Grenfell Everything I Have Learned On The Causes Behind The Inferno.
[ Posted by Lara Keller 5/8/17 Updated 7/7/19 ] Blog Table Of Contents
1. Introduction: Grenfell Tower Fire
Grenfell Tower fire in the early hours of the 14 June 2017, spread with incredible speed spreading from the 4th to the top 24th floor in 20 minutes. Over 80 men, women and children has been killed by toxic fumes and heat, and their remains cremated. It really is not too strong to describe this as the most horrific fire in the UK since the Second World War. To keep any respect for democracy and government regulation in the UK, the residents of Grenfell Tower need to get full justice, and there needs to be urgent action to make residential housing (social, student or private) safe in the UK. On a more personal level I was shocked by the death of Mohammed al-Haj Ali a young civil engineer who had been forced to flee the malevolent state of Assad’s Syria, only to become a victim of the indifferent UK state. It is a bitter irony that indifference for life does not respect borders.
The thing that marks this Tower Block fire out from previous ones is the very rapid speed of the fire spread, and the high number of deaths and injuries. The main issue in getting justice is understanding the “rapid fire spread”. In this article I ignore the cause of the kitchen flat fire and sprinklers. I ignore even the response of the fire services, fire stopping around windows, fire doors, and the quality of the escape routes and fire advice (“Stay Put Policy”). The reason is that the authorities will argue if Grenfell had been a normal fire these systems would have worked. I will also ignore the disgraceful treatment of the residents before and after the fire by the super wealthy penny pinching social cleansing Kensington and Chelsea Council. This is an important issue, but who is actually directly responsible for the crime of refurbishing Grenfell Tower to make it a death trap needs to be established, and this means looking deeply at the cause.
In the weeks following the fire, there were several fire engineering experts giving concrete informed opinions on the fire in the UK media. This from the Daily Telegraph http://www.telegraph.co.uk/news/2017/06/24/grenfells-unusual-design-led-blaze-spread-say-investigators/ it quotes a “well places source” putting focus on voids in the cavity behind the rainscreen (termed “cladding” by media) on the Tower Blocks columns (some architects term these “fins”).
Early video of the fire appeared on BBC Newsnight, showing the two columns near the 4th floor (close to the North-East corner) flat where the flat started, on fire from the 4th to the 24th floor, the entire height of the building is in flames in 20 minutes.
2. Cladding and ventilated cavity barriers.
So it seemed from several sources that fire stopping in the cladding cavity on the columns was the key area to look at. The media at this time were still just reporting that the Reynobond PE rainscreen used, had a flammable Polyethylene core, and this was the nearly the sole cause of the rapid fire spread. Obviously the Polyethylene is a major cause of the fire, but is it like the “bullet” with the botched fire stopping in the cladding as the “gun”.
Cladding is fitted on to a 1970s Tower Block like Grenell Tower, to bring the insulation up to modern standards, stop the degrading of the concrete walls by weather, stop damp getting into flats and to improve the outside appearance. It is fitted as a “cladding” system like this:
The word Cladding is often used to refer to all the layers, with the outer one being called Rainscreen. There was a standard 50mm cavity gap between the Rainscreen and the Insulation at Grenfell. This allows the insulation to stay dry, as it is only effective if dry. This ventilated cavity is essential so that changes in air pressure between the inside and outside of the building will not draw rain into the cavity. The problem with this, is that fire can get into the cavity gap and spread unseen between floors.
To stop this and to still allow ventilation, at each floor level a complete line of Ventilated Cavity Fire Barriers is installed, which goes from the layer of the Concrete Wall to the Rainscreen. Steel brackets hold them firmly in place. A 25mm air gap is left between the Cavity Barrier and the Rainscreen. The Cavity Barrier contains a layer treated with an Intumescent material, that expands when exposed to heat. So when the cavity reaches typically 180 C, the line of Cavity fire Barriers should have fully expanded to close the 25mm air gap left in the cavity. Each floor level section of cavity in the cladding is then sealed from the floor above and below.
There are lots of companies who make Ventilated Cavity Fire Barriers, and I do not know who makes the best ones. Here is an informative piece from a company called Tenmat http://www.tenmat.com/passive-fire-protection/ventilated-fire-barriers :
I do not have had connection to people who make any of these cladding products, my only concern is that Grenfell Residents get full justice, which means looking into the construction details and so trying to get at the root cause of the rapid fire spread.
3. The Chimney Effect.
The Chimney Effect refers to how vertical cavities without fire barriers, can act like chimneys, which can create very powerful upward drafts. Obviously this updraft extends flames upwards, and carries molten droplets of plastics used in Rainscreen (like Polyethylene) upwards. This updraft also effects how semi combustible “charring” cladding products like PIR insulation burn. Here is a graph showing the “chimney effect”. It’s force is driven by the height of the chimney and the temperature of the fire at the bottom of it. The updrafts are then compared to the pressure you would feel from outside wind speeds on the Beaufort scale. At Grenfell the distance from the 4th floor flat to the top flat is around 45m.
[Important Note: The cladding added to tower blocks is ventilated along its entire height. Therefore this is a “leaky” chimney effect. The updraught figures above are the upper boundary of what is possible and refer to a non-leaky chimney. ]
To know how semi combustible products used on Grenfell Tower would react to a “chimney effect” in a 45m cavity, requires looking at the fire behavior of these materials.
4. Tests of the individual cladding materials used.
At end of June 2017 the BRE (UK Building Research Establishment) were testing small samples of materials cut from Grenfell and other Tower Blocks. The Guardian newspaper reported that these tests were less than transparent and not to standard procedures: https://www.theguardian.com/uk-news/2017/jun/26/tower-block-cladding-tests-after-grenfell-fire-lack-transparency-say-experts . It was being reported that the samples of rainscreen and insulation from Grenfell both failed these fire tests: [Det Supt] McCormack said: “Preliminary tests show the insulation samples collected from Grenfell Tower combusted soon after the tests started. The initial test on the cladding tiles also failed the safety tests.” (https://www.theguardian.com/uk-news/2017/jun/23/grenfell-tower-fire-police-considering-manslaughter-charges).
I looked at previous fire tests for the the Rainscreen (Reynbond 55 PE, 0.5mmALU-3mmPE-0.5mmALU) and the Insulation (Celotex rs5100, which is 100mm depth version of rs5000). According to various brochures for these materials they are fire rated in the UK as Class 0. This translates to Euroclass B as the table below shows:
Euroclass ratings use the Single Burning Item (SBI) Test, which uses a 30KW Propane Burner to heat a sample of the material in a special high tech room where the heat produced and combustion products are analyzed, to determine how much energy the sample has released and what gases, smoke and droplets are formed. (http://www.sp.se/en/index/services/firetest_building/firetest_bu%C3%ADlding/EN_13823_SBI/Sidor/default.aspx). Two sheets of the material (1m x 1.5m and 0.49m x 1.5m) are mounted on a trolley in a corner configuration. The burner is mounted at the base of the corner, and heats the material for 21 minutes.
[Important Note: Since this was orginally written Celotex rs5100 has been withdrawn. It has reclassified from UK specification Class 0 to a Class 1 material. A test by Exova Warringtonfire WF 383671 on 17/8/17. The THR(600) ie 10 minute heat release in the SBI Test for the rs5100 (100mm) PIR product was 5.26 MJ. The change in classification may be due to the Fire Growth Rate (FIGRA) which measures the peak heat release. Still uncertain at this time. 7/7/19]
Celotex rs5100 is rated as Euroclass B, which means the heat energy produced by the burning sample in 10 minutes is less than or equal to 7.5 MJ (THR600s <= 7.5 MJ). The 30KW propane burner supplies 18 MJ to the sample over 10 minutes, and the sample produces 7.5 MJ or less. Do not have the exact THR600s for rs5100, only that <= 7.5MJ. The sample is 2.24 m2 of rs5100. Rs5100 weighs 3.38 kg/m2 and PIR produces approximately 26 MJ/Kg when burnt. So the sample could produce 197 MJ if completely burnt. Therefore less than 4% of the heat energy contained in the insulation is released in this test.
This PIR type foam, is a modified rigid thermoset type of polyurethane, that consists of rings of molecules strongly cross linked, as opposed to thermoplastics formed of loosely bound chains of molecules. When heated with a blow torch PIR burns, forming a char layer, but goes out when the torch is removed. There is also an adhesive layer of relatively thick aluminum foil attached to the PIR to resist ignition.
There is a test for the Reynobond 55 PE panels (0.5mmALU-3mmPE-0.5mmALU) (CSTB Report No. RA11-0032 9/2/2011) that gives THR600s as 2.2 MJ. These panels consist of two sheets of 0.5 mm Aluminum (the 55 refers to this) bonded to a 3mm largely unmodified polyethylene core. Basically the Aluminum prevents very much of the core being exposed to oxygen, where heat can be applied to it to make it burn. So is you supply 18 MJ of heat with a propane burner to a sample of Reynobond 55 PE panels, you will only generate 2.2MJ of heat from areas of the panels that have been burning. This fire is therefore not self sustaining.
The sample is 2.24 m2 of Reynobond 55 PE. The 3mm polyethylene core provides 0.003 m3 per m2 of panel. Polyethylene weighs 950Kg/m3. So each m2 of panel provides 2.85 Kg of polyethylene. Polyethylene produces approximately 43 MJ/Kg when burnt. So the sample could produce 275 MJ if completely burnt. Therefore less than 1% of the heat energy contained in the rainscreen is released in this test.
It is difficult to know what tests and standards the BRE (UK Building Research Establishment) were using when they did their single item burning tests on samples of rainscreen and insulation. The London Metropolitan Police did say in a press conference the Rainscreen was harder to burn than the Insulation. The media reporting about the Rainscreen had given the impression that if a corner of it was lite with a burner the whole cladded building facade would go up in flames. The reporting of these first small scale BRE tests was very limited , and tended to reinforce this impression that “cladding materials only” was the reason for the rapid fire spread.
5. Overview of “natural” wall and chimney fires.
When a solid item burns, fuel vapour is released from the item as it heats up. A buoyant plume is formed, which in natural fires of sufficient size is turbulent (has eddies) and contains unburnt fuel vapour. The energy from the burning vapour produces heat radiation (infrared waves) and heats the air pulled into the plume as it rises (“entrainment”).
The heat supplied (whose density over an area is called “heat flux”) to an inert wall from a fire at its base, can be due to radiation or convection (direct transfer from hot gases). The density of the heat flux and the proportion of radiation to convection changes with height and how much soot is in the plume (sooty flames produce move infrared radiation). The lower part of the plume that contains flames corresponds to the section that contains unburnt fuel vapour. Above the flame height the plume spreads out and dilutes its energy.
If the fire is below an inert chimney, the infrared radiation produced by the flames must all be directed towards the walls of the chimney, and the hot gases of the fire are squeezed into a smaller faster moving cross section so increasing convection to the walls. Updraft pulls more oxygen into the fire, and increases the rate of burning and the speed of plume. When compared to wall fires the plume will supply more heat energy to the walls of the chimney, and the flame height will be much greater.
If the walls of the chimney are not inert, then a growing positive loop develops as the burning walls of the chimney create more updraft. This is why stopping the chimney effect in the gap between the rainscreen and insulation is absolutely critical with cladding materials which are in in any degree combustible.
5. How “chimney effect” would have effected Grenfell cladding materials.
Taking the case of the PIR Insulation Celotext rs5000 used at Grenfell. A vertical wall of this material is going to burn more fiercely if trapped in a cavity behind rainscreen. Much more of the heat energy from an area of burning insulation is going to directed back to the area of insulation above it, rather than being lost to the open air. As PIR insulation burns it forms a solid char layer. Rather than the hot plastic foam breaking up into a gas that more easily mixes with air. A strong updraft forces oxygen into the char layer and so helps it burn. The same idea applies to the uses of the bellows in a blacksmith’s charcoal forge.
So in the scenario above a section of vertical semi-combustible cladding (rainscreen, cavity, insulation) is subject to an intense fire in an iron crib at the bottom:
On the Left the rainscreen near the crib fire burns, and cannot enter the cavity because cavity fire barriers have closed it. The insulation nearest the crib fire burns most intensely, but much of the resulting heat energy is lost to the open air. The sections of insulation above this burn less intensely as the amount of supplied heat diminishes. This is more like the fire behavior of PIR insulation in a room corner “single item burning” test.
On the Right there is the same set up, but with no (effective) cavity fire barriers. Fire from the crib enters the cavity and is drawn up like a chimney. A strong updraft and the more efficient movement of heat energy up the surface of the burning insulation means the whole surface can burn more fiercely and rapidly. This is very different fire behavior than the room corner “single burning item” test. It also means that it needs a smaller total quantity of combustible material to create a serious fire within a cavity. It has been reported that in the rainscreen and the insulation, when added together, there was the calorific equivalent of 15 tonnes of combustible material on the outside of Grenfell Tower.
Also the Rainscreen would burn differently in the two scenarios above:
On the Left the Rainscreen will only burn to a certain height and then stop. The single item burning tests show it needs external supplied heat to burn, and this falls rapidly with distance from the crib fire. The intensity of the Insulation burning is falling off with height, as heat is lost, and so cannot supply heat to the Rainscreen.
On the Right the Rainscreen burns due to the crib fire in the same way, but the fire moves up the cavity like a chimney. The insulation burns fiercely, and the trapped heat also heats the Rainscreen. The polyethylene core softens and the bond with the outer aluminum sheets weakens. This means when the liquid polyethylene is drawn into the cavity and ignites it will burn much more fiercely, than in the corner of the room “single item burning” test.
It seems highly probable that the sudden bursting into flames of sections of Reynobond PE Rainscreen on the columns at the Grenfell Tower did not occur when the Rainscreen was cold, but when an existing fire behind the Rainscreen in the Cladding Cavity had preheated the Rainscreen to a critical temperature. This is illustrated here:
6. Taplow Tower Fire 2012.
There was an intense flat fire at Taplow Tower in Camden London in 2012 that broke through a window, where the cladding consisted of Reynobond PE Rainscreen and Mineral Wool insulation (non combustible). The flat was gutted and put out by the fire brigade from the inside. The external cladding only burned up to the next floor level.
This shows that the rainscreen Reynobond PE does not burn uncontrollably if subjected to an intense fire source. It is assumed here that the cavity fire barriers worked at Taplow Tower, and the fire was prevented from entering or going far into the cladding cavity.
7. Cladding system test of materials used on Grenfell Tower by BRE ~ July 2017.
By July 2017 the BRE (UK Building Research Establishment) were carrying out a “full scale” test on the combination of cladding materials used at Grenfell Tower (as well as othe combinations) using a BS 8414 rig set up, that is used to determine if external building envelope systems (ie cladding) will contain a fire within a compartment (ie flat) for 15 minutes. It is assumed that all the components of the cladding system are installed correctly, including rainscreen, cavity fire barriers and insulation. To pass the test, the internal thermocouples in the cavity and in the insulation on level 2 (which models the next floor up from the fire source), must not reach 600 C for 30 seconds in the 15 minutes after the “crib fire” (which models fire source, peak output 3MW, approx 250Kg of dry pine) starts.
The BS 8414 test on the Grenfell cladding materials was reported by the media as an “absolute fail”. This is illustrated below:
The intense crib fire burned away a section of the Reynobond PE Rainscreen nearest the source. Flames were observed reaching up to the top of the test rig and sometimes beyond it (flames in natural fires obviously fluctuate). The test was finished after just 7 minutes. The actual results of the fire on the test rig looked like this:
7. What would a BS 8414 style test with 20 levels look like?
This BRE BS 8414 test does not explain why 20 floors of cladding at Grenfell Tower was on fire in the first 21 minutes. This test proves that a powerful large dry pine crib wood fire can melt the polyethylene core of panels directly above, and the combined fire can burn a large hole in the Reynobond PE Rainscreen. It does not show that under normal fire conditions that the Rainscreen will continue to burn without a large external heat supply (like the wood crib).
Imagine if the BRE had been able to build a 20 floor test rig, the results would look like either the second or third case illustrated in the image below:
If the BRE had built a 20 floor tall test rig, then after 30 minutes the crib fire would use up all the wood fuel, and the damage to the rainscreen and insulation could potentially extend to a few floors. This is shown in the second (middle) case in the image above.
The first (top) case in the image above shows how the actual BRE test results would have looked on this enormous test rig.
The third (bottom) case in the image above, shows what actually happened at Grenfell Tower, with the fire racing up 20 floors in 21 minutes. This could only happen, from all the information I and others have seen, if the cavity fire barriers were ineffective AND combustible cladding materials were used, and the fire raced up a 45m chimney behind the rainscreen.
8. Defective design and installation of Cavity Barriers on Grenfell Tower.
There is some evidence that there were gaps in the horizontal cavity fire barriers under the rainscreen at Grenfell Tower. This uses technical planning drawings submitted to the local council and high resolution photographs of the exposed cladding when the refurbishment was finished. This picture below shows gaps in the cladding ventilated cavity fire barrier line on a Grenfell Tower column:
The rainscreen panels are hung on the small rods bolted across the vertical black aluminum channels, that cut through the horizontal cavity fire barrier line. There is no evidence of fire stopping in these aluminum channels. Also almost all of the height of the columns has a ribbed and grooved surface in the concrete. There is no evidence that these ribs and grooves were firestopped.
I then transposed these air gaps onto a technical drawing of a Grenfell column cross section :
You can see the ribs and grooves in this concrete column cross section. These air gaps are shown in yellow, and assume to be unprotected. Using the cladding photographs above there are yellow air gaps around the aluminum channels that support the rainscreen panels, these are assumed to be unprotected. Also shown in blue is the 25 mm protected air gap between the ventilated cavity fire barrier and the rainscreen, which will be closed when the barriers expand with heat (fully closed at 180 C).
The image was created as accurately as possible using a detailed technical drawing to scale. A graphics program scanned the image for yellow and blue pixels. Scaling this up to the real actual scale, an air gap 250 cm squared was protected by cavity fire barrier, while an air gap of 270 cm squared was unprotected, using the assumptions given. If this situation was repeated on other Grenfell cladded columns, then the cavity fire barriers would be essentially useless, as fire bypassed them to flow through unprotected air gaps. [2019: submissions to the Grenfell Inquiry in 2018, by Dr Barbara Lane exposed many more serious defects in the cavity barrier coverage, installation and product selection on Grenfell Tower. In some cases cavity barriers were installed upside down or even the wrong way round]
The picture below shows cladding under a window in a horizontal cladding section between the columns:
Gaps have been cut in the horizontal cavity fire barrier line to take the vertical aluminum channels.
There also appears to be no vertical firestopping between the vertical cladded columns and the horizontal cladded sections around the flats’ windows, that should have stopped the fire moving sideways from the columns [2019: submissions to Grenfell Inquiry stated only half of the necessary vertical fire barriers around the columns were installed].
9. Defective design and fire-stopping around Windows on Grenfell Tower.
It has also been reported that air gaps were left around the flats’ windows, allowing smoke and fire easy access to enter flats ( https://www.channel4.com/news/grenfell-new-revelations-did-window-renovations-contribute-to-spread-of-fire ). This excellent piece of reporting [2019: confirmed by 2018 Grenfell Inquiry submissions] suggests that the windows of the flats were moved forward by the refurbishment from being in the concrete walls to being in the same vertical plane as the insulation. This allowed smoke and flames direct access to flats by going through the burning PIR insulation. There was no fire stopping around the UPVC windows to stop his happening.
10. Defective internal fire-stopping and smoke extraction in Grenfell Tower.
Internal firestopping at Grenfell Tower needs to be examined as closely as external firestopping. The refurbishment involved removing and installing new pipes in the vertical service risers, that runs vertically through the flats over the entire height of the building. An application was made to remove the internal vertical fire-stopping between flats, and it is not known if this was effectively replaced after the refurbishment ( http://www.insidehousing.co.uk/fire-safeguards-in-grenfell-were-temporarily-removed-during-refurbishment/7020464.article ).
The lift lobbies and the protected stairwell quickly filled up with smoke, preventing many people from escaping. The last resident to escape reported being forced by smoke in his flat into launching himself into the thick smoke outside his flat, and down the stairwell. He thought he was treading on lots of fire hoses, but they turned out to be dead people who had collapsed and died on the stairwell from toxic smoke. The doors of the flats and the doors to the stairwell should have been able to contain fire and smoke for 60 minutes.
There should have been a system to extract smoke from the lobbies and the stairwell. None of this happened. Defective fire stopping from fire doors is a common and very serious problem in social housing tower blocks in London ( https://www.ifsecglobal.com/london-tower-block-fire-councils-social-landlords-ignored-warnings-years ). The refurbishment was completed in 2016, and these new fire doors should have all been installed to the correct standard. Residents of Grenfell Tower were ignored when they complained about botched internal refurbishment of their flats. Was this lack of care also extended to fire doors?
11. Serial botchers of refurbishment of social-housing.
In March 2015 Grenfell residents formed an action committee, and started refusing to allow Rydon contractors access to their flats, due to the unprofessional standard of the work being done ( https://grenfellactiongroup.wordpress.com/2015/03/19/who-you-gonna-call-rydonbusters/ ).
A similar refurbishment project in 2014 also carried out by Rydon at the North Myatts Field Estate in Lambeth in London resulted in a whistle blower inside Rydon writing to the residents association ( https://www.theguardian.com/society/2017/jul/21/the-real-cost-of-regeneration-social-housing-private-developers-pfi ):
This is about another project by the same company, and does explicitly mention the issue of fire-stopping, fire assessments, emergency lighting and smoke alarms. It suggests a chronic failure of respect for health and safety regulations by Rydon management and its sub contractors.
12. Only non-combustible cladding is adequately safe in the real world.
Currently the UK BRE [July 2017] are doing full scale tests on combinations of rainscreen and insulation. Without effective fire stopping (cavity fire barriers) no combination is safe for use in high rise residential or office buildings. Unprotected cavities in cladding provide a route for fire to travel between compartments (ie flats).
Only Limited Combustibility Rainscreen (ie Fibre Cement Panels) and Mineral Wool Insualtion should be allowed on high rise buildings. Other combinations are much too dependent on adequate fire stopping. This is too much “core fire safety”, to hang on correctly installed fire-stopping, which is often hidden, not inspected and botched by low integrity contractors (whose obsessive focus are on costs and profit rather than safety).
These other cladding combinations need to be stripped out and replaced. The choice of cladding materials allowed by building regulation needs to reflect the reality of the lack of respect for UK government regulation among many in the construction Industry ( https://www.theguardian.com/uk-news/2017/jun/15/long-builder-chain-for-grenfell-a-safety-and-accountability-issue ).
13. Without fire-stopping tower blocks are human incinerators.
The Grenfell Tower would not have been turned into an absolute fire death trap, if fire-stopping was not systematically botched. This central issue must be forcefully pursued by fire engineers reporting to the public inquiry. However this seems very unlikely if construction industry friendly experts are appointed by the inquiry. [2019: Fire-stopping and cavity barriers were investigated and reported on to the inquiry in 2018. The well known fire science experts appeared to avoid meaningful conclusions, for some reason.]
This is how the Rapid and Deadly Spread of the Grenfell Tower Fire could have happened:
14. The legal aspects.
Lastly on the legal side, the UK 1972 Defective Premises Act states that builders have a duty of care to residents of the buildings they build or alter ( http://www.legislation.gov.uk/ukpga/1972/35 ). The Building Regulations 2010 Act in Schedule 1 sets out the requirement for buildings to adequately resist internal and external fire spread ( http://www.legislation.gov.uk/uksi/2010/2214/schedule/1/made ). The UK government has given guidance on how this can be achieved in Approved Document B ( https://www.gov.uk/government/publications/fire-safety-approved-document-b ). There is also the UK Building Control Alliance guidance on the “Use of Combustible Cladding Materials on Residential Buildings” ( http://theriveroflife.com/wp-content/plugins/BCA-Technical-Guidance-Note-18.pdf ), which gives this very relevant advice:
“Within the confines of a cavity, the flame will also elongate up to ten times its length as it searches for oxygen. Hence, the need for robust cavity barriers, restricted combustibility of key components and the use of materials with a low spread of flame rating is necessary, particularly given the delamination and spalling [breaking into fragments] nature of some of the components when heated”
It should be possible to pursue those responsible for the Grenfell Tower Fire disaster for corporate manslaughter (senior managers whose policies led to death and injury, penalty=fines) and gross negligence manslaughter (against individuals whose actions resulted in a failure of duty of care leading to death and injury, penalty=imprisonment).
What is really needed is more whistleblowers. Now is your time.