Is 'the impossible turn' possible?

[Reluctant737]

Hi,

Excuse me if I'm being brief, I'm just on my way out the door, but this question has just popped up in my head -

I am wondering if anybody here knows if there's ever been a study into the so called impossible turn following an engine failure (SEP) after takeoff.

I guess what I'm looking for primarily is the relationship between varying performance figures of different aircraft. For example, given a Piper PA28, with a typical climb angle of, say, 5 degrees in still wind, if the pilot were a computer, able to perform 'to perfection', is the turn possible?

Of course there would be numerous variables, such as W/V, aircraft climb rate and speed (climb angle), L/D ratio, aircraft maneuvering performance, surrounding terrain etc.

Really it would be good to get different peoples' points of view on this and different experiences, first or second hand.

Obviously we all know that following an EFATO in an SEP, it's second nature to disregard the possibility of returning immediately, but yep, just curious,

Cheers. :ok:

[Old Smokey]

SOME people have got away with it, MOST DON'T!

The 3 biggest factors in being LUCKY (and I mean LUCKY) are -

(1) Very prompt turn-back without any waste of time,

(2) Takeoff into a strong Headwind, which becomes a strong Tailwind for the return, and (most importantly),

(3) Takeoff on a very long runway well in excess of your requirements, such that you can still make the "unused" runway beyond your Takeoff point.

You would need all 3, or at least 2 of these things going for you to contemplate a 180 degree turn-back. In all other circumstances (99% of the time), DON'T EVEN THINK ABOUT IT!

[Clandestino]

Even prompt turn won't save you if you're too low to complete 180° and alignment turns.

People who got away have probably flown SLMGs. With PA-28 or C-172, the impossible turn is really impossible.

[sapperkenno]

I know of two instances (both in Supercubs, after a loud bang and loss of power) where they successfully 180'd and landed, and both were at less than the recommended 500ft (land-ahead) height when they lost power. Both pilots were very experienced instructors, and knew (in hindsight, from what everybody is taught) that at that altitude they shouldn't have attempted such a maneuver, but it felt right at the time, and ended up with a safe outcome.

The FAA has some kind of graphical demonstration in their textbooks, showing that if you make a standard turn you end up 800' under the runway after doing a 180 below 500ft, or words/pictures to that effect. As a student pilot, they don't want you to have any idea that such a turn might be possible, and quite rightly, as the statistics show that this has led to a lot of accidents in the past. So, statistically it's safer not to try it, and that's why the FAA preach it that way. They don't like the idea of low altitude stall/spin accidents.

A quick google gave this http://jeremy.zawodny.com/flying/turnback.pdf (which hopefully closely answers your question) and there were loads of other results which may explain your answer, so have a very good look there as it seems like a rather in-depth study from a reputable source.

I think John Farley has done a piece on this somewhere (which I recall reading in Flyer mag a while ago), and mention of that is probably elsewhere on this site, as I'm sure it's been covered before, but I don't want to be the smart-ass pointing it out to you, and can't be bothered looking right now.

The only studies I know of, are the FAA's and the link I've given you, which I've only just found and haven't finished perusing yet, due to the off-putting formulae within the first few pages! I know of similar "studies" regarding canyon turns, and all the comparisons of rate/radius at different speeds with different flap settings, slipping/skidding turns and all kinds of weird and wonderful things, so that might be of value.

You could always try it yourself, at a safe height, but it sounds like you really want to get technical with it, so there would far be too many variables involved to satisfy such a need for perfection.

I'd say that it's definitely possible from the right altitude, and I'd hedge my bets that we'd all probably be surprised that this "magic altitude" for any particular SEP, on any particular day would be significantly lower than 500ft AGL.

[Fair_Weather_Flyer]

I had a moment myself where I managed to do a 180 and return in a 172, although the engine was still producing some power and that was enough!

I did see an FAA/AOPA DVD on this subject. They calculated that from about 700ft (can't remember the exact height) it was possible to make a 180 and land. The key was using the appropriate bank angle which was between 45 and 60 degrees. This was based on calculations that a steep turn would have used up less altitude than a shallower turn by the time the 180 was complete. Was a bit of an eye opener.

[bookworm]

Of course there would be numerous variables, such as W/V, aircraft climb rate and speed (climb angle), L/D ratio, aircraft maneuvering performance, surrounding terrain etc.

The most obvious parameter is the height at which the engine failure is presumed to occur. If you climbed straight ahead to 2000 ft and the engine stopped, you wouldn't ignore that big emergency landing site behind you in favour of a 150 m field that happens to be in front of the nose, would you. Similarly, if the engine stopped at 200 ft, I don't think many of us would contemplate a turnback. So the question to be asking is about the height at which landing again at the departure airport should be contemplated.

[Hot 'n' High]

I believe the RAF have looked into this on several occasions; perhaps someone can shed some light on their findings. While I don’t have their detailed conclusions to hand, I recall that the consensus was that it was fraught with danger – even when flown by experienced Instructors who were expecting the manoeuvre. They tended to support Old Smokey!

Of interest, one very significant issue is that of speed control as the aircraft rapidly turns from a headwind to a tailwind at very low level. The resulting overwhelming visual cue from the rapidly increasing groundspeed as the turn takes place at very low level causes an instinctive tendency for pilots to reduce airspeed to compensate. Given that such turns often lead people to fly at significant angles of bank, the danger is obvious. This illusion is not helped by the fact that their attention is more on the runway that they are trying to get back to, rather than the ASI which would reveal the extent of the illusion. There is a school of thought that there is a significant risk that many aircraft would be lost in a low level "turnback" due to a stall/spin at low level.

As a glidist, we had a simple rule for a winch launch. The only time a “turnback” was flown was from 700ft. At that height, you actually flew a normal circuit joining the downwind leg midpoint at fairly well a normal height for that part of the circuit anyway. All the rest were modified “land aheads” which were designed to keep the glider inside the airfield boundary. Obviously not possible for powered aircraft but the concept is sound. If you are high enough to fit in to a traditional "glide circuit” go for it. If not, pick a field ahead. In reality, I think the Space Shuttle would be the only craft which could meet the criteria!

For those odd airfields where there is literally nowhere to go (ie a takeoff path over a large built-up area with no obliging playing fields or rivers under the climbout) you may have to consider a "turnback" but, forget the runway. Just head for the first available space behind you. If taking off in such a situation, the pre-take-off brief must specifically deal with this possibility each and every time!

[Capt Pit Bull]

I believe the RAF have looked into this on several occasions; perhaps someone can shed some light on their findings.

Can't speak for the current policy but in the 80's turnbacks were taught and practiced (on the Bulldog in the UAS world), to students with low 10's of hours.

There was an envelope for suitable wind component. The procedure required a steep gliding turn which needed a pronounced nose down attitude. Definitely achievable, have practiced it myself and also seen one happen 'in anger' due to an EFATO.

However, certainly a lot of scope for things to go wrong (but on the other hand, if the donkey has quit, things have already gone wrong.)

That said, in the absence of a definitive study, by appropriately qualified individuals for a specific aircraft type, I wouldn't touch it with a bargepole.

i.e. the fact that I know its doable in a 'dog doesn't mean I'm dumb enough to claim its doable in anything else.

With regards to Hot n Highs second paragraph, the points are valid, but trainable.

[Miserlou]

Who says you have to land back on the runway you just left anyway?

Alignment was mentioned earlier and it is a red herring. Often a cross runway will be just as suitable because many fields have preferential runways due to local noise restrictions. I have a hard enough time persuading a couple of radio operators that I will not be using the runway they advise for this reason. I even caught one lyinig on air that a runway was Notamed closed when it wasn't.

Anyway, in the commercial aviation world we are advised to crash on an airport if possible. An airport has many advantages for crashing on including a minimum response time for emergency services and usually loads of obstacle free space.

When I suffered an engine failure I chose to land on a big open grass area of the airfield, NOT on the runway which had a crosswind. Admittedly, I was insured in that I was very high at the time of the failure but this, perhaps, emphasizes the point. You don't have to land on a runway.

[RMC]

The RAF fly the Grob 109...if the headwind component is over 15 knots a turn back is recommended if you have 300'. That said it has a glide ratio of 29:1 as opposed to 10:1 for most light aircraft.

This SLMG had to perform this manoeuvre on many occasions as the RAF version had a major problem with EFATOs.

As with most things in aviation very few are black and white. Agreed in general terms it is not a good idea to turn back ...but it is situational.

As bookworm? pointed out if you have the height and have taken off into a headwind a turn back may well be a more sensible option. I once had an instructor who said at 2,500 feet it would be better to crash land in the built up area straight ahead than to turn back.

He had taken the impossible turn argument married it with a picture he saw of a guy who crashed on a house and walked away!

[triton140]

Under 500 feet, forget it - choose best glide and take yer chances ahead ....

Circuit height - go for it.

Let's face it - you don't have a whole lot of time to make up your mind (speaking from a perspective of a real EFATO) - so decisiveness is the key .....

[Hot 'n' High]

Hiya Capt PB,

With regards to Hot n Highs second paragraph, the points are valid, but trainable.

Quite agreed but, my assumption is that we are talking about your average PPL who flies maybe twice a month, so I would think such precession flying is probably best avoided from the outset as such an evolution would require fairly constant practice to keep things safe. I have checked out the odd PPL for whom the taxy to the holding point is something of an adventure! :O Low level steep turns? Mmmmm. Even I would not consider it unless I was totally out of other options.

Miserlou has a good point with the benefit of landing back anywhere on the airfield – oh, and I would have told ATC exactly where they could have stuck their NOTAM! Think my Mayday would sort of invalidate that bit of paper they were waving at you – particularly as it was a fake. The Gimli Glider springs to mind! Where I have had to return rapidly to the field for rough running engines etc, ATC have always said words to the effect of “Plonk it down anywhere you want Mate!” Thankfully, always made it back to the runway but it’s good to have such positive support from ATC in the hour of need!

RMC, at 300 ft in a PA-28 you will be flaring the aircraft before you can say “It’s all gone a bit quiet up front!” LOL! But, as you say, if there is no-where to go ahead, you have to give it a go to the side or behind. Having a plan up your sleeve is definitely the order of the day and the concept of getting to anything large and flat is what you are after. If it is grey and has numbers painted on it – who cares. Green and flatish would do me! Doesn’t even need to be within the airfield boundary! Just as long as it’s flatish. As you can tell, H‘n’H is not really that picky! :}

I guess, at PPL level, what they are formally trained to do (and checked during routine 6 monthlies etc) is to make a good go of landing in a field +/- 30 deg to the track if at all possible, or, where they know this is not possible, they can tell me exactly what they would do. After all, all you are doing is utilising the last few hundred feet of a normal glide landing approach (which is not far removed from a normal powered landing) so stick with what they practice each time they fly. KISS definitely is the name of the game. No low-level semi-aeros for me!

Anyway, just my views – an interesting Thread this!

[IO540]

Here is a nice reference on the feasibility of turning back (http://www.nar-associates.com/technical-flying/impossible/possible.html)

[bfisk]

Well, above 500' I've gotten away with it several times. Used to practice this regularly in 152s and 172s, throttling smoothly back to idle. Sometimes we would land, sometimes we would see that this was not possible today -- go around! After doing this some tens of times you kind of get a gut feeling about how it will work out. I say, better to have tried it knowingly and with an out, than to find out in dire straits...

[soay]

There's a good article about this here (http://www.avweb.com/news/pilotlounge/pilots_lounge_121_big_silence_after_takeoff_196805-1.html) on AVWeb.

The NTSB report of a Cirrus SR20 that crashed killing an instructor and his student while practicing this manoeuver can be read here (http://www.ntsb.gov/ntsb/brief.asp?ev_id=20060118X00087&key=1). It determined the probable cause to be:

"The student pilot's failure to maintain an adequate airspeed while maneuvering, and, the flight instructor's inadequate supervision of the flight. A factor in the accident was the strong tailwind encountered as the airplane turned from an upwind to a downwind during the teardrop maneuver."

[Flash2001]

I seem to remember that Neville Shute Norway described F/Lt George Stainforth executing this maneuver on the maiden flight of an Airspeed somethingorather. It was only at this point that the Airspeed design staff understood why Stainforth had spent 4 hours before takeoff on repetitive cockpit drills.

After an excellent landing you can use the airplane again!

[Hot 'n' High]

Ladies and Gents,

Earlier today, sadly, there was a fatal accident involving a light aircraft on take-off from Fenland, UK.

While the cause is not officially known, the following has been reported by bose–x on the Private Flying thread:

This was a sad case of someone attempting a steep turn at low level after what looked like an EFATO to land back on the 36 runway. I was in the overhead at the time joining for fuel and another of our guys was on base as the accident happened.

Whatever the case, our thoughts go out to the family and friends of the pilot at this dreadful time. If, and it is only “if”, this accident is as reported above, given what we have been discussing, it is very sobering indeed.

A very sad day – my sincere condolences to all affected by this tragic accident.

[barit1]

40 years ago - in a tired Cessna 140, on a 6000' runway -

At about 50' altitude the cowl popped open, due to a faulty latch. I had at least 4000' runway left ahead of me, and no desire to burn up an engine in the pattern, so it was an easy choice to close the throttle and land straight ahead. IIRC I slipped a bit to try to close the cowl to help visibility.

No harm done except to my self-esteem for accepting marginal hardware on preflight. :uhoh:

But I still had power as long as I kept CHT under redline. If it had happened at 500' I could have flown a tight pattern at part power, I guess, with similar outcome...

[RMC]

H n H ...I agree in a PA28 or anything similar not only is 300 ft a no go but I wouldn't fancy 900 ft. I make the reference to the SLMG just to show one extreme end of the turn back spectrum (there was also a question as to the RAF's view).

[Checkboard]

The "old" CAA in Australia used to produce a magazine know as "The Aviation Safety Digest", and there was an article published in that called "The Impossible Turn" which went through all of the mathematics of the turn at various heights and so on. I have some old copies lying around, but no scanner.

[IO540]

Maybe something like this (http://jeremy.zawodny.com/flying/turnback.pdf) ?

[Reluctant737]

IO540,

That is exactly what I was looking for, I speak Mathematics better than English!

Thank you to everyone who's replied, I hope you're all finding it as interesting as I am. Please keep them coming - I'll give a longer reply tomorrow but I need my bed now, just finished a five day stint!

See you all tomorrow and thanks :ok:

[helldog]

Thanks for trying to make my head explode IO540.

[Capt Pit Bull]

I agree entirely.

[Romeo India Xray]

Who says you have to land back on the runway you just left anyway?

Alignment was mentioned earlier and it is a red herring. Often a cross runway will be just as suitable

Agreed! I was faced with this very situation as an instructor in a 152. EFATO with at a couple of hundred feet above the upwind end of the runway. Not the best place for it when you have nothing but trees ahead. 30 degree ofset runway was positioned such as to not require alignment, only a 210 degree turn.

An airport has many advantages for crashing on including a minimum response time for emergency services

Exactly what went through my head in the milisecond I had to make the decision and simultaneously take control. Even if you have only enough height to complete the 180 to arive at the threshold of the intersecting runway, you still have the rudder to take care of some of the alignment issues. In my case it was not required because after a few seconds of nothing (felt like a lifetime), the engine decided to play ball with enough rpm to get me around an abridged circuit, however by the time I got some power back, we were already most of the way round to the intersecting (disused) runway which would have been reachable although not fully aligned. The abridged cct was decided on because from this position I was able to fly it all within gliding distance of a suitable runway or clear grass apron - better than landing unannounced across the active at a very busy, uncontrolled field.

I think that in his "The Proficient Pilot" books (not sure which one, but I will check and edit it in tomorrow), Barry Schiff dedicates a chapter of one book to this very maneuver. If I remember, his thinking is the same as others (including myself from experience), tight turn (remember stall speed will increase), remember wind direction and turn upwind.

Last point - I spent the rest of the day trying to convince the student that this is something he should never attempt. I wouldn't recommend it for anyone who is out of practice or low hours.

[RatherBeFlying]

There's three recent fatal accidents due to failed turnbacks.

In a glider with 30:1 glide ratio, turnbacks from 300' are something that students get to do before being sent solo. There are glider schools that do simulated rope breaks from 200' in low perfpormance gliders.

If you're one of these people who limit bank to 15 degrees, ain't no way.

Once I've turned crosswind, I might think about it, especially if there's a crosswind runway available, but I'd be a lot more inclined to land upwind.

[Romeo India Xray]

Once I've turned crosswind, I might think about it, especially if there's a crosswind runway available, but I'd be a lot more inclined to land upwind.If you are an experienced pilot (which I assume you are), you should not be so quick to be certain of such things. Circumstances on the day could dictate the best course of action being a 180 (or partial 180 in my case).

If the choice is between turning back downwind and (maybe) crashing on a well equipped airfield, I will take being close to the airfield any day in preference to eating pine cones straight off the trees.

My Schiff book detailing this is in the office so shall collect it tomorrow to post the reference, but basically the 180 is both theoretically and practically possible, dependant on a number of circumstances which will only be aparrent to the pilot on the day. For a low hours pilot, the experience to make the best judgement call may not be available for decission making in the available time (although your decission making does become mighty quick in these circumstances), and for these pilots, land ahead HAS to be the best option.

It would be good to get more input on this thread from pilots who have suffered REAL EFATOs in SE AC. I noticed on another thread that things are descending into FS2004 on this topic :ugh:

Edit: The Schiff book which dedicates a chapter to the turnback is The Proficient Pilot Volume 1, Chapter 34. The Schiff books are all well worth a read.

[RatherBeFlying]

In at least two of the recent three fatal turnbacks, the pilots were highly experienced. At least one was an instructor.

However one should note that the three a/c were homebuilts: Kitfox, Falco (possibly factory built) and the latest an owner design.

It takes some gumption to put a spamcan down into poor landing terrain; even more when it's your baby that you built yourself over several years.

Perhaps for homebuilts there's even more reason to find an airport with long runways and/or landable terrain off the runways.

[werbil]

I believe that for ASEPTA approval (single engine turbine IFR charter ops) in Australia turn backs have to be demonstated. I understand that the procedure is a 45o angle of bank turn on the dials, and there is minimum height that it can be considered from. I think the reason for doing the turn on the dials is that pilots execute the maneuver more accurately without the distraction of the ground swinging around.

[deltahotel]

It's not impossible at all BUT (there's always a but)........

Need:
to have practised it
to get nose down and speed up
to turn into xwind if poss
to have pre planned/pre briefd
to have clear parameters eg "above x' i will turn back l/r"
to know that achieving the runway is good but not vital - flat green airfield is good
to know that fields ahead are probably better
to only do it for real if there is no other option

On the Bulldog never served at an airfield where it was necessary. Camp at Woodvale was interesting. Lee on Solent - the 90deg turn for the beach better than the water. For the students, keep it simple eg "once yr half way round the upwind turn you may continue the turn to the a/f"

too many people have been hurt in turnbacks, inc a good mate at woodvale some yrs back.

just my thoughts - hope it helps the debate

[Mark1234]

Regarding the turnback, I have nothing to add, I'm not someone who should be trying it (in my opinion)...However I'm bemused by the obsession with runways.

IF one has to turnback, AND it's a bit marginal, surely you stick it anywhere you like on the damn airfield? Who needs a runway - anywhere's probably better than the average field, and beats the hell out of dying trying to align some tarmac - even if you have to bounce across some taxiways, demolish some cones etc.

[TruBlu351]

We'd practice turnbacks in the RAAF quite often in the PC9. Students weren't allowed as it does take a little practice and they can get a little busy. Forget doing it in the CT4!!...it drops like a brick and you'll spud in, so just pick a paddock ahead :)

You have to be very aware of the wind vs height vs direction of turn etc etc.