Grenfell Tower Disaster, Part II: What Happened? Could it Happen Here?

The specifications that I wrote for aluminum composite panels (ACP) were very detailed. They were also very specific that in that whatever was proposed by the contractor, it would have had to have passed the testing required by CAN/ULC-S134 for:

1. Flame Spread Distance: flaming on or in the wall test assembly could not spread more than 5.0 metres above the opening during the test procedure, and
2. Heat Flux: during the flame exposure on the wall assembly could not be more than 35 kW/m² measuring 3.5 metres above the opening during the test procedure.

Now this is getting a bit technical, but I can assure that if this was the case for Grenfell Tower, the fire would have been limited to one or two storeys.

Offshore design for oil and gas production platforms represents almost half of my career. There was only one form of thermal insulation that we ever used in this context. It was mineral wool made from rock slag. It does not burn and comes in various densities from batts that can be formed around steel angles and stiffeners to rigid board that can be adhered to flat bulkhead surfaces. So when presented with on-shore health care design, and later in my career, high-rise tower design, there was only one form of building insulation I would consider – that was mineral wool. It is environmentally friendly and simply cannot combust.

I noted the use of foamed plastic insulation by others. It was cheap, and possessed superior insulation value for the thickness as compared to mineral wool. I often used the foamed plastic, but only under the specific circumstances permitted by the code. I was usually forced into this situation because of cost. My preference was always for mineral wool.

Soooo…

When the Grenfell Towers in North Kensington (west London) caught fire and burned killing seventy-one people in June of 2017, our architectural firm was glued to the live feeds. We guessed what was happening based on the footage. We then waited for aftermath photos to be posted in order to do our own post-mortem assessment. At twenty-four stories, Grenfell Tower was twice as high as anything we had ever designed, but we expected there would be lessons to be learned from this, and dearly hoped that any buildings we had designed or retrofitted with composite aluminum were not comparable.

Based on the live news feeds, we were certain that the problem was polystyrene or polyisocyanurate insulation. We were correct – at least partly. We never dreamed that aluminum composite could burn as it did on this building, but then again, we had never specified or accepted the non-fire-rated version of this product.

A judicial inquiry into the fire found that the polyisocyanurate insulation was installed flat against the original pre-cast concrete façade. Based on the thickness used, this would have more than tripled the insulation value. An air gap (standard construction technique) was created with ACP installed as the exposed decorative finish.

First of all, we were stunned that foamed plastic insulation could be accepted on a high-rise building. Not permitted in Canada. In the Grenfell Tower fire this material would propagate rapidly and never self-extinguish. It would be a ferocious killer.

Secondly, we were equally floored that a substitution was made for the ACP type. An inferior product was used that was never fire tested. Only ACP that meets CAN/ULC-S134 could be used on a building like this in Canada. A fire-rated aluminum composite panel would have prevented the ignition of foam insulation on subsequent upper stories.

Thirdly, our National Building Code requires firestopping at every floor level. This is essentially a non-combustible barrier that eliminates continuous air cavities that act as chimneys in the event of a fire. Firestopping would have acted in tandem with a fire-rated ACP to retard the spread of flame up the exterior of the façade.

Finally, Grenfell Tower had only a single stair tower and was not equipped with sprinklers. No building design in Canada (regardless of scale) would permit only one means of egress. This is where our life safety philosophy for exiting diverges completely with the European approach. And of course, sprinkler installation has been a major source of scrutiny (particularly with older building stock) since the 1996 version of the National Building Code that essentially made sprinklering more economical than previously accepted alternatives.

Therefore, to the question: Could a Grenfell fire happen in Canada?

Based on the assumption that similar buildings here have been designed according to code, the answer, ostensibly, is no. But the answer, when specifically directly towards aluminum composite, is maybe. Unrated, untested ACP is as saturated in the market here as anywhere in the world. The manufacturers don’t care. It is up to architects to specify the correct product and to exercise considerable diligence in seeing that it is installed.

The difference is not lost on many building authorities. The City of St. John’s have been particularly keen on enforcing CAN/ULC-S134. Its plan review staff are among the very best, due in large part to the leadership of the (now-retired) former manager, Sylvester Crocker. But I cannot be sure of what happens outside of St. John’s.

I note that the provincial Department of Transportation and Works Master Specification for Composite Wall Panels does not refer to CAN/ULC-S134, and does not even bring up the subject of fire testing. This is astounding. Combined with government policy of accepting the lowest bids for architectural services, this makes for an interesting recipe indeed. And municipalities who hire so-called project management consultants to facilitate capital building projects: How likely are they to catch items such as this?

ACP has certainly been used prodigiously outside the City. This is the authoritative realm of Service NL – the same provincial body that, in late 2016, declared that building designs could be prepared by engineers as well as architects. The TW Master Specification (promulgated by TW engineers) will form the basis for ACP installations on these buildings paid for with taxpayer dollars.

Sadly, I don’t know of any bureaucrats or local engineers familiar with CAN/ULC-S134 and how it should be applied. I could be wrong.

I almost certainly will be, after they read this.