5} pages 3-5, paragraphs 2.4, 2.6 and 2.9. said they thought that the rainscreen had in fact been a non-combustible fibre cement board.⁴⁶⁶⁴⁶⁶ Pargeter {KIN00020824/100} page 100, paragraph 10.57; Pargeter {Day 83/175:11-19}; Heath {KIN00008834/96} page 96, paragraph 11.53; Brazier {KIN00008828/59} page 59, paragraph 11.37. None of the latter group had had any involvement in the test.⁴⁶⁷⁴⁶⁷ Pargeter {KIN00020824/100} page 100, paragraph 10.55; Heath {KIN00008834/96} page 96, paragraph 11.52; Brazier {KIN00008828/58} page 58, paragraph 11.31. Their belief that fibre cement boards had been used, which was expressed with varying degrees of confidence, was based solely on investigations carried out by Kingspan in 2016 and 2019.⁴⁶⁸⁴⁶⁸ In 2016, while working on the development testing of a lower lambda insulation product called K115, Kingspan began preparations to replicate the test in accordance with BS 8414 carried out in May 2005, leading it in 2016 and again in 2019 to try to ascertain the precise nature of the board used as external cladding in that test. Heath {KIN00008834/96} page 96, paragraph 11.53; Pargeter {KIN00020824/100} page 100, paragraphs 10.55-10.57.Show full note No BRE witness was able to recall the precise nature of the board beyond the description given in the test report.⁴⁶⁹⁴⁶⁹ Clark {Day96/121:19-24}; Colwell {Day232/205:16-17}; Baker {Day100/50:6-11}.

¶22.13 Mr Heath and Mr Meredith both told us that the components used in the 2005 test rig had been chosen in discussion with and on the advice of BRE.⁴⁷⁰⁴⁷⁰ Heath {Day78/204:17-22}; {Day78/207:9-12}; {Day78/208:18}-{Day78/209:7}; Meredith {Day75/130:23-25}. Ivor Meredith identified Dr Sarah Colwell,⁴⁷¹⁴⁷¹ In 2005, Dr Sarah Colwell was the joint Business Group Manager for the BRE's Passive Fire Team, Colwell {BRE00047571/3} page 3, paragraph 12. David Hoare⁴⁷²⁴⁷² From 2005 to September 2015, David Hoare was the BRE's Business Group Manager in relation to cables. He also had some involvement in the testing of composite panel systems and in the approval of fire test reports, Hoare {BRE00005622/2} page 2, paragraph 5. and Philip Clark⁴⁷³⁴⁷³ Philip Clark was the Burn Hall Manager at BRE between 2004 and 2017. Mr Clark was involved in BS 8414 testing at BRE from 1999 onwards and was part of the team that developed Parts 1 and 2 of BS 8414, Clark {BRE00005768/3-5} pages 3 and 5, paragraphs 13 and 22. as the BRE staff who were involved in discussions with Kingspan both before and after the test.⁴⁷⁴⁴⁷⁴ Meredith {Day75/59:8}; Colwell {Day75/20:24}; Clark {Day75/63:8-20}; David Hoare, the Senior Consultant who approved BRE's test report). See BRE Report {BRE00002511/2}. Philip Heath described the test as a "pilot test", with the particular assembly suggested by BRE.⁴⁷⁵⁴⁷⁵ Heath {Day78/191:14-17}. It seems clear that the test had been intended to be the first in a series of large-scale tests.⁴⁷⁶⁴⁷⁶ Meredith {Day75/59:21-25}; Heath {Day78/202:23}-{Day78/203:4}.

¶22.14 Mr Heath and Mr Meredith also told us that BRE had suggested that if Kingspan tested a system with a non-combustible rainscreen which met the criteria in BR 135, it could use the results to produce a report indicating that K15 could be used in other systems incorporating non-combustible rainscreens.⁴⁷⁷⁴⁷⁷ Heath {Day78/191:18-24}; {Day78/209:1-7}; Meredith {Day75/58:21}-{Day57/59:6}. Ivor Meredith described that as a "scope of application"⁴⁷⁸⁴⁷⁸ See for example, Meredith {Day75/92:24}-{Day75/93:2}. or "expanded scope of application"⁴⁷⁹⁴⁷⁹ Meredith {Day75/99:19-21}. report. He said that obtaining a report of that kind from BRE was the main purpose of carrying out the test. He understood that a report of that kind based on just one test might render K15 acceptable for use on buildings over 18 metres in height with all non-combustible rainscreen materials.⁴⁸⁰⁴⁸⁰ Meredith {Day75/59:3-6}. In an internal activity report written on 31 May 2005, the day of the test, Mr Meredith told his managers that a successful result that afternoon would remove any limitations on selling K15 into the high-rise market.⁴⁸¹⁴⁸¹ {KIN00021738} first paragraph.

¶22.15 It was for that reason, as Ivor Meredith and Philip Heath agreed, that the boards chosen as the rainscreen for the system tested in May 2005 were not of a kind that would ever be installed in practice, either then or now. They were intended to represent a non-combustible rainscreen of any kind.⁴⁸²⁴⁸² Meredith {Day75/69:20}-{Day75/70:24}; Heath {Day78/207:9-15}. Dr Colwell confirmed that Kingspan had given her to understand that its aim had been to test K15 with a generic rainscreen.⁴⁸³⁴⁸³ Colwell {BRE00047571/48} page 48, paragraph 299; Colwell {Day232/214:3-5}. We are satisfied that Kingspan's intention in choosing cement particle or fibre cement boards for the test was simply to represent as wide a range of non-combustible rainscreen panels as possible. Since the witnesses agreed that in either case the boards used had been non-combustible, it is not necessary for us to reach a decision on precisely which kind was used in the test.

¶22.16 The test took place at BRE's burn hall on 31 May 2005 and was successful in that BRE expressed the view that the system appeared to have met the criteria in BR 135.⁴⁸⁴⁴⁸⁴ Heath {Day79/8:6-13}. The test report, written by Philip Clark and approved by David Hoare, was issued on 9 December 2005.⁴⁸⁵⁴⁸⁵ {BRE00002511}. A separate report classifying the system in accordance with BR 135 was issued more than ten years later, on 28 September 2015.⁴⁸⁶⁴⁸⁶ Report P101812-1000 {KIN00000134}.

¶22.17 In the event, no "scope of application" report was produced by BRE on the basis of the test. Mr Meredith's evidence, which was consistent with Mr Heath's understanding of what had happened, was that after the test had taken place, BRE had changed its position by retracting its previous suggestion that it could produce a report providing support for the use of K15 with other types of non-combustible rainscreen materials.⁴⁸⁷⁴⁸⁷ Meredith {KIN00022312/5} page 5, second paragraph; Meredith {Day75/130:21}-{Day75/131:4}; Heath {Day78/192:20}-{Day78/193:1}; {Day78/209:16}-{Day78/210:3}. In his evidence, Mr Meredith described a heated meeting with Dr Sarah Colwell and an unidentified director of Kingspan Off-Site Ltd⁴⁸⁸⁴⁸⁸ Kingspan Off-Site Ltd, previously known as Kingspan Metl-Con Ltd, was part of the Kingspan group. At the time it manufactured and sold pre-fabricated building and facade systems: Stevens {MET00080972/3} page 3, second paragraph. at some stage after the test during which she had made it clear to him that no wider report could or would be produced by BRE.⁴⁸⁹⁴⁸⁹ Meredith {KIN00022312/5} page 5, first, second and third paragraphs; Meredith {Day75/130:10}-{Day75/133:24}. He recorded those events in the background section of an internal summary to his managers on 7 January 2008, describing both BRE's initial advice and its later retraction.⁴⁹⁰⁴⁹⁰ {KIN00008847/5} page 5, first and second paragraphs.

¶22.18 Dr Colwell denied that she had been aware of any discussions with Kingspan of the kind described by Mr Meredith either before or after the test in May 2005.⁴⁹¹⁴⁹¹ Colwell {Day232/214:16}-{Day232/217:4}. She said she had not given any advice to Kingspan on the assembly it should use in the test⁴⁹²⁴⁹² Colwell {Day232/205:18-21}. or on how to satisfy the criteria in BR 135.⁴⁹³⁴⁹³ Colwell {Day232/196:24}-{Day232/198:16}; {Day232/200:23}-{Day232/201:7}. Dr Colwell was shown an internal Kingspan document in which Mr Meredith had reported to his managers that, following a Single Burning Item test on K15 in March 2004, she had told him that K15 had a good chance of passing the BS 8414 test if the right detailing was adopted.⁴⁹⁴⁴⁹⁴ {KIN00021657} ninth paragraph. In the same document Mr Meredith had recorded that in October 2004 Dr Colwell had suggested that Kingspan should carry out a preliminary "naked" test (i.e. with no rainscreen) to help it pass when it was tested in accordance with BS 8414-2.⁴⁹⁵⁴⁹⁵ {KIN00021657} twelfth paragraph. Dr Colwell told us that Mr Meredith was mistaken when recording those comments⁴⁹⁶⁴⁹⁶ Colwell {Day232/197:9-12}. and that she thought he was putting a particular slant on their discussions.⁴⁹⁷⁴⁹⁷ Colwell {Day232/198:12-16}.

¶22.19 On 14 December 2004 Mr Meredith sent Dr Colwell an email in which he said that Kingspan had decided to follow her guidance and therefore wanted to perform a full test on K15 in accordance with BS 8414-1.⁴⁹⁸⁴⁹⁸ {BRE00047572/1}. When asked about it, she said that she could not understand what he might have meant.⁴⁹⁹⁴⁹⁹ Colwell {Day232/200:10-18}. She said that her discussions with Kingspan at that time about testing K15 in accordance with BS 8414 had been of a general nature,⁵⁰⁰⁵⁰⁰ Colwell {Day232/198:18-22}. and had concerned the programme of tests it was planning to carry out. They had not extended beyond exploring how BRE could meet Kingspan's required programme.⁵⁰¹⁵⁰¹ Colwell {Day232/201:2-7}.

¶22.20 We did not find Dr Colwell's evidence on these matters very convincing and have been left with the clear impression that she and other employees of BRE were doing much more to assist Kingspan than she was willing to admit. Moreover, although she initially denied that she had known anything about discussions between BRE and Kingspan during or after the May 2005 test,⁵⁰²⁵⁰² Colwell {BRE00047571/47} page 47, paragraph 295. Dr Colwell was constrained to accept that, as reflected in an email exchange between BRE and Mr Meredith in March 2006,⁵⁰³⁵⁰³ {KIN00005165/2} "Further to our meeting last month". she had attended a meeting in February 2006 with him and an employee of Kingspan Off-Site, as he had described, during which they had discussed the test in May 2005. She said she could not recall any details of that discussion.⁵⁰⁴⁵⁰⁴ Colwell {Day232/210:16-25}.

¶22.21 Dr Colwell's recollection of events during the period was poor and fragmented and tended to shift in response to the documents. In contrast, Mr Meredith's account was detailed, clear and essentially unchanging. His recollection was also supported by a number of documents, including contemporaneous correspondence.⁵⁰⁵⁵⁰⁵ See, for example, the report of January 2006 {KIN00020718/2} first bullet point under "Projects"; {KIN00004795}; {KIN00005165/1}. We are satisfied that in the period leading up to the test in May 2005, he thought that a successful test could lead to some kind of generic approval for the use of K15 on buildings over 18 metres in height, even if limited to other systems incorporating non-combustible rainscreens. It is more likely than not that his belief was influenced to a substantial degree, if not founded upon, advice received from Dr Colwell or others at BRE.

¶22.22 It is not clear exactly when BRE decided that it would not produce a report on K15 that would support its use in conjunction with any non-combustible rainscreen. Philip Heath could not recall⁵⁰⁶⁵⁰⁶ Heath {Day78/212:20}-{Day78/213:3}. and Ivor Meredith could only say that it had occurred shortly after the test.⁵⁰⁷⁵⁰⁷ Meredith {Day75/130:14}. Dr Colwell denied that any retraction had occurred. In April 2006, Kingspan was still chasing Dr Colwell for an assessment⁵⁰⁸⁵⁰⁸ {KIN00005163}. and was clearly still expecting it to be one that would confirm the suitability of K15 for use on buildings over 18 metres in height generally and could be sent to all its customers.⁵⁰⁹⁵⁰⁹ {KIN00004795}. Although we do not consider that the precise date of BRE's change of heart is of particular importance, it is of note that by the date of his report on 7 January 2008 at the very latest (and probably considerably earlier),⁵¹⁰⁵¹⁰ Ivor Meredith's report of 7 January 2008 {KIN00008847/5} contained an explanation of the circumstances in which Kingspan had begun a programme of BS 8414-2 testing, the key driver being BRE's refusal to issue a report supporting a broad scope of application. The first test in the programme had taken place on 20 December 2007 but planning for it had begun early in 2006. On 7 April 2006, Dr Malcolm Rochefort noted in his diary, "K15 – do we need another 8414 test?" {KIN00024441}.Show full note Mr Meredith had understood and had made his managers aware that any claims Kingspan made about success in the test carried out in May 2005 could apply only to external walls constructed using the same components in the same configuration.⁵¹¹⁵¹¹ {KIN00008847/5} second paragraph. In his report, he informed his managers that the "BRE are adamant that approval only applies to what is tested".⁵¹²⁵¹² {KIN00008847/5} third paragraph.

Kingspan's presentation of the 2005 test: 2005 and 2006

¶22.23 Ivor Meredith knew that a test in accordance with BS 8414 could not produce a result that would cover all possible uses of K15.⁵¹³⁵¹³ Meredith {Day75/71:20-23}. With the exception, again, of Dr Rochefort,⁵¹⁴⁵¹⁴ Rochefort {Day80/20:22}-{Day80/21:23}; {Day80/25:3-13}; {Day80/35:16}-{Day80/36:12}; {Day80/82:13}-{Day80/83:19}; where Dr Rochefort cited the fact that BS 8414 is a system test as the basis for his lack of action following a later unsuccessful BS 8414-2 test. every Kingspan witness from whom we heard confirmed their understanding that BS 8414 was a system test that had no generic application, still less any application for one particular component of the system tested.⁵¹⁵⁵¹⁵ Pack {Day86/64:11-19}; {Day86/140:5-17}; Meredith {Day75/85:1-19}; Heath {Day78/196:14}-{Day78/197:2}; Mills {Day77/25:11-24}; Millichap {Day81/32:25}-{Day81/33:24}; Pargeter {Day83/113:21-25}-{Day83/114:1-3}; Burnley {KIN00022662/3} page 3, paragraph 1.10. Nonetheless, following the test in May 2005 Kingspan proceeded to claim precisely that in all its marketing and sales literature relating to K15, as well as in its correspondence with customers.

¶22.24 By October 2005, in emails advising customers about the use of K15 on buildings over 18 metres in height, Kingspan was using variations of standard wording produced by Ivor Meredith which he acknowledged to colleagues at the time was possibly "a bit heavy in some areas".⁵¹⁶⁵¹⁶ {KIN00003687} third paragraph. Kingspan always told customers that K15 had passed a test in accordance with BS 8414 and sometimes told them that any material claiming to be fire safe should have passed that test.⁵¹⁷⁵¹⁷ {KIN00002575/2} third paragraph. The text produced by Mr Meredith included a statement that "To fall within the requirements of Approved Document B Kooltherm K15 can continue above 18 metres"⁵¹⁸⁵¹⁸ {KIN00003687} sixth paragraph. and statements were also made to customers that K15 met "the criterion [sic] of BR 135 and thus the 18m rule".⁵¹⁹⁵¹⁹ {KIN00002577/2} While we note that Mr Meredith does refer in this email to a system, he included in that system only K15 and the particular cavity barrier product used in the May 2005 test, manufactured by Promat. He represented that those two products together met the criteria in BR 135, as part of an ultimately unsuccessful attempt to encourage Promat to make representations to the market about the compatibility of the two products. See also {KIN00005172/10} first paragraph for the claim that K15 "easily meets" the criteria in BR 135.Show full note Both claims were false.

¶22.25 In January 2006 Kingspan began sending out to customers⁵²⁰⁵²⁰ {KIN00005133}. a promotional document entitled "What's lurking behind your façade?"⁵²¹⁵²¹ {KIN00005350}. which came to be referred to within Kingspan as the K15 or BS 8414 "flyer".⁵²²⁵²² For example, see {KIN00003688}; {KIN00005179}. In it Kingspan stated that K15 had been tested and "awarded certification to BS 8414-1" and had been "assessed and approved in accordance with BR 135". Both statements were false. BS 8414 is a test method which does not provide the basis for certification of any kind, much less the certification of a particular product, and a single component of a system tested in accordance with BS 8414 could not be assessed or approved in accordance with BR 135 because the performance criteria in BR 135 were applicable only to the system as a whole.

¶22.26 Kingspan withdrew the flyer after it came to the attention of BRE,⁵²³⁵²³ No BRE witness was able to assist us with any details of the complaint. Despite an email copied to Dr Colwell in June 2006 in which Ivor Meredith referred to discussing the complaint with her {KIN00005180}, it was her evidence that she had no recollection of it: Colwell {Day232/223:4-15}, or indeed of ever discussing any Kingspan marketing literature with BRE colleagues {Day232/218:3-8}. who at some stage before 8 June 2006⁵²⁴⁵²⁴ The date of Ivor Meredith's internal email informing technical teams that the flyer had been withdrawn {KIN00005179}. complained to Kingspan about the inaccuracy of its contents.⁵²⁵⁵²⁵ {KIN00005179}; {KIN00005180}; Meredith {Day75/121:10-12}. It does not appear that there was any further complaint from BRE about Kingspan's broader misrepresentation of BS 8414 as a product test. Although the words "certified" and "approved" were eventually removed from the flyer, avowedly to "steer clear of the BRE's wrath",⁵²⁶⁵²⁶ {KIN00005206}. the misleading assertion that K15 as a product had been tested and was acceptable for use above 18 metres "when assessed in accordance with BR 135" persisted in the revised text that Mr Meredith told Kingspan's technical advisors to use thereafter.⁵²⁷⁵²⁷ {KIN00005179}; See draft amended text at {KIN00005194}.

¶22.27 Meanwhile, in 2007 Kingspan began to write what Mr Meredith called "letters of suitability" for the use of K15 on high-rise projects.⁵²⁸⁵²⁸ Meredith {Day76/81:15-23}; {KIN00005308}; In February 2006 Joel Clarke circulated the original BS 8414 flyer to technical staff, commenting that it would be better to send the flyer out than "write tedious letters". He recommended that the flyer should be attached to any correspondence that was being sent to sizeable architects and designers "just in case they've been fooled by the lies Celotex have been spreading about how it doesn't comply": {KIN00005179}.Show full note In emails and letters to customers Kingspan recommended K15 as a suitable product for use on buildings over 18 metres in height. That was done either by asserting that K15 was suitable for use in the particular system proposed by the customer or by representing that K15 was suitable for use in all systems on buildings over 18 metres in height, or both.⁵²⁹⁵²⁹ {KIN00008954/2}; {KIN00005203/2}; {KIN00002576}; {KIN00005216/1}; {KIN00005240}.

¶22.28 Occasionally, Kingspan represented to customers not only that the use of K15 was allowed in any cladding system on a building over 18 metres in height in accordance with Approved Document B, but also that the existence of the data from the BS 8414 -1 test in May 2005 constituted an alternative route to compliance for K15 as a product.⁵³⁰⁵³⁰ {KIN00002577/1}. It was also said that choosing a product that had "passed" a test under BS 8414 might in fact be a more onerous (and, by implication, safer) "route" than choosing a product that was non-combustible or of limited combustibility.⁵³¹⁵³¹ {KIN00002577/1}. In other instances, Kingspan asserted that although K15 was not a non-combustible material, it had "been proved to perform the same as a non-combustible material" by testing in accordance with BS 8414.⁵³²⁵³² {KIN00002583/1}. Again, none of those statements were true.

¶22.29 The text drafted by Ivor Meredith and used by Kingspan's teams in that correspondence was checked by Philip Heath, who admitted not only that he had provided supervision and guidance but also that he had had the final say on the wording.⁵³³⁵³³ Heath {Day79/109:5-19}; Meredith {Day76/82:6-8}. False claims about K15 were therefore not being made in error or by rogue junior employees but with the knowing approval of a senior manager.

¶22.30 Indeed, Philip Heath told us that at that time Kingspan often found itself in the position of having to explain to customers Approved Document B's alternative route to compliance through the BS 8414 test.⁵³⁴⁵³⁴ Heath {KIN00020709/31-32} pages 31-32, paragraph 4.50; Heath {Day78/182:9-11}. Kingspan was keenly aware that there was a lack of detailed knowledge about BS 8414 and BR 135 among its customers, many of whom were unfamiliar with both. It relied on the fact that an unsuspecting market was itself highly likely to rely on it for advice and to accept its claims about K15's suitability for use as reliable. Kingspan was reluctant to send out its one BS 8414 test report⁵³⁵⁵³⁵ BRE Report {BRE00002511}. to any of its customers (it had nothing else to send), but was lucky in that they sometimes accepted its word.⁵³⁶⁵³⁶ {KIN00005380/2}; Meredith {Day76/82:11}-{Day76/83:4}.

¶22.31 Kingspan's strategy was to go after every job.⁵³⁷⁵³⁷ Meredith {Day76/83:3-4}. Mr Meredith's detailed report to senior managers in January 2008⁵³⁸⁵³⁸ {KIN00008847/5} third paragraph. contained an important summary of the strategy at the time, namely, that Kingspan promoted the fact that K15 had "the ability to pass" (referring to the successful May 2005 test). A similar claim appeared in various Kingspan articles for industry journals⁵³⁹⁵³⁹ See for example {KIN00005198/3} third paragraph; {KIN00005219/3} second paragraph; {KIN00005262/3} second paragraph. It is not known if these articles came to be published. and in the BS 8414 flyer,⁵⁴⁰⁵⁴⁰ {KIN00005350} first paragraph. in all of which Kingspan warned readers to question the use of any external cladding products that "do not have the ability" to pass the BS 8414 test. That was linguistic sleight of hand. It was a cynical misrepresentation of the nature and purpose of BS 8414 and BR 135 and of the provisions of Approved Document B relating to them.

¶22.32 Mr Meredith told us that the strategy had been agreed between him and his senior managers as offering Kingspan's only opportunity of moving forward once it had become clear that BRE was not prepared to provide a generic report on the suitability of K15 based on the BS 8414 test in May 2005.⁵⁴¹⁵⁴¹ Meredith {Day75/117:17}-{Day75/118:24}. The other obvious option, namely to behave honestly and with integrity by avoiding false statements, does not appear to have been considered by anyone at a senior level.

Fire Performance claims for new technology K15

¶22.33 In the period between 2005 and 14 June 2017, two important claims were made by Kingspan in relation to the fire performance of K15. The first was that K15 had achieved a classification of Class 0 through testing to Parts 6 and 7 of BS 476. The second was that K15 had been tested to BS 8414 and had met the criteria in BR 135. Kingspan gave both claims prominence, both in its marketing literature for K15⁵⁴²⁵⁴² A classification to Class 0 for K15 was claimed in all K15 product literature from 2001 onwards; reference to BS 8414 was first made in K15 product literature in May 2007 and continued thereafter. March 2001 {KIN00020720}; January 2002 {KIN00009173/1}; November 2002 {KIN00008018/1}; March 2003 {KIN00002579/1}; November 2005 {KIN00022668/1}; June 2006 {KIN00005371/1}; September 2006 amended "What's lurking behind your façade?" {KIN00000055}; May 2007 {KIN00002580/6}; November 2008 {KIN00009703/6}; March 2011 {KIN00003545/1}; July 2016 {KIN00000070/7}.Show full note and in its correspondence with customers. They also appeared in the certificates relating to K15 published by the British Board of Agrément (BBA)⁵⁴³⁵⁴³ {BBA00000038/5}; {BBA00000037/5}; {BBA00000036/6}; {BBA00000040/5}. and Local Authority Building Control (LABC).⁵⁴⁴⁵⁴⁴ {KIN00005705/4}; {KIN00009547/2}; {NHB00000798/1} (Class 0 claim absent from this certificate), {LABC0000997/3}; {LABC0001842/2}; {KIN00016968}; {KIN00000076}.

¶22.34 As explained above, the transition to new technology K15 involved various changes to the chemical composition and manufacturing process, as well as the introduction of perforated foil facers.⁵⁴⁵⁵⁴⁵ {KIN00022307} Major Changes, Row 7; Rochefort {KIN00008838/6} page 6, paragraphs 3.7-3.10; Rochefort {Day80/50:6}-{Day80/55:1}. As a result, new technology K15 had different physical properties as well as a different chemical composition. The K15 product that Kingspan sold after September 2006, although still a phenolic foam, was without question a different product from that which it had previously been selling.

¶22.35 Following the introduction of new technology K15 in 2006, Kingspan continued to make the two principal claims about the product's fire performance without any reference to the changes that had been made to the product, which were not disclosed to the market. In relation to new technology K15 those claims were entirely false. The true position was that no test in accordance with BS 8414 had been carried out on a system incorporating new technology K15 or was carried out until 23 March 2015 and no classification in accordance with the criteria in BR 135 had been made or was made until June 2015.⁵⁴⁶⁵⁴⁶ 23 March 2015 BS 8414-2 test. Test report 302995 dated 14 May 2015 {BRE00002494}; BR 135 Classification Report P100769-1000 dated 8 June 2015 {BRE00002495}. Moreover, new technology K15 did not achieve results when tested in accordance with BS 476 parts 6 and 7 (which was a product test but, under Approved Document B, largely irrelevant to insulation) such as to justify a rating of Class 0 at any stage before the Grenfell Tower fire.⁵⁴⁷⁵⁴⁷ Pargeter {Day84/72:13-16}. That came to light only as a result of our investigations. At no time did Kingspan inform the market that the particular fire tests on which it purported to rely had in fact been carried out on a product that was no longer available and had not been sold since 2007 at the latest.

¶22.36 For many years Kingspan marketed and sold K15 relying heavily on tests which had been carried out on a different product. That was not the result of a mistake, as Dr Rochefort claimed,⁵⁴⁸⁵⁴⁸ Rochefort {Day80/68:9-16}. or misunderstanding; it was done deliberately. The relevant test and classification reports were not withdrawn until October 2020, following the Inquiry's investigations.

Class 0

¶22.37 As has been noted, a classification of national Class 0 was claimed for K15 in all promotional literature from 2001 onwards. Some of Kingspan's marketing brochures claimed that both the K15 product and, separately, its core, were "Class 0 fire rated".⁵⁴⁹⁵⁴⁹ See for example March 2011 {KIN00003545/6}. That was a nonsensical assertion when considered alongside the test standards used to achieve it. Although none of its literature ever disclosed the fact, in order to substantiate its claim, Kingspan had carried out tests in accordance with BS 476 Parts 6 and 7 at various times between 2004 and 2016 not on K15 boards as supplied to the market but on the foil facer of the product alone.⁵⁵⁰⁵⁵⁰ {KIN00022205}. That was done by attaching the foil to a calcium silicate or other non-combustible substrate for the tests.⁵⁵¹⁵⁵¹ See for example {KIN00000256/5}. Kingspan sought to justify its actions by reference to the wording of Approved Document B, which stated that Class 0 was achieved if a material or, in the case of a composite product, the surface of the product met certain criteria when tested in accordance with BS 476 Parts 6 and 7.⁵⁵²⁵⁵² {CLG10000007/122} Appendix A, paragraph 13.

¶22.38 Both Adrian Pargeter and Dr Rochefort told us, and Kingspan in its closing statement submitted, that that was a legitimate practice on the basis of the wording of Approved Document B.⁵⁵³⁵⁵³ Pargeter {KIN00020824/8-10} pages 8-10, paragraphs 3.11-3.18; Pargeter {Day84/69:16-19}; Rochefort {Day80/89:7}-{Day80/90:1}; Kingspan Module 2 closing submissions {KIN00025944/32} page 32, paragraphs 84-87. Mr Pargeter defended it as a technical but common-sense interpretation of the wording in Approved Document B.⁵⁵⁴⁵⁵⁴ Pargeter {Day84/82:23-25}. We disagree. In our view such an interpretation of Approved Document B is over-literal, artificial and has at least the potential to lead to absurd results. Kingspan's Arron Chalmers, who was a candid, if at times lurid, internal commentator on Kingspan's style of doing business, described it as "a bit of a cheat".⁵⁵⁵⁵⁵⁵ {KIN00004168}. Neither Dr Rochefort nor Mr Pargeter had checked the test standards to ensure that it was legitimate for Kingspan to claim that K15 had achieved Class 0 by testing in that way.⁵⁵⁶⁵⁵⁶ Pargeter {Day84/84:12-15}; Rochefort {Day80/89:17-21}. Had they done so, they would have discovered that the practice was at odds with both the purpose of BS 476-6 and its wording, both of which make clear that it is designed as a test of a product, not part of a product.⁵⁵⁷⁵⁵⁷ {BRE00005557/18}. Kingspan had sought the advice of Frans Paap at Exova on the subject. He apparently told it that the provision could be interpreted in that way⁵⁵⁸⁵⁵⁸ {KIN00004168/1}. but said that it should test the complete product before making any claim about Class 0 because the foam behind it was likely to affect the performance of the facer.⁵⁵⁹⁵⁵⁹ {KIN00004168/2}. Adrian Pargeter accepted that the interpretation was neither conservative nor ideal but said that Kingspan was prepared to adopt a literal interpretation of Approved Document B that might be less safe but would facilitate sales of K15.⁵⁶⁰⁵⁶⁰ Pargeter {Day84/118:10}-{Day84/119:6}.

¶22.39 To the extent that the tests relied on to support the claim that K15 was rated Class 0 had been carried out on the foil facer alone, that claim was false. Understandably, there was evidence of some disquiet about it even within Kingspan's own ranks.⁵⁶¹⁵⁶¹ Meredith {Day76/181:4}-{Day76/182:8}; {KIN00004168/1}. Other than for one specific and limited purpose relating to the spacing of cavity barriers,⁵⁶²⁵⁶² Lane, Phase 1 Report Section 11 {BLAR00000006/41} paragraph 11.11.18; Millichap {Day81/42:13-16}. Class 0 was not relevant to the suitability of insulation for use on buildings over 18 metres in height, as Kingspan knew. No Kingspan witness claimed otherwise. Nonetheless, in much of its correspondence Kingspan either treated Class 0 as evidence of the suitability of K15 for use on buildings over 18 metres in height or treated it as the equivalent of limited combustibility. For example, it sometimes told customers in very general terms that Class 0 was the "highest possible product fire performance classification", using language which effectively elided the two concepts. It also made recommendations that K15 be used for projects involving buildings over 18 metres in height, immediately followed by a statement that K15 had achieved Class 0, as though the two were related.⁵⁶³⁵⁶³ See examples at {KIN00005363/8}; {KIN00005218/2}; {KIN00009014/1}; {KIN00002583/2}; {KIN00002882/1}.

The fire performance of new technology K15

¶22.40 At the time of the change from old to new technology K15 in 2006, and for some time after, Philip Heath and Dr Rochefort considered the properties of the two products to be very similar.⁵⁶⁴⁵⁶⁴ Heath {KIN00020709/9} page 9, paragraph 3.4, "…basically the same chemistry but different production technology..."; Heath {Day78/227:2-4}; Heath {Day79/11:20-25}; Rochefort {KIN00008838/7} page 7, paragraph 3.10; Rochefort {Day80/54:23}-{Day80/55:1}. Dr Rochefort explained that the similarity between them on a chemical level was such that he had not expected their fire performance to differ significantly.⁵⁶⁵⁵⁶⁵ Rochefort {KIN00008838/7} page 7, paragraph 3.10; Rochefort {KIN00008838/50} page 50, paragraph 11.22; Rochefort {Day80/67:19-22}. He did not regard the introduction of perforations to the foil facers as significant,⁵⁶⁶⁵⁶⁶ Rochefort {KIN00008838/7} page 7, paragraph 3.10; Rochefort {Day80/54:18-25}. because the composition was the same⁵⁶⁷⁵⁶⁷ Rochefort {Day80/53:13-17}. and the perforations represented less than 0.5% of the surface area of the board.⁵⁶⁸⁵⁶⁸ Rochefort {Day80/54:1-5}. However, in an email to Philip Heath on 7 July 2005 Lamtec, the manufacturers of the foil facers, had alerted Kingspan in the clearest terms to the fact that the fire performance of the foam would be significantly different once the facers were perforated because the perforations would potentially allow gases produced by combustion to reach the atmosphere and burn.⁵⁶⁹⁵⁶⁹ {MET00081264}. The warning appears to have been ignored.

¶22.41 Despite the apparent similarities between the old and the new products, Kingspan's Product and Process Development System was engaged in order to assess any differences between various aspects of their performance.⁵⁷⁰⁵⁷⁰ {KIN00022307} Major Changes, Row 7, Column N; Heath {Day79/16:7}-{Day79/17:14}; Rochefort {KIN00008838/9} page 9, paragraph 3.16. Following its completion new technology K15 was able to enter the market.⁵⁷¹⁵⁷¹ Rochefort {KIN00008838/9} page 9, paragraph 3.16.

¶22.42 Dr Rochefort had revised the Product and Process Development System at an earlier stage to ensure that the programme for the change to new technology would include fire testing.⁵⁷²⁵⁷² {KIN00022307} Major Changes, Row 7, Column N; Rochefort {KIN00008838/9} page 9, paragraph 3.16. Responsibility for fire testing and certification fell to Philip Heath,⁵⁷³⁵⁷³ {KIN00022003/2}; Heath {Day79/22:21}-{Day79/23:7}. who did not sign the section off as complete⁵⁷⁴⁵⁷⁴ {KIN00022307} Major Changes, Row 7, Column N. until 21 February 2008.⁵⁷⁵⁵⁷⁵ Rochefort {KIN00008838/9} page 9, paragraph 3.16. He and Ivor Meredith explained that the "sign off" of any particular section of the Product and Process Development System did not in fact indicate completion but rather that something was "in hand" and confirmed that fire testing had continued for many years after the new technology had become available to the market in 2006 and the process had been signed off by Mr Heath in February 2008.⁵⁷⁶⁵⁷⁶ Meredith {Day75/39:3-13}; Heath {Day79/18:14-25}. In those circumstances, it is difficult to see what real purpose Kingspan's Product Process and Development System served in this case.

¶22.43 Philip Heath was not able to explain why Kingspan had started to sell the new form of K15 years before fire testing on it had been completed.⁵⁷⁷⁵⁷⁷ Heath {Day79/21:5}. Dr Rochefort told us that it was difficult in practice to have everything signed off before a new product was sold and that as long as testing was still being carried out without any obvious problems that had been considered good enough.⁵⁷⁸⁵⁷⁸ Rochefort {Day80/58:5-10}. He assumed at the time that new technology K15 had achieved European Class C for the purposes of CE marking,⁵⁷⁹⁵⁷⁹ The letters "CE" are the abbreviation of French phrase "Conformité Européenne" which literally means "European Conformity". It is a marking which certifies that a product has met EU health, safety, and environmental requirements. which he had understood (wrongly) to equate to national Class 0.⁵⁸⁰⁵⁸⁰ Rochefort {Day80/60:13}-{Day80/61:1}. Mr Meredith thought that testing to European standards had been undertaken before new technology K15 had been made available to the market and that a Class C had been achieved,⁵⁸¹⁵⁸¹ Meredith {Day75/155:6-9}. but he could not recollect any other testing.⁵⁸²⁵⁸² Meredith {Day75/40:10-16}.

Testing in accordance with BS 476 parts 6 and 7

¶22.44 Neither Mr Meredith nor Dr Rochefort nor Mr Heath could recall the results of tests carried out in accordance with BS 476-6 or BS 476-7 on the new K15 product that was actually produced by Kingspan at Pembridge.⁵⁸³⁵⁸³ Meredith {Day75/41:10-15}; Heath {Day79/25:12-13}; {Day79/28:21}-{Day79/29:24}; {Day79/77:12-16}; Rochefort {Day80/89:1-16}. Indeed, neither Dr Rochefort nor Mr Heath could tell us clearly what fire tests had been carried out on new technology K15 or whether any difference in fire performance between the old and the new technology products had been discovered. That was despite Mr Heath's acknowledging that he, together with Research and Development Manager Vincent Coppock⁵⁸⁴⁵⁸⁴ Heath {KIN00020709/15} page 15, paragraph 3.24. and Technical Processing Manager Gwyn Davies,⁵⁸⁵⁵⁸⁵ Rochefort {KIN00008838/11} page 11, paragraph 3.22. had been principally responsible for the testing⁵⁸⁶⁵⁸⁶ Heath {Day79/24:18}-{Day79/25:9}. and despite Dr Rochefort's acceptance that as Technical Director (to whom Mr Heath, Mr Coppock and Mr Davies all reported)⁵⁸⁷⁵⁸⁷ Rochefort {Day80/4:23}-{Day80/5:6}. responsibility for ensuring that products were fit for market, including in terms of fire safety, ultimately fell to him.⁵⁸⁸⁵⁸⁸ Rochefort {Day80/8:23}-{Day80/9:2}.

¶22.45 Quite apart from that unsatisfactory evidence, we were struck by the fact that there appeared to be no composite record of the fire testing carried out on new technology K15 or of any reported difference in fire performance between the old and the new technology formulations. Although it was difficult to discern what fire testing had actually been undertaken, what did become clear was that there had been no comprehensive programme leading to a proper understanding of the fire performance of new technology K15 before it entered the market. Mr Heath admitted as much and accepted that that should have occurred.⁵⁸⁹⁵⁸⁹ Heath {Day79/21:6-12}.

¶22.46 A detailed analysis of the contemporaneous documents, including notes from internal meetings and reports written by Mr Meredith, revealed that from 2005 onwards a number of small and medium-sized tests had been carried out on the new technology K15. That included calorimetry and single burning item tests as well as repeated testing to BS 476 Parts 6 and 7. The documents also revealed that none of that testing appears, in Kingspan's view at the time, to have brought positive news about the fire performance of new technology K15. In particular, numerous internal documents between 2006 and early 2009 reveal discussions about Kingspan's inability to obtain a classification to national Class 0 for the new product and details of its continuing efforts to do so over the course of many years. The documents were widely circulated among technical staff, including Dr Rochefort and Mr Heath. During the same period, however, Kingspan claimed for all external purposes that K15 had achieved that classification.

¶22.47 In a half-yearly report prepared for Mr Heath in July 2007, Ivor Meredith identified one of the low points of the period as a failure to verify basic properties for new technology K15, including its rating as Class 0.⁵⁹⁰⁵⁹⁰ {KIN00005292/11} penultimate bullet point. In the same report, he referred to numerous failed attempts to pass the BS 476-6 test and said that repeats of tests that had narrowly failed and the addition of fire retardants were being considered.⁵⁹¹⁵⁹¹ {KIN00005292/7} first paragraph.

¶22.48 In an internal report sent to Dr Rochefort and Mr Heath on 7 January 2008 that referred to the European single burning item test, Mr Meredith reported that the introduction of perforations in the foil facers of new technology K15 had led to a reduction in the European classification. He recommended the removal of the perforations and the addition of a fire retardant, noting that the latter could also help to achieve Class 0.⁵⁹²⁵⁹² {KIN00008847/3} eighth paragraph; {KIN00022466/1} paragraph 1.1, third bullet point "BS 476 Part 6 still elusive for certain formulations…".

¶22.49 Dr Rochefort told us that Kingspan's fire focus group had done quite a lot of work to establish the effect on fire performance of the newly introduced perforations in the foil facer.⁵⁹³⁵⁹³ Rochefort {Day80/64:12-17}. Mr Meredith told us that the matter had been discussed at monthly technical meetings; his hypothesis was that the perforations allowed the blowing agent in the foam to ignite more readily,⁵⁹⁴⁵⁹⁴ Meredith {Day75/155:10}-{Day75/156:9}. causing a deterioration in fire performance. We note that Professor Bisby's findings about the effect of the perforations tend to support that hypothesis.⁵⁹⁵⁵⁹⁵ Bisby, Phase 2 Experiments Work Package 2 Report {LBYWP200000001/120} paragraph 635. Dr Rochefort did not say what work had actually been carried out and we found no documents that shed light on any such assessment, other than a reference in the minutes of a meeting on 25 July 2008 to plans to perforate only one side of the facer. We note that at that meeting, which Dr Rochefort attended, someone stated that perforations had been shown to reduce fire performance considerably.⁵⁹⁶⁵⁹⁶ {KIN00022467/1} paragraph 1.2. In any event, new technology K15 was available for sale before any such assessment had been undertaken, as Dr Rochefort acknowledged.⁵⁹⁷⁵⁹⁷ Rochefort {Day80/64:18}-{Day80/65:7}.

¶22.50 In a monthly technical report produced in April 2008, Mr Meredith reported indicative calorimeter tests comparing old and new technology K15 in which the latter had showed a faster time to ignition and double the heat output (or rate of heat output, as we understand it).⁵⁹⁸⁵⁹⁸ {KIN00003698} third paragraph.

Testing under BS 8414

¶22.51 By the end of 2005, BS 8414 Part 2 had been published, containing the large-scale test method for external cladding systems supported by steel frames rather than masonry.⁵⁹⁹⁵⁹⁹ October 2005 {BSI00000097}. As a result, challenges began to arise in relation to Kingspan's use of test data from its BS 8414-1 test in 2005 to support the use of K15 at height on steel-framed buildings, for which no BS 8414 test data existed.⁶⁰⁰⁶⁰⁰ Heath {Day79/36:18-25}; See for example {KIN00005243/2}. In an internal report in January 2008, Ivor Meredith noted that despite Kingspan's fighting the case for each project, the absence of any BS 8414-2 test data was causing it to lose sales on a daily basis. He considered it essential to obtain such data to support the sale of K15 for use on buildings over 18 metres in height.⁶⁰¹⁶⁰¹ {KIN00008847/2} fourth paragraph; Meredith {Day75/146:3-8}.

¶22.52 In 2007 and 2008, four tests under BS 8414-2 were carried out on systems incorporating new technology K15. Two of them, on 20 December 2007⁶⁰²⁶⁰² Test 239825{KIN00008847/2}; {BRE00032344/1-3}; {BRE00032344/121-124}; {BRE00032344/170-171}. and 24 April 2008 respectively,⁶⁰³⁶⁰³ Test reference unknown. See {BRE00032344/183}; {BRE00032344/187-190}. were carried out on systems using solid aluminium rainscreen panels⁶⁰⁴⁶⁰⁴ Eggington {SOT00000029/2} page 2, paragraph 9. and were undertaken in partnership with Sotech Ltd, a rainscreen manufacturer, and Voestalpine Metsec plc, a supplier of structural steel frame systems. The other two, on 9 April⁶⁰⁵⁶⁰⁵ Test 241103 {BRE00011552}. and 6 June 2008 respectively,⁶⁰⁶⁶⁰⁶ Test 241105 {BRE00011552}. were carried out in partnership with Kingspan Off- Site and incorporated Kingframe steel loadbearing systems⁶⁰⁷⁶⁰⁷ Stevens {MET00080972/4} page 4, final paragraph. and rendered rainscreen boards.⁶⁰⁸⁶⁰⁸ {BRE00011552/1}. All four systems failed to meet the performance criteria in BR 135.⁶⁰⁹⁶⁰⁹ Heath {Day79/35:7-10}; Rochefort {Day80/70:4-8}. Note that during Test 241105 on 6 June 2008 (Kingspan Off Site) the heat source was extinguished early, thereby terminating the test, due to the failure of the hearth. The rig was allowed to burn for the full duration of 60 minutes for observations to be made. See {BRE00011552} paragraph 1, lines 2-3, "... the full requirements of the standard were not however complied with..."; Wilkinson {MET00080973/6-7} pages 6-7, fourth paragraph.Show full note At all times, Kingspan concealed from its three co-sponsors, including Kingspan Off-Site, the fact that its previous test in accordance with BS 8414-1 in May 2005 had been carried out on a system that incorporated the old formulation of K15.⁶¹⁰⁶¹⁰ Dr Rochefort told us that he did not know whether Kingspan Off-Site had been told that the 2005 test in accordance with BS 8414-1 had been carried out on a different, old version of K15: Rochefort {Day80/118:7-11}. All the other witnesses who were involved in this testing confirm that Kingspan Off-Site had not been told: Stevens {MET00080972/12} page 12, second paragraph; Wilkinson {MET00080973/8} page 8; Heath {Day79/90:10-12}; Meredith {Day75/171:20}-{Day75/172:4}; {KIN00020917/1} penultimate paragraph.Show full note

¶22.53 In January 2008, after the first test on 20 December 2007, Ivor Meredith (who had not been present at the test) sent Mr Heath, Dr Rochefort and others a report containing his detailed observations and an analysis of what had occurred.⁶¹¹⁶¹¹ {KIN00008847}. Describing the result as a "raging inferno", he explained that the top fire barrier had been breached 17 minutes into the test, that the system had failed the temperature criteria in BR 135⁶¹²⁶¹² {KIN00008847/2} penultimate paragraph. and that the test had been terminated early because there had been a danger that the laboratory would be set alight.⁶¹³⁶¹³ {KIN00008847/3} first paragraph.

¶22.54 Mr Meredith also recorded specific observations about the performance of the K15 insulation, as well as offering views (both his own and those of BRE) on the performance of the new K15 compared with the old, which had been tested in 2005. He included photographs of that earlier test in his report in order to demonstrate the differences he was describing.⁶¹⁴⁶¹⁴ {KIN00008847/21-22}. He said that the new K15 had burnt ferociously "on its own steam" and that it was, in his words, "... very different in a fire situation to the previous technology ...".⁶¹⁵⁶¹⁵ {KIN00008847/3}.

¶22.55 Mr Meredith recorded what he described in the report as "official" and "unofficial" comments from BRE.⁶¹⁶⁶¹⁶ {KIN00008847/3}. He thought they had been provided by Philip Clark,⁶¹⁷⁶¹⁷ Meredith {Day75/160:3-13}. who had been present at the test together with Dr Colwell and Tony Baker⁶¹⁸⁶¹⁸ "Test witnesses" {KIN00008847/2}. and had been the most approachable on the subject of testing.⁶¹⁹⁶¹⁹ Meredith {Day75/160:5-7}. The "unofficial" comments were that the insulation had been fully involved in the fire, that the foam core had continued to burn after the flame source had been extinguished, that surface spread of flame on the K15 was apparent and finally, that BRE had not remembered the product performing like that in the previous test. The "official line" was that the system as a whole had failed and that responsibility for the failure could not be attributed to any individual component.⁶²⁰⁶²⁰ {KIN00008847/3}.

¶22.56 All three BRE witnesses denied that they had made "official" and "unofficial" comments in the way described by Mr Meredith and insisted that that was not something BRE would do.⁶²¹⁶²¹ Clark {Day96/166:16-19}; {Day96/168:4-7}; {Day96/173:18-19}; Colwell {Day232/228:18}-{Day232/229:11}; Baker {Day100/59:8-10}; {Day100/60:4-8}. Dr Colwell acknowledged that it would have been wholly inappropriate for BRE to do that.⁶²²⁶²² Colwell {Day232/228:8-20}. However, Mr Clark told us that comments to that effect could have been made by Dr Colwell or Mr Baker⁶²³⁶²³ Clark {Day96/165:1-11}. and went on to say that although he could not recall making them himself, he might have commented on how bad the burning was over the surface of the K15 and that he could not rule out having made them.⁶²⁴⁶²⁴ Clark {Day96/166:2-7}. We think it unlikely that Mr Meredith had invented comments of that kind or in that detail. We prefer his evidence and are satisfied that Mr Clark did make them and that Dr Colwell was probably aware that he had done so. That is supported by Mr Meredith's contemporaneous report and by a number of other contemporaneous documents. One particularly striking example is an email sent on 9 January 2008 by Mr Meredith to Dr Colwell in which he asked her to confirm that BRE would not deviate from the "official statement" and would not state in writing that the failure was due to the flame spread across the K15 insulation.⁶²⁵⁶²⁵ {KIN00003693/1} final paragraph; {KIN00020908/1} first paragraph and penultimate paragraph.

¶22.57 Both Mr Heath and Dr Rochefort confirmed that they had read Mr Meredith's report at the time,⁶²⁶⁶²⁶ Heath {Day79/38:1-7}; Rochefort {Day80/70:13-20}. including his strongly worded observations about the performance of K15 in the test in December 2007, in which he had stated that the way in which the K15 had burned was of the "most concern".⁶²⁷⁶²⁷ {KIN00008847/3}. Mr Heath told us that he had been concerned by the report.⁶²⁸⁶²⁸ Heath {Day79/53:6-9}. Indeed, he said that he had already been aware that Mr Meredith had concerns about the fire performance of new technology K15 and had agreed that he should write the report in order to draw the attention of more senior Kingspan personnel to them.⁶²⁹⁶²⁹ Heath {Day79/38:2-13}. Save for conveying those concerns to Dr Rochefort and offering support to Mr Meredith, however, Mr Heath did not take any action.⁶³⁰⁶³⁰ Heath {Day79/56:17}-{Day79/57:2}. Dr Rochefort told us that he too had been disturbed by the report⁶³¹⁶³¹ Rochefort {Day80/76:10-14}. and that he had understood that Mr Meredith was attempting to make him and others sit up and take notice that there was reason to be worried about the fire performance of the new K15 product.⁶³²⁶³² Rochefort {Day80/77:14-25}.

¶22.58 Nonetheless, the report was effectively ignored by Dr Rochefort, despite his claim to the contrary.⁶³³⁶³³ Rochefort {Day80/81:11-16}. Echoing BRE's "official line" as recorded in Mr Meredith's report,⁶³⁴⁶³⁴ {KIN00008847/3}. both Mr Heath and Dr Rochefort attempted to explain Kingspan's inaction by saying that BS 8414 was a system test and that it was possible that components of the system other than the K15 might have been responsible for the failure.⁶³⁵⁶³⁵ Rochefort {Day80/83:4}-{Day80/83:19}; Heath {Day79/36:14-17}. Apparently for those reasons, neither Dr Rochefort nor Mr Heath gave consideration to suspending the sale of K15.⁶³⁶⁶³⁶ Rochefort {Day80/82:2-10}; {Day80/91:17-24}; Heath {Day79/68:4-20}. Dr Rochefort said that after the test in December 2007 he had considered moving back to the old technology and that he had discussed that option in a couple of internal emails,⁶³⁷⁶³⁷ Rochefort {Day80/91:24}-{Day80/92:8}. but his recollection was not supported by any documents we have seen. It was also inconsistent with the minutes of the many technical meetings held at that time,⁶³⁸⁶³⁸ 30 January 2008 {KIN00022466}; 21 February 2008 {KIN00023870}; 25 July 2008 {KIN00022467}; 22 August 2008 {KIN00023871}; 22 October 2008 {KIN00022468}. during each of which Kingspan's plans for dealing with the fire performance of the new technology were specifically discussed, often as a "high priority" or as "urgent" business. In none of those minutes was there any mention of reverting to old technology K15.⁶³⁹⁶³⁹ {KIN00022467/1} paragraph 1.2; {KIN00022468/1} paragraph 1.2, fourth bullet point. We think that Dr Rochefort's memory was unreliable on that point.

¶22.59 On 9 April 2008, a second test of a system incorporating new technology K15 was carried out in accordance with BS 8414-2 but again it failed to meet the performance criteria in BR 135, with temperatures exceeding the required levels at around 16 minutes into the test.⁶⁴⁰⁶⁴⁰ {KSO00000099/6} final paragraph; {KIN00008844/1} sixth paragraph. That test was organised by Kingspan Off-Site, whose Project and Development Engineer, Aidan Wilkinson,⁶⁴¹⁶⁴¹ Wilkinson {MET00080973/1} page 1, final paragraph. witnessed it. He described the system as having performed "abysmally".⁶⁴²⁶⁴² Wilkinson {MET00080973/4} page 4, third paragraph. There was then a third test on 24 April 2008, which was the second test carried out in partnership with Sotech and Metsec. Various modifications had been made to the system tested in December 2007 to try to improve its performance, but again it failed to meet the performance criteria in BR 135.⁶⁴³⁶⁴³ Eggington {SOT00000029/3} page 3, paragraphs 17-18. John Eggington was the Managing Director of Sotech at the time of these tests in 2007 and 2008. Despite that record of persistent failure, neither Dr Rochefort nor Mr Heath nor anyone else at Kingspan was moved to take any further action.

¶22.60 On 6 June 2008 Kingspan Off-Site carried out its second test at BRE in accordance with BS 8414-2 (the fourth in the series). After the test, Aidan Wilkinson, and Mark Stevens, his manager and Technical Team Leader,⁶⁴⁴⁶⁴⁴ Wilkinson {MET00080973/1} page 1, final paragraph; Stevens {MET00080972/3} page 3, first paragraph. contacted Kingspan to share their observations about the performance of K15 during both tests and to raise questions about it. In particular, Mr Stevens and Mr Wilkinson had observed that the K15 had continued to burn for more than 30 minutes after removal of the source of ignition.⁶⁴⁵⁶⁴⁵ {KIN00009031/1}; {KIN00003714/1}. Kingspan Off-Site questioned Kingspan several times both by email and by telephone about whether that was normal or to be expected for K15; they also asked Kingspan for help in identifying what else might have caused the prolonged burning.⁶⁴⁶⁶⁴⁶ {KIN00003704}; {KIN00020917}. That was a pressing matter for Kingspan Off-Site, which had already arranged to carry out a series of further tests of systems incorporating K15.⁶⁴⁷⁶⁴⁷ Email from Mr Stevens to Mr Meredith {KIN00009031/1}.

¶22.61 Rather than telling Kingspan Off-Site what all three by then plainly knew about the worrying fire performance of new technology K15,⁶⁴⁸⁶⁴⁸ Rochefort {Day80/121:15-16}; Heath {Day79/88:5-8}; {KIN00003704/1} "The question of K15s bad fire performance is no longer just an internal one". on 2 July 2008 Dr Rochefort, Mr Heath and Mr Meredith worked together to craft a carefully worded response.⁶⁴⁹⁶⁴⁹ {KIN00003714/1}. They purported to express surprise at the poor performance of the product, given what Mr Heath described as Kingspan's previous success with BS 8414, and focused on the possibility that particular details of the test rig or inconsistencies in fire testing generally might have been the cause of the problem.⁶⁵⁰⁶⁵⁰ {KIN00020916}.

¶22.62 Having responded in those terms, Mr Heath sent an email to his Kingspan colleague Gwyn Davies in which he said, "I'm spinning so much I'm dizzy".⁶⁵¹⁶⁵¹ {KIN00009031/1}. Mr Heath, to his discredit, denied before us that that was effectively a boast about being less than truthful,⁶⁵²⁶⁵² Heath {Day79/92:2}-{Day79/93:19}. but that is obviously what he meant. He later accepted that there had been gaps in the information he had provided and said that he had been trying to delay matters so that Kingspan could develop a better product for testing.⁶⁵³⁶⁵³ Heath {Day79/92:6}-{Day79/94:7}. At one stage, Mr Meredith seems to have been unsure of the account he was to give Kingspan Off-Site, asking Philip Heath by email, "Are we saying that the product supplied which failed their test is the same product that we got to pass the test?".⁶⁵⁴⁶⁵⁴ {KIN00020917/1} penultimate paragraph; See also text in red added by Mr Meredith to {KIN00008844/2-3}. It is clear to us that both Philip Heath and Ivor Meredith were, and knew that they were, answering Kingspan Off-Site in terms that were not simply vague but actively designed to conceal and mislead.

¶22.63 After further questions from Kingspan Off-Site on 16 July 2008, Dr Rochefort suggested that Kingspan should say that it ought not to carry out further testing as planned while Kingspan investigated possible reasons for the failure, on the basis that the "same batch…may suffer from a similar problem".⁶⁵⁵⁶⁵⁵ {KIN00003714}. As Dr Rochefort acknowledged, Kingspan well knew that the problem was not limited to any particular batch.⁶⁵⁶⁶⁵⁶ Rochefort {Day80/121:9-16}.

¶22.64 The simple truth behind Mr Heath's "spin" was that Kingspan did not have a version of K15 which it considered capable of performing well enough within a system to enable the criteria in BR 135 to be met;⁶⁵⁷⁶⁵⁷ Heath {Day79/83:8-13}; {Day79/92:14-16}. Kingspan was working to understand why, as Mr Meredith put it in an email to Mr Heath on 3 July 2008, old tech worked and new tech did not.⁶⁵⁸⁶⁵⁸ {KIN00020917/1}. Having carefully considered all the evidence, we are in no doubt that it was that fact that drove Kingspan to conceal from Kingspan Off-Site its serious reservations about the fire performance of new technology K15, rather than reasons of confidentiality, as Mr Heath suggested,⁶⁵⁹⁶⁵⁹ Heath {Day79/90:13}-{Day79/91:17}. or a desire not to cause alarm, as Dr Rochefort claimed.⁶⁶⁰⁶⁶⁰ Rochefort {Day80/121:22}-{Day80/122:10}.

¶22.65 The internal correspondence from July 2008 revealed numerous discussions between the three (and others) about trials, modifications and additives which might improve the fire performance of new technology K15.⁶⁶¹⁶⁶¹ {KIN00008844}. The contemporaneous documents show that Kingspan was becoming desperate by this time to find answers to its predicament. Mr Meredith described his request for samples of a particular version of the foam for further testing as "business critical".⁶⁶²⁶⁶² {KIN00003704/1}.

¶22.66 In an internal email on 9 June 2008, Dr Rochefort wrote to Mr Meredith and others saying, "Ivor, if you're confident that old process will pass the test, we may have no choice but to provide old process K15 for Offsite until the FR [fire retardant]⁶⁶³⁶⁶³ Rochefort {Day80/110:10-16}. issue is sorted out", adding, " ... we can't carry on providing something that we know fails a fire test".⁶⁶⁴⁶⁶⁴ {KIN00003704/1}. Dr Rochefort told us that despite saying "for Offsite", he meant that Kingspan might have to revert entirely to producing and supplying old technology K15 across the market.⁶⁶⁵⁶⁶⁵ Rochefort {Day80/110:3}-{Day80/111:21}. That is not what either Mr Heath or Mr Meredith understood him to mean, however, and neither of them thought that changing back to old technology K15 for the market generally had been considered or discussed as an option at the time.⁶⁶⁶⁶⁶⁶ Heath {Day79/81:22}-{Day79/82:2}; Meredith {Day75/197:3-7}. We saw no other document that contained a reference to any consideration of reverting to old technology K15 for general manufacture and sale and we do not believe that Dr Rochefort, or anyone else at Kingspan, gave serious consideration to stopping production of new technology K15. Instead, Kingspan simply hoped to improve the product's fire performance while continuing to sell it and claiming compliance with Approved Document B on the basis of the 2005 test.

¶22.67 In July 2008, Mr Meredith reported that a fire retardant version of K15 was being sent to BRE for indicative calorimetry testing in the hope of finding a lower calorific output and slower time to ignition, characteristics that he said were required to maintain Kingspan's position in the high rise facade market.⁶⁶⁷⁶⁶⁷ {KIN00008848/2}. He told us that this testing was being carried out in an attempt to find small scale tests in which new technology K15 demonstrated a fire performance similar to that of old technology.⁶⁶⁸⁶⁶⁸ Meredith {Day75/199:7-11}.

¶22.68 Mr Meredith said that he felt a tremendous responsibility to ensure that the product Kingspan was selling was fit for the market.⁶⁶⁹⁶⁶⁹ Meredith {Day75/157:20-23}. It is clear to us from the evidence that by July 2008 he was worried about the performance in fire of the K15 then being sold. He made repeated and strongly worded attempts to make it clear to his managers how grave his worries were, but his impression at the time was that even if they were listening to him, they were neither responding to, nor acting on, what he was telling them.⁶⁷⁰⁶⁷⁰ Meredith {Day75/157:24}-{Day75/159:1}.

¶22.69 Both Dr Rochefort and Mr Heath told us that they had taken Mr Meredith's concerns seriously.⁶⁷¹⁶⁷¹ Rochefort {Day80/130:8-14}; {Day 80/132:19-23}; Heath {Day79/56:10-16}. They each said that they had been concerned by what they had read in the report he had made in January 2008 and by the performance of K15 in the BS 8414 tests carried out in April and June 2008.⁶⁷²⁶⁷² Rochefort {Day80/76:10-14}; {Day80/109:22}-{Day80/110:2}; Heath {Day79/53:6-9}; {Day79/77:10-16}. Philip Heath accepted that he had been aware in 2008 that there was a "paper trail"⁶⁷³⁶⁷³ {KIN00020913/4}. showing considerably better performance in fire for old technology K15 than for new.⁶⁷⁴⁶⁷⁴ Heath {Day79/82:25}-{Day79/83:6}. Despite their professed concerns, however, they appear to have taken no action. In particular, no consideration was given to suspending or withdrawing sales of K15, either then or at any time.⁶⁷⁵⁶⁷⁵ Rochefort {Day80/82:2-10}; {Day80/91:10}-{Day80/92:4}; Heath {Day79/65:25}-{Day79/66:23}.

¶22.70 Mr Meredith said that Kingspan's reaction to the problem had amounted to no more than hope and the suggestion that further testing might provide a better result.⁶⁷⁶⁶⁷⁶ Meredith {Day75/158:11-16}. That is consistent with the evidence of Dr Rochefort, who said that Kingspan's response to the concerns expressed by Mr Meredith was that a lot of work was being done to see whether the problems could be solved.⁶⁷⁷⁶⁷⁷ Rochefort {Day80/132:21-23}. It is also clearly borne out by the documents, in particular the records of Kingspan's technical meetings, which show that efforts to improve the fire performance of the new technology foam continued throughout 2008 and into early 2009,⁶⁷⁸⁶⁷⁸ Minutes of meetings on 22 August 2008 {KIN00023871}; 22 October 2008 {KIN00022468/1} paragraph 1.2; 25 February 2009 {KIN00022475/1-2} paragraph 1.2. ceasing only on the instructions of Mr Heath following the publication of the LABC Type Approval certificate for K15 in May 2009.⁶⁷⁹⁶⁷⁹ {KIN00005382}. Kingspan continued to sell K15 to an unsuspecting market without interruption or warning while hoping that a positive test result might turn up.

The first BBA Certificate: October 2008

¶22.71 The first BBA certificate for K15 was issued on 27 October 2008⁶⁸⁰⁶⁸⁰ Certificate 08/4582 Issue 1 {BBA00000038/1}. after several years of work by Kingspan in support of its application.⁶⁸¹⁶⁸¹ Meredith {Day75/41:7-8}. Those involved from Kingspan were Senior Technical Adviser Gareth Mills⁶⁸²⁶⁸² Mills {KIN00022329/11} page 11, paragraph 36; Mills {Day77/5:13-24}. assisted by Ivor Meredith.⁶⁸³⁶⁸³ Meredith {Day75/201:17-24}. Both were supervised by Philip Heath until 2010 and thereafter by Tony Millichap, who replaced Mr Heath as Head of Technical when Mr Heath moved to another part of the business at the end of 2009.⁶⁸⁴⁶⁸⁴ Millichap {KIN00020821/3} page 3, paragraph 2.4; Mills {Day77/4:12-19}.

¶22.72 Kingspan was well aware of the importance of BBA certificates in achieving acceptance of construction products by architects and building control bodies.⁶⁸⁵⁶⁸⁵ Heath {Day79/132:12}-{Day79/133:15}. Mr Mills told us that he had regarded a BBA certificate as a marketing tool that would give customers greater confidence in the product;⁶⁸⁶⁶⁸⁶ Mills {KIN00022329/11} page 11, paragraph 39; Mills {Day77/33:15}-{Day77/34:20}. Mr Meredith described a BBA certificate as a "green light of acceptability".⁶⁸⁷⁶⁸⁷ Meredith {Day75/201:8-11}. We agree with that characterisation. The process of obtaining the first BBA certificate and deciding its final content involved various discussions between the BBA and Kingspan, principally Mr Mills, as well as comments on drafts being sent back and forth between the two companies. Kingspan circulated drafts internally to a team of proof readers for checking and comment. Mr Mills collated the comments and sent them back to the BBA.⁶⁸⁸⁶⁸⁸ Mills {KIN00022329/15} page 15, paragraph 45(H); Mills {Day77/51:12}-{Day77/52:10}.

"Will not contribute to the development stages of a fire … "

¶22.73 The first BBA certificate contained three important statements about the fire performance of K15. The first, which appeared on the first page under the heading "Key Factors Assessed", was that "… the boards will not contribute to the development stages of a fire or present a smoke or toxic hazard".⁶⁸⁹⁶⁸⁹ Certificate 08/4582 Issue 1{BBA00000038/1}. No witness from Kingspan or from the BBA was able to offer a coherent explanation of the meaning of, or the basis for, that sentence or how it could properly be said to apply to K15, a combustible polymeric foam. Mr Mills did not know what it meant but said he had had no basis for disagreeing with it.⁶⁹⁰⁶⁹⁰ Mills {Day77/43:1-12}. Mr Meredith agreed that it was very vague but that he had believed it was a plausible statement.⁶⁹¹⁶⁹¹ Meredith {Day75/202:23}; Meredith {Day75/204:6-13}. Mr Heath could not tell us what it had been based on but said that he thought it had been intended to reflect the behaviour in fire of thermoset materials.⁶⁹²⁶⁹² Heath {Day79/139:11-15}. Christopher Hunt of the BBA, who had approved the certificate,⁶⁹³⁶⁹³ Christopher Hunt was Head of Approvals (Physics) at the BBA and signed the certificate: Hunt {BBA00011087/3} page 3, paragraph 7; Certificate 08/4582 Issue 1 {BBA00000038/1}. told us that he was not sure what the scientific or evidential basis for the statement might have been.⁶⁹⁴⁶⁹⁴ Hunt {Day108/174:21-25}. He acknowledged that it was "slightly ambiguous"⁶⁹⁵⁶⁹⁵ Hunt {Day108/170:8-13}. and more confusing than helpful.⁶⁹⁶⁶⁹⁶ Hunt {Day108/175:2}. John Albon, the BBA's Head of Approvals from 2013 to 2019,⁶⁹⁷⁶⁹⁷ Albon {BBA00010723/3} page 3, paragraph 13. told us that he had been concerned by the statement when he read it,⁶⁹⁸⁶⁹⁸ Albon {Day110/112:2-21}. describing it in his evidence as an opinion that could not be substantiated and was not appropriate.⁶⁹⁹⁶⁹⁹ Albon {Day110/112:20}-{Day110/115:1}. He therefore removed it.⁷⁰⁰⁷⁰⁰ Albon {Day110/112:14-18}. However, that was not until October 2015, when the fourth version of the certificate was issued.⁷⁰¹⁷⁰¹ Certificate 14/5134 Issue 1 {BBA00000040/1}.

Class 0

¶22.74 The second significant statement in the BBA certificate about the fire performance of K15 was that it was classified as Class 0 and could therefore be used in accordance with the provisions of paragraphs 12.5 and 12.6 and Diagram 40 of Approved Document B.⁷⁰²⁷⁰² Certificate 08/4582 Issue 1{BBA00000038/5-6} paragraph 7.2. As we have explained, that was untrue. In October 2008 Kingspan had not been able to obtain a Class 0 classification for the product it was selling and continued, internally, to discuss and report on its attempts to do so throughout 2008 and the early part of 2009.⁷⁰³⁷⁰³ {KIN00022205}; {KIN00022467/1} paragraph 1.2; {KIN00022468/1} paragraph 1.2; {KIN00022475/1-2} paragraph 1.2; {KIN00022476/1} paragraph 1.2.

¶22.75 Although the BBA had requested reaction to fire and surface spread of flame test data from Kingspan at an earlier stage in the preparation of the certificate in September and December 2004,⁷⁰⁴⁷⁰⁴ {BBA00011093}. it appears that the only data of that kind provided to the BBA related to tests undertaken between 1991 and 2003 on various products, some produced in Kesteren, rather than at Pembridge, but none of them tests on the new technology K15 that Kingspan was currently selling.⁷⁰⁵⁷⁰⁵ See summary at Moore {Day107/175:5-24}. Adrian Pargeter told us that investigation of historic records in preparation for the Inquiry had revealed that after the change to new technology in 2006, Kingspan could only have been relying on tests carried out on separate and different products⁷⁰⁶⁷⁰⁶ Pargeter {Day84/40:13-20}. These were reports of tests carried out on Kesteren-produced products: BS 476-6 on DL 2000 {BBA00011291}; Part 7 on DL3300 {BBA00011292}; Parts 6 and 7 on DL3300 {BBA00011293}. to support its claim that K15 had achieved Class 0. (It also relied on the results of tests carried out on the foil facers alone but they were not provided to the BBA).⁷⁰⁷⁷⁰⁷ Pargeter {KIN00022610/29} page 29, paragraph 3.40; Pargeter {Day84/39:6-24}. Ivor Meredith and Philip Heath, as they acknowledged in their evidence, had been aware of that at the time.⁷⁰⁸⁷⁰⁸ Meredith {Day75/41:10-15}; {Day79/28:21}-{Day79/29:24}; Heath {Day79/25:10-13}.

¶22.76 On 1 July 2008, Gareth Mills wrote to the BBA⁷⁰⁹⁷⁰⁹ {BBA00000244}. proposing the inclusion of the statement that K15 was classified to Class 0. He sent a document listing Kingspan's comments⁷¹⁰⁷¹⁰ {BBA00000250/1} fifth row; Mills {Day77/57:16}-{Day77/59:12}. on a BBA draft of the certificate.⁷¹¹⁷¹¹ {BBA00000249}. The text he proposed was an extract from a Kingspan product marketing brochure for K15.⁷¹²⁷¹² Mills {Day77/58:19-24}. Mr Mills did not send the BBA any data derived from tests carried out in accordance with BS 476-6 or BS 476-7 but said that he had thought that such data existed and that it might have been sent to the BBA at some other time by someone else.⁷¹³⁷¹³ Mills {Day77/59:7-12}; {Day77/60:1}-{Day77/61:23}. Whether or not that was his true understanding at the time, it was not correct. In effect, by dint of Mr Mills' efforts, the BBA did no more than recycle Kingspan's own false marketing literature.

¶22.77 Neither at the time the first certificate was published nor at any time thereafter did the BBA have any evidence at all on which to base a statement that K15 had achieved Class 0, as its own witnesses accepted⁷¹⁴⁷¹⁴ Albon {Day110/156:19}-{Day110/157:7}; Moore {BBA00011097/4} page 4, paragraph 15. and as Adrian Pargeter effectively confirmed.⁷¹⁵⁷¹⁵ Pargeter {Day84/44:6-9}. Kingspan thus persuaded the BBA to state that the product being sold as K15 had achieved a Class 0 rating (just as it was itself doing in its marketing literature)⁷¹⁶⁷¹⁶ {KIN00002580/6}. knowing that statement to be false.

The BS 8414 test in 2005

¶22.78 The first issue of the BBA certificate also referred to a test carried out in accordance with BS 8414-1. It provided a fairly lengthy description of the system tested but did not state the date on which the test had taken place and was imprecise about the details of key parts of it. In particular, the certificate made no mention of the masonry backing for the test rig and instead used the phrase "non-combustible substrate". Most significantly, the description of the system ended with the words, "Within the stated test time the temperature at the level 2 thermocouples did not exceed 600°C, therefore displaying limited fire spread away from the fire source and that the product meets the criteria stated within BRE 135".⁷¹⁷⁷¹⁷ Certificate 08/4582 Issue 1{BBA00000038/5} paragraph 7.1. That assertion was fundamentally misleading, since the criteria in BR 135 relate to the performance of the system as a whole, not that of any particular component.

¶22.79 When asked about that wording, Mr Heath agreed that it was incorrect⁷¹⁸⁷¹⁸ Heath {KIN00020709/48} page 48, paragraph 7.35. and sought to explain it away as an oversight on the part of Kingspan.⁷¹⁹⁷¹⁹ Heath {Day79/140:22}-{Day79/141:1}. Mr Meredith at first gave a similar explanation, suggesting that Kingspan had had only a limited time to check the accuracy of the BBA's draft.⁷²⁰⁷²⁰ Meredith {Day75/205:14}-{Day75/206:13}. In fact, however, it is clear that drafts of the certificate had been passing between the BBA and Kingspan for several months from June 2008 onwards before the certificate had been issued.⁷²¹⁷²¹ Kingspan provided detailed comments on at least three BBA drafts: 1 July 2008 {BBA00000250}; 15 July 2008 {BBA00000211}; 3 October 2008 {BBA00002874}. The correspondence reveals that the statement that the product had met the criteria in BR 135 had been proposed by Gareth Mills on 3 October 2008 at a fairly late stage in the process.⁷²²⁷²² {BBA00002874/1}. That was contrary to the evidence Mr Mills had given in his witness statement, which was that the BBA, rather than Kingspan, had suggested the wording.⁷²³⁷²³ Mills {KIN00022329/14} page 14, paragraph 44(L).

¶22.80 The BBA's first draft of the certificate relating to K15 dated 4 June 2008 was sent to Kingspan for comment. It included a statement in paragraph 4.7 that K15 was combustible but that it could be used in various circumstances "… where no storey is at a height of more than 18m above ground ... ".⁷²⁴⁷²⁴ {BBA00000249/5} paragraph 4.7. In an email of 27 June 2008 to Mr Mills and Mr Heath, Mr Meredith expressed the view that publication of that wording would be a "disaster".⁷²⁵⁷²⁵ {KIN00024474}. In the same email, he urged caution over the plan to "slip the existing BS 8414 data in at the last minute so they didn't have much time to look over it and pick holes ...", warning that it might be foolish to think that the BBA would do anything other than go over the data with a fine-tooth comb. Mr Meredith need not have worried. The BBA did no such thing and inserted Mr Mills' suggested text verbatim into the final version of the certificate.⁷²⁶⁷²⁶ {BBA00002874/1}; Certificate 08/4582 Issue 1{BBA00000038/5} paragraph 7.1.

¶22.81 Mr Mills said that he could not remember who had proposed the change from the BBA's original description of the system tested in accordance with BS 8414 as based on a "blockwork" substrate⁷²⁷⁷²⁷ {BBA00000219/7} paragraph 7.1. to one with a "non combustible" substrate.⁷²⁸⁷²⁸ {BBA00000211/7} paragraph 7.1. However, correspondence within Kingspan reveals that the idea had come from Philip Heath, who suggested it on the basis that "it might allow us a little spin in the future".⁷²⁹⁷²⁹ {KIN00024476/1}. Kingspan had been using similarly inexact language in its correspondence with customers for some time⁷³⁰⁷³⁰ {KIN00005240/2}. and went on to make use of it more widely in its K15 technical bulletins and product literature following the publication of the BBA certificate in October 2008.⁷³¹⁷³¹ {KIN00002607/3}; {KIN00003545/6}; {KIN00009585/2}.

¶22.82 Ivor Meredith understood that the data obtained from the BS 8414-1 test in 2005 related to a system with a masonry substrate and did not apply to a system with a steel framed structure⁷³²⁷³² {KIN00020913/4}; Heath {Day79/43:21}-{Day79/44:2}; {KIN00003701}. and had made his managers aware of that. Kingspan sought to recommend K15 for use on high-rise buildings by stating (or at any rate implying) that the 2005 test data applied equally to masonry and steel-framed structures as long as the latter were sheathed with a non-combustible board.⁷³³⁷³³ {KIN00002607/3}. As the correspondence reveals⁷³⁴⁷³⁴ {KIN00024476}. and as Mr Meredith and Mr Mills acknowledged,⁷³⁵⁷³⁵ Meredith {Day76/72:23}-{Day76/73:7}; Mills {Day77/82:10-22}. that wording was consciously designed to conceal the absence of any data derived from a test in accordance with BS 8414-2, which by 2008 had become a pressing problem for Kingspan.⁷³⁶⁷³⁶ {KIN00008847/2}; {KIN00020913/4}; {KIN00003701/1}; Meredith {Day75/146:3-8}. Kingspan's advice to customers that K15 could be used on buildings over 18 metres in height with any non-combustible substrate was not based on any test evidence or the advice of any suitably qualified expert.⁷³⁷⁷³⁷ Heath {Day79/47:18}-{Day79/48:6}.

"Advice should be sought from the certificate holder"

¶22.83 A few lines below the description in paragraph 7.1 of the system tested in accordance with BS 8414 the BBA certificate contained paragraph 7.3, which told the reader that, "In buildings with a floor more than 18m above ground level, advice should be sought from the certificate holder."⁷³⁸⁷³⁸ Certificate 08/4582 Issue 1{BBA00000038/6} paragraph 7.3. Mr Mills told us that he could not remember how that paragraph had come to be included.⁷³⁹⁷³⁹ Mills {Day77/49:7-9}. He remembered that Kingspan had asked the BBA to remove from an early draft a passage that explicitly restricted the use of K15 to buildings under 18 metres in height.⁷⁴⁰⁷⁴⁰ Mills {KIN00022329/13} page 13, paragraph 44(A); Mills {Day77/50:2-16}. When questioned about that, he accepted that he had proposed the removal from the summary on the front page of the certificate of the words "up to 18m" on the basis that the BS 8414 test report demonstrated that the product could be used on buildings over 18 metres in height.⁷⁴¹⁷⁴¹ {BBA00000219/1}; {BBA00000211/1} text in red. He said that he had offered to provide the test report to the BBA in order to support his proposal.⁷⁴²⁷⁴² Mills {Day77/31:8-13}; {Day77/49:10}-{Day77/51:11}; {Day77/67:2}-{Day69/69:2}. It could do no such thing, of course, for reasons of which Kingspan was well aware.

¶22.84 The section of the BBA certificate that advised readers to contact Kingspan when considering the use of K15 on a building over 18 metres in height was of substantial benefit to Kingspan. When potential customers sought its advice, Kingspan almost invariably advised them that K15 was suitable for such use, either in a generic sense or for a particular proposal. When doing so, it referred them to the BBA certificate, which directed them straight back to itself. Mr Meredith acknowledged that the clause was very welcome, because it meant that customers would contact Kingspan, giving it an opportunity to sell K15 for use on their projects.⁷⁴³⁷⁴³ Meredith {Day76/6:13-19}. In later internal correspondence from January 2013, it was referred to as a "get-out" clause⁷⁴⁴⁷⁴⁴ {KIN00005572/4}. that Kingspan tried to avoid losing.

¶22.85 The self-serving nature of the provision thus allowed Kingspan to give direct advice to potential customers about whether to use K15 on any particular building. The plain and obvious conflict of interest which that created appears to have passed unnoticed by the BBA. NHBC, the largest private building control body in the country, continued to treat that advice as unremarkable until as late as the autumn of 2013, when the significance of the sheer volume of K15 on high-rise buildings began to dawn on it.

What the BBA did not know

¶22.86 Although the BBA was not aware of the fact,⁷⁴⁵⁷⁴⁵ Albon {Day111/28:10}-{Day111/29:7}. the description in the certificate of the BS 8414 test carried out on K15 rested on a fundamental falsehood. Despite a term of the contract that obliged Kingspan to notify the BBA of any changes to the product or of any new or additional information it had acquired about its suitability for use,⁷⁴⁶⁷⁴⁶ {BBA00010745/3} paragraph 7(g). Kingspan did not tell the BBA that the test in 2005 had been carried out on a version of K15 that was by then obsolete and was no longer being manufactured.⁷⁴⁷⁷⁴⁷ Heath {Day79/137:3-7}; Millichap {Day81/194:13-17}. When asked about that, Mr Mills and Mr Heath told us that they had thought the BBA was aware of the change to new technology and the foil facers⁷⁴⁸⁷⁴⁸ Mills {Day77/41:4-14}; Heath {Day79/136:16}-{Day79/137:17}. because of the audits and quality plans in which they had been recorded.⁷⁴⁹⁷⁴⁹ Heath {Day79/136:16}-{Day79/137:17}; Mills {Day77/40:3-9}; {Day77/41:4-12}.

¶22.87 We found no reference to the changes in the composition of K15 or the introduction of perforated foil facers in any BBA document. In any event, it is difficult to see how any quality plan or audit could have recorded the change to new technology. K15 had not been subject to BBA certification at the time of the changes and there would therefore have been no audit or quality plans in which to record that information.⁷⁵⁰⁷⁵⁰ Meredith {Day75/40:21}-{Day75/41:3}; Albon {Day111/28:2-7}, "Once we have accepted a report, certified the product, and then have the production under surveillance, we continue to accept that data as being representative of the performance of the product on the understanding that the formulation and specifications are, from that point onwards, unchanged". Moreover, it became clear that no initial on-site assessment of K15 had taken place before the BBA issued the first certificate, apparently because several other Kingspan products had already obtained BBA certificates and the BBA was satisfied with its quality standards.⁷⁵¹⁷⁵¹ {BBA00000161/3} final three paragraphs; Albon {BBA00010751/22} page 22, paragraph 74; Albon {Day110/162:11} -{Day110/164:24}; Hunt {Day109/39:12}-{Day109/40:1}.

¶22.88 Moreover, Kingspan did not tell the BBA that four tests on systems incorporating new technology K15 had failed to meet the criteria in BR 135⁷⁵²⁷⁵² Mills {KIN00022329/13} page 13, paragraph 43(H); Heath {Day79/138:5-9}; {Day79/142:12-16}. or that, even in October 2008, it had misgivings about the fire performance of the product.⁷⁵³⁷⁵³ Heath {Day79/137:18}-{Day79/138:4}; {Day79/142:17-21}. Philip Heath told us that Kingspan did not consider those matters to be relevant.⁷⁵⁴⁷⁵⁴ Heath {Day79/138:1-4}; {Day79/143:1-4}. Mr Mills also told us that the four BS 8414-2 test results were not relevant to the scope of approval Kingspan was seeking from the BBA.⁷⁵⁵⁷⁵⁵ Mills {Day77/38:13}-{Day77/39:12}. We disagree. They were plainly relevant, since they were the only tests in accordance with BS 8414 that had been carried out on systems incorporating the product that was the subject of the certificate.

Amendments proposed by the BBA: December 2008

¶22.89 On 24 December 2008, less than two months after the BBA certificate had been issued, George Lee, the Project Manager who had drafted it,⁷⁵⁶⁷⁵⁶ Lee {BBA00010794/1} page 1, paragraph 1; Albon {BBA00010751/8} page 8, paragraph 31. sent an email to Kingspan in which he set out a number of proposed amendments to the part dealing with behaviour of the product in relation to fire. He explained that the BBA had received a number of comments on the clarity of the wording of that section. In particular, he proposed that the section on the first page of the certificate headed "Key Factors Assessed" should be amended by adding the following wording: "The product has been tested to BS 8414-1 for a specific construction on masonry walls".⁷⁵⁷⁷⁵⁷ {KIN00009103}.

¶22.90 Eventually, on 5 March 2009, Philip Heath responded. He expressed concern about the proposal to make changes to the wording of the certificate so soon after it had been issued and warned that there might be cost implications for Kingspan, which he would need to assess before he responded. Two minutes later, he sent the exchange to Ivor Meredith, Gareth Mills and Andrew Pack, then Technical Services Manager.⁷⁵⁸⁷⁵⁸ Pack {KIN00008702/4} page 4, paragraph 2.8. His email said simply, "Let the file gather dust, guys". Plainly, as Mr Pack and Meredith admitted (but Philip Heath would not),⁷⁵⁹⁷⁵⁹ Heath {KIN00020709/81}, page 81 paragraph 11.47; Heath {Day79/154:20}-{Day79/155:16}. He characterised it as an instruction "not to proactively respond". that was an instruction to put off responding to the BBA's proposed amendments for as long as possible.⁷⁶⁰⁷⁶⁰ Pack {Day86/147:7-11}; Meredith {Day75/211:2-12}. Mr Meredith told us candidly that this instruction had been given because it was in Kingspan's interests for the existing BBA certificate to remain in circulation. He said that Kingspan had not been prepared to allow the BBA to make it clear to the public that there were restrictions on the use of K15 on buildings over 18 metres in height because it could limit sales.⁷⁶¹⁷⁶¹ Meredith {Day75/212:4-12}. Mr Mills admitted that that had possibly been the reason for the instruction to delay.⁷⁶²⁷⁶² Mills {Day77/102:23}-{Day77/103:4}. Mr Heath denied that, stating that, although Kingspan was keen to extend the scope of the approval, he had simply been concerned about the implications of the amendments for marketing materials, which by that time had been produced.⁷⁶³⁷⁶³ Heath {Day79/154:20}-{Day79/155:24}.

¶22.91 We do not think that Mr Heath was telling the truth. The explanation given by Mr Meredith was strongly supported by the contemporaneous internal correspondence,⁷⁶⁴⁷⁶⁴ {KIN00024474}; {KIN00024476}. which, in particular, included Kingspan's comments on drafts of the certificate. They show that it was trying to persuade the BBA to certify that K15 was suitable for use on buildings over 18 metres in height in a general and unlimited sense and that K15 as a product met the criteria in BR 135.⁷⁶⁵⁷⁶⁵ {BBA00000250/1-2}; {BBA00000211/1}; {BBA00000211/7}; {BBA00002874/1}.

Marketing and challenges: 2008–2009

¶22.92 Throughout 2008, and in the early part of 2009, Kingspan continued to market K15 as a product which had met the criteria in BR 135 and was therefore suitable for use on buildings over 18 metres in height, both on masonry and metal-framed substrates. Its marketing literature included drawings of both types of construction without qualification. Kingspan's promotional literature referred to the 2005 test without giving a date and without referring to the fact that it was applicable only to masonry substrates. It made no effort to draw the attention of the reader to the fact that BS 8414 is a method for testing a system as a whole rather than an individual component or product.⁷⁶⁶⁷⁶⁶ {KIN00009703/6}; {KIN00002607/3}. Under the supervision of Philip Heath⁷⁶⁷⁷⁶⁷ Heath {Day79/109:5-19}; {KIN00003723}; {KIN00003718}. it did the same in its responses to customers who asked for advice about the use of K15 at height and continued to write "letters of suitability" representing that the 2005 test data was "directly applicable" to various proposed constructions when it plainly was not.⁷⁶⁸⁷⁶⁸ {KIN00002589/3-4}; {KIN00009042}; {KIN00002594/2}. Once the BBA certificate for K15 had been issued, Kingspan relied on the sections describing the BS 8414 test and the suitability of the product for use on buildings over 18 metres in height which it had shaped so carefully, sometimes referring to it as constituting the BBA's approval of the product.⁷⁶⁹⁷⁶⁹ {KIN00002592/3}; {KIN00003723/1}.

¶22.93 Despite those efforts to conceal the true position about K15's performance in fire, the period was marked by challenges from astute designers. One such was Wintech Group Ltd, a facade engineering consultancy,⁷⁷⁰⁷⁷⁰ Taylor {WIN00000002/1} page 1, paragraph 1. which in 2006 had begun to ask questions about the suitability of K15 for use in high-rise projects and the relevance of Kingspan's 2005 test to steel-framed structures.⁷⁷¹⁷⁷¹ {KIN00005243/2}. Both Mr Heath and Mr Meredith acknowledged that the questions it raised were valid,⁷⁷²⁷⁷² Heath {Day79/115:22}-{Day79/116:9}; {Day79/120:22}-{Day79/121:7}; Meredith {Day76/86:19-24}. but at the time Kingspan fought off the challenges as best it could. Mr Meredith told other designers that Wintech was taking a precautionary approach to following the guidance in Approved Document B.⁷⁷³⁷⁷³ {KIN00024478/1} penultimate paragraph.

¶22.94 Wintech's persistence led members of the technical team to seek advice from Mr Heath. On one occasion Mr Meredith asked him for a standard answer because it was becoming difficult for him to know what he could write without risking a legal confrontation. Mr Heath's response was "Wintech can go f#ck themselves, and if they are not careful we'll sue the a#se of them".⁷⁷⁴⁷⁷⁴ {KIN00005363/1}. Mr Heath forwarded the exchange to a friend outside Kingspan, calling the situation a "nightmare" and writing "I'm trying to think of a way out of this one, imagine a fire running up this tower !!!!!!!!!!!!!!!!!!".⁷⁷⁵⁷⁷⁵ {KIN00009066/1}. Even if we were to accept Mr Heath's explanation that the emails were an attempt to make light of a situation which was causing him "frustration",⁷⁷⁶⁷⁷⁶ Heath {Day79/118:4-25}; {Day79/124:5-19}. we think, despite his denials,⁷⁷⁷⁷⁷⁷ Heath {Day79/124:11-19}. that they expose a casual disregard for public safety at a senior level in Kingspan, a determination to defend K15's position in the market at all costs and a keen awareness on Kingspan's part that it needed to find a way out of the situation it had created through its own mendacity.

The LABC Certificate 2009

¶22.95 As has been explained elsewhere in this report,⁷⁷⁸⁷⁷⁸ Chapter 23. in May 2009 Kingspan obtained an LABC System Type Approval Certificate for K15. That certificate, obtained by Andrew Pack and Philip Heath, stated that K15 could be considered a material of limited combustibility and that it was suitable for use in all situations shown in Diagram 40 of Approved Document B, including on buildings over 18 metres in height.⁷⁷⁹⁷⁷⁹ {KIN00005705/5}. Philip Heath's jubilation at obtaining the certificate was clear from his contemporaneous correspondence, in which he said the document could be "GOLD".⁷⁸⁰⁷⁸⁰ {KIN00024991/1}. He celebrated the formal confirmation of its issue on 7 May 2009 with the single word "FANBLOODYTASTIC".⁷⁸¹⁷⁸¹ {KIN00024993/1}.

¶22.96 On 7 May 2009, Mr Heath circulated the "GREAT NEWS!" to various internal Kingspan mailing lists, offering special thanks to Mr Pack, summarising the benefits the certificate would bring and highlighting in bold in his quote from the certificate the words "material of limited combustibility". In the same email, Mr Heath said that the certificate had completed the "tool kit" for K15.⁷⁸²⁷⁸² {KIN00005383}. The toolkit, he explained, was the marketing information used to sell the product.⁷⁸³⁷⁸³ Heath {Day79/189:8-12}.

¶22.97 In response to a question from Dr Rochefort asking what test results had been used to obtain the certificate, Mr Heath said that Kingspan had sent so much fire data to David Jones, the certificate's author, that it had "probably blocked his server" and that LABC had convinced itself that K15 was the "best thing since sliced bread" without even having "to get any real ale down him".⁷⁸⁴⁷⁸⁴ {KIN00020714}. Dr Rochefort never received a proper answer to his question.⁷⁸⁵⁷⁸⁵ Rochefort {Day80/146:15}-{Day80/147:7}. His suggestion that testing of which he had been wholly unaware might have been carried out in Australia, China or Singapore was simply not credible.⁷⁸⁶⁷⁸⁶ Rochefort {Day80/144:23}-{Day80/146:14}. The reality is that there was no answer. As Mr Pack accepted, Mr Heath's claim was nonsense.⁷⁸⁷⁷⁸⁷ Pack {Day86/135:5-16}. No fire test data at all had been provided to LABC.

¶22.98 On 11 May 2009, in an email to Mr Meredith, Dr Rochefort and others, Mr Heath repeated that the LABC certificate satisfied the requirements for K15 to be installed on buildings over 18 metres in height and gave instructions that no further fire testing was to be carried out on K15.⁷⁸⁸⁷⁸⁸ {KIN00005382}. He told us that the decision to stop testing and to rely on the LABC certificate to promote the suitability of K15 for use on buildings over 18 metres in height was a collective one.⁷⁸⁹⁷⁸⁹ Heath {Day79/193:11-13}. During a meeting of Kingspan's Fire Focus Group on 18 May 2009, attended by Dr Rochefort and Mr Meredith but not by Mr Heath, it was agreed that work would continue on the development of a K15 product that would be able to pass a BS 8414-2 test as a back up.⁷⁹⁰⁷⁹⁰ {KIN00008845/1-2}; Meredith {Day76/58:21}-{Day76/60:20}. However, Mr Heath then reversed that decision, confirming in an email to Mr Meredith on 17 June 2009 that no further facade testing involving K15 would be carried out and that the LABC certificate would be relied on "until we are challenged".⁷⁹¹⁷⁹¹ {KIN00005387}. It is therefore clear that at that time Kingspan accepted that it had no K15 product capable of forming part of a facade system that would meet the criteria in BR 135 when tested in accordance with BS 8414-2⁷⁹²⁷⁹² Rochefort {Day80/155:7-20}; {Day80/143:10-20}. and that it made a calculated decision to use the LABC certificate to gloss over the absence of relevant test evidence.

¶22.99 Both Mr Heath and Mr Meredith accepted that LABC's central statement about K15's suitability for use on buildings over 18 metres in height (described in an internal presentation to Kingspan's sales teams as "the highlight")⁷⁹³⁷⁹³ {KIN00009178/5}. was false and thoroughly misleading. Mr Heath accepted that he had expected it to be challenged.⁷⁹⁴⁷⁹⁴ Heath {Day79/207:11-15}; Meredith {Day76/24:24}. Mr Meredith, who said that at the time he had been "drowning in inquiries" about the use of K15 on buildings over 18 metres in height, told us that he was simply following suit from his line manager that, in the absence of any relevant test evidence, the LABC certificate was the solution.⁷⁹⁵⁷⁹⁵ Meredith {Day76/35:12-24}. He too expected it to be challenged. Once the certificate had been obtained, the strategy was to send it out to customers and let it do the talking.⁷⁹⁶⁷⁹⁶ {KIN00005385}.

¶22.100 By August 2009, that strategy appeared to be working. In an email sent that month to others in the technical and sales teams (including Mr Heath), Mr Meredith noted that questions about the use of K15 at height had died out since obtaining the LABC System Type Approval.⁷⁹⁷⁷⁹⁷ {KIN00009135/1}.

Amended BBA certificates 2010-2013

¶22.101 Meanwhile, dust had been gathering on the amendments to the certificate relating to K15 that the BBA had proposed in December 2008.⁷⁹⁸⁷⁹⁸ {KIN00009103}. Although the next version of the certificate bears the date 6 April 2010,⁷⁹⁹⁷⁹⁹ Certificate 08/4582 Amended Issue 1{BBA00000037/1}. that was said to be the date on which it had been amended; it is clear from later correspondence that it was not the date on which it was issued.⁸⁰⁰⁸⁰⁰ {BBA00002644/7}. No Kingspan witness was able to identify the date on which it had been issued. John Albon and Jon Denyer⁸⁰¹⁸⁰¹ Denyer {BBA00010780/4} page 4, paragraph 13. thought it had been issued on 12 July 2013, although there was some uncertainty about it.⁸⁰²⁸⁰² Denyer {BBA00010780/10} page 10, paragraph 38 (publication date inserted by Jon Denyer); Albon {Day110/109:9‑25}. We have seen no evidence to confirm or contradict that later date.

¶22.102 The amended certificate contained three significant changes to the text relating to the performance of K15 in fire. First, the section on the front page headed "Key Factors Assessed" contained the additional words that Mr Lee had suggested, namely, "The product has been tested to BS 8414-1:2002 for one specific construction on masonry walls".⁸⁰³⁸⁰³ Certificate 08/4582 Amended Issue 1{BBA00000037/1}. The second change was to the description of the system tested in accordance with BS 8414 in 2005, which now stated explicitly that BS 8414-1 was a test for masonry substrates and omitted the previous assertion that "... the product meets the criteria stated within BRE135".⁸⁰⁴⁸⁰⁴ Certificate 08/4582 Amended Issue 1 {BBA00000037/5-6}; See previous wording {BBA00000038/5}. Last, in the passage immediately following the statement that

¶K15 was rated national Class 0 (which had previously said that the product could be used in accordance with the provisions of paragraphs 12.5 and 12.6 of Approved Document

¶B)⁸⁰⁵⁸⁰⁵ Certificate 08/4582 Issue 1 {BBA00000038/6}. the amended version said that the product could also be used in accordance with the provisions of paragraph 12.7 of the guidance.⁸⁰⁶⁸⁰⁶ Certificate 08/4582 Amended Issue 1 {BBA00000037/5};

¶22.103 That proposed draft of the amended certificate was sent to Kingspan on 8 April 2010 by Mr Hunt, who explained the reason for the amendments and sought Kingspan's agreement to them.⁸⁰⁷⁸⁰⁷ {BBA00002644/7}. In his response, which was not sent until 21 June 2010, Mr Mills effectively proposed reverting to the original, i.e. the complete removal of the amended wording on the first page ("for one specific construction on masonry walls") and the addition at the end of the section describing the system tested in accordance with BS 8414 the words "… therefore meeting the criteria in BR 135".⁸⁰⁸⁸⁰⁸ {BBA00002644/6}. After that, no further progress was made for many years. Kingspan failed to respond to another email from Mr Hunt on 9 July 2010 and the BBA decided that the certificate would need to be re-issued rather than amended.⁸⁰⁹⁸⁰⁹ {BBA00002644/1-5}.

¶22.104 Discussions resumed in January 2013 as part of what appear to have been steps leading to the eventual issue of a new certificate in July 2013. Mr Mills, Mr Meredith and a senior technical adviser at Kingspan, Joel Clarke,⁸¹⁰⁸¹⁰ Clarke {KIN00008790/4} page 4, paragraph 2.6. considered how best to prevent the inclusion in the certificate of wording that would indicate any restriction on the use of K15 on buildings over 18 metres in height.⁸¹¹⁸¹¹ {KIN00009353}; {KIN00005570}. Tony Millichap oversaw those discussions, which showed that Kingspan was now conscious of its waning influence over the BBA in relation to the contents of the certificate. Indeed, in an email sent to the other three on 9 January 2013 Mr Clarke commented that he thought Kingspan might be "pushing our luck" in seeking to restore the advice to contact the certificate holder⁸¹²⁸¹² {KIN00005570/2}. and said that at a meeting the day before people had said that the BBA was "onto us".⁸¹³⁸¹³ {KIN00005570/1}. That may have been right. When the amended certificate was issued Mr Mills' comments appear to have been ignored and Kingspan's attempts to ensure that the certificate should include wording to the effect that "the product" had met the criteria in BR 135⁸¹⁴⁸¹⁴ Email from Gareth Mills to the BBA dated 31 May 2013 {KIN00002145/4}, "... for the section relating to BS 8414, we would want the general description of the construction tested included, i.e. similar to that is (sic) our existing certificate, along with some text stating that the result achieved met the criteria from BRE 135...". had not been successful.

¶22.105 Despite the BBA's evidence⁸¹⁵⁸¹⁵ Albon {Day109/56:13}-{Day109/57:2}. and Kingspan's arguments to the contrary,⁸¹⁶⁸¹⁶ {KIN00025944/34-37} paragraphs 92-101. we think that the BBA's statement that K15 could be used in accordance with the provisions of paragraph 12.7 of Approved Document B was dangerously misleading because it suggested that K15 was a material of limited combustibility, which was not the case. Apart from Mr Mills, who seems to have thought that the statement was accurate,⁸¹⁷⁸¹⁷ Mills {Day77/109:6}-{Day77/115:4}. all the other witnesses who were asked about it, including Mr Meredith⁸¹⁸⁸¹⁸ Meredith {Day76/3:17-20}. and Mr Millichap,⁸¹⁹⁸¹⁹ Millichap {Day81/80:8-20}. acknowledged that the reference to paragraph 12.7 in those terms was inaccurate and misleading. Mr Albon also acknowledged in his evidence that it was misleading.⁸²⁰⁸²⁰ Albon {Day110/179:15}. In its closing statement for

¶Module 2, however, Kingspan went to some lengths to persuade us otherwise.

¶22.106 As we have explained in Chapter 48 of this report, the purpose of paragraph 12.7 was to restrict the use of combustible insulation materials at height. It did so by advising that only insulation materials of limited combustibility should be used on buildings over 18 metres in height, except where they form part of a cavity wall construction and are enclosed by two leaves of masonry.⁸²¹⁸²¹ {CLG10000007/96}. The exception for masonry cavity wall construction reflects the fact that the insulation is protected from fire on each side by a thick layer of non-combustible material. It was clearly intended to deal with a narrowly defined type of construction wholly unlike those for which the BS 8414 test and the BR 135 criteria were designed.

¶22.107 The passage in question has to be read and understood in the context of the certificate as a whole, which states that it relates to K15 "for use to create a warm frame wall construction or for use against masonry substrates, with a ventilated cavity and a weatherproof cladding system / protective rainscreen". The accompanying illustration depicts its use with a steel frame construction. Other sections of the certificate, such as paragraphs 1.1, 4, 5.2, 8.2, 8.5 and 12, suggest that it was not directed to the use of the product as insulation inside an enclosed masonry cavity wall and we have seen no clear evidence to suggest that either Kingspan or the BBA had such a use in mind. However, the real vice of the statement lies in the apparently innocuous word "therefore", which provides the link between the phrase "Class 0 or 'low risk'" and the statement that the product may be used in accordance with the provisions of paragraph 12.7 of Approved Document B. It is clear that insulation used within a masonry cavity wall construction does not need to be Class 0 or 'low risk' or of limited combustibility. The causal link can therefore be relevant only to the first part of paragraph 12.7, thereby suggesting that K15 is of limited combustibility. We therefore think that the overall thrust of the certificate was clear in that respect and that it was both erroneous and misleading.

¶22.108 On 17 December 2013, the BBA issued a third version of the certificate for K15.⁸²²⁸²² Certificate 08/4582 Issue 2{BBA00000036}. It significantly tightened the previous language in relation to the use of K15 on buildings over 18 metres in height, in particular by explicitly stating that one specific build-up had met the criteria in BR 135 through a test in accordance with BS 8414 Part 1. In a footnote, it repeated that K15 had been tested in accordance with BS 8414 only in one specific construction and added that a separate test would be required to establish the performance of any other combination of materials. It also removed the statement containing the reference to paragraph 12.7 of Approved Document B and the section directing readers to contact Kingspan for advice on the use of K15 on buildings over 18 metres in height.⁸²³⁸²³ {BBA00000036/6}.

¶22.109 In recognition that its continuing efforts to prevent any "restrictive" wording about the use of K15 at height were not going to be successful, Kingspan took comfort from the fact that those restrictions were printed in "the smallest possible font and buried deep in the certificate itself". That was celebrated internally as "significant progress" from the position in a previous draft, as was the fact that the explicit restriction no longer appeared on the front page of the certificate.⁸²⁴⁸²⁴ {KIN00005870/1}.

¶22.110 Mr Meredith and Mr Mills both agreed⁸²⁵⁸²⁵ Meredith {Day76/14:5-9}; Mills {Day77/131:21-24}. that they, Mr Clarke and Mr Millichap had been working to ensure that the certificate imposed the minimum restrictions on the use of K15.⁸²⁶⁸²⁶ {KIN00005570/1}. Mr Millichap told us that his team might have been asking for more than was appropriate,⁸²⁷⁸²⁷ Millichap {Day82/13:3-10}. but that they had simply been keen to ensure that K15 was presented in the most positive light and had tried to extend the scope of the certificate in a way that was accurate.⁸²⁸⁸²⁸ Millichap {Day82/7:12-22}. Mr Meredith went further, however, admitting that inaccurate assertions made by Kingspan had not been due to an excess of enthusiasm but to a conscious attempt to persuade the BBA to accept claims that would maximise sales of K15, whether they were true or not.⁸²⁹⁸²⁹ Meredith {Day76/12:16-20}.

¶22.111 Again, we think that the evidence of Mr Meredith was more reliable. We are left in no doubt that Kingspan took advantage of the BBA's initial failure properly to investigate the characteristics of K15 and thereafter did all it could to ensure that the certificates did not include wording that restricted the suitability of K15 for use on buildings over 18 metres in height, both by putting forward wording it knew to be inaccurate and by failing to correct the BBA's own inaccurate statements. Loose language or not,⁸³⁰⁸³⁰ Millichap {Day82/7:4}. Kingspan's long-running internal discussions about what it could get away with⁸³¹⁸³¹ {KIN00009353/1}. betrayed no concern whatsoever for accuracy.

LABC, NHBC and Wintech: 2012-2013

¶22.112 By May 2012, the LABC System Type Approval certificate issued to Kingspan in May 2009 had lapsed,⁸³²⁸³² {KIN00005705/1} "Conditions of Certification", second bullet point. leading to what Mr Meredith described in an email sent to Tony Millichap on 15 October 2012 as a "major headache and potentially a lot of lost work" in relation to high-rise buildings. He suggested that without the LABC certificate new data would probably be needed and that pressure should be put on the production teams to develop a form of K15 that could be expected to enable a system to meet the criteria in BR 135.⁸³³⁸³³ {KIN00005552/1}; Meredith {Day76/37:17}-{Day76/39:2}. The exchange reveals the degree of reliance that had been placed on the LABC certificate and confirms that at that time Kingspan thought that K15 was still not capable of forming part of a system that could be tested successfully in accordance with BS 8414.

¶22.113 In late 2012, Kingspan thought about continuing to rely on the certificate issued in May 2009, but its lapse had not escaped the notice of Wintech, which continued to raise questions about the suitability of K15 for use on buildings over 18 metres in height.⁸³⁴⁸³⁴ {KIN00005554/2-3}; Meredith {Day76/39:5}-{Day76/41:23}. In the event, Kingspan did obtain further LABC certificates for K15, the next being published on 28 August 2013.⁸³⁵⁸³⁵ {LABC0001000}; {LABC0001001}. We deal with those later certificates in Chapter 23.

¶22.114 By 31 October 2013, Wintech's efforts to draw attention to Kingspan's apparent lack of test data for K15⁸³⁶⁸³⁶ {KIN00005683}. had culminated in a meeting between them attended by (amongst others) Dr Rochefort, Peter Wilson, the managing director of Kingspan,⁸³⁷⁸³⁷ {KIN00022334/17}. and Paul Savidge, the managing director of Wintech.⁸³⁸⁸³⁸ Savidge {WIN00000004/2} page 2, paragraph 5. At the meeting Dr Rochefort maintained his view that K15 was suitable for use on buildings over 18 metres in height,⁸³⁹⁸³⁹ {KIN00003823/2}; Rochefort {Day80/172:4}-{Day80/176:21}. but accepted that

¶Kingspan did not have any BS 8414-2 test data and told Wintech that such tests would be undertaken and made available within six months.⁸⁴⁰⁸⁴⁰ Rochefort {Day80/181:22}-{Day80/186:4}; Savidge {WIN00000004/25} page 25, paragraphs 63-67; Taylor {WIN00000002/26} page 26, paragraphs 57-61. In an email sent after the meeting,

¶Paul Savidge warned Dr Rochefort that if the data were not made available within that period Wintech would publicise its concerns about the use of K15 on high-rise buildings to the industry at large.⁸⁴¹⁸⁴¹ {KIN00003823}; Meredith {Day76/139:7-21}.

¶22.115 The contents of the LABC certificates for K15 had also come to the attention of NHBC. In November 2014, it described the expression "K15 can be considered a material of limited combustibility", as "all garbage" and the approval of K15 on buildings over 18 metres in height as "an accident waiting to happen".⁸⁴²⁸⁴² {NHB00000810/1-3}. In Chapter 26 of this report we describe the discussions between NHBC and Kingspan during the period from late 2013 onwards in relation to the use of K15 on buildings over 18 metres in height. Mr Meredith told us that NHBC had started to ask difficult questions in late 2013.⁸⁴³⁸⁴³ Meredith {Day76/50:7-10}; {NHB00000645/2}. Certainly, by February 2014, and following unsuccessful efforts by NHBC to obtain test evidence from Kingspan that demonstrated the suitability of K15 for use on buildings over 18 metres in height (in particular on steel frame systems), NHBC had asked Kingspan to provide written approval for every project of that kind on a job by job basis.⁸⁴⁴⁸⁴⁴ {KIN00005894/1}. The evidence that Mr Meredith had repeatedly assured NHBC he would send to support the suitability of K15 for use at height proved not to exist. Consequently, the pressure on him to resolve the situation increased.⁸⁴⁵⁸⁴⁵ Meredith {Day76/97:11}-{Day76/99:11}; Evans {Day219/218:25}-{Day219/219:21}. On 16 June 2014 NHBC wrote to Kingspan saying that unless test evidence supporting the use of K15 in any construction other than that tested to BS 8414- 1 in May 2005 had been provided by 30 June 2014 NHBC would reconsider its acceptance of K15 as suitable for use on buildings over 18 metres in height.⁸⁴⁶⁸⁴⁶ {NHB00000757/3}. In the event, however, neither occurred.

BS 8414 test: 2014

¶22.116 In January 2014, Kingspan embarked on a programme of testing at BRE in accordance with BS 8414-2, the first since June 2008. The tests were arranged by Mr Meredith under the supervision of Mr Millichap and Dr Rochefort.⁸⁴⁷⁸⁴⁷ {KIN00022334/17}. The first test in the series, held on 7 January 2014,⁸⁴⁸⁸⁴⁸ Test 291718; no report produced. incorporated Trespa high pressure laminate rainscreen boards.⁸⁴⁹⁸⁴⁹ {KIN00000673}. We were not able to establish whether the test in January 2014 had incorporated K15 insulation boards of the kind that were then being sold commercially. Mr Meredith told us that he thought it had,⁸⁵⁰⁸⁵⁰ Meredith {Day76/110:4-7}. but Mr Millichap thought otherwise⁸⁵¹⁸⁵¹ Millichap {Day82/27:1-8}. and an internal Kingspan presentation from May 2015 referred to a version of K15 with an unperforated foil facer.⁸⁵²⁸⁵² {KIN00021945/17}. (By January 2014, K15 was being manufactured and sold with perforated foil facers.) A spreadsheet produced by Kingspan for the Inquiry described the type of K15 used in the test as "unknown".⁸⁵³⁸⁵³ {KIN00022357/3}. Whatever kind of K15 it contained, however, the system did not meet the criteria in BR 135, the temperature exceeding the relevant limit after just under

¶15 minutes, with flames breaching the top of the test rig before 30 minutes had elapsed.⁸⁵⁴⁸⁵⁴ {KIN00000673/1}.

¶22.117 Following that test, Kingspan worked on changes in advance of the next test. It is clear that Ivor Meredith, Dr Rochefort and Tony Millichap (among others) were closely involved in the discussions about possible changes. They included a suggestion made by Mr Meredith in an email dated 13 January 2014 to produce for testing a K15 product modified with an alternative blowing agent called 'solstice'.⁸⁵⁵⁸⁵⁵ {KIN00021095/2} fifth bullet point. Dr Rochefort did not agree with changing to a solstice-blown foam and asked Mr Meredith to speak to him about it. Mr Meredith subsequently told Mr Millichap that he thought Dr Rochefort would be concerned about putting too much in writing.⁸⁵⁶⁸⁵⁶ {KIN00021095/1}.

¶22.118 It is apparent from the internal correspondence in January 2014 following the first test⁸⁵⁷⁸⁵⁷ {KIN00021095}. that at that point Dr Rochefort, Mr Millichap and Mr Meredith all realised two important things. First, that at that time they did not think that Kingspan had a K15 product that was capable of being used in a system that, when tested in accordance with BS 8414, would meet the criteria in BR 135. That included the K15 that Kingspan was then selling. Secondly, that as a result, the next test or tests could not be carried out on the standard K15 product.⁸⁵⁸⁸⁵⁸ Millichap {Day81/122:12-25}; Millichap {Day82/27:16-19}.

¶22.119 Whatever reservations Dr Rochefort might initially have had about changing to a solstice-blown version of K15, the next test on 19 March 2014⁸⁵⁹⁸⁵⁹ Test 293940. Test report dated 26 June 2014 {BRE00002516}. was carried out on a system incorporating a 15 millimetre Trespa decorative rainscreen board and a "trial" version of K15 made using solstice as the blowing agent with a 50 or 100 micron unperforated foil facer.⁸⁶⁰⁸⁶⁰ {KIN00022357/3}; {KIN00021945/17}. Dr Rochefort, Mr Meredith and Mr Millichap were all aware at the time of the test that those were the characteristics of the insulation tested,⁸⁶¹⁸⁶¹ Rochefort {Day80/203:5-12}; Meredith {Day76/124:23}-{Day76/125:9}; Millichap {Day82/30:23}-{Day82/31:1}. although the K15 on the market in March 2014 had a 25 micron perforated foil facer and did not use solstice as the blowing agent.

¶22.120 Despite the use of a modified K15 product, by 31 March 2014 Kingspan had learnt from BRE that the system tested on 19 March 2014 had not met the criteria in BR 135.⁸⁶²⁸⁶² {KIN00021907/2}. Towards the end of the test flames had breached the top of the rig and the test had therefore been effectively terminated straight away.⁸⁶³⁸⁶³ {KIN00010461/3}; {BSI00000097/11} clause 8.5. Kingspan immediately set about challenging BRE's decision, first by making a formal complaint⁸⁶⁴⁸⁶⁴ {KIN00010461/1}; {BRE00003576}. and then by warning BRE that it was consulting lawyers.⁸⁶⁵⁸⁶⁵ {BRE00015592/2}. Based on its analysis of a video recording of the test, BRE stood firm and did not produce a BR 135 classification report. Mr Meredith nonetheless made good his initial promise to Dr Rochefort and others that he would get something out of the test⁸⁶⁶⁸⁶⁶ {KIN00021907/2}. by obtaining a test report.⁸⁶⁷⁸⁶⁷ {BRE00002516}. The resulting test report dated 26 June 2014 was later relied on in a large number of desktop assessments produced to demonstrate that systems incorporating K15 would comply with the criteria in BR 135.

¶22.121 Despite being aware from 31 March 2014 that the system tested on 19 March 2014 had not met the performance criteria in BR 135,⁸⁶⁸⁸⁶⁸ {KIN00021907/2}. Kingspan represented on a number of occasions,⁸⁶⁹⁸⁶⁹ {NHB00000703}; {KIN00002186}; Millichap {Day82/107:5-24}; {Day82/134:2-23}. even as late as February 2015,⁸⁷⁰⁸⁷⁰ {KIN00002199/1}. that the test supported the suitability of K15 for use on buildings over 18 metres in height. That involved a double falsehood: the insulation tested in that system was not K15 in the form in which it was being supplied to the market and the system had not met the criteria in BR 135.

¶22.122 The next test took place on 7 July 2014. Dr Rochefort directed that terracotta tiles be used as the rainscreen.⁸⁷¹⁸⁷¹ {KIN00021907/2}. There was some discussion in late March 2014 between Mr Meredith, Mr Millichap, Dr Rochefort and others at Kingspan about the product that should be used in the next test, standard, solstice or phosphoric.⁸⁷²⁸⁷² {KIN00021907/2}. The first was the K15 that was being sold commercially, the second was the trial version which had been tested in March 2014 and the third contained certain other chemical modifications.⁸⁷³⁸⁷³ Millichap {Day82/54:10-13}. On this occasion, they decided to use a solstice-blown product with a 50 micron foil facer, perforated on one side and unperforated on the other.⁸⁷⁴⁸⁷⁴ {KIN00022357/3}; {KIN00021945/17}; Meredith {Day76/158:19-21}. Mr Meredith told us that the decision to test that form of the product was taken by all those who had been involved in the discussions,⁸⁷⁵⁸⁷⁵ Meredith {Day76/130/10-19}; {KIN00021907}. who included Dr Rochefort, Mr Millichap and Peter Wilson.⁸⁷⁶⁸⁷⁶ {KIN00021907}. The system tested met the performance criteria in BR 135.⁸⁷⁷⁸⁷⁷ Test report 297099 dated 10 September 2014 {BRE00002656}; BR 135 Classification report 291642 dated 5 March 2015 {BRE00002514}.

¶22.123 There was evidence that Mr Meredith had taken steps to ensure that labels stating "trial" were removed from the insulation supplied to BRE for the test.⁸⁷⁸⁸⁷⁸ {KIN00021904/3}. He agreed that that might have been to ensure that BRE did not become aware that a trial version of K15 was being tested.⁸⁷⁹⁸⁷⁹ Meredith {Day76/161:4-20}. Neither the test nor the classification report produced by BRE made any reference to the fact that the insulation was a trial version, both simply referring to it as "K15 Kooltherm insulation board".⁸⁸⁰⁸⁸⁰ {BRE00002656/6}; {BRE00002514/6}. As a result, the reader would not have been able to tell from the reports that the insulation was anything other than the K15 then being sold by Kingspan.

¶22.124 Mr Pargeter, Mr Meredith, Mr Millichap and Dr Rochefort all admitted that they had been aware at the time that the test on 7 July 2014 had been carried out on a system incorporating a trial product,⁸⁸¹⁸⁸¹ Meredith {KIN00022312/58} page 58, paragraph 120(a); Millichap {Day82/64:9-15}; Rochefort {Day80/216:13-17}; Pargeter {Day83/191:7-16}. a fact clearly borne out by the documents. Mr Millichap⁸⁸²⁸⁸² Millichap {Day82/76:15}-{Day82/77:3}. and Mr Meredith⁸⁸³⁸⁸³ Meredith {Day76/157:16}-{Day76/158:2}. told us that it had been common knowledge within Kingspan.

100

¶22.125 The test result was immediately celebrated internally within Kingspan,⁸⁸⁴⁸⁸⁴ {KIN00010825/2}. with jubilation expressed at the effect it would have on competitors and open discussion of the pressure Kingspan had been under to achieve it.⁸⁸⁵⁸⁸⁵ {KIN00010825/1}. It does not appear to have occurred to any Kingspan employee, other than Mr Millichap, whose objections appear to have been roundly dismissed⁸⁸⁶⁸⁸⁶ {KIN00010825/1}; Millichap {Day82/75:21}-{Day82/76:13}. and must have been short-lived, to question any aspect of the "success" being celebrated. The test result was swiftly also celebrated externally. Kingspan informed NHBC, building contractors, designers and the wider market that it had carried out a successful test to BS 8414-2 as part of its "test portfolio".⁸⁸⁷⁸⁸⁷ {KIN00002186/2}; Millichap {Day82/104:23}-{Day82/105:4}; {KIN00011049}.

¶22.126 Thereafter, in its correspondence, technical bulletins and promotional literature, Kingspan relied on the result of that test to support the use of standard K15 on buildings over 18 metres in height and that remained its position until it withdrew the test in October 2020.⁸⁸⁸⁸⁸⁸ {KIN00024104/1}. Kingspan's witnesses were asked why it had relied on a test carried out on a different product, given that all those involved (including Mr Pargeter, who was promoted to Head of Marketing in November 2014)⁸⁸⁹⁸⁸⁹ Pargeter {KIN00000494/4-5} pages 4-5, paragraphs 2.7 and 2.8. knew that the test had not been carried out on K15 of the kind being sold in the market. There was a conflict of evidence between Mr Meredith, who did not attempt to deny that Kingspan had done so knowingly and deliberately, and Dr Rochefort, Tony Millichap and Adrian Pargeter, who did.

¶22.127 Dr Rochefort told us that at the time he had had no idea that the test carried out in July 2014 was being used by Kingspan to support the sale of standard K15 for use on buildings over 18 metres in height. When faced with documents that tended to show that he had been aware of how the test was being used,⁸⁹⁰⁸⁹⁰ {KIN00020736/2} paragraph 6; {KIN00021904/1}. he put forward various unconvincing explanations while resolutely denying that Kingspan had deliberately relied on it to promote or sell standard K15.⁸⁹¹⁸⁹¹ Rochefort {Day80/222:9}-{Day80/224:3}; {Day80/217:9-19}; {Day80/224:17-22}. In our view the evidence clearly demonstrated that he was aware that the marketing and technical departments were making use of the test in that way.

¶22.128 Mr Millichap agreed that it was wholly inappropriate to use the July 2014 test to support the sale of standard K15⁸⁹²⁸⁹² Millichap {Day82/66:12-15}. and said that he had not known that that was being done.⁸⁹³⁸⁹³ Millichap {Day82/64:19}-{Day83/65:6}. However, having seen Mr Meredith's email of 21 August 2014,⁸⁹⁴⁸⁹⁴ {KIN00020736}. he accepted that he had known that letters confirming success in the test that had been sent in response to questions from contractors⁸⁹⁵⁸⁹⁵ {KIN00020736/2} paragraph 6. can only have referred to the test in July 2014.⁸⁹⁶⁸⁹⁶ Millichap {Day82/68:17}-{Day82/69:6}. Moreover, faced with the record of a meeting held on 15 July 2014, at which he had been instructed to produce reports and letters announcing the result of the test for use as a marketing tool,⁸⁹⁷⁸⁹⁷ {KIN00020736/3}. he admitted that he had been actively involved in using the test to support the sale of standard K15.⁸⁹⁸⁸⁹⁸ Millichap {Day82/72:18-24}. He also confirmed that he had known throughout his time as Head of Technical that the product Kingspan had tested in July 2014 had never been produced or sold commercially.⁸⁹⁹⁸⁹⁹ Millichap {Day82/64:9-15}.

101

¶22.129 In the summer of 2014 Adrian Pargeter was Kingspan's Product Development Manager; he was promoted to Head of Marketing in November 2014. It is clear from the documents to which we have referred that he was closely involved in the testing of K15 and the attempts to produce a product that was capable of forming part of a system that would satisfy the criteria in BR 135 when tested in accordance with BS 8414. He could therefore be expected to have had a detailed knowledge of the differences between the various forms of K15 that were produced for testing during that period and, towards the end of 2014, of the steps being taken by the marketing department to promote and sell the product.

¶22.130 Mr Pargeter told us that he had not known that Kingspan had relied on the test in July 2014 in order to market and sell standard K15 until February 2019, when he had discovered it in the course of preparing to give evidence to the Inquiry,⁹⁰⁰⁹⁰⁰ Pargeter {KIN00022610/56-57} pages 56-57 paragraph 7.16; Pargeter {Day83/191:17}-{Day83/192:9}; {Day84/4:14-25}. although he had known that a non-standard version of K15 had been tested in July 2014. In effect, his evidence amounts to saying that until early 2019 he had thought that two separate tests had been carried out in July 2014, one on the standard product and one on a trial product.

¶22.131 Mr Pargeter told us that he had not been involved in any discussions or decisions to use the report on the test carried out in July 2014 to support the sale of standard K15.⁹⁰¹⁹⁰¹ Pargeter {KIN00022610/28} page 28, paragraph 3.37. However, his evidence was not consistent with a series of documents created in 2014 and 2015 that he had either written, approved or read, all of which contained statements that specifically relied on that test to support the use of K15 on buildings over 18 metres in height. One of them was a technical bulletin published in August 2015 entitled "Routes to Compliance",⁹⁰²⁹⁰² {KIN00000086}. which Mr Pargeter confirmed he had read in full (including the appendices and diagrams) thoroughly⁹⁰³⁹⁰³ Pargeter {Day84/15:17-24}; {Day84/17:14-15}. before he had approved it.⁹⁰⁴⁹⁰⁴ Pargeter {Day84/12:21-25}. The document contained pictures of the systems used for the three BS 8414 tests on which Kingspan was relying by August 2015.⁹⁰⁵⁹⁰⁵ {KIN00000086/17}. One of them was the test carried out in May 2005; another was the test carried out in July 2014. Mr Pargeter said that he had not realised at the time that it was a picture of the test carried out on development material in 2014. There was a series of other documents dating from 2015 and 2016, all of which Mr Pargeter confirmed he had read, in which the test carried out in July 2014 was described or referred to as a test carried out on K15 and was being relied on in support of the promotion, certification or sale of K15.⁹⁰⁶⁹⁰⁶ {KIN00002289}; {NHB00001109/2} second bullet point; {BBA00000200/5} paragraph 8.2, {BBA00000201/5} paragraph 8.2.

¶22.132 Given the positions he had occupied within Kingspan, between 2014 and 2019, it would be surprising if Mr Pargeter had not been aware of the nature of the product involved in the July 2014 test or of how his department had been marketing a product as important as K15. We found his attempts to distance himself from the activities of the marketing department dishonest and entirely unconvincing. We have concluded that Mr Pargeter was fully aware, both before and after he became Head of Marketing, that Kingspan was relying on the results derived from a test incorporating a form of K15 that was not being produced commercially to sell standard K15 and that he actively condoned it at a very senior level.

102

¶22.133 We think it highly likely that Kingspan had always intended to rely on the first successful test it could achieve, whatever form of the product had been used in the tested system. Kingspan's reliance after July 2014 on a test carried out using a development product was part of a carefully planned, carefully concealed and long-running deception.

Conclusions

¶22.134 The story of the development and marketing of K15 for use on buildings of over 18 metres in height between 2006 and 2019 is one of deeply entrenched and persistent dishonesty on the part of Kingspan in pursuit of commercial gain coupled with a complete disregard for fire safety. Unfortunately, Kingspan's dishonesty was facilitated, albeit inadvertently, by serious incompetence on the part of two bodies, the BBA and LABC, to which the industry looked for confirmation that K15 was suitable for use on buildings over 18 metres in height. Both those bodies, although supposedly independent, compromised their independence by entering into negotiations with Kingspan over the wording of their certificates and agreeing to include language that was inappropriate and in some cases misleading. They both failed to examine rigorously the material on which Kingspan's applications were based or to require the production of test data that supported its claims.

¶22.135 The effect of Kingspan's dishonest marketing of K15 was to create a spurious market for a polymeric insulation product suitable for use on high-rise buildings generally, which drew in Celotex as a competitor. Celotex found it impossible to create a similar product using polyisocyanurate foam and could not understand how Kingspan had been able to produce an organic polymeric insulation board that apparently enabled designers to follow the guidance in Approved Document B relating to buildings over 18 metres in height. It therefore embarked on its own campaign to break into the market by dishonest means. Kingspan cannot be blamed for Celotex's dishonesty, which was the choice of Celotex itself, but it did create the conditions that encouraged it and in which it was able to flourish.

103