On 5 September 2017, the Government published data for the number of high rise residential buildings in England (as of 31 August) with various types of ACM cladding, and with various types of insulation:
The number of buildings accounted for in this table is:
81 + 107 + 8 + 21 + 12 = 229.
The total number of buildings in England with ACM cladding (including those stripped post-Grenfell) is given in paragraph 5 (and also in paragraph 8):
Since 173 + 16 + 89 = 278, it is clear that 278 is the number of buildings as well as the number of samples.
The same figure of 278 is given in paragraph 28, in both text and table:
In the right hand columns all 278 buildings are counted as being associated with one of the seven combinations of ACM and insulation for which large-scale tests were carried out. It will be remembered that the total for these 7 tests given in the previous table was 229 only. Footnote 5 explains that there are 49 buildings for which the insulation is unknown, and that these have been counted with the combinations that were tested and failed.
229 + 49 = 278
and so the discrepancy is resolved. So we have:
229 buildings with 1 of the 7 combinations that were large-scale tested; and
49 buildings with unknown insulation; giving the total of
278 buildings.
This means that there were no buildings known to have one of the two combinations that were not tested. We already know from paragraph 5 that there were no buildings with phenolic insulation and A2 ACM, since all ACM samples failed the small-scale combustibility test, and A2 ACM passes that test by definition of the A2 grade.
But what is extraordinary is that the DCLG was claiming to have been unaware of any buildings with phenolic insulation and PE ACM cladding. Let us examine the first table again:
It appears from columns 1 and 3 that PE ACM had been much more commonly used than FR ACM. 81 buildings had PE ACM with PIR, only 8 had FR ACM with PIR. 107 buildings had PE ACM with stone wool, only 12 had FR ACM with stone wool.
It appears from row 2 that phenolic insulation has been more commonly used than either PIR or stone wool. 21 buildings had phenolic with FR ACM, only 8 had PIR with FR ACM, and only 12 had stone wool with FR ACM.
If PE is more common than FR, and phenolic more common than PIR or stone wool, then PE plus phenolic might be expected to be the most common combination. But of course there could be some reason why the two were not used together. So we turn to other evidence.
‘The most common product seen on site’
Steve Evans, who has served as Chair of the Building Control Alliance Technical Committee, stated in a presentation that a change in a BBA Certificate for the
most common product seen on site
in late 2013 was one of the factors that led the BCA to issue Technical Guidance Note 18. I established through a correspondence that this product was Kingspan’s Kooltherm K15 phenolic insulation board for rainscreen cladding systems, and I obtained copies of its 2008 and 2010 BBA Certificates. I have shown previously that the 2010 BBA Certificate had a major loophole that apparently allowed the product to be used above 18 metres without the large-scale BS 8414 test that is required by Approved Document B2 to be carried out on the system in which it is to be incorporated.
It looks very much as if the closure of the loophole in the 2013 BBA Certificate prompted a search to open a new loophole, followed by the creation of the new Desktop Study ‘option’ by the BCA in its TGN 18. The BCA had no more authority to bypass the statutory guidance of the Approved Documents than the BBA had done, but as the representative organ of the building control bodies, it is easy to see how its guidance began to be followed by building control officers in place of that of the Approved Documents.
It seems to follow that considerable use was being made of the BBA Certificate loophole prior to its closure in December 2013. Otherwise, why would the loophole’s closure have led to a search for a new way to use the product on high rise buildings in systems that had not undergone a BS 8414 test?
What then was the loophole? The 2010 BBA Certificate stated that K15 could be used
in accordance with … [paragraph] 12.7
of Approved Document B2, even though it did not meet the limited combustibility requirement of that paragraph! The reason given is that the product is classified as Class 0, even though that classification has nothing to do with combustibility!
Since paragraph 12.7 applies only to buildings over 18 metres, K15 could therefore, according to the Certificate, be used under the Linear Route on high rise buildings, with no requirement for a system test. It follows that, under the Certificate, it could be used in combination with outer cladding that met the requirements of 12.6.
12.6 requires that the external surfaces of the external wall construction should be national Class 0 or Euro Class B above 18 metres. The Reynobond PE ACM panels used on Grenfell Tower were classified as Euro Class B in a BBA Certificate issued in 2008, as well as in a CSTB Certificate issued in 2011. Alucobond PE ACM panels were classified as national Class 0 in BBA Agrément Certificate 05/4214, issued in 2005. Larson PE panels were classified as national Class 0 in Warrington Fire Certificate of Approval CF 5223 in January 2014. The certificate was signed by Sir Ken Knight, now Chair of the Expert Advisory Panel established by the Government ‘to make buildings safe following the Grenfell fire’:
The Certificate states clearly that because it is Class 0, the material may be used on
the external surfaces of multi-storey buildings
including residential buildings and institutions:
There are thus at least three PE ACM products that have been certified to national Class 0, and at least two – Alpolic PE and Reynobond PE – that had been certified prior to the closure in December 2013 of the loophole in the 2010 BBA Certificate.
It is thus easy to see the potential for cladding systems with Kingspan’s phenolic Kooltherm K15 insulation boards and PE ACM panels to be approved between April 2010 and December 2013. The insulation boards could be considered to be covered by the 2010 K15 BBA Certificate, and the ACM panels under AD B2 12.6 through a national Class 0 or Euro Class B classification.
Was the 2005 K15 BS 8414/BR 135 test pass misused?
In late 2005, a system incorporating K15 successfully completed a BS 8414-1 test, satisfying the criteria which had been set in the 2003 second edition of BR 135 Fire Performance of External Thermal Insulation for Walls of Multi-Storey Buildings. The following diagram from a 2015 Kingspan Technical Bulletin represents the build-up:
and K15’s 2008 BBA Certificate 08/4582 provides some extra detail:
The outer cladding is described in the diagram as ‘non-combustible cement board’, but in the Certificate as ‘cement particle board’, a term which normally refers to a combustible product, with a substantial wood particle component. The ventilated cavity was 40 mm deep, and there were two horizontal cavity barriers. The system is included in the post-Grenfell release of information about cladding systems that have received BR 135 Classifications from the BRE:
A very misleading advertisement by Kingspan, August 2006
Kingspan placed a full page advertisement in the August 2006 edition of International Fire Protection:
The last thing you need behind rainscreen cladding, Kingspan proclaimed, was
flammable insulation
particularly in a multi-storey construction:
Since Kingspan were certainly not intending to advertise their own product as ‘the last thing you need’, it follows that they were claiming that it was not flammable. But is this true?
Tom Clarke of Channel 4 news put Kooltherm K15 to the test in an independent laboratory in early July 2017 (at 1 minute 30 seconds and following):
It burst into flames:
and kept burning:
If this is not a flammable product, I do not know what is. Shall we conclude then, with Kingspan, that Kooltherm K15 is
the last thing you need
behind a rainscreen cladding system?
Andrew Chapman
