Hacker News Reader

Comment by u1hcw9nx | Hacker News Reader

Turbine disk failure is one thing that can down a plane, and you can only reduce the probability so much.

Fan blades and turbine blades can be contained by containment cases if they break off, but turbine disks can't. When a disk breaks, it's an uncontained engine failure. A 20 kg disk fragment flying at Mach 1 goes through everything.

Airframes are have a 30-degree fragment spread cone around every disk stage. Airframes are designed so that no single fragment can destroy all redundant systems at once in this zone (flight controls, fuel lines, and hydraulic systems). Jet fuel cannot be stored in the wings in this zone.

View on HN

I admit that I glossed over the section on the machining of the oil pipe, but the facility appears to bear most of the blame.

These were pulled and scrapped from the whole A380 fleet.

"Investigators also criticized the culture within the Hucknall facility that manufactured the HP/IP bearing hubs, identifying signs of complacency and widespread procedural non-compliance. A paperwork review showed that the required signatures were missing from 131 out of 138 retrospective concessions issued between 2009 and 2011, and a large number of minor non-conformances had not been properly handled through the normal chain of command."

That's an oversight failure as much as a failure of that plant.

The oversight was at the plant.

Yes, but there are levels of oversight and this kind of thing should not be possible to hide. Someone on the official side of this messed up as well.

But it seems to me that the whole failure could have been prevented with a bunch of temperature sensors spread throughout the engine, and a good anomaly detection system.

Don't know why you're getting faded out, because having worked on exactly this engine, software mitigations were put in place. That's exactly your `anomaly detection system`

It was a belt-and-braces fix, because the primary fix was of course to manufacture the engines correctly in the first place.

The "lots of temperature sensors" thing is actually easier said than done; jet engines already have lots of temperature sensors, and they will turn the engine off if a huge overtemperature is detected, but it's not a free win to add more of them. It increases the risk of a faulty sensor shutting down a perfectly good engine in flight, which from a safety-critical perspective is a borderline disaster condition all by itself.

(technically you usually cross-correlate temperature sensor readings with pressure sensor readings to prevent sensor faults turning engines off, but still, extra sensors isn't always a free win.)

Though in this case the software mitigation was turbine overspeed detection, not overtemperature detection.

I like to sit away from the wings to have a better view out the window, but not sitting in the fragment spread cone is probably also a good idea ;)

Could either marketed as a discount seats or “excitement seats”.

It is worse than even that:

> Within the space of four seconds, the energy from the annulus gas flow accelerated the IP turbine past its critical speed, until centrifugal forces exceeded the ultimate strength of the nickel alloy disk. The red-hot, wildly spinning disk instantly fractured into several sections, which rocketed outward in multiple directions at incomprehensible speed.

“It goes without saying that if any of the turbine fragments had entered the passenger cabin, there would have been injuries, if not fatalities, even if the plane later landed safely.”

They got lucky.

The fact that it's assumed to have infinite energy when planning for this is crazy. Engineers saying yeah there is no material in the world that can stop it.

Infinite is a bit poetic there. For large enough n, n inches of (say) armor steel could stop it. But that suggestion got Sales talking about "commercially viability", and Finance sided with them, and we all know how that story goes.

It's not (purely) the sales and finance people, planes need to be able to get off the ground to be useful at all. I doubt even an A380 would still be able to lift off if you strap a dozen inches of armor plating to every part of the plane that could possibly be reached by shrapnel.

You'd put the containment armor directly around the engine. It wouldn't take dozens of inches. Whether you use Newton's rule of thumb (penetration ends when the mass of armor displaced equals the mass of the shell that hit it) or more modern rules - a failing turbine disk is neither a precision-engineered anti-armor weapon, nor a heavy battleship shell.

I'd bet the plane would still be able to take off, if it made it that far. But it never would - with the extra weight guaranteeing that no airline would buy it, Finance would never sign off on funding to even complete the design work.

(Yes, the reality behind my quips about Sales and Finance is more complex:)

Full containment is probably possible, but it would render the aircraft nearly useless, not just economically uncompetitive.

A 70kg chunk of nickel alloy is very dense, and at 800m/s , that’s 20 MJ. Thats about the same as 5kg of TNT, but focused on a small area.

For comparison, a modern 120mm artillery shell fired from a tank or similar field artillery has 10-20 MJ of kinetic energy.

A couple thousand kg of armour could likely contain that kind of energy, but it would also translate that energy to the engine mounts, wing, spar, and fuselage if it was not going to become a projectile itself. You could easily stop a fragment and critically damage a spar attachment, rupture a wing fuel tank, or end up with translational shock damage to any number of critical systems and structures. You’d probably have to double or triple your engineering margins of the entire aircraft, probably doubling the weight of the airframe.

So yeah, possible probably. But not on an aircraft designed to carry passengers or freight. It would be the biggest 6 seat aircraft ever built lol.

Interestingly, if you work backwards from that cargo capacity, you can arrive at a much smaller aircraft that could possibly feature full containment, because engine diameter and energy goes way down and containment starts to look much more practical. I would not be surprised if containment could be achieved in the small biz jet scale, with just significant sacrifices in cargo or range.

OTOH even the A10 warthog does not have full disk containment, and it’s probably the strongest candidate ever fielded for such a project… so, ymmv.

> A 70kg chunk of nickel alloy is very dense, and at 800m/s , that’s 20 MJ.

> For comparison, a modern 120mm artillery shell fired from a tank or similar ...

Yes. But when the goal is kinetic penetration of modern armour, they don't use artillery shells. Instead -

https://en.wikipedia.org/wiki/Armour-piercing_fin-stabilized...

- which bear no resemblance to a failed turbine disk. And have about twice the muzzle velocity, which makes huge difference in penetrating a modern armour systems. Ditto their far greater density, vs. nickel alloy.

OTOH, your "A couple thousand kg of armour could likely contain..." seems too charitable. High-bypass jet engines are big. Beyond that, I'm more optimistic about mitigating shocks and such.

A critical parameter here - https://en.wikipedia.org/wiki/A380#A380F The freight variant was to have a 6,400 mile range carrying 150t of payload. So we could squander 25t per engine on disk containment and shock mitigation, and still have a 50t payload. Or a 3-digit passenger headcount.

Bottom line, there is no real-world difference between your "nearly useless" and my "disastrously uneconomical". Neither one will be designed, let alone built.

For the A10, I'd say military priorities would always block full disk containment - because every pound used for that is a pound they don't have for more weapons or munitions or range or armour or fuel or runway length or electronics or whatever. And over most of the 360 which disk fragments could exit a warhog's engines, there's nothing critical which they could hit.

Good counterpoints all, I think. I’m impressed that we could write off 50t and still have an airplane. That seemed so preposterous that I did not consider it, honestly. Turns out big stuff is big.

Now we can go to the running of the bulls with our respective napkin backs, and see who attracts the bull!

>>I’m impressed that we could write off 50t and still have an airplane.

100t actually - 25t per engine, of which there are four. And the plane could still carry 50 tonnes of cargo/people.

That’s wacky, for lack of a better word. It’s like a whole flying apartment building.

> Now we can go to the running of the bulls with ...

Thanks, but I think I'll be running with the chickens. ;)

That sounds interesting. Not necessarily better. Who was it that said quantity has a quality all its own?

Yes but on the warthog the discs are very far away from the pilot. Or pretty much anything else that matters. The fuselage can take a lot of a beating.

https://en.wikipedia.org/wiki/File:A10Thunderbolt2_990422-F-...

A failed disk from engine #1 is just a few feet from engine #2.

That's a good point. Somehow I didn't think of that.

I know we all like to hand wring about stuff like this but realistically it's "infinite" to prevent the same genius textbook engineers who cooked up the VMC procedure and the CMM procedure from cooking up equally half baked "with a plate of this thickness at this angle it will remove X energy from anything that penetrates it therefore we are good to put some super critical thing right in it's path" solution.

Maybe not stop it, but perhaps they can design it so that if it happens it's more likely to be directed away from the fuselage? Like with armor that does not block or absorb but deflect, which requires less energy. Perhaps even a small reactive armor strip? (dangerous as though that would be on a plane).

But then again I'm sure aviation engineers would have thought of this and rejected it for good reason.

> Jet fuel cannot be stored in the wings in this zone.

TFA describes a fuel leak resulting from a piece of the turbine disk passing through a fuel tank located within the wing:

"When the fragments of the IP turbine disk passed through the wing and belly of the A380, they caused considerable secondary damage along the way, not only to the №2 engine but also to the left wing fuel tank, which sprang a leak"

It also includes a photo of the inside of the damaged wing tank.

Was the requirement you describe added after this incident?

Seems close to impossible for existing designs. The wings are commonly almost entirely full of fuel at max fuel, and I doubt that taking the capacity hit is an option even in newly designed planes.

As it sounds like you have some knowledge about these engines, from the article:

“These [combustion] gases spin the [turbine] disk, which is attached by a drive arm to its associated drive shaft. The shaft then transfers the turbine’s rotational energy forward to the corresponding compressor at the front of the engine.”

I'm probably misunderstanding, but it sounds like the disk is attached to the shaft by a single so-called “drive arm”, which could be imagined to span maybe 10° on the circumference. I would have thought there'd be at least 3 such arms at 120° intervals, if not a continuous full 360° connection, which then I wouldn't call an arm. Could you clarify this point?

What's a turbine disc?

It’s explained in the article

Reply on news.ycombinator.com