Over and over again, the runway turnback maneuver comes up several times a year in accident reports, as it did in our coverage last week. Those who argue against it, often call it “the impossible turn.” But any reasonable analysis proves that turning back to the runway is anything but impossible. But it is risky and pulling it off without prior practice would be difficult. This video addresses some of the elements of making the turnback work.

37 COMMENTS

  1. There was a failed turn back attempt two weeks ago when a Bonanza lost power and tried to turn back to North Perry airport, Pembroke Pines, FL. They almost made it, but crashed short and three people died. A security camera filmed the crash, showing the aircraft nose down in a 60 degree left bank (in a full stall) when it hit a car going down the road. The Cub demonstration is flawed for two reasons, one, it’s a Cub, and two, the engine is operating, not a true high drag windmilling prop. A true test would be to kill the engine and allow it to keep windmilling, high drag, but the loss of a very nice Cub just wouldn’t be worth it.

    • Cub or Bonanza is irrelevant. You need to work the numbers for your airplane and the conditions that day. Plenty of people get killed in Cubs because they don’t run the numbers before doing a maneuver either.

    • Turnback or not, all bets are off when you snag wires like the North Perry Bonanza (my on scene shock of the fatal results of a Baron hitting wires decades ago formed this opinion).

      If you don’t have the altitude to push and unload the wing, don’t make the turn. The wing can’t stall in zero G, but a steep turnback while pulling positive g + probable uncoordinated flight (slip, skid) is an invitation to a departure from controlled flight. I might turnback in my friendly J-3, but higher performance airplanes will not forgive a moment of inability to fly the wing.

    • There was a character ( Jim Hazleton ) who ran a flying business but prior to that spent years after WWII flying super drops, ie dropping super phosphate from modified Tiger Moths on to pastures. Well over and over hundreds of times his turn to start a new run would be slow climbing wing over, sort of a pre stall with turn. So what happened years later when his Piper Cherokee engine quit on take off, the exact same wing over with landing in the opposite direction.

      • Exactly. I learned ag. flying with Max Hazelton and we trained in C180’s doing figure eights off the company’s airstrip, landing and taking off each way. Especially when flying a Beaver on top dressing, you pull up, jack on flap as the speed washes off then let the nose go, but ONLY after seeing where the nose is going to go. As the nose drops you jack off the flap and do your run. You do it 100 times a day…
        You soon learn whether you’d get back to the strip or not if the noise stopped. Basically it’s all about understanding your aircraft.

  2. John King’s cameo is a brilliant touch, Paul. When teaching simulated rope breaks in gliders I had to overcome the hard-wired response (in this country, at least) of my students always to turn to the right when the towrope releases. At Cub/glider speeds, it’s critical to turn *into* whatever crosswind may exist, to reduce the cross-track distance you need to make up to get realigned with the runway…if there’s only one runway, that is. Your point about considering the turnback decision factors before advancing the throttle on takeoff is paramount.

    By the way, 45° is a good rule of thumb: I did the math once and learned that 42° of bank yields the minimum loss of altitude per degree of heading change in any aircraft flying at a constant airspeed in a glide. What that airspeed should be for this maneuver is another topic for discussion, “no faster than necessary” being the correct, if nebulous starting point.

  3. RIP to the 3 people that died in the Bonanza accident. There is no comparison between a Cub and Bonanza in this situation. The Cub is light, agile, and basically can turn on a dime, whereas the Bonanza is heavy, sluggish at best, and will sink like a rock without power. The Bonanza pilot probably made the best decision under the circumstances.
    If I had to give any advice I would say, go find a good aerobatic instructor to teach you how to do aerobatics. The best money I ever spent was during a week in Texas with Duane Cole. The training taught me how to handle myself under extreme situations, reverse course fast, it gave me lots of confidence, it saved my life later when a student and myself got hit by a dust devil and rolled inverted as we were about to touch down. Airline, ‘Flight Deck’, pilots take note-have you ever flown upside down or read, “Fate is the Hunter’?

  4. Right on, Paul!

    I wish people would stop referring to the “Possible Turn” as the “Impossible turn.

    If you experience engine failure after take-off, there is an altitude/airspeed/weight/pilot combination that will allow you to complete a ‘Safe Turn’ back to the departure runway.

    It was always a given that if you had the above-noted combination, you could always return to your departure runway if you lost your engine.

    The trick was to pin down the exact combination of the four items listed above that would allow safe execution of the manoeuver.

    The “land straight ahead or (say) thirty degrees either side of departure path” came about because engine failure in the first few seconds of take-off put the pilot in an extremely tenuous position: no speed or altitude to trade for turning performance. Any attempt to do so resulted in a broken airplane at best, and broken people at worst.

    Hence, the “never turn back” dictum.

    I view the present push to train for the “Impossible Turn” with trepidation: unless the pilot

    1. Is going to maintain currency (minimum 75-100 hours/year),
    2. Ascertain minimum altitude for procedure commencement with dead and wind milling prop (more drag with dead engine),
    3. Routinely practice the manoeuvre (which I doubt – most don’t even practice stalls/circuits),

    then I don’t hold out much hope for the success in the event of an actual EFATO.

    There’s nothing wrong with telling pilots to find the minimum altitude at which they KNOW they could execute a 210 degree turn back to the runway. Then add a fudge factor up to fifty percent for screw-ups.

    Don’t become a test pilot on your first engine failure after take-off; half-way around the turn back is not time to find out you’re out of airspeed, altitude, and ideas…

  5. “The Cub demonstration is flawed for two reasons, one, it’s a Cub, and two, the engine is operating, not a true high drag windmilling prop.”

    50 years ago, I was a flight instructor at a busy flight school. We would usually fly until close to last light (8 p.m. here in Minnesota in the summertime). The three of us would take the 65 hp. Champs up to 2000′ AGL–shut down the engine (no restart probable–you had to dive to over 100 mph and work the rudders to get it turning again) and spot land–furthest from the hangar had to buy the beer. I don’t know about a Cub, but found that the prop stopped rotating between 55 and 60 mph. We felt that the stopped prop INCREASED the glide, vs. the drag of a windmilling prop–much like feathering the prop. Paul–have you ever tried it in the Cub? Is the glide better or worse?

    For all of the talk about a sudden and catastrophic engine failure–most failures just get rough–still producing SOME power. Consider the options below:

    For all of the “possible/impossible” discussion, I haven’t seen a mention of the length of the runway. Obviously, taking off from a 1200′ runway, you aren’t coming back if the failure is right after departure. At 400′–you may make it–but then you are landing downwind on a short runway. Again–pre-takeoff planning is in order. With a Cub, taking off from a 5000′ runway, you can climb to 200′ AGL and STILL land on the remaining runway by employing one of those nifty slips that Berge writes about. Much higher than that, you’ve got a good chance of getting back.

    Finally (whew!) we’ve considered the crosswind runway, but how about taking it back all the way around to land on the runway you just departed? Consider that example of a failure at 400′. Turns to crosswind in the traffic pattern usually start at 400′ AGL–if you are ON the crosswind, and there is any wind at all, you may be better to simply turn an abbreviated downwind and land on the departure runway, instead of a 270 degree turn back (or if you have a crosswind runway, you are already on downwind leg for IT). Think about it–from 200′, land straight ahead on the runway. FROM A 400′ CROSSWIND LEG, the left turn to downwind on your departure runway (or simply turning base on the cross runway) may be your best bet. If the failure happens between that short time between 200 and 400 feet, well, “I guess it just isn’t your day!”

    Lots of better variables, but the “impossible turn” is just ONE of them. Let’s change the “impossible turn” to what it REALLY is–pre-takeoff planning.

  6. In 2010 I gave up the idea of the turn,except really in superbly impossible cases (say the only option is landing into a group of fuel tanks), or when I have SO much altitude that I have to really bad to fail.

    In Portland Maine two friends died trying the turn after the YAK 57 apparently died. The people involved were two of the most experienced pilots one could find. One of them was not only an accomplished military pilot but a test pilot, and the other an experienced aviator that worked as a tower controller.
    In short, if they tried and failed, I do not see it as an option for mere ~~100hr/yr like myself.

  7. The Cub is a relatively easy turn-back airplane: it’s slow. A lot of LSAs should be even better: steep initial climb, low flying speed. Indeed, they’re quite likely to run into the “overshoot the runway going downwind” problem.

    The altitude lost, and the lateral displacement you have to adjust for, vary as the square of the airplane’s minimum sink speed (which is increased by about 21% in a 45-degree bank). A Bonanza is a very different machine, in a turnback, than a Cub. Unfortunately, the Bonanza also has much worse odds for going straight ahead: the sudden stop in the Bonanza is going to be drastically worse than the sudden stop in a Cub.

  8. Another great film, Paul. The John King cameo was a great touch. Knowing the political blow back would be formidable, it nonetheless seems that ‘best land/crash zones’ (or worst) could be marked on airport diagrams. The repetition of every pilot attempting to determine this for himself at the beginning of every takeoff and the associated errors of reasoning that are naturally introduced could be greatly mitigated. Certainly no-one wants the pin dropped on their outhouse, but I’m just saying. If this information is already in ForeFlight or elsewhere then please pardon the idea, thank you.

  9. “Part of the discussion for complex airplanes is ti pull the prop control to low rpm, high pitch. That will make a big difference in the drag.”

    This is something I’ve advocated for years. We teach “engine failures” by pulling the power off, but don’t teach pulling the prop back in an aircraft with constant-speed prop. We do it on twins (feather) but not on singles. It made the difference on one of my engine failures–(old Cessna 210) over the bogs in Northern Minnesota. When the engine quit (cylinder replaced less than 4 hours before–broken rod) I switched tanks, checked mags, checked fuel–just like a twin–then brought the prop back and the prop stopped. My Dad asked “are you going to put it on that trail?” I replied, “No, that’s Canada–you have to haul the wreckage out–I think I can make the airport.” We did–but barely–bringing the prop back made the difference. I told him he was lucky–“what are the chances of having an engine failure and you’re riding with a glider pilot?” (smile)

  10. Paul: NAFI did a nice webinar on this. You left out a big point–try climbing to 700-1000 ft at Vy or greater. Many aircraft will not make it back to a moderate-length runway without a 10-15 knot headwind. The climb angle is shallower than the glide angle. You can make the turn and line up, but you won’t make it back to the opposite threshold. Oops. I have tried this with several of our flight-school aircraft, and it is correct.

    My Maule M7-235 has the opposite problem. If I do a Vx climb and turn back at 500 ft (no wind), I’ll be lucky to hit on the runway because I am so high!! So much for STOL. But I can S-turn a bit and make it work.

    The solid advice is to not try the turn unless you have practiced it in the make/model (including engine horsepower) that you are flying. The NAFI webinar also shows that the fatality rate is WAY higher for turnbacks than for ahead landings after an engine failure.

    Best

    Vince

    • All due respect, I don’t think the data is good enough to support this conclusion. It’s likely to be biased against the turnback because when they go bad, they *always* get into the accident database. The successful ones–however many there are–may not. The same could be said of the straight-aheads, but I know of at least three were were not counted. There are so few of either that noise in the data could contaminate any conclusion.

      Having said that, I’m not a proponent of either. My gut tells me the turnback is marginally higher fatal rate. But twice as high? I’m skeptical.

  11. Many decades ago I was very thankful a vetern instructor didnt land straight ahead into high tension lines and four way traffic,after an engine failure.My only trophy was a stripe from the shoulder harness.Practising some engine outs after reading about this accident,found that i dont mind the 180 if neccesary,but there are a few variables to think about regarding making it back to the runway

  12. I love this argument. Of course it’s possible to make the turn under a certain set of circumstances. Yep, go up and practice to your hearts content. Make sure you practice every week, in every aircraft you fly, including practicing all your other maneuvers and emergency procedures. If you actually have an engine failure, and you didn’t practice it in this aircraft last week, go straight ahead. I have lost too many friends to the “maybe-possible turn.” Even our old friend BS, who is a huge proponent, had two actual tries at it and crashed both times, Does that tell you anything? Oh, he thinks that now he could do it. Well, maybe; maybe not. There are way too many variables – only some of which can be assessed before takeoff. I would never let a student entertain the idea of trying it. I think it’s irresponsible for anyone to tout the merits of a maneuver that has killed and maimed so many pilots. If you can do it, great – I hope you never need to test your theory.

  13. https://data.ntsb.gov/carol-repgen/api/Aviation/ReportMain/GenerateNewestReport/102636/pdf
    Paul,
    The NTSB article linked above is an NTSB aviation accident preliminary report for a crash that occurred on Feb 16, 2021. The pilot(s) attempted to return to the airfield with fatal results. Both pilots had CFII certificates and were twin engine rated. One of the pilots held an ATP rating.
    What a pilot does when experiencing engine failure is the PIC’s decision to make. For me, landing straight ahead into the wind is far superior than attempting to return to the airfield with fatal results. Countless numbers of pilots have failed when making the attempt to return to the airfield. It is time to stop this unnecessary loss of life.

  14. Apropos of Mr. Hanson’s comments, shouldn’t more (or all) pilots be trained to fly gliders, so that in an emergency involving a heavier-than-air craft, they will have more options available, provided the wind cooperates, and other elements remain constant for the duration of the flight? Early aviators did much the same, first out of necessity, then out of habit, and eventually by choice. I imagine that bush pilots often face the same dilemma, even today. Perhaps we can learn from their example, since they must cope with short runways, imposing terrain and bad weather, often in aging or underpowered machines that compel them to rely on agility, thus to allow air currents to bring them down safely, amid distress.

  15. Haven’t watched the video yet (I will soon), but it has been proven multiple times that the “impossible turn” is not so impossible, as long as certain starting conditions are met. It is a high-risk maneuver though, but sometimes a straight-ahead power-out landing over a highly-populated area is also high-risk. But one thing I have learned from helicopter flying is that much of helicopter training is also high-risk, yet obviously it can be done with reasonable safety. It all comes down to understanding the risks before you even get in the aircraft.

  16. Good review, and at an important time, post Beech A36 North Perry Airport Hollywood, FL caught on a RING camera.. While the investigation stills plays out, it brings back the forever question. Straight ahead, or turn back..? No doubt, each instance is unique, to location, piloting skill, proficiency, and aircraft specific.. Its all up to the PIC. 91.3.. Whatever you way you choose.. Just don’t screw it up..

  17. What is theoretically possible is one thing, as is practicing the maneuver at a “safe” altitude. IMO, there are at least two factors that add significant risk to performing this maneuver in a real emergency: (1) the fact that it is a real emergency will add stress. Sufficient confidence in your skill to perform this maneuver may somewhat mitigate the effect of this stress, but a pilot needs to evaluate himself ahead of time in this regard (2) the visual impact at low altitude will be completely unlike any drill performed at a “safe” altitude, and the potential for distraction to alter the precision of the maneuver will be significant. Furthermore, even if you regularly practice idle power precision approaches within the normal pattern, the turn back maneuver will IMO be very different. Aside from these issues, the departure runway length and your particular aircraft performance will have a significant effect on whether or not you are even in a position to make it back to the airport should you succeed in reversing course while maintaining control. What do I mean? Just this, without complicating the issue, let’s say your goal is simply to return to the same runway surface. I fly a 2 seat, single engine GA airplane. Based on calculations for my plane, I believe I would need an absolute minimum of 900 ft. AGL to safely maneuver to land on the reverse course to that from which I departed. But, the kicker is this. Notice I did not say “land on the runway”. That’s because unless I takeoff from a runway that is at least 4000 feet long, by the time I reach 900 feet AGL (1000 is better), even if my turn back is successful, I will land short because the runway will be too far behind me by the time I start the maneuver. So, if I’m taking off from a runway shorter than 4000 feet, there is no consideration whatsoever to turning back, at least with the thinking that I’ll make the runway. If this is something you fail to consider, you may attempt a maneuver that is pointless to risk. You should always do your best to familiarize yourself with what is ahead and to the sides of your departure path and, IMO, already know what your likely course of action will be before you put the airplane on the runway. All of this is my opinion. Do your own calculations and analysis and take the action you feel is best.

  18. I apologize for not asking the above question without additional explanation.
    On the back of the cabin over Paul’s shoulder there are a couple of notes which appear in some scenes then disappear in others. One is “Your message here.” Just wondering what the were.

  19. Only General Aviation (GA) perpetuates this myth. The U.S. Military has been training the “Possible Turn” for decades. Excellent analysis done by the U.S. Naval Academy Engineering Department and astronaut Brett Jett. https://www.nar-associates.com/technical-flying/jett/jett_wide_screen.pdf
    Know your options, practice-practice-practice with a CFI. If you can learn to land a plane, you can learn the “Possible Turn.” – R. Elves, USAF (ret), ATP, “Gold Seal” CFI

  20. Many years ago I got my motorglider ticket at Sportys. On my check ride the FAA examiner reached over and brought the throttle to idle at just 300′ and said “land it”. This was a practiced procedure, immediate nose down for speed, 45-55 degree banked turn and slip it in for a landing. As John King commented, overshooting the runway is a big concern.
    Would I ever consider doing this at 300′ with a Bonanza, NWIH! So as Paul said it depends on equipment and conditions. Figure it out in advance.

    • As most CFI-Gs point out, hitting the fence at 3 mph if you overshoot is a lot better than hitting the ground at 60 mph at the near end. If the far end of the runway is really bad, you could always ground loop it during your roll out.

  21. I know most pilots are probably not capable but why not use the stall turn or immelmann to immediately reverse direction and put the nose down to gain airspeed and then land. Would to much altitude be lost?

  22. Rick Marshall’s exposition of the math and data my be helpful – climb rate versus sink rate for any particular aircraft, configuration, weight and performance – and the yes it can or no it can’t result, regardless of how far out you have got (slopes don’t lie) – “http://www.inflightmetrics.com/”, “http://www.eaa.org/videos/6193816644001”, “http://www.eaa.org/SpecialUseComponents/newslettersmagazines/SADownload/?pdfID=e85fe188-b2c5-4f8b-afc8-272d1ffd1b3f”.

  23. Condensed version: Impossible except when it isn’t. And vice versa.

    Never had a power failure on T.O. and don’t care to – too many variables to consider in too little time.
    I suspect I’d do what most do, make a snap decision that could have just as well been a coin toss.

  24. I think the turn back is a great example of what should be vs what is, when the engine fails on takeoff

    What should be be: Every pilot, every time immediately smoothly pitches down to the glide and rolls in 45 deg of bank in a nice coordinated turn.

    What is.: The shock of this cannot be happening to me as the speed bleeds off, then a panic roll without any rudder, then the unconscious pullback as the windshield is filled with the ground rushing up….then the stall and spin and a smoking hole with dead bodies.

    The accident stats are clear if the airplane is under control, wings level, level pitch attitude and not accelerating the crash is almost always survivable. There are almost no examples of unlandable areas.

    Finally between one half and three quarters of all engine failure are caused by the actions or inactions of the pilot. The best way to deal with the engine failure scenario is to not have the engine fail in the first place.

  25. Pilots have a chance to address this on every landing. While on downwind, base or final “If the engine quit now, will I make it to a runway?” From my experience, quite a few pilots will not. On most GA planes on final you would have to be at least 300 ft AGL a half mile out, and with no head wind, to maybe make the runway threshold if the engine quit. Low, flat finals with power won’t make it.

    A thought experiment: Imagine somewhere on takeoff your engine quits. If you were able to immediately spin your plane around at that exact location and head back to the airport, would you make it? Now add it a few seconds for reaction time, a 180 degree turn descending (at about 600 – 1000 fpm) and you’re back to that same distance from the runway, would you make it?

    Engines are more reliable than tow ropes, which is why glider pilots, including 14 year olds preparing for solo, practice this maneuver with an instructor. Even though 200 ft AGL is the “standard”, instructors often minimize risk pulling the release at 300+ AGL.

    On your next landing, pull back the throttle and see if you can deadstick it to the runway. Get a feel for how high you are and the distance from the runway.