Impossible Turn Tested at Altitude

Jaybird180 said:
Ask the dead pilot that crashed into the house in Chicago.
Click to expand...

I didn't say it was a guaranteed survival, I said your odds are better than trying a low energy turn back. This is aviation, sometimes you're going to die no matter what.
 
I did it at 600' AGL in my CTSW. Total non-event with 200' under me as I came over the numbers. That said, I know my engine and prop are making a little bit of thrust at idle and I'd not try a return with less than 600' if the engine were stopped.
 
As someone who accidentally killed his engine on a go around, I am here to say that at least two seconds is about right. I was surprised how long it took me to realize that the throttle was not working right.
 
ChrisK said:
As someone who accidentally killed his engine on a go around, I am here to say that at least two seconds is about right. I was surprised how long it took me to realize that the throttle was not working right.
Click to expand...

I would like to hear this story.
 
ChrisK said:
As someone who accidentally killed his engine on a go around, I am here to say that at least two seconds is about right. I was surprised how long it took me to realize that the throttle was not working right.
Click to expand...

Did you time it, or did it seem like 2 seconds? Because in these situations your perception of time is likely not accurate.
 
Henning said:
Did you time it, or did it seem like 2 seconds? Because in these situations your perception of time is likely not accurate.
Click to expand...

This is an interesting point.
There was a study that showed the brains ability to process information increases when stressed.

I vaguely recall the test it was something like this:
They gave people a large digital watch that flashed numbers by in a sequence over and over and had people try and determine the sequence. They were flashing fast and they were not able to get the sequence by watching it.

Then they shoved them out of an airplane while wearing the watch.
On the way down they were to try and determine the sequence. Everyone improved many times over.

After watching this, I decided that just like phones, CD, ears have sampling rates (16k, 22k, 44k) I think all of our senses do and our brain as well. It takes a snapshot of the data 1000s of times per second rather than fluid incoming stream. Speed the brain up (Adrenaline) and Time slows down which is what you hear all the time.

I was hit by a car once and the process of seeing the car coming at me until I was on the ground felt like about 2 minutes.

I also think this is why time goes by really fast when you get hypoxia the sampling rate slows down but life happens at the same speed.

Makes me think when you remember an event are you pulling up 100s o snap shots and loading them into your mental media player to assemble them?
 
It's probably more like processor management. You always have all these processes happening in the brain running in the background using up processor resources slowing the process of the 'open page'. When things are going wrong putting us at risk, the brain suddenly shuts down all unnecessary processes and clears the processor and boosts the speed of the primary process.
 
I lost several cylinders once on departure and had to make an emergency return. IMO no matter how many times you train for it when it happens for real there is always a period of surprise. In the case of partial power the surprise takes a little longer to kick in since there is no obvious event. My wife actually said, "Is something wrong with the engine?", before I fully realized I had a problem.

The only way to truly find out how we would react in training would be to have someone unexpectedly reach over and pull the power when you least expect it and even then you would instantly know the drill. Of course we don't do that because people would die. Kind of ironic that we advise people to do something we are scared to train for because it could easily kill us even in a controlled training environment.
 
Alexb2000 said:
IMO no matter how many times you train for it when it happens for real there is always a period of surprise...
Click to expand...
That's not just your opinion, it's recognized and engine-out procedures are developed using that assumption. Time compression is usually considered to be a factor *after* a critical condition is identified, *recognition* of the condition is always assumed to have some delay. The Navy flight test standard for engine failure recognition when testing for Vmca(dynamic) is no recovery inputs until one of the following is reached: 1 sec elapsed since throttle chop, 20 deg of bank angle change, or a predetermined sideslip limit is reached. I recall using between 2-4 sec for some special cases in the past but in some of those cases stall AOA was reached before Vmc.

Nauga,
the standard bearer
 
If you're going to play these silly games, don't forget that the 180 degree "impossible turn" is comprised of more than 180 degrees of heading change - it takes two turns to get back to and lined up with the departure runway totaling somewhere around 240+ degrees of heading change.

I wouldn't leave out the "confusion delay", they put it in there for the guys doing V1 cuts in their jets. Is it going to be any less of a surprise to you than it is to the pros when it happens?

You mentioned half flaps so I'm assuming that your normal takeoff configuration is with half flaps, it not, you should practice using your normal takeoff configuration. Also, don't forget to retract your landing gear if you're flying something with retracts. (The point is to make this as realistic as possible.)

The problem with all of this is that it falls into the category of "Nice to know, but so what?" The collective wisdom of generations of pilots has labeled this turn impossible and for a reason. Of course, we can all cite examples of pilots who have pulled it off and for each of those, we can find examples of good pilots who weren't quite so lucky.

The safe, time proven, bet is to avoid the urge to make that turn back to the departure runway.
 
Last edited:
The delay factor is something that can be managed in your favor by your mindset and thought process taking off. If you brief for and expect an engine failure climbing and waiting for it to happen until you hit 700', you can minimize the reaction delay.
 
Dead Stick said:
If you're going to play these silly games, don't forget that the 180 degree "impossible turn" is comprised of more than 180 degrees of heading change - it takes two turns to get back to and lined up with the departure runway totaling somewhere around 210 degrees of heading change.

I wouldn't leave out the "confusion delay", they put it in there for the guys doing V1 cuts in their jets. Is it going to be any less of a surprise to you than it is to the pros when it happens?

You mentioned half flaps so I'm assuming that your normal takeoff configuration is with half flaps, it not, you should practice using your normal takeoff configuration. Also, don't forget to retract your landing gear if you're flying something with retracts. (The point is to make this as realistic as possible.)

The problem with all of this is that it falls into the category of "Nice to know, but so what?" The collective wisdom of generations of pilots has labeled this turn impossible and for a reason. Of course, we can all cite examples of pilots who have pulled it off and for each of those, we can find examples of good pilots who weren't quite so lucky.

The safe, time proven bet, is to avoid the urge to make that turn back to the departure runway.
Click to expand...


Right. That's what I did (half flaps, climb at Vy). The point of the exercize was well one, to go flying and do something but also find out what my min was.

People throw around 1000 feet AGL. I now know that would have killed me.
I had to do 180, then an additional 45 then an additional 45 back to line up.

I think I did everything the way I was supposed to.
I really don't think it was silly to do this.
 
SixPapaCharlie said:
Right. That's what I did (half flaps, climb at Vy). The point of the exercize was well one, to go flying and do something but also find out what my min was.

People throw around 1000 feet AGL. I now know that would have killed me.
I had to do 180, then an additional 45 then an additional 45 back to line up.

I think I did everything the way I was supposed to.
I really don't think it was silly to do this.
Click to expand...

try wearing a clown nose while doing this, then it would be silly. :lol:
 
Purely a thought exercise, since the dead don't talk, but...

I wonder if you could ask, let's say, 100 or 1,000 pilots who failed to make the impossible turn, "What were you thinking?" if a pattern would result.

Most popular?

"I had tried it so many times at altitude, I could have SWORN I could do it."

"I totally KNEW I would never attempt it, but something clicked and the siren song of the runway took over."

"I dunno. Just seemed like a good idea at the time."

Any other candidates?

And does anyone know of any account of a failed impossible turn survivor? That might be interesting, though probably not many of those around.
 
Dead Stick said:
It is a good exercise, but if it leads someone to attempt the impossible turn then it is a fool's errand.
Click to expand...

I agree.

realistically, my entire state is a huge field but I kinda always had that 1000 ft rule in my head. And I probably could do it in 1k ft but I have to learn how to do it. Until I do, my limit is 1800 ft.
 
Henning said:
The delay factor is something that can be managed in your favor by your mindset and thought process taking off. If you brief for and expect an engine failure climbing and waiting for it to happen until you hit 700', you can minimize the reaction delay.
Click to expand...
In jet's it's briefed for each and every takeoff. It still is a factor and must be considered. In fact, a few years back, the delay built into V1 cuts was increased by a couple of seconds if I recall correctly.
 
Have you not tried it in the pattern at different altitudes on a quiet day?
 
SixPapaCharlie said:
I agree.

realistically, my entire state is a huge field but I kinda always had that 1000 ft rule in my head. And I probably could do it in 1k ft but I have to learn how to do it. Until I do, my limit is 1800 ft.
Click to expand...
This is a tender subject for me. I lost a good friend trying to make it back to the airport in his homebuilt. The guys was as good of a stick as they come. (Flew airshow routines in a T-28 and held a low altitude aerobatic waiver.) It didn't keep him from making a bad decision that continues to reverberate through his family 25+ years later.

You've got to have the mindset that if that engine quits on you during certain windows of exposure the title has passed on to the airplane's new owner - the insurance company - and your only concern from that moment forward should be to simply be able to walk away from it after everything has come to a full and complete stop.
 
Sometimes there aren't any good options straight ahead... KCDW [link] RWs 4 and 28 aren't looking good for straight ahead, 22 is questionable, 10 is rarely used.

What would you do here?
 
FastEddieB said:
Purely a thought exercise, since the dead don't talk, but...

I wonder if you could ask, let's say, 100 or 1,000 pilots who failed to make the impossible turn, "What were you thinking?" if a pattern would result.
Click to expand...

Right on. The theory of the turnback is solid.. slow down near stall speed and whip it around. The reality, though, is hundreds of destroyed planes and dead pilots.
 
igorek82 said:
Sometimes there aren't any good options straight ahead... KCDW [link] RWs 4 and 28 aren't looking good for straight ahead, 22 is questionable, 10 is rarely used.

What would you do here?
Click to expand...
Nobody is forced to use any given airport. Just saying.

You can play the odds all you want, but a smart pilot always hedges his bets.
 
Not that there's any way of knowing, but I wonder what the historical percent chance of success is on <1000 ft. turnback attempts. I know a lot of pilots make it just fine, but many... don't.
 
igorek82 said:
Sometimes there aren't any good options straight ahead... KCDW [link] RWs 4 and 28 aren't looking good for straight ahead, 22 is questionable, 10 is rarely used.

What would you do here?
Click to expand...

:confused: There's thick stands of fluffy trees everywhere, way, way, way, better off going for the tree tops under control just above stall. People walk away from tree top landings all the time, I'd even put $100 on more people walk away from tree landings than die in them I'd say less than 5% of turnarounds from under 400' are successful. Nothing says straight ahead, a 120° arc is no less safe.
 
Last edited:
I have only attempted the turn in two different airplanes, and all the attempts, save two, ended up with a successful landing on the same runway from which I departed.
The successfull were at an altitude of 500'. Some with crosswind, some with wind straight down the runway, and others with no wind at all.
The airplanes used were a Cherokee 140, and a Cessna 150. Me and CFI in both. Cherokee with full fuel, and C-150 about 3/4 full.
The two unsuccessful attempts ended with me landing on the crosswind runway once, and the instructor telling me that we just crashed, and to apply full power.
My instructor never had me to practice at altitude, but only on takeoff, under actuall conditions, except the engine is idling, rather than dead.
proper pre-flight planning, has a great effect on how one handles such a manuver. An engine failure on takeoff should not be the "un-expected event", but should be planned for, as an "It's gonna happen" kind of thing. So you'll know well in advance what you're gonna do and when. Then, there's only the <1sec. that Henning speaks of, to realize, then react.
But, you have to practice, practice, practice. And with an instructor who can teach the proper method. Practicing at "safe" altitude will give you insight as to how the airplane may behave, but not the visual cues, or the sense of urgency.
I wouldn't attempt a turn back in a Mooney from 500', but I'm not saying it can't be done, But I know I can do it in a pa28, and a c150. I would think that a 172 could be done as well, with similar loading.
 
bluerooster said:
I have only attempted the turn in two different airplanes, and all the attempts, save two, ended up with a successful landing on the same runway from which I departed.
The successfull were at an altitude of 500'. Some with crosswind, some with wind straight down the runway, and others with no wind at all.
The airplanes used were a Cherokee 140, and a Cessna 150. Me and CFI in both. Cherokee with full fuel, and C-150 about 3/4 full.
The two unsuccessful attempts ended with me landing on the crosswind runway once, and the instructor telling me that we just crashed, and to apply full power.
My instructor never had me to practice at altitude, but only on takeoff, under actuall conditions, except the engine is idling, rather than dead.
proper pre-flight planning, has a great effect on how one handles such a manuver. An engine failure on takeoff should not be the "un-expected event", but should be planned for, as an "It's gonna happen" kind of thing. So you'll know well in advance what you're gonna do and when. Then, there's only the <1sec. that Henning speaks of, to realize, then react.
But, you have to practice, practice, practice. And with an instructor who can teach the proper method. Practicing at "safe" altitude will give you insight as to how the airplane may behave, but not the visual cues, or the sense of urgency.
I wouldn't attempt a turn back in a Mooney from 500', but I'm not saying it can't be done, But I know I can do it in a pa28, and a c150. I would think that a 172 could be done as well, with similar loading.
Click to expand...
The problem isn't that it can't be done - it certainly can, but perhaps we ought to rename this maneuver the "improbable turn". The problem is being able to recognize when you are "in the window" where you can do it - the point where are you in a place (altitude, distance from runway, etc.) where it is even possible. The problem most pilots have is that they simply have no idea where that window is and how dramatically it can change from flight to flight so they end up guessing or coming up with some arbitrary number like 300' AGL or 1000' AGL or 1500' AGL whatever. Where the wind is coming from is also a big factor regardless of whether it's a LH or RH pattern, which way would you turn after an engine loss on this particular takeoff, left or right? Without that understanding and the discipline to apply it, it is a total crapshoot. Some guys are lucky, some aren't. Even when you know where the window is, doesn't mean that you'll be in it when the engine calls it a day. For many of us, the safest advice is to limit our turns to 45 degrees or so of runway heading, take what your get and fly the airplane until it stops moving and unfortunately the statistics bear that out.

I've found that the more decisions that I can make on the ground, in the safety and comfort of my leather recliner, the better I off I usually am. When it comes to the "impossible turn" I think you need to ask yourself what would constitute a successful outcome when it's your turn in the barrel. The answer you give to that question will determine the approach you take to attempting the "improbable turn". If you define a successful outcome as being able to walk away, under your own power, from the insurance company's newly purchased airplane you'll have a different view of the 180 degree turn. If success to you is making it back to the runway, then that 180 degree turn is more important. The time to try and figure all of this out is not immediately after your engine has quit. It's when you're under absolutely no pressure, in the comfort of your easy chair. You should also reaffirm your decision immediately prior to each and every takeoff.
 
Alexb2000 said:
The only way to truly find out how we would react in training would be to have someone unexpectedly reach over and pull the power when you least expect it and even then you would instantly know the drill. Of course we don't do that because people would die. Kind of ironic that we advise people to do something we are scared to train for because it could easily kill us even in a controlled training environment.
Click to expand...

This is my point. We don't train realistically.
Back in the '60s, during the Vietnam war, I trained initial Army pilots in multi-engine in a B-55 Baron.
We routinely practiced an engine out during T.O. at 3 critical points:
First, right before Vr. Second, right after lift-off before gear up with an abort and landing.
Third, after gear up/ no remaining runway with clean up and continue flight.
We did it because the student was going to get one of those scenarios on his checkride.

And we didn't do it by reaching up and pulling a throttle back.
I held my left hand over the mixture controls while wiggling my right hand under my left hand, and snapping the mixture back when it was appropriate.
Sometimes, I/we would sneak a fuel valve off while the student was completing pre-takeoff checks.
We never had any accidents because of it, because we were trained.
Of course, it was an up-front drill with plenty of briefing and dry cockpit practice at first, but when the student gained basic proficiency, we were continued to vary the sequence and timing to be as realistic as possible.
It was going to be on his checkride that way.

It was after this Army experience that I learned that civil pilots don't do such things, but it is because you can't get there with quickie 5 or 10 hour courses.
Well, maybe some could, but the community is convinced it would be too dangerous.
Like spins.
I also taught Primary in the L-19 Birddog and proficiency in a precision 3-turn spin was required before solo, and that was also going to be on his checkride.
 
SixPapaCharlie said:
I would like to hear this story.
Click to expand...
There is honestly not much to it. I finished my tailwheel training in a cub over the summer.
cub-kovacs.jpg

My young instructor liked to make sure that I got a turn hand propping the aircraft, but he always flew from the back and I flew from the front (until I was signed off and then we switched seats). When I hand propped, we used the "solo key" which can operate the mags from the back:

solo_key.jpg


On one of my final landings before getting signed off, I decided I was not happy with my sink rate, so I elected to go around. In the cub, one holds the throttle with the left hand and the stick with the right. I was accustomed to reaching with my left hand over to the carb heat to push it in and move it quickly back to the throttle on a go-around. So, left hand to left side of cabin, carb heat in, left hand quickly back to throttle. Carb heat, throttle, Carb heat, throttle.

inside2+017.jpg


CubThrottle.jpg


Halfway through the process, I bumped the key. Remember the key? Go look at it again.

Henning said:
Did you time it, or did it seem like 2 seconds? Because in these situations your perception of time is likely not accurate.
Click to expand...

Upon reflection after the flight, my instructor and I figured we were hanging above the runway at ~40 MPH with our speed decaying waiting for the power to come in, which never came. It is of course possible that it wasn't 2 seconds. It seemed like 30 seconds...

My instructor yelled "MY AIRPLANE!!!!", pointed the nose down, and pulled up at the last second, bouncing it on the runway. We stopped at about 45° to the runway. I got out and hand propped it again and we went around a couple more times.

He signed me off shortly after that and threw the key in the trash.
 
Henning said:
The turn is best made right above stall speed.
Click to expand...

True - But not the one you're thinking of. Minimum turn radius is just above stall speed *for the bank angle used*. So if we look at the common C172N (Vy= 73 KCAS, Vg = 66 KCAS, Vs1 = 50 KCAS), If we're starting at Vy, we're at 1.46Vs1, a 60-degree bank will have us stalling at 1.414Vs1 or 71 knots. As long as you can manage to lose less than two knots in the recognition phase, you can crank her around right at that speed - Though since you'll be slowing down, you'll need to either allow some downward acceleration (ie don't pull quite so hard, which will work against your turn radius), or lessen the bank angle through the course of the turn.

The only method I found that I could reproduce a <250' loss result was was to immediately pull for 3gs to transfer any excess speed I had into ballistic energy
Click to expand...

Hmmm.

On the Citabria, you have Vs1 = 58 mph CAS, Vg = 70 mph CAS, Vy = 74 mph CAS.

So if you're climbing out at Vy (74) and you pull 3G, your stall speed will be 100 mph CAS, 26mph faster than you're going.

Something tells me this approach is not optimum. :no::nono: I'm sure it's fun, but it's definitely not optimum. The most G's you can pull at Vy in a Citabria without stalling it is about 1.63.
 
flyingcheesehead said:
True - But not the one you're thinking of. Minimum turn radius is just above stall speed *for the bank angle used*. So if we look at the common C172N (Vy= 73 KCAS, Vg = 66 KCAS, Vs1 = 50 KCAS), If we're starting at Vy, we're at 1.46Vs1, a 60-degree bank will have us stalling at 1.414Vs1 or 71 knots. As long as you can manage to lose less than two knots in the recognition phase, you can crank her around right at that speed - Though since you'll be slowing down, you'll need to either allow some downward acceleration (ie don't pull quite so hard, which will work against your turn radius), or lessen the bank angle through the course of the turn.



Hmmm.

On the Citabria, you have Vs1 = 58 mph CAS, Vg = 70 mph CAS, Vy = 74 mph CAS.

So if you're climbing out at Vy (74) and you pull 3G, your stall speed will be 100 mph CAS, 26mph faster than you're going.

Something tells me this approach is not optimum. :no::nono: I'm sure it's fun, but it's definitely not optimum. The most G's you can pull at Vy in a Citabria without stalling it is about 1.63.
Click to expand...

You do stall, and start a spin, that's why you have to immediately unload the wing pretty much immediately. You're yanking back as hard as you can to transfer the energy impact to the highest ballistic arc you can, then unload the wing to reduce stall speed and kick the tail around, you spin if you don't unload enough, if you do it just right the wing drop takes you through 90° before you start losing altitude, then you kick in the opposite rudder as you start accelerating and pull out best that you can without a secondary stall. You aren't 'flying' from the time you yank to the time you recover, and it's sketchy because you have absolutely no spare energy.

That was the point of the whole exercise and came about after a discussion of the modern 'tumbling' aerobatic routines that use ballistic inertia to suspend them where aerodynamics wasn't that were starting to come about at the time. I wondered out loud if that could be applied to an "impossible turn" scenario, and Jack being the CFI he was said, "Let's go find out."

The result was "Yes it could, but it would be nuts to try."
 
bluerooster said:
I wouldn't attempt a turn back in a Mooney from 500', but I'm not saying it can't be done, But I know I can do it in a pa28, and a c150. I would think that a 172 could be done as well, with similar loading.
Click to expand...

Au contraire - I'd rather do it in a Mooney. Retractable gear = better glide ratio. This guy successfully did it in his Mooney, even with sub-optimal execution:


The engine failed about 26 seconds after takeoff in an M20C, so almost certainly below 500 AGL.
 
flyingcheesehead said:
Au contraire - I'd rather do it in a Mooney. Retractable gear = better glide ratio. This guy successfully did it in his Mooney, even with sub-optimal execution:


The engine failed about 26 seconds after takeoff in an M20C, so almost certainly below 500 AGL.
Click to expand...

:confused: The Mooneys I flew typically beat 1200-1500fpm on takeoff. I imagine the big bores do better.
 
nosehair said:
This is my point. We don't train realistically.
Back in the '60s, during the Vietnam war, I trained initial Army pilots in multi-engine in a B-55 Baron.
We routinely practiced an engine out during T.O. at 3 critical points:
First, right before Vr. Second, right after lift-off before gear up with an abort and landing.
Third, after gear up/ no remaining runway with clean up and continue flight.
We did it because the student was going to get one of those scenarios on his checkride.

And we didn't do it by reaching up and pulling a throttle back.
I held my left hand over the mixture controls while wiggling my right hand under my left hand, and snapping the mixture back when it was appropriate.
Sometimes, I/we would sneak a fuel valve off while the student was completing pre-takeoff checks.
We never had any accidents because of it, because we were trained.
Of course, it was an up-front drill with plenty of briefing and dry cockpit practice at first, but when the student gained basic proficiency, we were continued to vary the sequence and timing to be as realistic as possible.
It was going to be on his checkride that way.

It was after this Army experience that I learned that civil pilots don't do such things, but it is because you can't get there with quickie 5 or 10 hour courses.
Well, maybe some could, but the community is convinced it would be too dangerous.
Like spins.
I also taught Primary in the L-19 Birddog and proficiency in a precision 3-turn spin was required before solo, and that was also going to be on his checkride.
Click to expand...

IMO that's exactly what we need to be doing today. Thanks for sharing.
 
You must log in or register to reply here.
Back
Top