• Ratty's Ramblings - The Impossible Turn

    Ratty's Ramblings

    Ratty's Ramblings - The Impossible Turn

    By Ian Radcliffe

    A few weeks ago, discussing forced landings, I mentioned "The Impossible Turn". Arguably the worst time to lose the engine in a single-engine plane is on climbout after takeoff. Low on speed, and just plain LOW, your options are severely limited. Since there's not a lot of time to troubleshoot, the usual recommendation is something like "lower the nose and land straight ahead", shutting off everything you can on the way down. But advances in aircraft and engine performance since the days when that mantra was born, have made another option available.

    Turning back to the field is a very tempting propositon. After all, you've just left a perfectly good landing site; just making it back inside the airfield perimeter guarantees you a relatively flat, smooth, unobstructed surface, one that could be infinitely more appealing than the area surrounding the airport. But the impossible turn gained that appellation for a reason: it mostly didn't work. In the old days, aircraft takeoff performance was often lackluster at best, but performance enhancements make the impossible more possible - in the right airplane, under the right circumstances. And, once again, with our flight simulator we can explore those circumstances without risking the nasty consequences of real-life miscalculations.

    Ratty's Ramblings 4

    It's not a bad idea to ask yourself during your takeoff rolls, "What would I do if the engine failed here...or here...or here? Clearly, if the engine dies during the takeoff roll or the initial part of the climb, continuing straight ahead, slowing, and avoiding obstacles is the best you can do. But at some point you will gain enough height to make a turn back viable. What that height is depends on a number of factors:

    Airplane. A July 2002 AOPA Pilot article, "Engine Out!" reported that a Cessna 172 required nearly 500 feet of altitude to return to the runway using an "aggressive" 45-degree bank and allowing the nose to fall "fairly dramatically" to maintain airspeed. The only way to find out what works for your plane is go out and fly it.

    Weight. Are you solo with half fuel, or is every seat full and the tanks topped off?

    Pilot skill and response time. Unless your immediate response is to lower the nose and begin the turn, you're probably not going to pull it off. Also, if you spend a lot of your time in straight and level flight, an "aggressive" bank and dramatic nosedrop will probably not come naturally. Practice helps; that's what flightsims are for. By the way, even with the throttle closed, idle thrust is just that - thrust. You won't have that if it happens for real. For added realism, pull the mixture.

    Headwind. Will make for a steeper climb, but becomes a tailwind on the way back, meaning higher landing speed and the chance of running out of runway. Crosswind angle is a factor too, affecting the choice of the best way to turn.

    Atmosphere and altitude. Air temperature and elevation both have an effect on climb performance. What you can pull off at sea level may not work at Denver International.

    Chosen climb profile. Did you opt for a cruise climb, best angle, or best rate?

    Runway length. How far behind you is it? And: is it long enough for you to land on and stop with a tailwind?

    All these variables ensure that there is no stock answer to the question "How high do I have to be?" Only experimentation can tell you that. Have fun!


    9 Comments
    1. dbauder's Avatar
      dbauder -
      Neat article! Another thing I read about the 180 degree turn when faced with an engine failure on takeoff. It's more than 180 degrees of turn. Let's say you lose the engine and make that 'about face'. The problem is, assuming no wind, is that you are not lined up with the runway after a mere 180 turn. You must continue the turn to finally line up with the runway, parallel taxiway or whatever you can find. Combine that with panic, slow airspeed and overuse of the rudder and you get a stall-spin. My virtual Piper Tomahawk can replicate it quite well!

      As for hangAr versus hangEr, you hit on one of my peeves as well
    1. revkev's Avatar
      revkev -
      A very interesting read especially for one (me) who has no real time flight training but plenty of aviation experience as a LAME. I also love simming. I do know that the turn back option is fraught with danger. When I was an Engineer in the Air Force we had an historic flight Tiger moth crash after take-off, killing the instructor. The student survived. They were practicing an engine failure after takeoff. Both were very experienced military pilots, the instructor had flown VIP's and MP's around and had flown in combat in Vietnam. The student was a captain on Orions. I flew with the captain on a number of occasions. My point is that the return to base after a failed engine, real or practiced, can really bite you in the posterior if you don't do it carefully. Vale to my friend who died.
    1. lnuss's Avatar
      lnuss -
      Nice choice of subject, and some very good information here. I'd add that different bank angles (giving a different rate of turn, of course, as well as a different sink rate) will give different results, but the 45º bank keeps the turn radius fairly small, compared to that at lower angles of bank, even though the descent rate is higher, and will give you a bit of advantage. And don't forget to keep that nose down.

      I would hasten to add that this is something for the sim, only, especially if you pull the mixture. I'd also add the caution that you should not rely on the sim's performance to learn what performance to expect in a real aircraft of the same type, partially because the sim rarely performs just like the real thing, but also because your performance sitting in a chair that is solid on the ground while performing this is not going to be the same as in a real aircraft that is moving around and putting Gs on you, perhaps bouncing in the bargain.

      Incidentally, you might want to start this procedure the first time at 800 or 1,000 feet AGL, discovering how much altitude you lose before getting back to the runway. Then you can start from a bit lower and check the results. Once you have established your best performance with no wind, add 10 knots of wind on the nose. Then 10 kts of wind from 30º, 45º and 60º to each side of the runway, then do all these with 15 kts and with 20 kts. You'll learn a lot. Some of what you'll learn can also be learned with ground reference maneuvers.

      For doing this is a real aircraft get a couple or three thousand AGL then try it. You might try that in a sim, too.

      Ian, you might consider an article sometime on doing ground reference maneuvers in the sim, such as S-turns across a road, turns about a point, etc. It would be very awkward doing view changes for this with a hat switch, but not too bad with something such as TrackIR. Of course folks could read up on that, too, in the FAA's Airplane Flying Handbook, Chapter 6.
    1. flightsimg's Avatar
      flightsimg -
      With my French PPL, I learned that any failure at takeoff is land straight ahead.
    1. ianhr's Avatar
      ianhr -
      Quote Originally Posted by lnuss View Post
      Nice choice of subject, and some very good information here. I'd add that different bank angles (giving a different rate of turn, of course, as well as a different sink rate) will give different results, but the 45º bank keeps the turn radius fairly small, compared to that at lower angles of bank, even though the descent rate is higher, and will give you a bit of advantage. And don't forget to keep that nose down.

      I would hasten to add that this is something for the sim, only, especially if you pull the mixture. I'd also add the caution that you should not rely on the sim's performance to learn what performance to expect in a real aircraft of the same type, partially because the sim rarely performs just like the real thing, but also because your performance sitting in a chair that is solid on the ground while performing this is not going to be the same as in a real aircraft that is moving around and putting Gs on you, perhaps bouncing in the bargain.

      Incidentally, you might want to start this procedure the first time at 800 or 1,000 feet AGL, discovering how much altitude you lose before getting back to the runway. Then you can start from a bit lower and check the results. Once you have established your best performance with no wind, add 10 knots of wind on the nose. Then 10 kts of wind from 30º, 45º and 60º to each side of the runway, then do all these with 15 kts and with 20 kts. You'll learn a lot. Some of what you'll learn can also be learned with ground reference maneuvers.

      For doing this is a real aircraft get a couple or three thousand AGL then try it. You might try that in a sim, too.

      Ian, you might consider an article sometime on doing ground reference maneuvers in the sim, such as S-turns across a road, turns about a point, etc. It would be very awkward doing view changes for this with a hat switch, but not too bad with something such as TrackIR. Of course folks could read up on that, too, in the FAA's Airplane Flying Handbook, Chapter 6.
      All great points, Larry. Thanks. The best one is the one I keep meaning to mention and keep forgetting - the Airplane Flying Handbook. A priceless source of information, along with the Pilot's Handbook of Aeronautical Knowledge. AND THEY'RE FREE!
    1. amcclymont's Avatar
      amcclymont -
      In gliders (which of course have a slow sink rate) the rule of thumb is you need 200 feet to turn back to runway. On PPL checkride the instructor will pull the cable release just above 200'. Our runway had a lake at the end, if you had a cable break below 200' you would have to do a gentle turn towards the closest lakeshore on the right. One thing that intrigued me about that recently (it is a long time since I flew for real) is whether for extending the glide it is better to fly near best glide speed or to dive into ground effect. I did some calculations, and ground effect is only effective within about a half-wingspan of the ground (or lake). So I think don't go for ground effect until you are getting down towards that limit.
    1. ianhr's Avatar
      ianhr -
      Quote Originally Posted by amcclymont View Post
      In gliders (which of course have a slow sink rate) the rule of thumb is you need 200 feet to turn back to runway. On PPL checkride the instructor will pull the cable release just above 200'. Our runway had a lake at the end, if you had a cable break below 200' you would have to do a gentle turn towards the closest lakeshore on the right. One thing that intrigued me about that recently (it is a long time since I flew for real) is whether for extending the glide it is better to fly near best glide speed or to dive into ground effect. I did some calculations, and ground effect is only effective within about a half-wingspan of the ground (or lake). So I think don't go for ground effect until you are getting down towards that limit.
      I love this stuff. I'll have to play around with that. The dive will increase drag and burn more energy, but then drag is less in ground effect so if you can fly for a while half a wingspan above the ground it might pay off.
    1. amcclymont's Avatar
      amcclymont -
      Quote Originally Posted by amcclymont View Post
      In gliders (which of course have a slow sink rate) the rule of thumb is you need 200 feet to turn back to runway. On PPL checkride the instructor will pull the cable release just above 200'. Our runway had a lake at the end, if you had a cable break below 200' you would have to do a gentle turn towards the closest lakeshore on the right. One thing that intrigued me about that recently (it is a long time since I flew for real) is whether for extending the glide it is better to fly near best glide speed or to dive into ground effect. I did some calculations, and ground effect is only effective within about a half-wingspan of the ground (or lake). So I think don't go for ground effect until you are getting down towards that limit.
      I should have noted in what I wrote above that in a glider on aerotow you are going much faster than best glide speed, so in fact on a rope break you pull the nose up rather than down, and can complete that 180 turn still close to 200 feet altitude. Also since you are turning back to the wrong end of the active runway, you would land alongside it on the grass, not on it, so don't need to turn more than 180.
    1. avallillo's Avatar
      avallillo -
      In an engine failure on takeoff situation, our objective is to save our lives, and save the airplane as much as possible, but IN THAT ODER! Many crashes after an engine failure on takeoff were due to the pilot trying (consciously or not) to save the airplane along with the people on board. First objective is maintain aircraft control, to prevent an out-of-control crash. Modern light airplanes are very good at preserving their contents relatively intact even in pretty severe impacts. The best way to lose control is to stall at low altitude. If you can avoid stalling you will probably survive.

      The straight ahead advice we used to get must be taken with a grain of salt. I remember a video on YouTube of a first flight in a homebuilt. The poor fellow lost his engine and proceeded straight ahead, right into the only tree anywhere near the area! He survived, but a heading change of even 10 degrees would have made for a better outcome. As you gain altitude your "window" of available landing spots widens considerably - by 300 feet or so you can probably get at least 60 degrees of heading change, and by around 1000 AGL you can turn back 180 degrees or a bit more if you want. You don't have to land straight ahead unless A: you are very low, or B: straight ahead is the best choice for other reasons.

      If you plan to make the turn back, gliding is typically not what you will be doing, at least not at "best glide" speed. The critical element is heading change - you want as much of that per foot of altitude as you can get. The way to change heading quickly is by putting as much of the lift vector off vertical as you can get away with, and put it to work pulling your nose around the turn. Steep bank, in other words. And G's. Now if you want to pull a few G's you need speed, significantly more of it than "best glide". You have altitude that you are willing to trade for heading change, and by lowering the nose well below best glide attitude you will ensure enough speed to pull a few G's without stalling. that is what might make the "impossible turn" possible

      As for diving into ground effect, definitely not. As you can see, that is not the reason we lower the nose in the impossible turn. Ground effect can, however, get you a few more feet of distance out of the very bottom of the glide, particularly if you delay flap deployment until the very end (around 100 AGL). But other than this one instance, you cannot manufacture distance by raising the nose in the glide.

      One last trick for getting the max performance in a glide - if you have a constant speed prop and enough windmilling oil pressure to do this, pull the prop knob full aft. This will remove a not inconsiderable amount of prop disk drag and extend the glide a bit. You can feel it. Likewise, if you need more drag, put the prop knob full forward. This is not in Cessna or Piper POH's, but it really works.
  • Recent Forum Activity

    Downwind66

    And then there was the Vought Corsair!

    Thread Starter: Downwind66

    What an aircraft! Rick :cool:

    Last Post By: DAVIDSTRAKA Today, 09:15 PM Go to last post
    Rupert

    Evening T&G and near Liverpool

    Thread Starter: Rupert

    Thought I'd do a little flying in the early evening in and around Liverpool.The weather was perfect! Sweet Baron is so smooth to fly. Click on...

    Last Post By: DAVIDSTRAKA Today, 09:14 PM Go to last post
    Raven214

    Known add-ons?

    Thread Starter: Raven214

    Has anyone compiled a list of known add-ons planned/in process for MSFS2020? I'm trying my best to keep myself occupied for the next 12 days, as I'm...

    Last Post By: PEIRascal Today, 08:08 PM Go to last post
    Elvensmith

    Simvation - Is The Site Safe?

    Thread Starter: Elvensmith

    I found a couple of helicopters I wanted to download via Google with links to the Simvation site. However when I clicked on the link I got a message...

    Last Post By: ftldave Today, 07:13 PM Go to last post