Pilot error
15 min read

A football metaphor works here…

It has long been standard to use the term “pilot error,” as in 86-percent of fatal GA accidents are caused by pilot error. But that it were so simple. To me it really starts with a pilot fumble. The ball is dropped, but the option is still there to recover. The error comes when the pilot fails to recover.

A recent tragic TBM 700 accident in Wisconsin prompted my thoughts on this, plus some thoughts on the use of pitch trim in such situations.

TBM crash in WI

A tragic accident – can we simply write it off as “pilot error?”

Witnesses said the approach for a landing did not look normal. The tower controller did not mention this and said that the airplane touched down at a normal spot on the runway. The pilot then reported that he was going around.

The touchdown was apparently pretty far from normal because there were 22 prop strikes noted along the runway, spread over 26 feet. That would have happened extremely quickly, as a result of a porpoise, which often occurs after a touchdown at a higher than normal speed. It is also worth noting that the number of prop strikes is not divisible by the number of propeller blades which meant that any prop damage was not equally divided among blades and could have led to roughness in addition to degradation of propeller efficiency during the attempted go-around.

The airplane pulled up in the go-around, started a turn to the left, apparently stalled, crashed and burned.

In the recently posted recap of 2013 accidents, it was noted that ten-percent of the fatal accidents occurred on missed approaches or go-arounds. Because a tiny percentage of the hours are flown while doing these procedures, this defines them as being quite hazardous and worth some careful study.

One more football metaphor and I’ll try to leave that to the Tigers and Razorbacks. Coaches teach it and players do it, usually for a while after an embarrassing fumble. It is the practice of holding onto the ball with both hands when a potential hazard looms.

In flying, we need a version of this. It needs to be a pilot operating mode that becomes active before a time of potential stress becomes critical. A go-around or missed approach certainly qualifies. That is definitely a time for a burst of brilliance and fancy footwork, but you have to be locked, loaded and ready to deliver. There is also a requirement to think carefully about how you use the elevator trim as the maneuver unfolds.

Go around

It looks so simple, but there’s a lot going on.

While contemplating this, I thought back to the night of December 20, 1986. I had a Mooney 252 in my hangar, leased as a replacement while my P210 underwent one of its not infrequent engine overhauls. The mission was to enable a cover photograph for the March, 1987 issue of FLYING. The related story was “Hardball IFR,” a night approach picture taken from the cockpit seemed appropriate, so we rolled the 252 out of the hangar and set out to get the picture.

Gordon Bowen of our staff was working the camera and we were flying at my Mercer County, N J. base. As always the tower was totally cooperative. Gordon had decided that a picture showing the illuminated panel with the runway and other ground lights in the windshield would work best. To get such a picture at that time, you actually had to do the deed and take the picture. Technology has changed all that.

First time around I only used partial flaps which would make the go-around easier. Gordon didn’t like that. He wanted the nose down more so the ground lights would be better framed in the windshield. So it was full flaps for the following approaches.

When he said we could go around and do it again, I came up with the power and started changing the airplane from the full flaps approach to the climb configuration. Any pilot familiar with these airplanes can tell you that this is a demanding maneuver.

With acceleration and flaps retraction, the pitch forces required to fly at the correct attitude in a Mooney are humongous to the point of being disconcerting. When flying airplanes for pilot reports, I always did a go-around to check behavior in that area and knew that the Mooney demanded aggressive control use in the maneuver but I had forgotten that it was quite so demanding.

Mooney 252 cockpit

Are you prepared to move those controls to full deflection?

After that first go-around, I moved my seat forward. I usually check before taxiing that the seat is adjusted so I can handily move all the controls to their maximum limits. I hadn’t done that and what seems like a natural setting for a Mooney seat doesn’t achieve that goal. To get there you have to be practically up under the instrument panel.

I did at least six approaches and go-arounds before Gordon was satisfied that he had his picture and we could head for a debriefing over adult beverages. I felt like I had worked my tail off pushing and pulling on those go-arounds in the 252 and had earned a few sips of the finest.

A bothersome thing was that I felt like I was getting the hang of it after six times. In actual combat, only the first time counts.

The certification rules on what they call “balked landings” relate to performance. The airplane has to be able to climb with the gear down and full flaps with an exception granted if the flaps can be retracted in two seconds or less with no loss of altitude and no sudden change in angle of attack. Further, required stick force in any maneuver can be as high as 50 pounds with one hand using a wheel or 75 pounds with both hands. Think about that the next time you grunt when picking up a 20-pound bag of charcoal that now really weighs only 16.6 pounds.

The point is, the rules aren’t written around pilots always being able to fly with gentle control pressures. Heavy-duty pushing and shoving is okay and this might be more apparent on a go-around than in most other places.

There is not much to flying go-arounds in relatively simple airplanes. A J-3 or Champ would be the simplest of all with no flaps and not too much power. They got a little more complicated when the basic Cessnas grew those big flaps, which initially could be extended to 40-degrees. The airplanes would hardly climb until the flaps setting was reduced so that was a first order of business. The good news was that the stalling speed didn’t increase much from 40 down to 20-degrees on the flaps. The bad news was that the pitch attitude of the airplane had to be changed by quite a bit.

Cessna landing

The more flaps you use, the more aggressive you need to be on a go-around.

In a move that was related more to improving stall characteristics than go-arounds, Cessna reduced the flaps travel to 30-degrees. This meant those ultra-steep approaches were off the table but that going around would be easier. The airplanes were also less prone to spin because with less flaps travel they could also have less up-elevator travel.

With more powerful airplanes, more effective flaps, and with the use of devices in the pitch control system to enable wider center of gravity ranges, the go-around becomes a lot more complex. So does the use of trim.

First, I’ll come clean on something. I never liked electric trim. It can be a trap for people who don’t understand that if, with the autopilot on, you exert force on the pitch system, against what the autopilot is trying to do, the trim system will run against your force. Likewise, a trim system can malfunction and run full trim in either direction. There is no certification requirement that the airplane be controllable with full trim so draw your own conclusion about the desirability of that.

There are ways to disable the electric trim and it’s necessary to know everything there is to know about trim malfunctions and how to deal with them.

Also, most people who have worked on certifying autopilots and electric trim systems often grumble about how slowly the FAA wants the trim to move. That simply means that the pilot is trimming and pushing or pulling at the same time. You might say that is chasing a moving target.

That why I like trim wheels. You move them however fast or slow you want to move them and are always in direct control of the trim. Most of the early jets had great old big trim wheels and I loved those. The smaller ones on GA airplanes are okay. I never had electric trim on my airplane because I just didn’t want it and when flying airplanes with both electric and manual trim I generally used the manual trim.

Cessna 172 trim wheel

When in doubt, use that big trim wheel.

Elevator trim is speed-related and an airplane will seek the speed for which it is trimmed in a specific configuration. Two things happen to this on a go-around. Flaps retraction, usually from full to takeoff initially, will result in a pitch change. This is quite strong in a Mooney. The trim was probably set for the approach speed in that configuration. The trim for the configuration and speed you want in a subsequent climb is likely different, perhaps substantially.

In a maneuver like that, I liked to leave the trim alone until done with the changes in configuration and until the airspeed is on the desired value. Then trim away the forces. I don’t mind the pushing and pulling and to me it is a continuous reminder of the difference between what the airplane wants to do and what I want the airplane to do. Once the differences are established they can be neutralized with trim.

If trim is used during times of change there’s no way to really know what is going on between you and the airplane.

So what does this have to do with pilots losing control of airplanes on go-arounds?

A pilot flying a go-around is at a disadvantage because, unless it is prompted by some conflict with other traffic, he has usually messed up. Key indications were ignored, bad decisions were made, and the airplane was flown into a difficult situation. It is hard to follow all that with a burst of brilliance.

Some feel that a go-around should be no problem because it is like a takeoff. That’s not the case, though, because a go-around usually starts with full flaps and with the pitch trim set for that full-flaps approach.

The usual go-around procedure is to go to full power and nose up, flaps to the take-off setting, and fly away. It is like a takeoff at that point except for a couple of things. The airplane will be out of trim, badly in some cases, and where the reduction in flaps might not cause a substantial change in stalling speed it will induce what feels like a sinking spell and will require a substantial change in pitch attitude with correspondingly high stick force required.

Something else I have noticed in flying go-arounds and missed approaches with other pilots is a reluctance to be aggressive with the power and with the controls. There is no question that this is a time for decisive action, which might not be a natural response after making the errors that got the airplane to such a bad place.

Pilot error

Yes, but why?

It is hard to rationalize how a pilot might have stalled and lost control of a 700 horsepower airplane operating at what must have been a relatively light weight, though some rationalization might come from the fact that the horsepower being applied went from zero to 700 pretty quickly. That would affect both pitch and roll and perhaps cause some confusion which is in itself dangerous.

Other factors could have come into play but all you can say for sure is that the ball was dropped and control was lost. When the NTSB puts a probable cause on an accident like that, they usually reference the pilot’s failure to maintain airspeed (or to manage angle of attack in today’s PC jargon) without offering thoughts on why he failed to maintain airspeed. It might well have something to do with the pitch forces encountered when transitioning from landing to climb mode or the use of trim.

In another recent accident a pilot apparently lost control of a turboprop Piper Meridian soon after takeoff. In the preliminary report, the NTSB noted that the elevator jackscrew corresponded with a 50-percent nose down trim setting. It’ll be interesting to see where they go with that in the final report.

Over the years I observed a lot of pilots who used pitch trim throughout the landing process. They felt that running nose up trim made it easier to land the airplane by keeping the pitch forces low. All I could say was that a pilot doing that would be at a big disadvantage should the landing have to be aborted.

Clearly, go-arounds should get plenty of attention in training and proficiency flying. I know I got checked out in a lot of airplanes without seeing a go-around and that was not good.

Another accident at about this time illustrated why we have to keep a basket of asterisks handy in this business. I and others have often said that once the wheels are on the ground a pilot is better off to leave it at that and make the best of the ground roll because it’s always better be in an accident while you are slowing down on the ground than one that comes when you are going too slowly in the air.

Phenom crash in England

When you’re Vref + 42 knots, this is the result – might a go-around have saved the day?

The asterisk in this case comes from a Phenom 300 light jet accident in England, the one that involved members of the Bin Laden family. That airplane was flying 42 knots faster than Vref when it crossed the threshold, it touched down with only 1,437 feet of the 3,474 foot long runway remaining, it went over an embankment, impacted some cars in a lot, and blew up. Clearly, that was just as bad as a screwed-up go-around.

In his excellent post on tips for instrument approaches, John Zimmerman pointed out the desirability of making sure all is proper when passing the final approach fix. The same thing applies to visual approaches. The simple fact is that a good approach is the primary determinant of success in what follows.

An old friend from years ago, Charlie Cannon, had flown off carriers in the Pacific in World War Two. After the war, when a local businessman asked Charlie if he could fly a 250 Comanche from the local 1,400 foot strip, he asked me about that because I had been using the little airport with a Comanche. I told him just to think about the airport as an aircraft carrier and have at it.

I talked to him later and all was going well. Then he said something I will never forget: “As long as everything is perfect when I get to the straight-away then I know it is going to work fine.” He is the only person I ever heard mention a “straight-away” and I don’t know whether the term was used regularly by Naval aviators of the day, but I knew what he meant and if you have ever watched movies of carrier landings you know too. They flew close patterns, starting the turn from downwind when abeam the carrier, with the length of the final approach set by how far the carrier traveled and how much the wind affected the aircraft as it made the continuous circle to final.

The “straight-away” was that relatively short final. If it didn’t look good there, it wasn’t going to work.

As a rule, GA pilots don’t fly close patterns and approaches to airports vary all over the lot. It really doesn’t matter how final approach is reached but there is a critical point on the final approach where a go-no go decision should be made. I always used 500 feet above the ground, which would be about a mile and a half out on a 90-knot approach as a place to take stock. If everything wasn’t in perfect order there or a good plan existed to put it in perfect order before reaching the threshold, then that is a nice comfortable place to begin a go-around and come around for a better organized approach.

Flying past that point with anything badly out of place can set the stage for really bad things. Aborting the approach at that point might be embarrassing but it is far better to turn your face red with a blush and come back around. Flying past that point with everything just right means, and I know you saw this coming, a good shot at a perfect touchdown.

Richard Collins
11 replies
  1. Stephen Phoenix
    Stephen Phoenix says:

    It seems to me that much of the problem with current training is the constant reinforcement to go around at any point things don’t seem right. That’s OK if that point is, as you say, 500′ AGL, but I think I just read a report of an Airbus going around after hitting short of the runway with a tail strike. I have to wonder if the possibility of significant airframe damage crossed that pilot’s mind before he launched into the air again. Better to roll out shedding parts than even just going around the pattern while shedding parts.

    The correct decision height for go around is probably a function of speed; ie a successful go around in a Cub can be instituted somewhat lower than a jet. I don’t know if that’s true or not. I’m pretty sure that the only time a go around should be undertaken after hitting the ground is when flying a glider.

    • Duane
      Duane says:

      Stephen – it is one thing to attempt a go-around in a heavy jet transport in the final seconds of the approach to landing … and it is quite another to attempt a go-around in a typical light aircraft. The latter is far more able to make emergency maneuvers close to the ground – in the hands of a competent and attentive pilot, of course – than the former.

      I don’t believe that current primary flight training is the problem at all. All primary flight training students are well prepared to manage go-arounds, stalls, steep turns, and other maneuvers. The accident rates of newly minted pilots, according to the most recent FAA accident analyses as well as according to long term accident analyses as well (it never really changes) is far far below the accident rates of well-experienced private pilots. It’s in the hundreds of hours following our check rides that most GA accidents occur – the peak being at around 500-600 hours. That happens to coincide with the pilot’s ability to develop bad habits that would never be tolerated in the cockpit by a primary flight training CFI.

      I’ve always made it a point to practice touch’n go and go-around maneuvers, and practice the stall series, steep turns and such. I also know that most of the private pilots I’ve known virtually never do that. The typical opinion expressed by many is that they think such training in basic flight maneuvers is done only to pass the check ride, and/or that they’re frightened by anything other than straight and level flight. One longtime Bonanza A36 owner and pilot I know (with well over 1,000 hours in his logbooks) told me that refused to ever perform stalls or steep turns in his Bo because he said “it frightens me”.

      It’s not primary flight training that is at fault. It is the lack of post-check ride training and practice, and the subsequent bad habits (including over-confidence and complacency) developed therefrom, that are the fairly obvious source of many if not most pilot errors.

      • Stephen Phoenix
        Stephen Phoenix says:

        Duane – The problem with practicing low go-arounds is that it then becomes reflex. This works fine with a Cessna 150 in Kansas; on a long runway. But when the day arrives that you find you’ve landed half way down a too short strip with trees or a hill off the end and it’s a hot day; the highly practiced go-around reflex will be the wrong action. I personally have witnessed three go-arounds that did not come out so well.

        When flying gliders they are not even an option, yet the glider business does not seem to suffer an unusually high fatality rate due to the lack of go-arounds.

        • Duane
          Duane says:

          Stephen – sure, practicing any maneuver over and over again makes it a reflex .. and that is a very good thing! It’s very difficult to do a maneuver for which you’re not proficient when your life and your passenger’s lives are at stake, and you don’t have but seconds to decide how to respond.

          Having a maneuver down as a reflex action, buried deep in both brain and muscle memory, means that the mental load required to execute the maneuver is minimized .. and that frees up your mind to more carefully consider the particular circumstances, such as length of runway left ahead, obstacles or hazards at the end of the runway, wind conditions, etc.

          I’d suggest that poorly executed go-arounds aren’t caused by going around per se, but by some combination of poor judgement and lack of proficiency in the maneuver.

          As someone who came of age in the US Navy on board a nuclear submarine during the Cold War (which was a hotter war than many think!), our crew and officers drilled and we drilled and we drilled and we drilled some more, for every foreseeable emergency, to the point that our captain drilled us to near exhaustion (I think he actually overdid it a bit on one of our deployments).

          But you know what, when we DID face a real emergency, it was a piece of cake for us to respond correctly, and it saved us from loss of mission, potential loss of life, and in one instance even potential loss of the vessel. The safety record for our nuclear submarines since the days of the USS Nautilus is superb, and the only two losses in the fleet were both due to unrecoverable failures of submarine and weapons systems, not crew errors, and both of those losses were back in the 1960s.

          In my own flying I have faced momentary emergencies also where my prior stall recovery and go-around practice almost certainly saved my life. You can’t convince me that practice is somehow a net negative. Even if practice is itself no substitute for good judgement of your present circumstances.

          There is no substitute for training and repetition thereof. It’s how the human brain works that requires this kind of preparation for emergencies.

  2. Duane
    Duane says:

    There are just so many variables in the scenarios that call for a go-around, it’s hard to come up with a useful universal thumb rule. Dick’s 500 ft AGL rule is fine for judging the overall quality of your approach to landing – if the approach is not going well at all by that point, by all means go around and not wait til you’re touching the wheels down to react.

    On the other hand, lots of things can and do happen below 500 ft that are not necessarily inevitable or predictable at 500 ft AGL.

    For instance, a badly-botched flare is something that only becomes apparent in the last few feet of descent, and if it’s not caught and corrected very quickly, you may well survive the outcome but you may have injuries and bent metal and very likely an unbudgeted engine teardown/inspection (i.e. overhaul) as a result of a prop strike (plus an overhauled or replace prop or props too). I have botched the landing flare (particular in my early flying days) and had the presence of mind to go around and try again before wheels or prop tips managed to contact mother earth. I credit my ability to avoid such a bad outcome to routine touch’n go and go around landing practice.

    There are also other last second contingencies that can arise below 500 ft, such as wind shears (in fact, it’s at or below 500 ft that wind shears usually occur) that can turn your headwind into no wind or even a tail wind, or a sudden gust of cross wind. Also, wild animals often decide to cross runways at the last instant of an aircraft’s approach to landing. Similarly, runway incursions by other aircraft or ground vehicles often happen at the last possible moment.

    So, yes, using 500 ft AGL to determine whether an approach needs to be retried is fine, but you still have to be able to react and go-around at the last possible moment in the approach to touchdown.

    As Dick says above, however, you don’t want that “last possible moment” touch’n go to be the first one you’ve ever attempted in the particular aircraft you’re flying. Indeed, I’m a believer in regular and routine practice of touch’n go landings and go-arounds (a “non-touch’n go”), just like we all did as primary flight students.

  3. Tom Curran
    Tom Curran says:

    Good article; I imagine the TBM accident might have been prevented if the pilot had established himself on a ‘stabilized’ approach early enough to realize it was not ‘normal’…for whatever reason. However, I’m wondering if the sheer violence of attempting a full power go-around in 740 shp turboprop with a mortally wounded, out-of-balance, 4-blade prop is what doomed this flight, regardless of how the flight controls/pitch trim were manipulated.

  4. Mike Sheetz
    Mike Sheetz says:

    Stephen Phoenix, I would not let myself land long enough down a short runway to not have the option to go around. The decision should have been made long before It got to that point. If it were a strip where a go around was not an option I would be sure to be on a stable approach long before that and do what would be necessary to assure the greatest chance of a favorable outcome. Good decision making is an attribute of good pilot, and that includes being timely.

    • Stephen Phoenix
      Stephen Phoenix says:

      Good, that’s what you should do and, I believe, that’s what Mr. Collins alludes to. But I personally saw a Cessna 206 loaded with 6 people do that very attempted go-around maneuver into the trees with very messy results. So, you cannot say it won’t happen. It will and it does continue to happen, so just maybe there is a flaw in the thinking somewhere. Just saying “it won’t happen to me because I am a superior pilot” probably goes through all pilot’s minds before the big day comes.

      • Duane
        Duane says:

        Stephen – you seem to have a “thing” about go-arounds because you saw one go badly one time. Many of us have seen takeoffs go badly, and/or have seen normal landings go badly … and/or have seen cruising flight go badly due to weather, mechanical issues, lack of fuel, whatever … none of those results are good cause to dismiss the requirement to use good piloting judgment and good piloting skills at all times.

        As with all things involved in flying light aircraft, there is always flight risk that has to be managed, and that risk has to be managed on every single flight. Not all pilots are created equal, some are better than others, though all of us are humans and therefore subject to human error.

        Some pilots rarely make big mistakes of judgment or competence, but the one time that they do, it’s deadly … while some pilots make dumb mistakes all the time and get away with it for years before the butcher’s bill finally comes due.

        Judging when to make a go around is an important pilot skill.

        Being able to execute the go around properly in the conditions in which you find yourself is an equally important pilot skill.

        One or a few pilots lacking either or both skills at a particular point in time doesn’t make those skills useless.
        Just the opposite is true, in fact. Which is the point of Richard’s post here.

  5. Jere Joiner
    Jere Joiner says:

    I like Dick’s final analysis. If at any point during the approach — and certainly during the final approach or straight away — there is uncertainty or things don’t feel right, abandon the approach and go around. There’s no shame in doing so — hey, call it practice! There’s lots of shame if things go to hell in a basket and you bend the plane.

  6. Brett Slade
    Brett Slade says:

    Off-topic post, here: I just happened on airfactsjournal. It is a pleasure to again have the sage advice of Dick Collins. I simply do not want to miss one of his posts. His opinion on modern IFR flying is a perspective that is sorely missed in the aviation press. Thanks, Dick!

Comments are closed.