Thursday, September 02, 2010

Uncontained Failure

A Qantas Airlines Boeing 747 jet climbing out of San Francisco on Monday experienced an uncontained failure of the number four engine. That means that the outside right-side engine failed so spectacularly that pieces of it penetrated the cowling, the case around the engine. The linked article has photos, and even though they were taken in the dark you can still see some of the turbine fan blades through the hole.

The pilots of course couldn't see the hole in the engine, only feel the vibrations through the airframe and see whatever indications the engine instruments reported. They certainly wouldn't have been normal. They shut the engine down, dumped fuel to get down to landing weight (you can see a couple of orbits on the Flight Aware track in the Flightglobal piece linked above), and returned for landing at San Francisco to investigate. Passengers saw flames and sparks coming from the engine; one of them took a video of the sparks streaking by, even after the engine was secured.

It's always interesting to see how pilots handle PAs in situations like this. (You can hear a little bit of the captain's announcement in the above video). Some companies give you a script, but the nature of emergencies is that they don't always suit the script. This captain emphasized that the crew train for this, and then I think he tried to throw in a bit of humour by saying, "in the simulator." He sounds calm, and I hope that calmed people down. Passengers tend to think they're all going to die whenever anything happens. The problem with reassuring people is that they always think you're lying to them, or at least that there's a reasonable chance of whatever you are saying won't happen, happening. My approach is to be honest, but a little bit vague about the problem and then distract them from worrying about imminent death by reassuring them that we will find them another airplane, and make sure they get to their destination. They can then worry about missed connections and lost baggage, which is not life threatening and returns them to the comfort of the habitual concerns of the air traveller.

This could have been worse if the containment failure had been on the left side of the cowling, (or the right side of an engine on the left of the airplane) because debris could have damaged the other engine or penetrated the cabin. Components leaving a turbine engine have a lot of energy. Those things rotate at over 10000 rpm.

LiveATC recording of the event from the perspective of the radio communications is available buried in this file. The Qantas pilots report their difficulties 18:20 into the recording. Eric M. has edited it down to the eighteen minutes directly concerned with the incident flight but you can only hear that version if you register (for free) at LiveATC.net. There's some confusion on the part of ATC, as Australia and the US seem to have different standards for the information an aircraft with an emergency must convey. Although the Qantas crew do deny they have an emergency. That means they don't believe they need any kind of priority to ensure safety.

An interesting detail about of the B747 is that it is equipped with a hardpoint for carrying a spare engine. Not connected up to fuel and power, just sitting there ready to be installed, like a spare tire on the back of your SUV. They don't carry one around all the time because they add a lot of drag and are very rarely needed. It's just a provision for transporting them. Another Qantas B747 will bring a spare in today for the one on the ground at SFO.

That's today's blog entry early, although I'll edit it to make it look like it appeared at 00Z as usual. Tomorrow's will be at the usual time.

20 comments:

D.B. said...

I was a passenger of a BA 747 from LHR to YYZ around 1980, as always in the window seat. About half way, I felt the aircraft yaw to the left, and noticed that I couldn't see the exhaust from #4.

Once on the ground at YYZ, the pilot finally told us that they had an engine failure.....

Of course, the time I was on a TWA L1011 climbing out of Philadelphia at night, going to LHR, and it had an uncontained #1 engine failure (just like this one) was more interesting. We dumped fuel and landed again, got shuttled to JFK, and stuffed into a B747, arriving rather late.

Both of those were over a span of about 2 years, which also included the Air Canada B747 that blew 4 tires landing in Montreal, and an Allegheny BAC 1-11 with brake failure. From then on, the next 3 million miles have been uneventful......

majroj said...

I hear the upcoming B747.1 will have a hardpoint to carry spare passengers under the other wing to aerodynamically balance the "spare engine" and make some money.

Frank Van Haste said...

Dear Trix:

From the position of the hole in the cowling, I'll bet on a first stage turbine disk rupture. Such things are usually tracked back to a latent manufacturing defect in the disk forging. Let's hope they have enough scrap metal left to do the forensic metallurgy.

Regards,

Frank

Rob42 said...

I listened to some of this recording played on local news radio this morning in Sydney. I had to chuckle at the pilot politely telling ATC that dispatch would get more information from them when they were good and ready. Obviously not responding to their datalink messages. Aviate, navigate, communicate...and communicate with dispatch last of all...

Brent said...

Even after engine shutdown and the end of all those sparks, I bet it took some guts turn that fuel dump switch.

mattheww50 said...

Fuel dump is actually pretty safe. The 'outlet' is actually almost at the wingtip. It is actually quite impressive to see 200,000+ pounds go overboard in a matter of minutes.

The secondary issue is that generally dump occurs at significant altitude. At the typical temperature at 25,000 feet and very low partial pressure of oxygen at altitude, the fuel is almost impossible to ignite anway.

Traveller said...

I saw the report on local news here in Melbourne.

Thanks for the detailed post. Yep, its a good thing the hole is away from the other engine and the fuselage.

Sarah said...

The ATC audio is fascinating. Fuel dump may have been routine, but they had to do it for 30 minutes in a hold ... with fireworks.

Aviatrix said...

I agree, Sarah. They must have been busy too, as they missed the waypoint the first time. I take it you figured out the LiveATC archive format? When you get to the end of the recording change the URL by 30 minutes and reload.

Sarah said...

LiveATC archive format? No, but thanks for the tip for future reference. I just listened to the 18minute excerpt you linked to above; it probably has every syllable uttered by Quantas 74.

After dumping they still had more than 20 tons of fuel. This sounds like a lot, but if I did my math right, full fuel is almost 200 tons, and 20 tons is about an hour remaining.

mattheww50 said...

MLW is 630,000 pounds, MZFW is 542,500 pounds. It is a safe bet that they were at or very close to MGTOW (875,000 pounds) on departure. Anyway getting the aircraft down to MLW wil leave a minimum of 77,500 pounds of fuel aboard (and probably a lot more than that). If you fill the tanks on a 747-400 you are going have be no more than 493,000 pounds for OEW,Pax,bags, freight etc.


They probably landed with more like 50 tons of fuel in the tanks, and that is closer to 4 hours worth at cruise (initial cruise burns about 28,000 pounds per hour, at MLW that figure is under 20,000 pounds per hour.

Sarah said...

Thanks for the hard numbers, Matthew50. Sounds like you know what you're talking about! However, the crew stated on the ATC recordings they had "20-something tons of fuel", I don't recall the exact number now.

Anyway, well done Quantas.

Joël Morin said...

I don't detect any confusion on the part of Oakland Center...
Aircraft will or will not declare emergencies after a system failure of some sort (even lost engines)and it's standard protocol to ask if the pilot's declaring an emergency when it's unstated. Mind you we can also provide emergency handling when ATC deems it to be required, requested or not.
This sounds like a properly handled "routine" emergency.
By the way, I never cease to be impressed by how long it takes for a heavy to dump fuel. I wonder what the dump rates are...

Aviatrix said...

I thought the controller didn't seem to understand when the crew tries more than once to say that they have no dangerous goods on board. Perhaps it's me not understanding the controller's response.

I certainly didn't mean that the controllers were confused in general.

Joël Morin said...

I've been listening to the raw feed and switched over to NORCAL.
ZOA was fine but you're right, the Arrival controller didn't get the dangerous cargo comment (wonder why.. accent??(she did miss another comment later on) it's a standard statement)
I heard 72 Tonnes of fuel remaining.
Arrival did upgrade the PAN PAN PAN to an emergency but QFA74 didn't object.
I'm surprised it took her so long to determine the distance to touchdown, but then again we have better tools.
I'm also surprised she didn't state the landing runway on initial contact.
Interesting how media almost never pick up on Canadian emergencies.
Cheers.

Aviatrix said...

I've seen some pretty minor ones make the papers. Even one I was on board where we had smoke in the cabin during descent (turned out we had overflown a forest fire and it was just smoke that had come in through the bleed air!) was duly noted with the boilerplate, "None of the 95 passengers was injured in the landing."

Sarah said...

And you're right matthew50 and Joël. My mistake. On review of the audio, they said "231 souls on board and 72 tons fuel remaining".

Robin Capper said...

Often hear "the pilots couldn't see" and wonder why aircraft don't have a few external web cam type cameras covering control surfaces, engines, undergear and airframe?

Aviatrix said...

Robin Capper: I guess the answer to that is it's more complicated than it sounds.

Drag is a huge concern for aircraft. The faster they go, the more effort goes into eliminating drag and something that stuck out enough to have a good vantage point, and built sturdily enough not to be obliterated by air pressure or flight through rain would likely produce unacceptable drag for its utility. It would have to be electrically heated to prevent ice from building up on it, too. Everything installed on or in an aircraft has to be certified for aviation, so it would have to be designed in, or be a very expensive upgrade.

There is a little bit of such technology used. There's a rear view camera on some large airliners for backing out of the gate area. I have a mirror on my left engine nacelle that shows me what my nosewheel is doing.

Ed said...

Saw a KLM 747 taxi in at Schiphol in the early 80s with five engines before I knew about the extra hard point. Very confusing.