Chapter 58: Window infill panels

Chapter 58: Window infill panels

Introduction

¶58.1 The spaces between the vertical columns of Grenfell Tower were filled by window assemblies which were constructed as a single unit and contained both glazed sections (with opening parts) and solid sections acting as walls. The solid sections were known as window infill panels.

¶58.2 Two types of window infill panel were installed during the refurbishment of Grenfell Tower, a large panel with an insulating core, which was set between the glazed windows and a smaller panel, also with an insulating core, which housed the kitchen extractor fan.^21522152 Grenfell Tower Inquiry: Phase 1 Report, Chapter 6, 6.29 and 6.31. They are shown in red and orange in the following figure taken from Dr Lane's Phase 1 report.

¶Figure 58.1 – Typical elevation for three-window configuration

¶58.3 After the fire it was found that both panels contained a core of Styrofoam, held between two sheets of aluminium.^21532153 Grenfell Tower Inquiry: Phase 1 Report, Chapter 6, 6.29 and 6.31. Styrofoam is a form of extruded polystyrene (often referred to as "XPS") and is combustible.^21542154 Lane, Phase 1 Report {BLAS0000008/61-62} paragraphs 8.10.36, 8.10.4 and 8.10.42. Although in his Phase 2 experimental work Professor Bisby found that those panels made the lowest contribution to the energy available in the external wall (only about 2–3% of the total energy available per floor),^21552155 Bisby, Phase 2 Experiments - Work Package 1 {LBYWP100000002/5} page 5, paragraph 32. he also found that they were relatively easy to ignite and, once ignited, had a comparatively high energy heat release rate per unit area.^21562156 Bisby, Phase 2 Experiments - Work Package 1 {LBYWP100000002/5-6} pages 5-6, paragraphs 33-34. He concluded that although they were not a primary cause of, or contributor to, the spread of fire across the external walls, they may have played an important part in the early development of the fire in and immediately outside Flat 16.^21572157 Bisby, Phase 2 Experiments - Work Package 1 {LBYWP100000002/5-6} pages 5-6, paragraphs 39-40. Accordingly, in this chapter we seek to explain how that combustible material came to be incorporated into the external wall of Grenfell Tower.

198

Specification at tender stage

¶58.4 In the NBS Specification, the final version of which was dated 30 January 2014, Studio E specified aluminium-faced window infill panels with insulating cores as part of the external wall. Although some thermal performance requirements were included, no particular material or product was prescribed at that stage.^21582158 {SEA00000169/145} clause 332. It was therefore left to others to select suitable products which met the stipulated requirements.^21592159 Crawford {Day9/112:14-22}; {Day9/147:5-8}; {Day9/154:17-18}. Studio E did not prescribe any minimum fire performance standards for the panels.

¶58.5 Studio E's tender drawings showed insulating panels between the areas of glazing, but they did not identify any particular product or type of insulation either.^21602160 {SEA00002499}; Hyett, Module 1 Report {PHYR0000029/77} page 77, paragraph 4.3.90.

Harley's specification

¶58.6 When Kevin Lamb read the NBS Specification he thought that Studio E had simply overlooked the need to identify the insulating material to be used.^21612161 Lamb {Day38/169:1-22}. On 25 September 2014 he produced a drawing of the window arrangement which was revised on a number of occasions culminating in Revision D issued on 3 March 2015.^21622162 {HAR00008886}. 25 September 2014 was revision A. "P1" appears in revision A. The final revision was D. dated 3 March 2015 in which both "P1" and "P2" appear. In his drawings the larger infill panels were designated "P1" and the smaller infill panels housing the kitchen extractor fans "P2". Mr Lamb also produced specification notes dated 15 January 2015 which identified the products he proposed to use.^21632163 {HAR00003866}. The P1 and P2 panels were both shown as comprising 2mm aluminium skins with a 24mm core of Kingspan TP10 rigid insulation. TP10 is a combustible insulation product manufactured by Kingspan.^21642164 Lane, Phase 1 Report {BLAS0000008/62} paragraph 8.10.39(b); BBA certificate {KIN00000276}.

¶58.7 When asked about his choice of insulating core material, Mr Lamb said that he had initially selected Kingspan TP10 for both panels because it was a standard product that he had used over many years and that he had put it into the specification as a starting point for discussions within Harley.^21652165 Lamb {Day38/173:9-10}; {Day38/173:12-18}; {Day38/178:25}-{Day38/179:2}. Based on a conversation he had had many years previously with a fabricator working for another client, he was confident that the result would be a "Class 0" panel.^21662166 Lamb {Day38/179:5-9}. However, he did not check the requirements of the Building Regulations or the guidance in Approved Document B before suggesting the use of TP10,^21672167 Lamb {Day38/174:2-5}; {Day38/175:13-17}. nor did he take any steps to investigate its fire performance, including its suitability for use on buildings over 18 metres in height. He did not discuss the matter with anyone else at Harley.^21682168 Lamb {Day38/178:20}-{Day38/179:18}.

199

¶58.8 The BBA certificate for Kingspan TP10 states that the product had in fact achieved a national Class 1 rating,^21692169 {KIN00000276/6} section 7. which is inferior to national Class 0.^21702170 See Chapter 5. National Class 1 means it has undergone testing in accordance with BS 476-7, not 476-6, which is also necessary for a national Class 0 classification. However, Mr Lamb did not look at the certificate; instead, he went ahead on his understanding that when fabricated the panel would achieve a Class 0 rating because the outer surfaces would be aluminium.^21712171 Lamb {Day38/181:18}-{Day38/182:14}.

¶58.9 Mr Lamb's evidence suggests that he was not familiar with the guidance in Approved Document B that insulation products used in the external walls of buildings over 18 metres in height should be of limited combustibility. Although he was aware that there was a difference between limited combustibility and Class 0, he had never considered what the difference was and simply assumed that materials with a Class 0 classification were acceptable for use.^21722172 {CLG00000224/96} clause 12.7; Lamb {Day38/83:9-17}. It did not occur to him that TP10 panels might contain combustible insulation and he therefore never considered whether combustible cores could be exposed in a fire.^21732173 Lamb {Day38/183:25}-{Day38/184:9}. That in turn demonstrated a lack of understanding on his part about how metal panels are liable to perform in a fire, a matter to which attention had been drawn in industry guidance long before 2015.^21742174 See the 2nd Edition of BRE 135 dated 2003 {BRE00005554/17-18} and the CWCT Standard for Systematised Building Envelopes, 2008 {CWCT0000046/11} section 6.3.

¶58.10 Mr Lamb's specification notes relating to the P1 and P2 panels were subsequently changed. Someone added handwritten annotations in red to amend the specification of the P1 and P2 panels to show 25mm of Styrofoam held between two 1.5mm skins of aluminium.^21752175 {HAR00003866}. That involved a reduction in the thickness of the aluminium skins and a corresponding increase in the thickness of the insulation, as well as a change of product. Mr Lamb said that the changes had been made by Mark Stapley, Operations Director at Harley.^21762176 Lamb {Day38/176:9-11}. We accept that evidence and think it likely that the changes were made by Mr Stapley after he had obtained the first quotation for the supply of panels.

¶58.11 Kevin Lamb produced a revised version of the specification notes,^21772177 {HAR00003869/1}. which reflected the changes made by Mr Stapley, save that the insulation in the P2 panel was still shown as Kingspan TP10.^21782178 Now increased in thickness from 24mm to 25 mm.

¶58.12 Kevin Lamb said that he had been told by Ben Bailey to obtain the infill panels from a company called Panel Systems Ltd, but that that company had been able to supply panels only with Styrofoam cores.^21792179 Lamb {HAR00010419/16} page 16, paragraph 63; Lamb {Day38/167:4} – {Day38/168:25}. Ben Bailey, however, said that Mark Stapley had ordered the panels some time before he had become project manager.^21802180 Ben Bailey {Day39/158:2-12}; {Day39/159:19-24}. We accept Ben Bailey's evidence on this point. It is apparent from the documents that Mr Bailey did not take over as project manager until 3 March 2015. Mark Stapley made his first inquiry of Panel Systems for the manufacture and supply of the window infill panels in January 2015,^21812181 {PAN00000012}; {PAN00000007}. and the first invoice for the supply of panels was dated 10 February 2015.^21822182 {MET00040281}; {MET00040279}. It seems clear, therefore, that the decision to obtain panels from Panel Systems Ltd was not made by Ben Bailey.

200

¶58.13 Mr Lamb's evidence was also at odds with the evidence given by Christopher Ibbotson, the owner and managing director of Panel Systems, which we have no reason to doubt. According to him, in 2015 the company manufactured composite panels using a range of materials and various insulation products.^21832183 Ibbotson {Day104/6:24}-{Day104/7:3}; {Day104/8:12-13}; {Day104/8:16-25}. If asked to do so, it would give advice on the most suitable kind of panel to meet the customer's needs. Any advice on fire performance of a product would be based on information obtained from the manufacturer.^21842184 Ibbotson {Day104/14:7-21}; {Day104/16:6-19}. It is worth noting that in September 2015 Ben Bailey asked Panel Systems for advice on the manufacture and supply of a panel for use in the mounting of an anti-arson letterbox at Grenfell Tower^21852185 {HAR00020436}; Ben Bailey {Day39/176:2}. and that two days later he was sent details of its Versafire Euroclass A1 rated non-combustible board.^21862186 {HAR00002853}.

¶58.14 If a customer did not specify a core material, Panel Systems offered Styrofoam as standard^21872187 {PAN00000017/2} page 2, paragraph 2.4. because of its thermal efficiency.^21882188 Ibbotson {Day104/20:15}–{Day104/21:2}. In those circumstances it would not generally inform the customer of Styrofoam's fire performance, nor would it tell the customer that it should check whether the use of that material in the particular circumstances they had in mind would comply with the Building Regulations.^21892189 Ibbotson {Day104/21:4-23}. Panel Systems marketed its insulating panels by reference to the fire rating of the core material, which was the primary factor in their performance. It therefore marketed panels with Styrofoam cores as European Class E.^21902190 Ibbotson {Day104/24:21}–{Day104/25:25}.

¶58.15 In the light of Mr Ibbotson's evidence, it is clear that panels with cores other than Styrofoam could have been obtained from Panel Systems and it follows that if Harley had wanted to do so it could probably have obtained panels with better fire performance which met the requirements for thermal efficiency. We are unable to say with confidence what led Mr Stapley to accept Styrofoam for the core, but the fact that it was a standard product combined with the reduction in the thickness of the aluminium skins leads us to think that considerations of cost may have played a part. At all events, there is no reason to think that anyone at Harley gave serious consideration to the fire performance of the panels.

¶58.16 On 19 January 2015 Mark Stapley asked Panel Systems to provide a price for the manufacture and supply of infill panels.^21912191 {PAN00000012}; Ibbotson {Day104/28:20-24}. For that purpose he sent a schedule of Harley's requirements^21922192 {PAN00000006/2}; Stapley {HAR00020574/8} page 8, paragraph 28; {HAR00010003}. and requested prices for aluminium spandrel panels with an overall thickness of 28mm, without specifying the core material to be used.^21932193 {PAN00000006/2}. Later that day Panel Systems gave Mr Stapley a quotation for 28mm thick Aluglaze panels with aluminium skins and 25mm Styrofoam cores.^21942194 {HAR00009866}; {HAR00009867}. On 20 January 2015, Mr Stapley sent a purchase order to Panel Systems for the supply of the panels set out in his original schedule.^21952195 Stapley {MET00040296/15} page 15. This was the same schedule as set by Mr Stapley on 19 January 2015 {PAN00000006/2}. There was no discussion about their fire performance.

201

¶58.17 Mark Stapley said that Panel Systems had specified the core material to be used in the P1 and P2 panels,^21962196 Stapley {HAR00020574/10} page 10, paragraph 33. but we think that unlikely, save in the sense that its quotation was based on Styrofoam because that was the default product. Harley had not specified any minimum fire performance for the core and there had been no discussion about what fire performance was required. As an experienced member of Harley's team Mr Stapley must have expected Panel Systems to quote for a basic product, unless asked to do otherwise. The responsibility for ensuring that the fire performance of the core was appropriate rested with Harley, not with Panel Systems.

¶58.18 Kevin Lamb accepted that he had never considered the fire performance of the window infill panels^21972197 Lamb {Day38/171:15-17}; {Day38/192:16-20}. or whether the use of Styrofoam in the external wall complied with the Building Regulations.^21982198 Lamb {Day38/187:1-9}. He was not aware that Styrofoam was not a product of limited combustibility, nor did he take any steps to assess whether the panel was suitable for use on high-rise buildings.^21992199 Lamb {Day38/188:4-9}. Mr Lamb assumed it was a product that Harley had used many times before.^22002200 Lamb {Day38/193:1-5}.

¶58.19 When asked why his specification notes showed Kingspan TP10 for the P2 panel instead of Styrofoam, as had been indicated in Mr Stapley's amendment, Kevin Lamb said that Panel Systems could not produce panels with inside and outside faces in different colours, as was required in the case of the P2 panels, so the specification had not been changed and the core remained Kingspan TP10. However, that does not strike us as very plausible. After the fire some surviving P2 panels were found to contain Styrofoam cores and there is no evidence that those or any other P2 panels had been obtained from a different supplier. We think Mr Lamb simply overlooked that part of the amendment and failed to make the required change.

Studio E's review of the specification

¶58.20 The revised version of Mr Lamb's specification notes was sent to Studio E for review and on 26 January 2015 Mr Crawford stamped it "Conforms to design intent", subject to the incorporation of certain comments, which did not include any comment on the window infill panels. Mr Crawford told us that he had not seen the handwritten annotations on the specification notes and that he did not think that he had been aware that Styrofoam had been substituted for Kingspan TP10 in the P1 panel.^22012201 Crawford {Day11/110:4-24}. He maintained that he had been commenting only on architectural intent and would not necessarily have been looking at the description of the materials in the specification.^22022202 Crawford {Day11/111:22}–{Day11/112:5}; {Day11/115:5:13}. Mr Crawford had no recollection of discussing the use of any particular insulation material for the infill panels with anyone at Studio E, Harley, or Rydon.^22032203 Crawford {Day11/112:15-25}; {Day11/114:23}–{Day11/115:4}. Mr Hyett was critical of Studio E for failing to identify the unsuitability of Styrofoam in the course of its design reviews.^22042204 Hyett, Module 1 Report (revised) {PHYR0000029/135} paragraphs 4.4.140 – 4.4.141.

202

¶58.21 On 15 July 2015, Kevin Lamb produced a further revision of the specification notes^22052205 {RYD00046822/1}. but the specification for the P1 and P2 panels remained as it had been in the previous version. On 17 July 2015 Mr Crawford stamped it as conforming to design intent. Again, he could not recall having noticed the specification of Styrofoam and Kingspan TP10.^22062206 Crawford {Day11/116:9-18}. He thought that it was for Harley to check that the use of the material complied with statutory requirements and that it was not Studio E's function to check Harley's work.^22072207 Crawford {Day11/117:4}–{Day11/118:23}; {Day11/120:2-13}. Mr Crawford also said that he thought that the Styrofoam would be treated with fire retardant to limit its combustibility, but he could not recall taking any steps to find out whether that had been done.^22082208 Crawford {Day11/116:22}–{Day11/117:3}; {Day11/119:6-12}.

¶58.22 We think it clear that Studio E had an obligation under its contract with Rydon to take reasonable steps to ensure that all designs, whether produced by itself or any sub-contractors, complied with the relevant statutory requirements, including the Building Regulations.^22092209 {RYD00094228/9} paragraph 8. Mr Crawford's understanding that it was not his role to check the materials specified by Harley to ensure that the external wall was compliant with the Building Regulations was therefore wrong. He ought to have applied his mind to the fire performance of the window infill panels, both as a matter of contractual obligation and in the exercise of reasonable skill and care,^22102210 Hyett, Module 1 Report {PHYR0000029/135} paragraphs 4.4.140-4.4.141. but he failed to do so.

Harley's understanding of the products

¶58.23 Harley placed a number of orders for aluminium panels with a Styrofoam core between February and June 2015.^22112211 Ibbotson {Day104/32:12-17}.

¶58.24 When Ben Bailey joined the project in March 2015 he was sent Harley's drawings and specification notes^22122212 {HAR00017738}; {HAR00003953}. but could not remember whether he had reviewed them.^22132213 {HAR00017762}; Ben Bailey {Day39/155:12-22}. He was not aware that Styrofoam was extruded polystyrene and he did not know who had specified its use in the P1 infill panels.^22142214 Ben Bailey {Day39/156:14-23}. He did not himself investigate whether it was appropriate to use panels with Styrofoam cores in the external wall.^22152215 Ben Bailey {Day39/162:21-25}; {HAR00009696}; Ben Bailey {Day39/163:14}-{Day39/164:17}.

¶58.25 Ben Bailey thought that the Kingspan TP10 product had been specified for the P2 panel by Kevin Lamb,^22162216 Ben Bailey {Day39/164:24}-{Day39/165:1}. but it is not clear why, since he had not discussed the choice of material with Mr Lamb.^22172217 Ben Bailey {Day39/165:14}-{Day39/166:1}. He did not look at the BBA certificate nor did he consider whether panels containing Kingspan TP10 would comply with the Building Regulations.^22182218 Ben Bailey {Day39/172:9}-{Day39/173:11}; {Day39/175:6-7}.

203

¶58.26 Ben Bailey also thought that the window assemblies (which included both glazed sections and fixed insulating panels) did not fall within the guidance in Approved Document B on the construction of external walls. He therefore rejected the suggestion that Harley should have taken steps to satisfy itself that the insulation in the window infill panels was suitable for use on buildings of more than 18 metres in height. Other witnesses thought differently, however. Although we recognise that a distinction can be drawn between the concrete structure (including the columns), which plainly forms the external wall, and the window assemblies containing some glazed and some solid panels, which fill the gaps between the concrete, we think it clear that the purpose of the guidance in paragraph 12.7 of Approved Document B is to reduce the risk created by the use of combustible insulation in the walls of high-rise buildings. We therefore think it clear that the reference to "external wall construction" in paragraph 12.7 includes solid window infill panels and we do not think that a competent specialist facade contractor could reasonably have thought otherwise.

¶58.27 Simon O'Connor, Rydon's project manager between May 2014 and August 2015,^22192219 O'Connor {RYD00094221/7} page 7, paragraph 14; O'Connor {Day26/5:4-7}. received Harley's drawings and specification notes on 6 March 2015^22202220 {HAR00003951}. but could not remember having looked at either the drawings^22212221 {HAR00003953}. or the notes.^22222222 {HAR00003955}; O'Connor {Day26/112:3}-{Day26/114:24}. He had not come across Styrofoam, extruded polystyrene or Kingspan TP10 before.^22232223 O'Connor {Day26/115:17}-{Day26/116:16}.

¶58.28 As we noted in Chapter 62, John Hoban, the building control officer at RBKC responsible for the Grenfell Tower refurbishment, also received Harley's specification notes on 6 March 2015.^22242224 {HAR00003955}. He did not notice that Styrofoam and Kingspan TP10 were being proposed as the insulating material in the window infill panels^22252225 Hoban {Day46/26:15}-{Day46/27:4}; {Day46/28:8-10}. He relied on Exova having looked at this specification. Hoban {Day46/27:16-21}. and did not look at the BBA certificates relating to either material; nor did he question the use of those materials.^22262226 Hoban {Day46/28:18}-{Day46/29:14}. He candidly accepted that he should have done so.^22272227 Hoban {Day46/29:15-17}.

¶58.29 On 24 August 2015, for the first time Ben Bailey asked Panel Systems to give him a price for supplying aluminium-faced panels with a core of Kingspan TP10.^22282228 {HAR00020331}; {HAR00020332}. On 27 August 2015, Panel Systems provided Harley with a quotation as requested^22292229 {HAR00018872}. but for panels with a PIR core, rather than Kingspan TP10. Michael Roache, a product manager at Panel Systems, said that he had advised Harley that Kingspan TP10 was not suitable for use as the core of a composite panel due to its manufacturing tolerance and uneven surface and that as a result a discussion had taken place (though between whom is uncertain) about an alternative PIR insulation.^22302230 Roach {PAN00000029/3} page 3, paragraph 14. Ben Bailey said that any such conversation was not with him. Ben Bailey {Day39/173:6}-{Day39/174:5}. Mr Ibbotson agreed that a TP10 core would not bond reliably to the aluminium skin. TP10 was therefore replaced with a generic PIR product.^22312231 Ibbotson {Day104/38:8-22}; {Day104/40:12-25}; {Day104/48:23}-{Day104/49:10}; Ibbotson {Day104/49:18-21}; {Day104/42:1-5}. Subsequently, Harley placed further orders with Panel Systems for panels with PIR cores.^22322232 {MET00040286}.

204

¶58.30 In total Panel Systems appear to have supplied a number of larger orders between January and June 2015 for panels with Styrofoam cores, and two smaller orders between August and October 2015 for composite panels with PIR cores.

¶58.31 Although Kevin Lamb had specified Kingspan TP10 insulation for the P2 panels^22332233 {HAR00003869/1}. and although panels with PIR insulation had been ordered by Ben Bailey,^22342234 {HAR00000445/6}. the Inquiry's experts discovered after the fire that some of the P2 panels had Styrofoam rather than PIR cores.^22352235 Lane, Phase 1 Report {BLAS0000008/62} paragraph 8.10.41. Mr Ibbotson denied that Panel Systems had substituted Styrofoam for PIR^22362236 Ibbotson {Day104/48:1-22}. and Ben Bailey could not shed any light on how some of the P2 panels had come to be supplied with Styrofoam rather than PIR cores.^22372237 Ben Bailey {Day39/167:12}-{Day39/168:21}. He said that most of the panels had been supplied directly to CEP for fabrication and insertion into the window frames, which might explain why the problem had not been picked up at the time.^22382238 Ben Bailey {Day39/168:25}-{Day39/169:25}; Ben Bailey {Day39/171:12-21}. However, he could not tell the products apart by their appearance.^22392239 Ben Bailey {Day39/172:3-7}. Neither the Rydon witnesses, nor Mr Lamb could explain how Styrofoam came to be substituted for PIR.^22402240 Lamb {Day38/177:7-12}; O'Connor {Day26/117:9-13}; Hughes {Day27/76:16}-{Day27/77:6}; Martin {Day30/54:10}-{Day30/55:4}. However, neither material was suitable for use in the external wall of Grenfell Tower. Given the lack of interest shown by Harley, Rydon and Studio E in the fire performance of the insulating cores of the window infill panels, we think it unlikely that any of them would have objected to the use of Styrofoam in the P2 panels.

205