The lessons of the King’s Cross Fire, or why building the LHC may be a bad idea
The King’s Cross Fire was a fire on an escalator in a subway in London in 1987. 31 people died, more than 60 were injured. It made the news of course, and I’m sure many of my older readers remember it. And if they were like me, they recall that the fire was on a wooden escalator. And if they thought much about it, they thought, well, what did they expect with a wooden escalator? Well, turns out they expected the fire to behave like hundreds of previous fires on these escalators, a bit of smoke for awhile, then the fire crew arrives and puts out the fire. So not only could the subway operators not have predicted this catastrophic occurrence, as it turns out, there was no way they could have predicted it. We’ll get to that in a moment.
OK, so we have two underground subway platforms connected by a tunnel with a WWII era largely wooden escalator in it. At some point a fire started under the escalator, almost certainly from a dropped match. Like so many other small fires before it, the staff simply closed off the escalator and called the fire department. The platforms above and below the escalator ramp remained open and were crowded with travellers, it was about 7:30 in the evening. Then, before the fire department arrived, and with no warning … the entire escalator above the fire burst into flame. Imagine over ten tons of bone dry wood and varnish igniting at once, survivors report that it was as if a giant blowtorch suddenly turned on. Everything flammable within about 40 feet of the tunnel top immediately burst into flame from the intense heat, including people. I will leave the horror and chaos to the gentle reader’s imagination.
So what the hell happened? How come this fire was different than the hundreds of others that preceded it? Investigators first thought it was possibly a bomb, the IRA was still active then. That was quickly ruled out by fire investigators examining the wreckage and debris. A train had arrived in the lower platform just before the flashover, and no doubt air being forced up the tunnel by the piston effect helped fan the flames, but it couldn’t account for the extreme intensity of the fire. So investigators made a sophisticated computer program to simulate the fire. And turned it on. The fire burned normally for a few minutes, then the smoke and flames from the fire lay down and hugged the floor instead of rising to the ceiling, heating the upper part of the escalator until a few minutes later it burst into flame at once, neatly replicating the fire. Smoke isn’t supposed to do this, it’s supposed to rise, not follow the floor. Investigators figured there must be something wrong with their program, and built scaled replicas.
And it did the same thing. To their surprise, investigators had discovered a hitherto unknown effect, now called the trench effect. In the right shaped tunnel with the right grade, smoke and flames from a fire in the right location will travel along the floor instead of rising to the ceiling. So while the fire was small and seemingly not dangerous, it was slowly but surely heating up the more than ten tons of wood and varnish above the fire. And when it got hot enough, it all caught fire at once. Highly flammable paint in the tunnel also burst into flame, contributing to the fire. And thus at the cost of 31 lives, science learnt something new about fires in tunnels. The remaining wooden escalators were of course removed, and highly flammable paint was no longer used on tunnels, as well as other changes to prevent a recurrence.
The lesson here? The map is not the territory. No one could have predicted this fire, because no one knew about the trench effect. And this is a point that I think a lot of people who believe in science forget. No matter how will understood something is, no matter how great our understanding of it, no matter how mundane and commonplace it is … reality bats last. There is always the possibility that humans are missing part of the picture. And this is especially true when humans do something that has never been done before. This is one of the reasons I am uncomfortable with the LHC (the world’s largest atom-smasher.) It’s also one of the reasons that building nuclear power plants all over the place may not have been such a grand idea, as Fukushima is now showing us.
In other words, no matter how smart the experts are that assure us that something is safe, that doesn’t mean it actually is safe. And when one factors the inevitable human error into the equation, it seems to me that “how bad could it get?” should be a far more important factor in safety calculations. Worst case scenarios sometimes do happen, including worst case scenarios like the King’s Cross Fire that even the experts couldn’t have predicted. Maybe the LHC is safe, I certainly hope so. I still would sleep better if we waited till we had the means to build it in orbit or on another planet though, building an experimental device that some people think could destroy the Earth strikes me as a bit foolhardy. Especially since the only real reason governments are spending piles of money on the LHC is because it may help build ever more fearsome nuclear weapons.
So we’re kinda screwed by the LHC no matter what happens. Sleep tight.
(The above image of the King’s Cross Fire aftermath is claimed as Fair Use under US copyright law. It’s not being used for profit, is central to illustrating the post, and its use here in no way interferes with the copyright holder’s commercial use of the image. Coming next, four epic fails of modern capitalism.)