Murphism and Airplanes

April 3, 2013 by Bruce Landsberg

pilotAny smart pilot or organization manager studies Murphy’s Law: life’s wonderful reality that if something can go wrong, it will! There are dozens if not hundreds of corollaries that remind us that complex airplanes are veritable breeding grounds for Murphy mischief.

In the March Issue of AOPA Pilot, I wrote about Air France 447, the Airbus 330 that went down over the south Atlantic after all three pitot tubes froze up. This caused the air data computer to leave the premises, and the automated systems, which the Airbus has raised to an art form, began cascading failures. The crew lost total situational awareness and the junior first officer, after some gyrations, pulled the big ‘bus into a stall and held it there for about three minutes for a final plunge from about 35,000 feet. Lots to learn from this one!

The saga of automation acting out continues. Here are two more incidents involving the Airbus where, once again, frozen pitot tubes discombobulated the autopilot, auto-throttle, and flight director. The crew managed to control the aircraft, but had to divert to an alternate airport because they couldn’t remain in reduced vertical separation maneuvering (RVSM) airspace and fly manually to the tolerance required. Guess keeping the tubes warm and dry is problematic.

The other incident involved a non-stabilized approach into Paris with another A340. This one had more help from the crew, with possible ATC assistance, in getting to the arrival gate not only high, but fast for a Cat III approach! The glideslope eluded them most of the way down. The aircraft at some point decided to go around, and a difference of opinion ensued between the crew and the ‘bus. It’s too long a story to relate here, but the crew and the automation never got on the same page, and let’s just say the outcome of the maneuver was seriously in doubt.

Before you think me anti-Airbus, recall that the Boeing Dreamliner, in pushing the performance envelope, has had a few miscues as well. Teething pains are one thing though, and ongoing incidents are something else—where one stops and the other begins is for your conjecture.

The principle concerning multifunctional devices states that the fewer functions any device is required to perform, the more perfectly it can perform those functions. Put more stuff on an aircraft, car, computer, home theater, or anything, and the reality is that it will break more frequently. DC-3s and Cessna 172s are prime examples of long-lasting simplicity.

Regarding automation for light aircraft, a certain amount is really good, and single-pilot flights in IMC without a simple and solid autopilot is much more work and distraction than it needs to be. However, remember Johnson’s First Law: When any mechanical contrivance fails, it will do so at the most inconvenient time.

My question is: How does it know?

Safe pilots are always learning, and the goal of the Air Safety Institute is to ensure pilots have a wealth of information to keep them flying safely. Our education programs are funded through donations from pilots dedicated to forwarding that mission. Show your support by donating to the AOPA Foundation today. 

 

Bruce Landsberg
President, AOPA Foundation

ASI Online Safety Courses  |  ASI Safety Quiz  |  Support the AOPA Foundation

4 Responses to “Murphism and Airplanes”

  1. Jerry Kaidor Says:

    So what is it with these pitot tubes? Airplanes have been flying at these altitudes for decades without falling out of the sky. Are they using some kind of newer fancier pitot tubes?

    I think authorities should withdraw certification for this deathtrap until Airbus fixes it.

  2. Joe Sass Says:

    Airbus had released a new design for the pitot tubes. If I remember correctly, the plane that went down was scheduled to have the updated pitots installed. Unfortunately it never got them.

  3. Joe St.Clair Says:

    Bruce Landsberg,
    President, AOPA Foundation

    Bruce:
    The Air France 447 accident of the Airbus 330 is very discomforting, to say the least! I have no experience in Airbus Aircraft, and have only read and heard of how their control inputs work, but I cannot understand the crews’ failures. I have spent thousands of hours behind EFIS run by AHRS black boxes, and the equipment is excellent, and certainly would be working fine under the circumstances of this crash. Further, there is no indication that the static system was not working, and also the angle of attack information, as descriptions via the black boxes had altitude and angle of attack information intact. Further, there would be GPS derived altitude, though not necessarily considered exact, would back up altitude awareness. How then can two, then three airline pilots, while looking at dual functioning (and presumably a standby emergency ADI) attitude and HSI presentations run by dual (perhaps 3 or more in an Airbus) AHRS, and altitude information, not simply maintain straight and level flight? Perhaps they would need to slightly reduce power % to avoid Mach limitation versus stall. And of course, why did this aircraft crew, unlike other aircraft crews in the area, fail to avoid this storm that caused these extreme conditions, with presumably operative weather radar?

    I, like I assume you, have spent many hours as a passenger over the Atlantic and Pacific, and many of those in the Airbus 330. It is absolutely terrifying to imagine sitting behind the likes of this crew in a Scary Bus. Despite your comments regarding the Dreamliner, I cannot but recite the old Mantra, “If it ain’t a Boeing, I’m not going”! I was thinking just that on my last trip to Europe last July, pleased that I had United crews flying a 777 over, and a 747-400 back.

    I would appreciate further comments from you addressing my comments.

    Thanks for all you do for GA,

    Sincerely,

    Joe St.Clair
    Captain Retired, SkyWest Airlines (Brasilia, CRT/EFIS)
    joej1@sbcglobal.net
    AOPA # 00406171

  4. Dan Winkelman Says:

    Let me preface this comment: I’m an aerospace systems engineer. I happen to work for a subsidiary of Boeing, though we have absolutely zilch to do with commercial air transports. I have held the views below since long before starting work for any Boeing subsidiary.

    The A330 and other Airbus incidents are evidence to back up my view that Airbus and Boeing have two dramatically different control philosophies, and Airbus is wrong. Airbus builds an aviation system that is designed to fly itself, and the “pilots” are actually systems operators. I believe that some of the airlines who rely on Airbus equipment agree with the Airbus philosophy, and they are no longer teaching people how to, first and foremost, be good PILOTS. As Mr. St. Clair pointed out, there was nothing wrong with the AHRS, there was nothing wrong with GPS. Pick a safe pitch attitude, pick a safe throttle setting that should maintain a reasonable airspeed between mach limit and stall, then pick a safe altitude outside of RVSM airspace (since you presumably cannot hand-fly well enough to be in RVSM space). Notify ATC, and make your way (carefully) to the nearest diversion airport.

    Instead, we ended up with three systems operators (not pilots) who could not figure out how to get the system to behave, and as a result many people died.

    Boeing, on the other hand, builds *airplanes*, not airborne systems. They give the pilot control of the airplane. The flight controls, displays, and automated assistants of a Boeing product have a fundamentally different philosophy: ease the pilot’s workload… but he or she is still the PILOT.

    Regardless, all this trouble simply reinforces my prejudice against computerized control systems… whether it’s “drive-by-wire” throttle in a car or fully automated flight controls in an airplane. Make it simple, make it robust, and train the operator how to fly or drive properly. I would like to see more drivers and pilots in the world, and fewer system operators at the controls. System operators do not have the foggiest idea what to do when their electronic nannies fail, and people are dying as a result.

Leave a Reply

*