Software Testing Blog

As many of you may remember, last month Qantas Flight 32 had a bit of a problem. The inboard left engine (#2) spontaneously exploded on takeoff – seriously damaging what was a brand new Airbus A380. This week, the Australian Transportation Safety Board released their preliminary assessment of the failure, pinning blame on a manufacturing defect in the Rolls Royce Trent 900 engine.

If you have a few minutes, check out this summary of the ATSB report. It’s fascinating reading (plus it includes some cool pictures), and you will learn a lot about how robust modern aircraft and piloting procedures really are. Take these details, for example:

• The A380 computer systems kept excellent track of the problems as they were unfolding, giving the pilots good information about the state of the aircraft
• Since the plane remained flyable, the pilots were able to spend 30 minutes following checklists that helped them make sure the plane could safely land
• This was the first time anything like this has happened to an A380, and both the pilots’ training and the plane’s engineering came together to create a safe outcome

With all that said, there were also a lot of close calls. Here are a few examples:

The engine explosion damaged one of the plane’s hydraulic systems. Fortunately, modern aircraft have redundant hydraulic systems, and after the crash of United Airlines 232, aircraft makers go to a lot of effort to route the systems in very different locations to prevent them from all being compromised.

The plane still held a lot of fuel, so it had to make a very heavy landing. It stopped only 150 meters before the end of the runway, using both the onboard brakes and the thrust reverser on Engine #3 to slow it down. Fun fact – the A380 only has two thrust reversers (located on the inboard engines), but Airbus was so confident that the plane’s brakes were sufficient for stopping that they nearly didn’t offer thrust reversers at all on the A380. Had Engine #3 not been able to reverse thrust, the plane would have overshot the runway.

After stopping, the brakes on the plane had superheated to a whopping 900 degrees C. Combined with a fuel leak from the left wing, there was a very real risk of a sudden fire starting underneath the plane. After shutting down the engines, the aircrew lost precious time trying to find the right radio frequency to communicate with the fire services. Fortunately, the crew was able to contact them in time to get the fire services to start foaming the runway before the fuel reached the brakes. (Want to know what happens when a plane’s brakes overheat? Check out this video of the emergency braking test for the Boeing 777.)

Did I mention that they shutdown the plane’s engines after landing? Well, about that. After the pilots kicked back and waited for the fire crews to do their thing, they got a curious radio call. The fire crews wanted to know why they hadn’t shutdown Engine #1 (the engine next to #2, which had exploded in flight). The pilots reported back that they had, and subsequently went through another shutdown procedure. Engine #1 still ran, and resisted all attempts at making it stop, including switching off the electric breakers that powered the thing. Finally, Qantas gave the fire crew permission to foam the running engine:

If you’ve ever wondered whether firing a massive water cannon at a jet engine will stop it, the answer is “yes.” Subsequent investigation revealed that the damage from Engine #2 had severed the control lines to Engine #1, making it impossible to turn if off.

Congratulations to the crew of Qantas 32 for great work, and to Airbus for building a solid plane. Of course, all this was possible because of the extensive planning, testing, and failure analysis that goes into building a new airplane. Airbus’s computer systems were able to keep the pilots and crew organized and help them solve the problems the best way possible. Good engineering all around.