A residential site in west London on a Wednesday morning, with a crane standing over half built blocks.
The mast is still upright, the cab is still there, and the counterweight has not moved.
The front of the machine is empty. The jib is on the scaffolding below.
Around a hundred people were on that site and every one of them walked off it.
Nobody was hurt, which is the whole story and also the odd part of it.
A jib does not usually come off at all.
Why a luffing jib is a different animal
A flat top crane carries a horizontal jib and moves the load in and out along it on a trolley.
A luffing machine does not. The whole jib pivots up and down at the front of the slewing platform, and the load moves by changing that angle.
Cities force that choice on developers rather than the other way around.
A luffing jib can be parked almost upright, so it sweeps a much smaller area over neighboring buildings and needs a smaller oversail agreement.
The trade off sits at the pivot. Everything the jib weighs, plus everything it lifts, passes through the connection where it joins the platform.
On a flat top the same loads are spread along a fixed structure. On a luffing machine they concentrate at a small number of pins.
Which makes those pins a single point where nothing may fail.
What happened on the site
The collapse came at around ten in the morning on the fourth of March, while the crane was in use.
The jib detached from its fixing and fell, landing across scaffolding on the building below rather than into the street.
The site was evacuated immediately and stayed shut afterward, with about a hundred people accounted for outside the hoarding.
The scheme is a housing development on Barlby Road in Kensington, built for the local council on land it already owned.
The crane was supplied by a hire firm rather than owned by the contractor, which is the normal arrangement in Britain.
No injuries were reported, and the damage stayed inside the site boundary rather than reaching the road outside.
Who investigates and what they look for
Two processes run in parallel after something like this, and they ask different questions of the same wreckage.
The national safety regulator investigates whether duties were met, and it can prosecute.
The crane owner runs its own examination with the manufacturer, and that one is about the machine.
Weather, load and the position of the jib at the moment of failure all get reconstructed from records rather than memory.
The physical evidence is the pivot itself. Investigators look at the pin, the lug, the retaining arrangement and the fracture faces on each.
A fatigue crack looks different from an overload break, and a pin that walked out of its housing looks different from both.
None of that is a quick answer, because the metallurgy runs on its own schedule.
What has actually been established
Very little, and that is worth saying plainly rather than filling the gap with plausible mechanisms.
What is confirmed is the date, the location, the detachment of the jib and the absence of injuries.
The regulator was notified and the site was closed, as the trade report recorded at the time.
The crane owner said an investigation had begun with the manufacturer and that a report would go to the industry.
No cause has been published since, and the later coverage adds nothing on that point.
So any account naming a specific pin is running ahead of the published record.
What the industry will change, if anything
Tower crane failures in Britain are rare enough that each one moves inspection practice across the whole hire fleet.
That is not sentiment. It is how a small industry with shared machines and shared parts actually learns.
The likely outcomes are narrow. A check regime for that pivot arrangement across the fleet, and a manufacturer bulletin if the parts are shared across models.
Erection and dismantling remain the highest risk moments in a crane’s life, and this machine failed during ordinary operation instead, which is the detail that will get attention.
Working over an occupied city is the constraint that shapes all of it, the same reason a helicopter set the last of 39 antenna sections on a Toronto tower rather than a crane.
Temporary works fail the same way when a single element is overloaded, as they did when a backpropping failure ran 105 feet across a slab.
Until the report lands, the useful statement is short. A tower crane lost its jib, and nobody knows publicly why yet.
