The AW169 Crash at King Power Stadium in Leicester: What Happened?

The investigation team recreated the Leicester AW169 scenario in the simulator with a test pilot: Every single attempt resulted in a crash 💥

If I asked you:

“What would you do if you suddenly lost tail thrust during take-off?”

You’d probably have a pretty good answer.

Lower the collective, manage the yaw, try to regain control, cushion the landing.

We’ve seen it, trained it, talked about it (many times).

But this accident forces a harder question:

👉 What if none of that actually works?

Because in this AW169 crash, the pilot did react.

But it wasn’t enough.

Let’s take a look at the AW169 crash at King Power Stadium in Leicester (UK), and what we can learn from it.

💥 Accident Overview

It’s 27th October 2018, we’re in Leicester, United Kingdom:

King Power Stadium is located just to the south of the city centre:

Earlier in the day, the crew of G-VSKP flew the aircraft from Fairoaks Airport at 1342 local to London Battersea, to pickup 3 VIP passengers.

At 1415 local, they departed with the passengers to leicester football club training ground.

The training ground was roughly a mile south of the stadium. After the match ended, the pilots flew to the stadium and arrived at 1847 local.

This is where the accident flight started. They departed with the passengers from the stadium at 1937 local, with the destination London Stansted Airport.

The AW169 climbed backwards (a manoeuvre called a “variable TDP departure”) to a height of 250 ft. At this point, the pilot started a 15 degree nose down rotation.

He called for the gear to come up, and the landing gear started to retract.

As the aircraft climbed through 300 ft, a small right bank was initiated. At this point, the helicopter experienced an increasingly developing right yaw movement.

The pilot applied full left pedal within about one second. The voice recording picked up from the cabin:

“Hey, hey, hey”

To which the pilot replied:

“I’ve no idea what’s going on”

About 4 seconds later after the initial yaw development, the pilot was heard uttering an exclamation.

A ROTOR LOW warning popped up.

At this point, the Radalt showed 430 ft. The descent started with a rotation rate of up to 209 degrees per second with pitch and roll oscillations at the same time.

At 75 ft the collective was raised to cushion the landing, and the helicopter struck the ground on a concrete surface, with a retracted gear, while still rotating.

It ended up on its left side and a post impact fire rapidly escalated.

No one was able to help the occupants due to the intensity of the fire. All 5 occupants on board were fatally injured.

We don’t publish all our Notes from the Cockpit (like this one) publicly, some are shared only by email. Get the next one sent straight to your inbox ⤵️

🔍 What Caused this Accident?

When this happened, there was a lot of speculation on whether or not the pilot adhered to the CAT A takeoff profile.

Let’s get 3 things out of the way first ⬇️

1️⃣ The investigation mentions that, while the rate of climb exceeded the stipulated 300 ft per minute, it:

“did not significantly influence the post-failure controllability of the helicopter.”

2️⃣ Also, the gear was raised below the stipulated point of Vy, but the investigation mentions:

“The investigation did not consider raising the helicopter’s landing gear before reaching climb speed to be a contributory factor in the accident or in its survivability.”

3️⃣ Finally, while a turn was also initiated before reaching Vy, the report states:

“The investigation concluded that the pilot choosing to enter a turn below VY was not of itself a factor in the accident.”

With that out of the way, let’s talk causes.

At the heart of the issue, we’re looking at a component called the tail rotor duplex bearing.

It looks like this:

Its job is to allow movement inside the tail rotor pitch control system while handling extremely high loads and constant motion.

During the accident flight, this bearing failed.

And once it did, the actuator shaft started rotating. After this, everything started to unravel incredibly quickly, let’s take a look at that.

So how did the bearing fail?

The failure was mainly caused by damage that had been developing over time inside the bearing itself.

The investigation found that the bearing was being exposed to:

🔸 Higher and more complex loads than expected
🔸 Dynamic axial and bending loads
🔸 Contact pressures high enough to break down lubrication inside the bearing

Once the lubrication film started failing, the metal surfaces inside the bearing no longer rolled smoothly over each other.

Instead:

➡️ The bearing balls started to slide across the race surfaces under a lot of pressure.

This created microscopic surface damage known as:

👉 Rolling contact fatigue

What happened after the bearing failed?

So, this is the course of events as per the investigation report, in chronological order:

Due to a right yaw pedal input, the tail rotor actuator control shaft started moving to the right under hydraulic pressure from the actuator.

Then ⬇️

The tail rotor duplex bearing seized. This caused the tail rotor actuator control shaft to start rotating.

The rotation of the shaft resulted in complete detachment from the control mechanism:

The result: complete loss of tail rotor control, and a tail rotor that’s now producing negative thrust, adding to the unopposed main rotor torque.

You can see an animation of this in this AAIB video (from 1:40) ⤵️

Finally, the investigation states ⬇️

“The unopposed main rotor torque couple and negative tail rotor blade pitch angle resulted in an increasing rate of rotation of the helicopter in yaw, which induced pitch and roll deviations and made effective control of the helicopter’s flightpath impossible.”

💡 What Can We Learn From This?

There are some valuable lessons we can learn from this horrible accident, let’s take a look ⬇️

Certification standards do not always reflect reality

It’s easy to assume that if something is certified… it’s been fully proven.

That it’s been tested against everything it might encounter in the real world.

This accident shows that’s not always the case.

Because the tail rotor bearing didn’t fail due to poor maintenance, misuse, or an obvious design error.

In fact:

🔸 The system met certification requirements
🔸 The testing complied with regulation
🔸 The failure mode had been considered during design

And yet… it still failed.

Why?

The investigation found that the bearing was exposed to:

➡️ Higher and more complex loads than the design had accounted for

And critically:

🔸 Real flight load data was not fed back into the design validation loop
🔸 Certification testing was not representative enough of operational demands
🔸 There were no explicit requirements to fully assess this type of fatigue behaviour

So on paper, the system worked. But in reality, it didn’t have enough margin.

After the accident:

The system was reviewed and improved:
🔸 Service Bulletins and Airworthiness Directives were issued
🔸 The understanding of loads, fatigue, and inspection requirements was updated

Some failures sit outside the envelope of pilot recovery

The investigation team carried out extensive testing in the simulator with experienced pilots, to replicate the event and to see if there were ways to assure a recovery.

The answer is no.

The report states:

“It was not possible, on any of the trial profiles, to reduce the yaw rate to a level where control of the helicopter’s horizontal trajectory could be established.”

And:

“Every simulated accident flight terminated in an uncontrolled touchdown which exceeded the simulator’s crash detection threshold.”

That tells you everything you need to know.

The reality is, some events are simply unrecoverable. If you ask an airline pilot to recover from a double engine failure over the middle of the Atlantic, you’re not going to get a great result. That’s just life, and aviation.

We’re simply on the edge of what is possible, and when things go wrong in ways that are the worst combination, there are versions of reality out there where there just isn’t an answer. This (unfortunately) is one of those.

Startle + Time Compression Is Brutal

No matter what unexpected situation we find ourselves in, we can’t escape some sort of processing time, and reacting to what we see.

Not to mention startle effect.

All of this works against us in a race against time.

The investigation report highlights some other really interesting points:

“The most significant influence on post-failure controllability was the response time for lowering the collective lever. When using a near-instantaneous response the TP was able to exercise a degree of attitude control as the helicopter descended but not enough to direct the flight path.”

And:

“When the reduction in collective pitch was initiated beyond two seconds the magnitude of pitch and roll instability became increasingly exaggerated and positive attitude control was not achievable.”

As well as:

“The greater the time between the injection of the failure and lowering the collective, the more unstable the helicopter became and the more difficult it was to control its attitude.”

We’ve covered the startle effect in more detail in this article on inadvertent IMC:

56 Seconds to Live: The Brutal Reality of Helicopter Inadvertent IMC

Single Points of Failure Still Exist (Even in Modern Aircraft)

While this is the case more so for helicopters than fixed wing, we can’t ignore that even in modern aircraft, there are single point of failures.

A bird hitting your disc cannot be recovered with a spare disc. If only!

Flying into a flock of geese can still result in double engine failure (as we’ve seen with the Hudson ditching). While yes, we have two engines, one event (geese) caused both to fail.

There are so many different examples of one event causing all redundancy to become, well, redundant.

But there are also more clear cut examples, like a single tail rotor drive shaft, a single tail rotor gearbox, a single tail rotor, etc.

One failure will result in a very, very bad day.

If you get a failure at the worst possible time, no amount of training will be able to get you out of it completely.

💭 Conclusion

This one’s uncomfortable to talk about.

Because when you go through it properly… you realise there isn’t much you would’ve done differently in that seat.

The pilot reacted, and put the correct inputs in.

But the helicopter was already doing something he couldn’t control.

And that’s a tough thing to accept.

We all like to think that if something goes wrong, we’ll handle it. That training, experience, and staying calm will get us out of it.

Most of the time, that’s true.

But… not always.

Sometimes things fail in a way that just moves faster than you can process.

And by the time you understand what’s happening… you’re already behind it.

That doesn’t mean we stop training or questioning things.

If anything, it’s the opposite.

You can find the AAIB report here.

Our best Notes from the Cockpit are only shared by email. Get them straight into your inbox, twice per month ⤵️