Skip to main content

View Diary: Flight Engineer - A Dying Breed (106 comments)

Comment Preferences

  •  Great reading about the 727 (7+ / 0-)

    Back a few decades ago I was a frequent flyer on Pacific Southwest Airlines. At the time I believe they flew nothing but 727s and I do recall that there were three up front in those days. Never really understood what the Flight Engineer did, but I can recall seeing that panel when I was boarding the aircraft and got a peek inside.

    As a former software engineer, I must say that I suspect that at this point nobody onboard really understands what makes these planes fly. I suspect that most of these systems are more or less fully automated under normal circumstances and I'm not that confident that the pilot and copilot really know how they work.

    The only trouble with retirement is...I never get a day off!

    by Mr Robert on Sun Jan 20, 2013 at 12:33:23 PM PST

    •  We have a basic knowledge of the systems (8+ / 0-)

      I know in principle how they work and I know enough to troubleshoot them when they break.

      There's only so much I can do in flight anyway. That's why so much redundancy is built into the aircraft. Every system has a backup and critical systems have at least two backups.

      For example, my current airplane the 757 has quadruple redundancy on the brakes:

      Normal, Alternate, Reserve and finally the residual pressure in the hydraulic accumulators will give you at least one good stop.

      If the pilot's good, see, I mean if he's reeeally sharp, he can barrel that baby in so low... oh you oughta see it sometime. It's a sight. A big plane like a '52... varrrooom! Its jet exhaust... frying chickens in the barnyard!

      by Major Kong on Sun Jan 20, 2013 at 12:52:17 PM PST

      [ Parent ]

    •  Not sure anyone understands (3+ / 0-)
      Recommended by:
      Mr Robert, KenBee, ER Doc

      the lately grounded Boeing 787. Maybe the quest for fully automated aircraft is a little ahead of the design curve.

    •  Old saying: (6+ / 0-)

      Pilots don't need to know how to build the airplane, only how to make it fly. The old days of knowing every component in a system and how it functions within the system are over. That's a good thing. The dinosaurs like the 727 needed to be understood at a deeper level than the new ones. There were some malfunctions that were tricky to diagnose. The symptoms between different problems could be similar with only a different light or gauge reading being the difference to fixing a problem or fixing the wrong problem and making the situation worse. It required knowledge, good procedures, and attention to detail. The last one could be extremely difficult in practice. I learned exactly how tired I could be and still function on the classic 747. Paying attention isn't cheap at the end of a 12 hour flight when you've been up 20+ hours at the end of a 12 day trip with six Pacific crossings.

      All of that has changed with technology integrated into the latest generations of airliners. Even Hong's 757 is thirty year old technology. It is a hybrid of "steam gauge" tech and computer monitored/controlled systems and displays. It is a monumental improvement over the 727 generation. The 777/787 and A320/330/340/380 generations are an even bigger leap forward.

      The latest jets have systems that monitor the primary systems like hydraulics, engines, fuel, A/C, etc. The systems themselves reconfigure for malfunction, identify the problem for the crew, and provide checklists to back up the automatic changes and/or direct crew action that the manufacturer thought was critical enough to require a human crew member's interaction. An example of this would be actually shutting down an engine that has had a problem that didn't make it stop running. The bottom line is that the crew's job has changed from monitoring system operation for proper function through indicators and gauges and diagnosing and correcting malfunctions to one of monitoring a system warning display and reacting to system generated messages. What has not changed is that the crew must still be aware of proper system operations. That is not an easy task for humans whether in an old jet or  the newest ones.

      The short answer to your concern about whether the crew knows how the system work is that they don't need to know. But they do need to be trained to recognize problems, be able to follow the procedures to correct or find a way to live with the malfunction, and be knowledgable enough of the system to know when things don't make sense or are the result of multiple or compound malfunctions. My airline career began with a bunch of trivia mixed with critical information about complex system where I didn't always have enough information to confidently solve problems and ended by having the skill and training to know how to read a display and follow a procedure. I

      t ended the way I prefer. I always approached flying with the attitude of help me identify what isn't right and tell me how to fix it or live with it until I can get on the ground. The new systems do just that and leave me free to use 37 years of experience and judgment to fill in the blanks.

      Time makes more converts than reason. Thomas Paine, Common Sense

      by VTCC73 on Sun Jan 20, 2013 at 02:58:08 PM PST

      [ Parent ]

      •  The rare case to watch out for (1+ / 0-)
        Recommended by:
        Mr Robert

        is then the unanticipated circumstance, what the designers who created these systems didn't think of. The unknown unknowns. Still useful to know as much technical detail as possible about what's going on behind the panels (hardware and software) to cover that extremely unusual circumstance. The low-hanging fruit of safety improvements are already done; now only the hardest things are left.

        Government and laws are the agreement we all make to secure everyone's freedom.

        by Simplify on Sun Jan 20, 2013 at 08:01:15 PM PST

        [ Parent ]

        •  I have to partially disagree. (1+ / 0-)
          Recommended by:
          Simplify

          I agree in that a pilot as to understand the general operation of the systems and the overall aircraft thought of as the whole system in itself. She has to know what the malfunction is and how it impacts the operation. But the modern aircraft already do most of that work and graphically display what is working and what is not. The warning system even tells the crew what they need to do to solve a problem and each airline backs that up with tailored procedures and manuals to further ensure successfully overcoming a problem.

          Where I disagree is that the modern system is so integrated that it is impossible to understand how each subsystem works to the extent we understood them in the days before the 757. The new generation of aircraft would require a pilot to not just understand a system but the way in which the computer monitoring and control works as well as what other system might be involved. I have to acknowledge that my airline did it right by deciding to forgo in depth knowledge that had the usefulness of trivia and focus on important things. The operating limits for example became is it green, good, yellow, maybe not so, or red, bad instead of a lengthy list of numbers that you no longer have the ability to monitor anyway. Further they were able to successfully transfer that philosophy to our other fleets including old generation airplanes. We went to far in a couple cases but that got corrected.

          I understand that most people are still concerned for their safety. Your ATM example shows that concern but I don't think you appreciate the level of redundancy in the designs, the maintenance programs, or extensive training and competency testing that goes into an airline operation. The point I'd like to make is that the fatal single point of failure is pretty much a thing of the past. There has to be multiple failures in the people, machine, and/or operation for someone to get hurt. A look at the safety record with appreciation for the amount of flying done speaks to a system that is not flawless but one that requires lots of people to have a very bad day for anyone to get hurt.

          Time makes more converts than reason. Thomas Paine, Common Sense

          by VTCC73 on Mon Jan 21, 2013 at 12:59:47 AM PST

          [ Parent ]

      •  Sorry, but as a retired software engineer (0+ / 0-)

        I'm not reassured because there is no way to insure that these systems are 100 percent free of bugs. And, I'm not quite ready to believe that manually overriding these systems will work in all circumstances.

        It's one thing when a bank's ATMs fail and people can't withdraw funds. It's quite another when people loose their lives.

        There are thousands of potential failure modes in these complex aircraft and there is no way to test for let alone train pilots to handle every situation.

        The only trouble with retirement is...I never get a day off!

        by Mr Robert on Sun Jan 20, 2013 at 11:32:32 PM PST

        [ Parent ]

        •  I might agree with you if (2+ / 0-)
          Recommended by:
          JimWilson, Mr Robert

          all potential failures were fatal to the system. A major design criteria is avoiding single point failures. Multiple redundancies are a long preferred method. Two engines, three hydraulic systems with wind powered ram air a turbine, four generators and a hydraulic backup generator operating multiple AC and DC busses, and seven flight control computers are but a few of the features of the A330 I flew for eight years before I retired. Remember that this is an aircraft that has computer controls for virtually every function from engine fuel controls to the passenger seat lighting controls. The more critical systems are built most robustly with multiple backups if multiple degraded modes all fail.

          Since you are interested from a software perspective I'll say a little more about the flight control computers. Seven of them are required for a fully functional system. Each is capable of providing sufficient control capability to successfully recover the aircraft. It isn't pretty depending on which one remains but there is full expectation of a successful recover of the aircraft on just one. Or none.

          What you say? None? Yeah, none. It is possible though very difficult to land with nothing but the backup rudder and engine power to land the airplane. It would be extremely difficult and there is no guarantee of success but that is the level of design redundancy.

          The most likely failures are various sensors or parts of a computer. The system is designed, and has always done so in my experience and knowledge, to reconfigure itself to operate with a failed or erroneous input, revert to a degraded mode, or disconnect the device. The pilots can also manually disconnect a box if needed. The probability of losing enough to make recovery not likely or impossible is extremely low. Not zero but low. I bet it is less than the probability of getting hit by lightning walking around in your yard.

          The greater threat is the operator not understanding the system or misunderstanding the ramifications of what he is doing. The A320/330/340/380 series of aircraft has never lost an aircraft due to system failure. Every single hull loss has been due to the pilots, test pilots too, not understanding the mode the aircraft was operating in or were unaware they were working against the system. Every aircraft was lost because the crew flew it into the earth. The biggest threats are the crew, the humans, just like it has always been since the first powered flight. The complex systems give us more information in a format humans can use easily and make it more likely they won't do something extremely stupid, always an option, always.

          The best advance I've seen in aviation safety for pilots is the move to threat and error management. My flying career began in the military where there was one way to do things. To err is human, to forgive is divine, neither of which is accepted policy. Thirty five years into that career the experts, aviators in this case, decided to acknowledge that crews make errors. Once you can accept the certainty of error you can move to analyzing the factors that lead to error and prepare yourself for when, where, and how they occur. Awareness of a threat makes it is much easier to prevent. This also creates an environment where a mistake can be caught and corrected sooner. The idea also encourages the crew to include others to participate which further increases the likelihood of preventing mistakes or catching them before there is a big problem. It is overall an exciting change that I think will lead to even safer air travel.

          Time makes more converts than reason. Thomas Paine, Common Sense

          by VTCC73 on Mon Jan 21, 2013 at 12:33:01 AM PST

          [ Parent ]

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site