I’ve taught aerobatic and upset recovery courses to many aviators over the years, and almost without exception am told at the conclusion of training that it represented the best investment of time and money they’d ever spent on improving their skills and confidence as a pilot.
In recent years, the corporate, charter, and airline pilots have begun seeking out this kind of skill set as well. It’s a good thing, because as the Department of Transportation recently reported, some of today’s pilots may not have The Right Stuff.
Where the cockpit is concerned, modern light GA aircraft have a lot in common with the latest crop of business jets and airliners. Under normal circumstances these advanced cockpits add to safety. But when things go awry? Well, as our airplanes become more advanced, they also become more complicated, and that can lead to situations which are not covered by handbooks, manuals, and type-specific training.
We’ve all seen the result of unexpected system failures which were not handled properly by the crew. In recent years, Air France 447 suffered from pitot icing which overcame the tube’s heating element and caused air data errors. During the resulting confusion, the crew entered a stall at 38,000 feet which did not end until the Airbus impacted the ocean. Last December, Indonesia AirAsia Flight 8501’s crew responded to a malfunction of the aircraft’s rudder limiter by pulling a Flight Augmentation Computer circuit breaker, which had the unintended consequence of disabling the autopilot. The pilots stalled the aircraft and it ultimately crashed into the Java Sea.
Just to show you that this isn’t something that only happens to “other people”, let me give you two examples of my own. I was flying a Gulfstream IV one afternoon when a wide variety of seemingly unrelated components began to fail. Over the course of 45 minutes or so, we lost air data computers, autothrottles, both autopilots, mach trim compensation, yaw dampening, pitch trim, the flight guidance panel, one altitude encoder, cockpit displays, a display controller, symbol generator, TCAS, an inertial reference unit, and many other elements.
Some of these items dropped offline completely. Others froze or began to malfunction. Some were annunciated on the Crew Alerting System, others were not. Now I knew these components were not on the same bus, nor did they have much in common except that they were electrically powered. Yet the electrical system appeared to be operating normally. We were in visual conditions and not far from landing, which added to the pressure. There’s no checklist for this situation, nor was it ever discussed or simulated during training. Do we land? The aircraft’s braking system is electrical. Should we hold?
Without getting into too much detail, this flight ended uneventfully, but by the time we did touch down, I was basically flying the world’s largest Piper Cub: nothing but a stick, throttle, a couple of analog gauges, and a window to look outside. And that was all I needed. As I recall, the failure was traced to a series of malfunctioning relays under the cockpit floor. Our success was a result of focusing on the basic task of flying the airplane. It’s easy to say, but much harder to do when you’re busy and unsure of what’s really going on with your (normally) trusty aircraft. Failures of this kind cause a rapid loss of confidence in the overall airplane. You’re constantly wondering what will fail next.
The second example was related by a friend of mine. After departure, she lost the #1 comm radio. Not a big deal — the jet as two of them. A little while later, that radio also failed. Over the next few minutes, the flight data recorder failed, followed by the slats, flaps, an AHRS, and other associated componentry. The crew was in instrument weather and flew according to lost communication rules, finally making a high speed, no flap/no slat landing at their destination. Their troubles were caused by a cracked potable water tank, which flooded an electrical equipment bay under the rear floor of the aircraft. Gravity being what it is, one might wonder why important circuit boards are located underneath a water tank… but that’s an issue for another day.
So what does this have to do with upset recovery training? Plenty. The odds of coming out of these scenarios in one piece is directly related to the pilot’s ability to retain control of a malfunctioning aircraft, and that’s when the workload falls heavily on his or her manual flying skills. Truth be told, today’s highly automated airplanes don’t help prepare us for situations of this kind. They do the opposite, physically flying the airplane for us most of the time.
You never know when sharp manual flying skills will pay off. In May of 2011, a Falcon 7X on approach into Kuala Lumpur experienced a rapid nose-up runaway trim condition which could not be stopped. The Falcon 7X was the first fly-by-wire business jet and had been in service for only four years, so this incident caught the attention of many people. It was serious enough that the entire 7X fleet was subsequently grounded. The final accident report was not issued until February of 2016, almost five years later, which should provide an indication of how complex the accident chain was on this event.
Oh, and the crew? They did it right, using a manual flying technique which, while it’s not taught in any type rating course I’m aware of, is taught by myself and others with an aerobatic background. In this case, the pilot learned it while flying Dassault’s other line of airplanes for the military:
While descending through 13000 feet, towards Kuala Lumpur, the elevator pitch trim began to move from neutral to the full nose-up position in 15 seconds time. This resulted in a sudden pitch up of the aircraft to 40° and the aircraft entering a climb. Initially both the captain (Pilot Monitoring) and the copilot (Pilot Flying) were both using the side stick in an attempt to regain control. The copilot then used the priority button to override the captain’s side stick inputs and asked him to stop. The copilot, a former military pilot with experience on Mirage IV and Mirage 2000 jets, then put the aircraft in a right hand bank to a maximum of 98 degrees.
Sudden, uncommanded full nose-up trim is about as bad as it gets when you’re talking about loss-of-control scenarios, yet the pilot was astute enough to remember that he could offset the unwanted lift by banking the jet. Have you been trained on this technique? The pilot had to deal with a beyond-knife-edge flight attitude, load factors as high as 4.6 G, and altitude which ballooned from 13,000 feet to 22,500 feet. What a ride that must have been!
I wasn’t able to locate an English version of the final BEA report, but the French original notes that “the Pilot Flying had performed this maneuver many times during his military career”. After 2 minutes and 35 seconds, the trim motor overheated and was finally cut off, allowing the crew to regain pitch control.
The investigation determined that a small soldering defect on one pin of a computer chip in the Horizontal Stabilizer Electronic Control Unit (HSECU) caused the nose-up instruction to be sent to the Tail Horizontal Stabilizer trim module. Think about the sheer volume of pins, solders, computer chips, and wiring in a modern airplane and you’ll start to realize that these aren’t far-fetched stories borne out of a science fiction novel.
As I said at the top, our aircraft are becoming more complex, and there’s no reason to expect that trend to change. This increases the likelihood of failures and scenarios for which we have not trained. If you’ll pardon the pun, when the chips are down, it’s usually the person behind the controls who determines whether the situation ends with a classic there-I-was hangar story or a fatal accident report.
Time and time again, we see that manual flying skills are as critical to safe flight as any powerplant or airfoil. Let’s keep ’em sharp.
Good write up Ron. Unloading the wing was a life saver for an airliner departing Lindbergh Field back in the 70’s as well. Same runaway nose up trim, military pilot at the controls, same result. I started doing aerobatics a few years ago and agree, it’s the most well spent money and time thus far in my career.
Thanks Steven. Speaking of the military, I had a conversation a while back with a former USAF fighter pilot which really reinforced how perishable those hard-won manual flying, aerobatic, and upset recovery type skills can be. He told me that despite years of high-g combat maneuvering, the past 15 years had seen him limited to straight-and-level corporate jet flying; as a result, he admitted now being uncomfortable with bank angles beyond 30 degrees or so.
You know I’m a believer! When Tom Woolf coined the term “The Right Stuff” he meant it to describe exceptional flying skills. It’s a pretty sad degradation, on both sides of the ledger, when you can use it to describe basic stick skills… I think we’ve been here before, but recovery training is a far better solution and investment for pilots than any mandatory 1500 hour rule – and will inevitably produce superior, more reliable results when things go pear-shaped. Plus it’s a hell of a lot of fun. Best flying money I ever spent, that’s for sure.
You’re right, in this context “The Right Stuff” refers to nothing more than basic aircraft control. But in an era when 99% of pilots spend 99% of their time in straight and level flight, perhaps the bar really is that low. I recall reading somewhere that we will probably not see the likes of Bob Hoover ever again. The argument goes: Hoover, while a very talented and hard working aviator, was also a product of a unique period in history which exposed him to an unbelievably wide variety of aircraft, flying experiences, and opportunities that are simply not available to today’s pilots.
As you noted, improving those skills involves doing some of the most enjoyable things you can do with an airplane, so I’ll never understand the hesitance to jump in with both feet.
Enter your comment here…whenever folks ask me why I don’t have this or that in my cockpit I tell them any “improvement” value is measured against the hit to the avgas budget in hours not flown…other than safety maintenance, few make the cut. Now if we could just keep the FAA from spending that avgas budget for us!
Great point, Rich. The average non-professional pilot flies something like 20 or 30 hours per year. Based on my experience as an instructor, I believe that to be a big part of the stick-and-rudder deficiency problem. You just can’t maintain proficiency on that kind of schedule.
Every mandate, inspection, and expense results in less money being available for actual flying, and that hurts safety. It’s not good for the aircraft, either. The FAA could make flying safer by getting people to fly more, and the best way to do that is to make it less expensive.
Ron,
I agree, as the aircraft gets more and more sophisticated and computer-controlled, the likelihood of out-of-control accidents will probably keep increasing..
I worked with a guy who flew the g450, and he said that during the type training the emphasis was on maximum use of automation vs manual flight (correct me if this isn’t accurate). I’m wondering if the trends will eventually reverse and the training will prioritize manual flying skills/the Right Stuff? What are your thoughts on this
I wouldn’t say the emphasis is necessarily on maximum use of automation, but rather intelligent use of automation. In a type rating course, the automation will be employed and emphasized because one of the purposes of the training is to teach the student how to use it. It can also make life easier when the instructor is throwing engine failures, hydraulic leaks, pressurization malfunctions, etc. at you. It’s a great tool and makes flying safer, but it can also be a crutch.
To put it another way, there are times to use the automation, and times you will want to (or *have* to) fly manually. A good pilot should be able to do both with equal acumen… but it’s hard to preserve those manual skills when they’re not used on a regular basis.
Business jets are designed to fly long distances on autopilot. They do that very well. But when automation fails, the workload falls — often quite heavily and without warning — on the pilot.
I’m not sure if the trends will reverse, but I hope they will. My prescription includes better simulation of automation failures, and regular exposure to tailwheel, glider, or aerobatic flying. Some of that is coming to pass through what is referred to as “Upset Prevention Training”. It’s not a panacea, but it’s a good start.
agree with the mindset that one must know and use the appropriate tools, skills and just as importantly the right headwork for the situation. With some of the poor decision-making I see that puts folks in situations beyond their and/or their aircraft’s capability I think it would be good to start addressing high risk IFR flying (flight into known icing, embedded thunderstorms, widespread low ceilings) with the same concern that we have for the minimally trained performing aerobatics. Without the right training, equipment, and decision-making skills both are extremely dangerous. Funny when I get remarks on the danger of aerobatics from folks who won’t think twice about their chances in high risk IFR.
A very thought-provoking comment, Rich. I always thought of instrument flight and aerobatics as being the two opposite ends of the aviation spectrum. Aerobatics is free-form and uses the full width and breadth of both the aircraft’s and pilot’s physical capabilities, whereas instrument flying is procedural, structured, controlled. There are so many rules for IFR flight — quite different from aerobatics, which has only a single regulation (91.303) controlling it. That’s probably why aerobatics is not permitted over people or built-up areas.
Having said that, I agree with your assertion that both are dangerous without the proper training, equipment, and decision making skills. Perhaps they have more in common than I first thought. It might also explain my attraction to those two specific areas of the Aviation Addiction Spectrum.
You probably get those comments about the danger of aerobatics because so many people kill themselves (or come close to doing so) in spectacular fashion. The other day a guy admitted doing barrel rolls in a Learjet in the past. His rationale at the time featured the Lear’s fighter jet progenitor, the Swiss P-16.
Anyway, I think you’re on to something here. The thing that gets pilots into trouble with both aerobatics and hard IFR flying is attitude and judgement. Those are hard things to teach, which might account for part of the reason we’ve done such a poor job of it.
I’m a supporter of hands-on manual flight, however I must note that often times we use the terms manual flying and aerobatic flying interchangeably. Aerobatic flight is always manual flight, but manual flight is aerobatic only at times.
Perhaps the message is to ensure that crews are comfortable with basic aerobatics? In my opinion, a crew member who is overall proficient at manually flying the airplane may still find himself in over his head in a 4.5G 40deg zoom climb into the flight levels, and may or may not have the presence of mind to roll the jet into a 90deg bank to control lift. The military pilot at the controls was capable of it, but I presume he had practiced that same maneuver in the fighter jet.
Yes, he had previous experience with that maneuver while flying Mirage fighters in the military. I wasn’t intending to use “manual flight” and “aerobatics” interchangeably. I use the 91.303 definition of aerobatics: “an intentional maneuver involving an abrupt change in an aircraft’s attitude, an abnormal attitude, or abnormal acceleration, not necessary for normal flight.”
Of course, you would not find me opposing the idea of ensuring crews are comfortable with basic aerobatics. Fifteen years of experience with aerobatics has convinced me that it does wonders for a pilot’s confidence and abilities, to say nothing of the fact that it’s also a lot of fun.
Count me as unconvinced that rusty manual flight skills are the probable cause of half a handful of notorious cruise flight LOC accidents by jet pilots. I am not a jet pilot, just a private pilot who owns and flies a little ole Piper Cherokee. I enjoy flying by hand while also appreciating that I have a good single axis autopilot to give me an extra pair of hands when I need it, given that I almost never have a qualified co-pilot beside me in the cockpit. I don’t have to worry much about being “over-automated” like the jet jocks may be.
It is true that having good manual dexterity, good hand-eye coordination, and a finely-tuned well-practiced touch on the flight controls can make the difference between a great pilot and someone who just gets by … but the accidents most oft-cited, and cited here again by Ron (particularly AF447) were not “failures of fine touch”. AF447 was a gross mishandling of a perfectly airworthy aircraft that did not need to be maneuvered at all when the autopilot kicked off that fateful evening over the Atlantic Ocean in the midst of the “tropical convergence zone” – a truly “fearful” flight environment full of nasty and dangerous thunderstorms, in some respects an airborne equivalent to rounding the Horn in ancient seafaring times.
The full and highly detailed story on AF447 is available in the final BEA accident report, which is available online in English, and I won’t go into all the hoary details here. However, it is necessary to correct a common misperception – the Airbus A330 never departed the normal flight regime until the Pilot Flying mis-manipulated the flight controls to force it into a steep climb (deck angles at various times recorded as high as 40+ degrees!) at near it’s maximum achievable ceiling (given that at that early stage of the flight, the aircraft was still heavily laden with fuel for its transatlantic journey). This was not a case of a pilot presented with an unusual attitude that he had to recover from – he created the unusual attitude himself, and never seemed to understand that his stick manipulations were the cause of this aircraft’s behavior.
Let me just say here that yanking the fly-by-wire side stick all the way aft and holding it there more or less continuously to the aft stops for several minutes, when one had been cruising straight and level at FL350 and suddenly lost the autopilot … is NOT evidence of one pilot being “a little rusty” on stick and rudder skills. No sane pilot would do that, ever. There is no circumstance in any fixed wing aircraft in which it is acceptable to yank the stick all the way aft to the stops and just hold it there til you die. That’s not a “rusty stick” – that’s a crazy stick!
Such action by the PF can only be explained by a pilot who either (1) never should have been allowed to fly solo in a light trainer by his CFI (which seems rather doubtful, given that he was a fully licensed professional pilot and qualified First Officer) OR there was some form or degree of fear-induced temporary insanity taking place on that flight deck. If the latter, it could have been brought on by the sudden startle factor, a bouncing flight deck in turbulence amidst thunderstorms, several alarms blaring simultaneously, and last but not least:
The stated desire of the pilot flying, expressed several times to his much more experienced captain, to climb the aircraft above a line of T-Storms painted ahead of the aircraft on radar as it penetrated the tropical convergence zone. Which his captain refused to allow him to do.
And not coincidentally, the accident sequence began when the captain who had refused the FO’s requests to climb the aircraft above FL350 was by then slumbering in the sleeping cabin aft of the flight deck.
People can and do do crazy things when we are already in a fearful state of mind, and then are startled and can no longer seem to think straight. Fear makes people act in ways we would never believe otherwise. This “panic factor” is extremely well documented in all manner of stressful, frightening circumstances, including warfare, ships in storms at sea, and air crew in storms in the air.
The relief Pilot Not Flying was likely unaware of the PF’s evidently fearful state of mind when the captain was relieved, and so he had no reason to expect the PF to do anything but fly the aircraft in a normal manner. The sidesticks of the two pilots were not interconnected in the Airbus 330 fly by wire system, so it was not obvious – visually or via physical feedback from his stick – that the PNF knew what the PF was doing with his sidestick. That seems to be a big design flaw that must also have contributed to the accident. A standard interconnected yolk pulled all the way aft to the stops would have been very obvious to the PNF, who could have pushed back or announced “I have the aircraft!”, and if need be, fought for control of the aircraft with the PF.
So no, I don’t buy that professional pilots simply don’t know how to fly their aircraft AT ALL because their stick and rudder skills are getting a little rusty. Letting those skills get rusty is, of course, not the way to go. Being a skilled stick is a great benefit to any pilot, whether flying a C-150 or an Airbus A330 or any other bird.
But I don’t know that there is any kind of programmed flight training one can do to calm a fearful, panic-stricken mind. It is a matter of psychology much more than it is stick-ology.
You’re right, the AF447 accident cannot be attributed to a minor nit in the pilot’s skillset. Like all accidents, it was caused by a chain of events. The flawed pitot tube, the ITCZ-related weather, the captain’s decision to leave the flight deck, the flying pilot’s fear, the Airbus’s design, and so on.
Having said that, I brought up AF447 because highly-automated airplanes can become incredibly distracting and disquieting when things aren’t working right — especially if the failure is related to the air data computer(s). Airspeed, altitude, vertical speed, autopilot, air data information, and more all disappear. The CAS fills with a dozen or more different warning and caution messages. It can be rather confusing and unexpected. You’re in IMC, getting beat up by turbulence, thunderstorms are in the vicinity, and suddenly the primary flight instruments are suspect, the airplane is yelling “stall! stall!” at you, and the bird has reverted to alternate law.
Should the pilots have been able to handle this scenario? Absolutely. But as one who’s encountered this sort of situation, I can understand the hazard it presents. The primary task of the pilot is to ensure he does not stop flying the airplane–ever. This kind of problem should be simulated in training, but it isn’t done well — if at all. Complex systems are prone to failure modes which have not been anticipated by training programs. The pilot must be able to keep a cool head, ferret out the good data from the bad, and keep on flying.
Ron – Pilots must be able to respond in such circumstances, but the only way to provide assurance that pilots can and will respond appropriately is to subject them to routine emergency drills in simulators. It appears that such routine training for many if not most airline pilots is limited, however.
Not being a professional pilot myself, I can’t comment on how the airlines and respective national flight regulatory agencies handle recurrent training for commercial air transport pilots. As a veteran US Navy member (submarines), I can say that we trained and drilled constantly at sea to deal with a wide variety of stressful simulated emergencies. The intended result of that training was that when the real thing hit, we wouldn’t “go crazy” in the midst of blaring alarm horns, warning lights, non-standard instrument readings, shouted orders, donning and wearing protective gear, all while under working under the watchful gaze of trained observers who are perfectly willing to “fail” an individual crew member or an entire crew, with potentially very serious career consequences for all involved including the Captain. It is stressful, it is not fun … but it is a large part of what keeps our Navy crews safe at sea in very challenging missions, in war or peacetime. Similar training and drills are practiced with civilian “first responders” and in relatively high risk industrial facilities such as nuclear power plants, chemical factories, etc.
In the civilian world, of course, military discipline is not available as a management tool, and perhaps airline owners and government check pilots may be less inclined to fail someone out of their job. But given that up to hundreds of souls on every airline flight put their lives in the hands of aircrew, it seems preposterous, given the consequences of a crew failure, that such high stress emergency drills are not routinely practiced and “passed” in simulators as a condition of employment. Not only would doing so improve the non-routine flight skills of pilots, it would also tend to weed out those pilots who are not psychologically fit to respond rationally in such circumstances.
We train for many emergency scenarios, ranging from engine failures to fires, depressurization, and individual system malfunctions. What we don’t train for are the sorts of integrated system faults that I encountered on that recent flight. It’s not clear that such things can even be simulated, because the systems, components, and failure modes are so diverse and complicated. The logical conclusion is that pilots are going to see things in the real world that were never seen in a simulator, and when that happens, manual flying skills will be required.
The advantage the military simulator has over my understanding of an airline sim is that the military test community spends a lot of time/effort/$$ understanding the whole V-n diagram (and beyond) of the aircraft response, including post-stall gyrations, spins, etc. Given that, you can do a lot more effective stall/spin recovery training in the sim. I doubt that will ever be available for airliners due to the expense and risk of mapping those areas in a real aircraft. While wind tunnel and modeling can fill some gaps, they can’t be verified for the same reasons.
Multiple cascading emergencies were routine in my mil sim experience (as both crew and imposed as sim evaluator) via creative scenario development…fail a gen on one side, fail the other engine, fail the RAT and bring the weather down to 0/0 and stand back…they weren’t always realistic, but part of it was to see how stress was handled…of course the most sadistic run was the one where you didn’t fail anything for an extended period.
Ooooh, that *is* sadistic. Even better if you build it up first by hinting at what a tough sim session they’re in for.
The great thing about manual flying is it’s fun! The ultimate reason why people learn to fly.
You hit the nail on the head, Bob. We learn to fly so we can fly. It’s so self-evident that it almost defies description, yet sometimes I think we lose sight of that simple truth.