Impossible turn may be possible after all

flhrci

flhrci

Final Approach
Joined
Jan 26, 2007
Messages
5,932
Location
Groveport, OH
Display Name
Display name:
David
I am doing my CFI renewal, ughhhh, through AOPA ASF and one of the safety trends slides has a video about the impossible turn. There is a maneuver to safely practice it and see what you and your aircraft are possibly capable of during an engine out.

This is new to me and I apparently missed it on AOPA Live when it was broadcast.

Watch for yourself and weigh in on what you think. Video works without beign a member of AOPA as far as I can tell.

http://www.aopa.org/AOPA-Live.aspx?watch={C2FBA3D9-09B3-4B4D-BD46-3DE691D77852}

David
 
Interesting. Wish I would of spent some time with my students performing this maneuver. I remember discussing the dangers of the impossible turn but I don't think we actually tried it.
 
It's only impossible if you lack the energy (speed/altitude) to complete it. The difficult part is knowing just where that line is in the aircraft and circumstances when it happens. And that's why Dave Rogers called his popular article on the subject "The Possible 'Impossible' Turn".
ftp://208.201.249.240/pub/users/joec/airplaneImpossibleTurn.pdf
Kinda like what Captain Tupolev's Chief Engineer told him about going to 105% on the reactor -- "Possible, but not recommended."

Make sure you have practiced this and know the minimum parameters for execution in your plane if you want it to be "possible", since crashing under control is usually more survivable than crashing out of control.
 
Last edited:
Even the CFI on board says he's going to stick with the 1000' rule. :dunno:

First attempt, knowing it was going to happen she lost 500', after 4 attempts she got it down to 300'.

Can it be done? Certainly, but a lot of people die trying.
 
Geico266 said:
Even the CFI on board says he's going to stick with the 1000' rule. :dunno:
Click to expand...
Unless you're way below any speed you should be going at 1000 AGL, that should work to prevent running out of air before you complete the maneuver for most any light single in most any wind condition. Further, since there are probably quite a few aircraft/conditions where it's beyond the average pilot's skill or the aircraft's capability below 1000 AGL, it's not a bad rule to use.
 
My assumption is that the 4-5 second delay between declaring engine lost and maneuvering the airplane in reaction to it is to simulate confusion, denial and similar reactions by the pilot. That may be valid if testing the general pilot population. But, would you hold a 4-5 second constant pitch after you lost your engine? I doubt if anyone reading this would do so. I would not. I would immediately pitch for best glide. One way to help oneself execute this immediate action drill is to brief the takeoff and departure, something I've seen in charter flying but seldom see in general pleasure pilots. A simple take-off brief may start as, "any problem on the ground we stay on the ground, any problem below XXXX feet we land straight ahead (modify that depending on the airport), any problem above XXXX feet we (complete to suit your situation). That was a sample only, your version may vary. The point is, we have prebriefed and "cocked the gun" to execute an action so that we are less likely to be a deer in the headlights for 4-5 seconds.

So, I wonder if a trained and even the simply aware pilot need not factor in the 4-5 second delay, especially if pre-briefed. There is absolutely nothing to be gained by an artificial delay in initiating best glide.

My second observation is that having demonstrated that the airplane would do the turn in 300' (even with the delay), then rationalizing to set the floor at 1,000', I think the discussion left the door open to further rationalization (alluded to in the video) of modification. Simply put, the pilot does a take-off from an airport where the departure is open and an off-airport landing straight ahead is feasible. The pilot knows there is little or no traffic, having just departed. The pilot has practiced a turn and accomplished it in say 400', but, following the advice of the video, has decided that a 1000 floor is better. My concern is that if this pilot lost an engine at 500 feet and had hopefully per-briefed, he would be powerfully motivated to execute the turn. That is a situation where I can see some delay in executing as the pilot goes through the "I'm not high enough" "yeah, but I did it before" dialogue in his mind.

So, perhaps we should be teaching ambush drill. Engine stops, immediate reaction is to establish best glide or flare for landing, whichever comes first. Next, maybe we should teach a hard floor. Gliders train to and do the turn at 200' - it's in the PTS. Maybe it should be added to the PTS so airplanes should do it per PTS at 800', 400', 1000' (whatever) and simply train to that and nothing else. That removes the questioning, which, we see, can be dangerous because it can add back in the time delay we trained to eliminate.
 
Going to go out and try it a few times,I think the1000 ft rule will work great for me.
 
I had a CFI that said his personal cut off point was about 800 agl for a turn around, but stressed it was his comfort level, and it may not be mine.
 
frfly172 said:
Going to go out and try it a few times,I think the1000 ft rule will work great for me.
Click to expand...

Your height at the time of engine failure is not the only issue. Depending on the length of the runway, if your climb angle is less than your glide angle, you may not make it back to the airport.
 
Wind is a big factor, too. Headwind on takeoff helps.... up to a point. Lots of variables. Heck, dude taking off behind you might be a factor. I seldom see that mentioned but add it to the list.
 
As a glider pilot and having practiced this maneuver I can see a few flaws in the maneuver they presented.

1. It isn't a 180 degree turn. It is normally a 225 degree turn followed by a 45 degree turn in the opposite direction. because you have to line up with the runway. To practice it properly you need to line up on a road and be lined up on the road when you finish the maneuver.

2. Not sure why they are holding attitude for the 5 count when the engine fails. The airplane is naturally going to enter a descending attitude, unless the pilot intentionally holds the nose up.

3.It totally ignores this issue of low level maneuvering. Last Months Soaring Magazine has an excellent article on the effects of Low Level Maneuver, especially when going downwind, i.e. the dreaded downwind turn. A couple paraphrased snippets from survivors of downwind turns, landings and maneuvering...
... It felt like the leans in instrument flying, if I wasn't constantly watching the Airspeed indicator it would start slowing down
... It felt like the elevator was stuck, I had to push exceptionally hard to put the nose down this close to the ground, The subconscious is a powerful thing.


In practicing these I have come to two conclusions...

1. If you haven't practiced turning back to a runway in an airplane similar to what you flying, don't try it in an emergency.

2. If you aren't already turning for your normal crosswind turn when the engine fails, then land straight ahead. (This can be airplane dependent, refer to conclusion #1)

Brian
CFIIG/ASEL
 
noobJohn said:
Wind is a big factor, too. Headwind on takeoff helps.... up to a point. Lots of variables. Heck, dude taking off behind you might be a factor. I seldom see that mentioned but add it to the list.
Click to expand...

Climb and descent angles can only be determined once the wind is known.
 
Runway configurations are also a factor...where I fly, there are two parallel runways, so the "impossible turn" is actually slightly easier to make...you turn towards the other runway and land on the parallel runway, without have to turn all the way back to the original.

But, yeah, not a bad maneuver to practice at altitude...climb in takeoff configuration, pull the engine, and see how much altitude you've lost when you can get straightened out going the other way. Practice it in a variety of wind conditions and get a feel for your and your plane's capabilities...with the added benefit that if it actually happens, you'll have less of an "Oh ****, now what do I do" because you've done it a bunch of times already.
 
I just think these sort of discussions are ridiculous and they just go on and on about the same things over and over. There is nothing impossible about turning back but there are so many different variables that make it silly to try and prefabricate a decision on whether or not to try it. There's nothing wrong with practicing such a maneuver but the fact is that the major cause of fatality in an engine out scenario is a stall/spin incident and not an off airport landing or even a crash landing if that's what it needs to be called. So obviously trying to perform a low level, engine out tight turning maneuver is going to increase the risk of that happening. That's the bottom line, period.
 
Ron Levy said:
I... Kinda like what Captain Tupolev's Chief Engineer told him about going to 105% on the reactor -- "Possible, but not recommended."...
Click to expand...

In the book they have a reactor meltdown.

That scene for the movie is likely on the cutting room floor.
 
My personal limit is turning back only if 1000 ft AGL, any less and it's a controlled crash into the nearest "suitable" landing area.
 
Not a case of the impossible turn, but I recall seeing a video awhile ago of Captain Sullenberger speaking (maybe at Oshkosh?) where he mentioned pushing forward on the yoke after the bird strike since airspeed had decayed.

It struck me because when I've practiced engine outs in a C152, it's always been at a high enough airspeed that I've needed to pull back and bleed off airspeed to get to to best glide 60 KIAS. Even Vy is 72 KIAS, so in a practice situation where I'm expecting an engine out (and even have the additional warning of my instructor leaning over to grab the throttle) I immediately pull back to get to best glide speed.

In a real emergency at low altitude climbing at Vx (54 KIAS), or even 72 KIAS plus a few seconds to react, I may end up below 60 KIAS by the time I react and will need to push forward , which goes against just about every engine out I've practiced. When I do my pretakeoff brief, I always take a moment to remind myself that.

Thanks for sharing. Will ask my CFI to practice these on my next flight.
 
Silvaire said:
I just think these sort of discussions are ridiculous and they just go on and on about the same things over and over...
Click to expand...

A lot of discussions are like that around here!
 
Lots and lots of variables. Am I going to turn around when there is a suitable field in front of me? Probably not. Am I going to do the whole 270 when I can land on the grass next to the airport? Probably not.

There is a lot more to this than altitude and can I do it. Way safer to land in a field then try such a maneuver without altitude to spare. One thing I ALWAYS do is brief the takeoff. Admittedly, there have been takeoffs where the safest thing would have been to turn around no matter the altitude. The closer to the airport I could get the less chance banging into something solid and unfriendly. But one should always fly the airplane. Better a controlled crash than an uncontrolled one.
 
JeffDG said:
Nope, in the book Red October rams them.
Click to expand...
I'm pretty sure that in both the book and the movie, Captain Tupolev's Alfa is sunk by its own torpedo launched at Red October and skillfully maneuvered back into Tupolev's face by the wily Captain Ramius in an undersea game of "chicken".
 
steingar said:
Lots and lots of variables. Am I going to turn around when there is a suitable field in front of me? Probably not. Am I going to do the whole 270 when I can land on the grass next to the airport? Probably not.
Click to expand...
For those wondering about steingar's geometry, that's 225 one way followed by 45 back the other. 225+45=270 degrees of turning. See Prof. Rogers' article linked above for more about that.
There is a lot more to this than altitude and can I do it. Way safer to land in a field then try such a maneuver without altitude to spare. One thing I ALWAYS do is brief the takeoff. Admittedly, there have been takeoffs where the safest thing would have been to turn around no matter the altitude. The closer to the airport I could get the less chance banging into something solid and unfriendly. But one should always fly the airplane. Better a controlled crash than an uncontrolled one.
Click to expand...
Agreed completely.
 
If your engine has a catastrophic failure on takeoff, and the aircraft is totaled in the off-airport landing will insurance pay for the hull value, or the hull value minus an engine?

After all, they do not insure against mechanical failure.
 
dans2992 said:
If your engine has a catastrophic failure on takeoff, and the aircraft is totaled in the off-airport landing will insurance pay for the hull value, or the hull value minus an engine?

After all, they do not insure against mechanical failure.
Click to expand...
If the airplane is totaled, they pay the agreed value on the policy. The only time they worry about what part is mechanical failure and what part is ensuing damage is when the airplane is not totaled. In that case, they'd take off the pro-rated value of the engine based on TBO.

Let's say the total repair bill including the engine was $50K, and the plane is worth enough that it isn't "totaled". The engine overhaul is $20K, and we'll say it had 1000 hours on a 2000 hour engine. They only pay $10K for the engine part of the bill, so you pay $10K and they pay only $40K, since you will at the end have achieved "betterment", i.e., a zero-time engine where you formerly had a 1000-hour engine.
 
JeffDG said:
Runway configurations are also a factor...where I fly, there are two parallel runways, so the "impossible turn" is actually slightly easier to make...you turn towards the other runway and land on the parallel runway, without have to turn all the way back to the original..
Click to expand...
You know, that brings up a strategy that I've never considered. Many of the paved runways around here have full length parallel taxiways that are amply wide enough to land on. Many are far enough off the runway to reduce the required total turn by at least 45 degrees. My home base, H88, included.

I will, from this point forward, as I'm taxiing onto the runway for take-off, make a mental note of which way I need to turn should I have a engine failure at a high enough altitude to allow a return.
 
Seems like losing the engine on takeoff rarely ever happens if you've already made it to at least 500' AGL or higher. They always tend to quit on takeoff before you've even made it to 200' AGL :hairraise:
 
Neal Howard said:
Seems like losing the engine on takeoff rarely ever happens if you've already made it to at least 500' AGL or higher. They always tend to quit on takeoff before you've even made it to 200' AGL :hairraise:
Click to expand...

Which is why I like doing short-field takeoffs on long strips- by the time you're out of usable runway, you're already passing 500AGL.
 
warthog1984 said:
Which is why I like doing short-field takeoffs on long strips- by the time you're out of usable runway, you're already passing 500AGL.
Click to expand...

Also why I like owning and flying an RV now. I'm already at or almost at pattern altitude by the end of the runway (without zoom climbing :D).

This NTSB report (http://www.ntsb.gov/aviationquery/brief2.aspx?ev_id=20031202X01976&ntsbno=FTW04LA028&akey=1) is a good friend of mine's crash. The night before the crash, I helped him perform a 10 minute long full power runup on the ground with the tail tied to a pickup truck to ensure the engine would make full power for long enough to get airborne, and that there were no leaks and that it would be ready for that first flight. The plane seemed like it was good to go, and was inspected and signed off by his AP/IA. He had assembled the fuel line that failed himself while working under the supervision of his AP/IA but apparently didn't use a mandrel to put the AN fittings on the ends, and there was the classic "rubber flap torn" inside the 303 hose (typical failure mode from not using a mandrel) which caused the blockage. Of course the loose flap of rubber inside the line stayed put during all the ground runups and decided to come loose somewhere during the takeoff roll, leaving just enough fuel in the carburetor bowl to barely get airborne.... Murphy's Law at work there for sure. His CFI buddy took the controls immediately after the engine quit and shoved the nose down and threw the plane into a steep right turn to try to make it back to one of our runways, else they'd have ended up into power lines trying to land straight ahead. We have a runway 13/31 (paved) and 04/22(turf) that form an X and he tried to get over to the turf runway 04 but didn't quite get there before running out of altitude and airspeed at about the same time, and with a sudden and strong tailwind, landed pretty hard nosewheel first in the rough plowed field between the runways. The nosewheel broke off and the plane bounced once back up into the air a few feet, and came back down onto the broken-off nose strut sticking out and flipped over hard. The airplane came to rest upside down, the engine and battery broke away from the firewall and were flung about 25 yards away, the whole aft fuselage ripped loose behind the back seat and was sitting almost 90 degrees to the side of what was left of the front cabin. Both fuel tanks were ruptured and avgas was all over the ground under the wreckage so it's a very good thing the engine and battery broke free and were flung too far away to ignite the gasoline. Both guys were dangling upside down by their shoulder harnesses and got out of the wreckage with cuts, bruises and concussions. My friend got some teeth knocked out and needed the back half of his scalp stitched back on, but they both pretty much walked away from what looked like it should've been a fatal crash. The Cessna cabin structure around the front seats held up with rollcage-like integrity but the rest of the plane was basically rolled up like a ball of crushed aluminum foil. I didn't witness the crash myself, but got to see the site up close in person several hours later after the authorities had left.
 
Last edited:
I have practiced turns at best glide to see for myself the altitude loss. The most effective way for me to lose the least amount of altitude, was to treat this maneuver like an instrument maneuver. I would focus on the attitude indicator and the airspeed indicator to keep the speed and bank angle spot on, only occasionally looking out the window. Using this method, it's surprising how little altitude you lose and if you have instrument training, how it's really not that big of a deal.

There's only one problem with my little scheme...

In the event of an actual total engine failure, my attitude indicator is likely to go tits up. Now I would have to focus my attention outside the cockpit and inside on the ASI. IMO, this is where pilots come to grief. Attention and focus goes on the outside and airspeed, the thing keeping you alive, gets ignored.

I hate to bring up this ol' saw again, but here is where I really think an AoA indicator would really help if it is installed correctly. In the interest of a clean, uncluttered panel, many pilots opt to put the AoA indicator in the panel, often next to the ASI. IMO, this is a big mistake. It needs to be on top of the glare shield and right in the pilot's view as they look out the window. Otherwise, it too will be ignored during this emergency turn.

I can't wait to get one installed and see how it helps.
 
Dav8or said:
I have practiced turns at best glide to see for myself the altitude loss.
Click to expand...
I suggest reading Prof. Rogers' paper again -- book best glide speed is way too slow for this maneuver. Vy is a lot closer for the maneuver done as he discusses.

I hate to bring up this ol' saw again, but here is where I really think an AoA indicator would really help if it is installed correctly.
Click to expand...
It would, indeed, provided you know the AoA for best glide, which does not change with g-load event though best glide speed does, and you will be somewhat g-loaded in this turnback maneuver.
 
I also think if you are going to use some hard AGL rule, you had better take into account those 100' plus trees that line the runway. That can really ruin an otherwise shaky plan to make it back to the runway. Or any other obstructions for that manner. A good reason to have a plan for engine silence every time you depart the ground.
 
Ron Levy said:
For those wondering about steingar's geometry, that's 225 one way followed by 45 back the other. 225+45=270 degrees of turning. See Prof. Rogers' article linked above for more about that.
Agreed completely.
Click to expand...

Actually its a 270* turn followed by an opposite 90* turn to return to the same runway. The reason for this is because you are on takeoff and not very far from the runway at all. Doing a 225* then a 45* will likely not leave you enough runway especially if you are talking about altitudes 500-1000ft. Again its all conditional on a LOT of things.
 
I don't know if I agree with those numbers. Neither turn seems like it would need to be a full 270 and then 90 degrees.... This isn't a pattern, it's getting as close as a straight line(as is possible) back to the runway as you can.
 
RBeaubien said:
Actually its a 270* turn followed by an opposite 90* turn to return to the same runway. The reason for this is because you are on takeoff and not very far from the runway at all. Doing a 225* then a 45* will likely not leave you enough runway especially if you are talking about altitudes 500-1000ft. Again its all conditional on a LOT of things.
Click to expand...
225/45 with a short straight between to line upwill lose less altitude than a 270/immediate 90. Prof. Rogers addressed this in one of his papers.
 
Ron Levy said:
I suggest reading Prof. Rogers' paper again -- book best glide speed is way too slow for this maneuver. Vy is a lot closer for the maneuver done as he discusses.
Click to expand...

I did just reread it and actually, if I read it correctly, he calls for the turn to be preformed at much less the best glide speed, not more. He describes this speed as Vturning. It is a calculation of stall speed clean, multiplied by 1.05 and then divided by cosine 0 or something. I am a failure at math, so I have no idea if I have stated this right, or how to do this calculation for my plane, but using his report and the numbers for their B33 Bonanza, it looks like this-

  • Clean stall = 72mph
  • Best Glide = 122mph
  • His calculated turning speed = 107mph
So the way his maneuver is achieved in this Bonanza is-

  • Climb out at best angle = 91mph
  • At engine failure, immediately turn to 45 degrees bank and accelerate to 107mph though out the turn.
  • Rolling out after 190-220 degrees heading change and accelerating to best glide of 122mph.
So, if I read it correctly, he does not advocate going faster than best glide, in fact the opposite. Personally, I think it's going to be tough to actually keep your mind set on 107mph like a hawk and then remember to go to 122mph in a crisis. I suspect the average private pilot is just better off focusing on one number, the number he already has ingrained in his memory rather than making more pitch changes and accept the altitude loss associated. Might not make the runway, but at least probably the airport.
 
Dav8or said:
I did just reread it and actually, if I read it correctly, he calls for the turn to be preformed at much less the best glide speed, not more. He describes this speed as Vturning. It is a calculation of stall speed clean, multiplied by 1.05 and then divided by cosine 0 or something. I am a failure at math, so I have no idea if I have stated this right, or how to do this calculation for my plane, but using his report and the numbers for their B33 Bonanza, it looks like this-

  • Clean stall = 72mph
  • Best Glide = 122mph
  • His calculated turning speed = 107mph
So the way his maneuver is achieved in this Bonanza is-

  • Climb out at best angle = 91mph
  • At engine failure, immediately turn to 45 degrees bank and accelerate to 107mph though out the turn.
  • Rolling out after 190-220 degrees heading change and accelerating to best glide of 122mph.
So, if I read it correctly, he does not advocate going faster than best glide, in fact the opposite. Personally, I think it's going to be tough to actually keep your mind set on 107mph like a hawk and then remember to go to 122mph in a crisis. I suspect the average private pilot is just better off focusing on one number, the number he already has ingrained in his memory rather than making more pitch changes and accept the altitude loss associated. Might not make the runway, but at least probably the airport.
Click to expand...
The numbers work that way for a very clean retractable like a Bonanza, but not for a fixed gear airplane. If you take a Cessna 172 with a clean stall of about 56 knots, and best glide of 65 knots, V(turning) comes to about 83 knots, which is well above best glide. Thus, in a 172, you'd make the turn at 83 knots, then decelerate to 65 knots for the level portion.

IOW, best glide speed is not a factor in computing V(turning), which is a function only of bank angle and stall speed. V(turning) may be greater or less than best glide (which is not a function of stall speed). So, best glide speed is irrelevant to V(turning), and Prof. Rogers never says otherwise.
 
You must log in or register to reply here.
Back
Top