Bjorn’s Corner: Flying revisited

Bjorn Fehrm has an aeronautical engineering background with patented devices on several aircraft.

By Bjorn Fehrm

April 09, 2015; I’m up for a challenge in the next weeks: I’ve been invited to fly an airliner. Having flown 14 aircraft types before it shouldn’t be so big news if it wasn’t for none of those types being close to the aircraft I will fly now, a modern civil airliner.

The previous types were military trainers, fighters and later civil sports and business aircraft. They are all more or less the same. Flying is like cycling and driving a car: it is something one learns and then doesn’t unlearn. So the flying part should be no problem.

It is not what makes me undertake weeks of preparations. It is that other thing, the aircraft’s computerized soul, that expects to be operated in a certain way. The buttons should be pressed and handles moved in the right order or the aircraft will tell me it doesn’t understand what I want.

So now I am reading through thousands of pages and flying civil airliner procedures day and night. We will dwell on how and why in a couple of Bjorn’s Corners.

The differences in flying

In the beginning of flying, it is all the same. You train in the theoretics of aerodynamics, aircraft systems, weather and flying traffic rules. Then on to the aircraft for hands on flying. The civil training aircraft are built to be straight forward to fly, robust and economical. If they are military trainers, you can add safe to take to and beyond normal flying limits to that. Why?

Military flying

Military flying, especially for fighters, is about flying on the limits. And to learn what to do when you have passed them. Therefore a lot of military basic training is about flying up to and beyond stall, with or without G-load. If you do it with G-load, the aircraft behaves differently, many divert violently in a quick-roll type stall and you are suddenly head down instead of up in a split second.

This is normal. You train it time and time again so that when it happens later in a dog-fight with a real fighter, it is no big deal–you grab the horizon (real or artificial if you stalled in a cloud) and get out of it. Push a little harder at low speed and you can end up in a spin or deep stall. You train for that as well.

Civil flying

Civil flying is different. If we start with sports or business flying, which is what I have done so far (after my military training), it was disappointing to see how shallow the stall training was. Straight stalls, no G’s, no crossed rudders (spin training) and you only went to beginning of stall and backed out immediately. This is no good base for someone that gets all crossed up in, e.g., a real stall in a cloud during bad weather flying.

The things that gets stressed in civil flying, especially when flying in accordance with Instrument Flight Rules, IFR, are the mechanics of flying. Procedures for take-off, climb, cruise, descent and landing. In good weather and bad.

Business flying is like airliner flying in that it follows the same IFR rules, airways and traffic control functions. The difference is that your aircraft is smaller, lighter and has few passengers. It is also less complex from a handling point of view, the check lists are for a simpler environment both system and regulatory wise. This would fit for all but the largest types, these are more like the aircraft we come to next.

Airline flying

Airline flying used to be like business aircraft flying, aircraft with rather simple system with a rather simple autopilot. The autopilot had control of all the aircraft’s axis but no auto-thrust; speeds were kept by the pilots. But the environment has changed. Civil airliner flying today is a highly sophisticated transportation business, with demands for precision and predictability in the finest detail.

Air traffic has and will grow at ever increasing rates. The figure shows a picture of the world’s air traffic during day time. It is of course highly zoomed but shows what we want: civil air traffic is mass transportation of people and goods in the most efficient and safe way possible.

Flightradar 24

Picture of the main areas of civil air traffic congestion. Source: Flightradar24.

To do that, everyone has to fly to the same rules and procedure, to the same speeds and with the same precision. Doing that hour in, hour out with a two pilot flight crew requires automation on a different level from other sorts of flying.

Initially the flight engineer was part of a team of three that shared the workload of achieving this kind of precision and safety. Then he and a lot of the original pilot tasks got moved to a computerized flight management system, the FMS. Today this is the heart of the flying airliner. It gets filled with key performance, route and weather data for the trip and then computes all the speeds, altitudes, routings and more…., ready to control the flight.

If your FMS is correctly programmed, you are fine to go, if not you will be asked to fly yourself. This is regardless if the aircraft is controlled in a conventional way or controlled by fly-by-wire. The FMS tells the flight director how he wants the aircraft to be flown and the flight director tells the autopilot what he wants to have done. The autopilot in turn tells the flight control system to move control surfaces so the aircraft flies the desired route. In case of engaged autothrust, it controls the engines for desired speeds in all this.

I list the chain of events because it gives a glimpse of the logical complexity of a modern civil airliner. Add now that each of these systems have different modes and if they get degraded back-up modes. It is complex. The question is if it is not too complex. This was all designed to help the pilots with the work load, the results is logical schemes which require long training to understand. 

When things go wrong

That something can go wrong is shown by the Asiana crash where mode confusion was the cause of the crash. The aircraft that I will fly has fly-by- wire in addition to the above. Fly-by-wire makes the aircraft easier and safer to fly as it hides aircraft handling complexities (aircraft quirks) and it adds protection from crossing flight borders. The cost is additional system complexity. New modes gets added to the above and there is more for the pilot to learn and keep track of. Once again the question is if it is worth it and how to design it so that it is.

The other danger is that the high level of automation makes one fly on needle and data fitting. In military training this is an absolute no-no; you do all actions on the horizon, never on a needle that is pointing in the wrong way. If you learned to fly on needle/data fitting it will not help you when you break away from a formation in cloud. You haven’t looked at your own instruments for minutes, you bank abruptly and pull 3G to make sure you are clear from the others, how shall you now react to an altitude indicator going spinning? The only way to avoid a mess is staring at the artificial horizon and flying the aircraft virtually only on that until you have a controlled level attitude, then you can start to worry about needles…

Yet the civil automated flight makes you check the needles and data all the time. It is part of your surveying the autopilot, hour in and hour out. Checking the aircraft’s attitude on the artificial horizon is a waste of time; it is always the same–the computer flies it to the millimeter. So when things go wrong, you are wrongly programmed; you fly on the needles/data and not the artificial horizon. This could be what happened at Air France 447…

It will be interesting to fly the airliner after all my training to master the complexity (we will look into this training in the next Corner). I then expect to be able to compare the more difficult aircraft to fly (the old fashioned one) but simpler to operate with the modern aircraft which seems to be the other way around. Only hands on experience will tell which one I will prefer and why.

16 Comments on “Bjorn’s Corner: Flying revisited

  1. That’s great, have fun!

    What everybody wants to know of course, what side of the ocean you fly from. Probably east, so 320/350 or ATR?

    • I can say as much that it will be a modern type and it will be great to experience all the advance technology and to compare with all I’ve tried before. And yes, it will be a flight in Europe.

        • We will have to see, what Bruce is testing is the Airbus fly by wire protection philosophy, that it works is exiting but to be expected.

          What interest me more is the total user experience and feeling of things. All airliners have reversion modes when things don’t work as expected, the interesting thing is how easy is it then to understand what is happening and what to do, ie what feel does the aircraft convey before and after functionality reversion and how will the non prepared pilot react.

          Also of course how smooth and satisfying is normal operation 🙂 .

          • Good luck and have fun! Looking forward to your report on the experience.

          • Thanks guys, yes it will be exiting, I will have it documented and it will be posted here 🙂

  2. Will see if this goes through as my IE11 will not work for comments (if anyone else is having that problem let Scott know) and when I fired up my old machine the keyboard would not work, sheese. Borrowed a keyboard from the new desk top

    Ok enough of me, while I am not in Bjorns leagues, I find it interesting that we share a lot of the same thoughts (I am titled an engineer but no letters, reality is a technician that crosses into engineering issues and questions for better or worse)

    That said, my flight training did not do spins and I got into one and only reading Wolfgang Langowitz (sp) Stick and Rudder got me out of it (once I remembered the right remedy)

    Also, extreme stuff does you no good in clouds in a light aircraft if the artificial horizon tumbles like (at least did) on them in my day (not sure what they do now, I think acrobatic had non tumbling but will stand corrected)

    And managing modern airliner seems to have almsot nothing to do with the old Stick and Rudder days

    But then can all pilots really handled that anyway? From what I have seen far too many cannot and I have no remedy.

    When they pay a regional pilot something like 15k a year for a skill that cost 100K to acquire ? rough go unless someone give it to you for free (like your parents)

    Frankly thats why I gave it up, I could not compete with the ones who had GI money (bless them) and did not want to starve for a ling.

    • That you got into a spin and got out of it without training is very interesting, what aircraft?

      Re artificial horizons, all modern glass ones should be without limits, I would be disappointed otherwise, and the air force ones were as well (of course), except for the primary trainer but we never flew IMC (non visual) with that.

      Re skills to handle an unforeseen situation like AF447, Mali MD80 this is the point. Airline safety is very very high considering the amount of airlines flying around the world. The goal must be to get it higher still and the question is how? One part must be to prepare pilots better to handle situations like the one you suddenly found yourselves in.

      The question is where does this road take us, more protections (probably) and simpler to understand fall back modes (probably). But that is the question.

  3. It was a Cessna 150 or 152. I had been told I was a bit sloppy on the rudder use in slow flight.

    I did a complete panick when it went into the spin as controls did nothing.

    I lost maybe 1500 to 2000 feet from 3000 when the aura of calm settled in, I knew I was going to die and then with the mind settled, I began to think and realize what it was and what the recovery was.

    Not as calm as I thought, when I made the rudder movement I probably put a dent in the floor. Bless the 150s, they were solid and it came out instantly at a steep nose down.

    I recovered at 800 or 900 feet AGL. Way too close.

    When my brother took up flying I told him to get spin training and he did.

    • That sound really terrible, as said trainers shall be safe to train things like stall and spin with so that a pilot know what do to when things go wrong. You were able to calm down and remember what you had read, how many would have.

  4. Bjorn,

    Again I know I am not anywhere near your degree of expertise in this, but I have posted a number of times my concerns on what I call mode confusion.

    My American reference is the “Almond Joy” syndrome. Sometimes I feel like a nut and sometimes I don’t.

    I also think there should be regulation in regards to it.

    Boeings auto throttle not coming back up when you are about to stall and the low air speed per the Asiana crash (and if I have it right Airbus would have). I think that is the right way to go as far as logic and it should be consistent.

    Each mfg does the automaton their own way and not to a standard. To me that is like Airbus decides that the wing break is at 125% and Boeing decides its 150%. I believe it was made at 150% as it should have been and everyone designs to that so you don’t have a race to the bottom cutting margin.

    Bells get ignored, the nagging voice gets ignored. If it does not work why are we still using it? I think it all should be researched and find out what works not what someone decided works but does not (I worked construction a lot and you ignore the backup horns after a bit)

    And are there pilots who simply cannot handle the unexpected? If so the testing should sort them out as hard as that is the consequences are harder still.

    I think we have gone into automation by accident not with a set of standards in mind, as its been possible, they implement it and without finding out what really works.

    When I read that when you loose the Pitots per AF447 there was a given thrust setting and angle on the artificial horizon that was safe it was “why would he automation not do that?

    And of course how badly eroded are your skills that you have full nose up and descending at 10,000 feet per minute and not understand its a stall?

    I do know that they are now working on unexpected incidents instead of rote training on takeoff and landings that are obvious something you do daily and its evident if you can’t handle those.

    And how do you get airlines to let pilots fly by hand even if its not as economical as the auto pilot and there is some passenger discomfort at times?

  5. I agree with the view of Sarah Kelman, the easyjet pilot who said that “Gliders are well suited to demonstrate what loss of control feels like”.

    Having flown gliders and power, starting with gliding is a considerable advantage in the basic stick and rudder skills, in things like low speed, high aoa flight, along with recovery from the spins which sometimes follow. Gliding also teaches energy management.

  6. Training has to be improved, that is clear and it has to be a part of your license to fly passengers that you shall be proficient to handle an unexpected situation, this can only be achieved with training.

    Aircraft piloting philosophie where Airbus and Boeings but also others have different paths is really tricky stuff. To understand more of this in a hands-on way is one of my main goals of the airline flying coming up.

  7. Very nice experience ahead !
    For sure, it will be interesting to read some of your reports !

    Concerning monitoring / automation , Eolien a former pilot has done some outstanding work figuring how hard it is to monitor thing when all is going wrong … (case study AF447… in French but with nice schemes: http://avia.superforum.fr/t1517-af447-un-accident-d-une-extraordinaire-complexite= )

    I also remember QF32 crew talking hours going through check lists before starting to think about landing

Leave a Reply

Your email address will not be published.