EFATO

[Big Pistons Forever]

Originally Posted by BackPacker

(And to put a bit of perspective to the debate, and how dependent it is on airplane capabilities and circumstances: In gliding we fly a full (although tight) circuit when the winch cable breaks at 300' or above. Yes, you read that correctly. A full circuit, with four 90 degree turns, to a normal landing in the normal landing area next to the take-off area, from 300 feet. But a winch launch cable break happens by definition *above* the airfield, not beyond the threshold, and any reasonably modern glider will easily beat a 1:30 glide angle.)

To a powered pilot 300 feet sounds incredibly low, but to put this in perspective to fly the pattern described above with the 4 turns and a landing back at the point of takeoff would require starting at least 1500 feet AGL if you were flying your average Cessna or Piper, and which is hardly an EFATO in the context of this thread.

[Piper.Classique]

The wind (assuming it is constant) has no effect on the flight dynamics - none!

True, but if there is any wind at all there will normally be a wind gradient and climbing through a wind gradient into wind and descending through a wind gradient downwind are not the same thing. Actually, I do practice turnbacks (because it's fun.......)
and it is a shed load easier with no wind at all....Depends much more on your rate of climb and glide angle than on what rate of turn you use. I can return no wind and land facing take-off direction in a super cub from 400 feet. I don't need to pull to the buffet (don't have a stall warner) ,just a brisk turn, then a genteel sideslip to line up to the landing area and deal with any excess height.

Takes me 600 feet in a DR 400. Gliders are another story again. 300 feet will indeed be enough for a brisk circuit off a wire launch. Aerotow, no, because of the lower rate of climb.

However, I suspect that if I have a real and unexpected engine failure that by the time I have got my arse in gear I probably will end up crashing straight ahead with the aircraft the property of the insurers who have so far made a large profit out of me

[foxmoth]

So I teach Vy climb to 1000 AGL as altitude is your friend so you want to get to an altitude where you have options as quickly as possible, and in the event of an engine failure below 1000, it is nose down to the glide and only turning enough to avoid major obstacles.

The examples I gave earlier (Southampton and Bembridge) have the problem that the "major obstacles" are all in front and to the side of you so the only real option may be a turn back (or at least a turn of over 90 degrees), if you CAN land straight ahead then I certainly agree with not turning back, but what to you do when that option is just not realistically there?

[BackPacker]

Foxmouth, if you are too low for a turnback (given your airplane performance characteristics, the environment and your own ability), and there's also no opportunity to land ahead, or left/right of your track, or anywhere else in reach, then what do you expect us to advice?

The best I can come up with is something I think the RAF was already saying in the 1930s: "Try to hit the softest object available with the least amount of speed."

[Big Pistons Forever]

Originally Posted by foxmoth

The examples I gave earlier (Southampton and Bembridge) have the problem that the "major obstacles" are all in front and to the side of you so the only real option may be a turn back (or at least a turn of over 90 degrees), if you CAN land straight ahead then I certainly agree with not turning back, but what to you do when that option is just not realistically there?

I was waiting for that question and it is a reasonable one. First off it speaks to the desirability of having a plan before the engine stops, not after, one of the reasons I make my students do a take off brief before every takeoff. This is especially important for difficult airports where a quick decision on angling the aircraft left or right after the engine failure may have a profound impact on the success of the manoever.

I think it is also important to point out that a uniform 9 Gee de-acceleration from 60 to 0 knots requires a ground run of about 25 feet or one fuselage length. The flying schools IMO do students a dis-service because they leave the impression that you have to have some nice long smooth field in order to have a "successful" forced approach. The reality is a survivable landing only requires that the aircraft be upright with and with an approximately level pitch attitude at a low airspeed and have at least a small run after touchdown to moderate the de-acceleration forces ( ie not a head on impact with a solid immovable object)

The best predictor of success for actual real world forced approaches has been that the aircraft was in control when it impacted the ground and that it hit at a spot of the pilots choosing, not a random area.

Finally do not forget that the accident statistics suggest that up to 80 % of all engine failures are directly caused by the pilots actions or inactions so the best way to deal with an engine failure is not to cause the engine to fail in the first place.

[BackPacker]

I think it is also important to point out that a uniform 9 Gee de-acceleration from 60 to 0 knots requires a ground run of about 25 feet or one fuselage length.

True, but you also need to consider that the aircraft brakes are NOT capable of supplying a 9G deceleration. If you want to stop the aircraft within one fuselage length, and live to tell the tale, you need something external to the aircraft to slow it down uniformly. The military does this with arrestor cables on aircraft carriers, but somehow I don't think that's an option in an EFATO scenario.

So if you are going to hit a concrete wall, it really doesn't matter how long the ground run preceding that hit is. Yes, if your ground run before hitting the concrete wall is one fuselage length, the *average* G force will be 9G or thereabouts: One fuselage length of almost zero G (horizontal deceleration), followed by one engine cowling length of 25+ G. Unfortunately your body is not going to care for the average G, but for the peak G.

You need to turn the reasoning around: If you have no place to go that resembles a runway or field, try to find something that gives you the most uniform deceleration along whatever your (crash) landing run is going to be. Wherever that may be. And if you can find something that is able to decelerate your aircraft more-or-less uniformly to a full stop over the length of at least one fuselage, you probably will not be killed by the horizontal deceleration.

As far as the rest of your post is concerned, I agree 100%.

[Big Pistons Forever]

Originally Posted by BackPacker

.

So if you are going to hit a concrete wall, it really doesn't matter how long the ground run preceding that hit is. Yes, if your ground run before hitting the concrete wall is one fuselage length, the *average* G force will be 9G or thereabouts: One fuselage length of almost zero G (horizontal deceleration), followed by one engine cowling length of 25+ G. Unfortunately your body is not going to care for the average G, but for the peak G.
.

True; which is why I made the comment

and have at least a small run after touchdown to moderate the de-acceleration forces ( ie not a head on impact with a solid immovable object)

The central point, which I think is a failing of todays flight training, is that the crappy patch of rough ground a few hundred feet long that is right in front you is likely to be discounted as a viable option because it doesn't meet the "right" criteria and so pilots are trained to turn towards the "good" field
which in the EFATO scenario is probably too far away or behind them.

[n5296s]

For landing on the crappy field, it all comes down to making an accurate touchdown. You can save yourself, if not the aircraft, by landing on a football field - as long as you land on the beginning of it. Land on the far end and hit the brick wall that (let's suppose) is there and things don't look so good. The trouble is that accurate touchdowns are pretty hard to get right and as far as I can see most people don't. I'm always a bit horrified by the number of people who touch down half way down our 2400' runway at Palo Alto. I really do TRY to get it on the numbers myself but I for sure don't always get within a football field of them - and that's an airport I know well, complete with windsock and known altitude.

When I did the 180 power off landing in my CPL I did actually get it on the numbers (phew), provoking a comment from the examiner that "you'd be amazed how many people get this badly wrong". And that's CPL candidates.

I'm extremely grateful for the huge salt marshes off both ends of my home airport!

[Big Pistons Forever]

Originally Posted by n5296s

For landing on the crappy field, it all comes down to making an accurate touchdown. You can save yourself, if not the aircraft, by landing on a football field - as long as you land on the beginning of it. Land on the far end and hit the brick wall that (let's suppose) is there and things don't look so good. The trouble is that accurate touchdowns are pretty hard to get right and as far as I can see most people don't. I'm always a bit horrified by the number of people who touch down half way down our 2400' runway at Palo Alto. I really do TRY to get it on the numbers myself but I for sure don't always get within a football field of them - and that's an airport I know well, complete with windsock and known altitude.

When I did the 180 power off landing in my CPL I did actually get it on the numbers (phew), provoking a comment from the examiner that "you'd be amazed how many people get this badly wrong". And that's CPL candidates.

I'm extremely grateful for the huge salt marshes off both ends of my home airport!

For Cessna singles retracting the flaps will make them land now if you are in danger of overshooting your touchdown point. However in every case you can make any aircraft land by smashing it into the ground at your chosen point.

Not pretty but much better then running into the brick wall. Remember the only part of the aircraft that has to be intact when you stop is the cabin.

[foxmoth]

The central point, which I think is a failing of todays flight training, is that the crappy patch of rough ground a few hundred feet long that is right in front you is likely to be discounted as a viable option because it doesn't meet the "right" criteria and so pilots are trained to turn towards the "good" field

Not talking about a "crappy patch of rough ground" but in one case, open sea, and the other, busy built up industrial areas with all manner of obstacles, and I would suggest if you only have ONE small crappy patch of rough ground surrounded by complete no go areas, the sort of pilot who would reliably get in there probably has the same level of experience as one that might be able to cope with a turnback!

[n5296s]

For Cessna singles retracting the flaps will make them land now if you are in danger of overshooting your touchdown point. However in every case you can make any aircraft land by smashing it into the ground at your chosen point.

Those are both really great points. We tend normally to be so focussed on a "good" landing that it's easy to forgot them - especially the second one.

Thank you.

[Big Pistons Forever]

Originally Posted by foxmoth

Not talking about a "crappy patch of rough ground" but in one case, open sea, and the other, busy built up industrial areas with all manner of obstacles, and I would suggest if you only have ONE small crappy patch of rough ground surrounded by complete no go areas, the sort of pilot who would reliably get in there probably has the same level of experience as one that might be able to cope with a turnback!

PPrune posters seem to very good at finding the one exception that will counter any argument.

Fine you win; a turn back is the only way to deal with an EFATO. Happy now

[Maoraigh1]

However in every case you can make any aircraft land by smashing it into the ground at your chosen point.

At near stall speed, I assume, or you'll still have the energy to get rid of.

[Big Pistons Forever]

Originally Posted by Maoraigh1

At near stall speed, I assume, or you'll still have the energy to get rid of.

Well obviously the lower the speed the better, but if the stone wall is looming in the windshield and you are still in the air then you may have to force the airplane to touchdown now in order to start de-accelerating. This may include full forward stick. This is obviously a desperation move but slamming the airplane on, busting off the nose wheel and a violent stop with the nose digging in will be much more survivable than hitting that stone wall head on at flying speed.


I was at the local flying club yesterday and as it happened the turn back issue got raised by another fellow. He said he saw a for real turn back to a narrow dirt strip by a C 185. The engine failed at about 600 feet AGL and the pilot got it back with a steep bank and a final wild skidding turn to get lined up. He said it was a very impressive example of flying skill. I then asked if he knew why the engine failed. His reply "Oh the pilot forgot to turn the fuel on and the engine died when the collector tank emptied"................

[Whiskey Kilo Wanderer]

Trying the Impossible Turn

I had a go at the Impossible Turn stuff today, albeit with a 2,000' AGL hard deck. The problem with being that high is trying to judge how far you have travelled horizontally for the loss of height. The trick I used was to zoom in the GPS (MemoryMap using CAA 1:250K charts) to maximum and start the climb when crossing a particular road.

Climbing to 300' and then simulating engine failure I couldn't make it back to the road without sinking through my 'hard deck'. I could do it with a 500' engine cut, so I guess this means that for an EFATO below 500' there’s no option but to land ahead. At 500' or above it might be possible to get back on to the field, probably with a down wind landing and only if the prospects of landing ahead were less inviting.

It was an interesting exercise, although I don't want to think about what would happen if a four second delay was introduced (to simulate the pilot’s recognition of the problem and to start to react).

Safe Flying,
Richard W.

[Big Pistons Forever]

Originally Posted by Whiskey Kilo Wanderer

It was an interesting exercise, although I don't want to think about what would happen if a four second delay was introduced (to simulate the pilot’s recognition of the problem and to start to react).

Safe Flying,
Richard W.

This is exactly the problem between what you can do in a practice session and what is realistically obtainable when the totally unexpected silence occurs. The pitch down to the glide attitude should be automatic but what to do next will always take a few seconds to process. Interestingly NASA research shows that even professional pilots who fly everyday typically have a 3 to 5 second reaction time when given an unexpected emergency.

That is why I fear that pilots who have practiced the turn back and say "well I have repeatedly practiced the turnback and can do it every time from an altitude of 500 feet" have perhaps an unrealistic appreciation of how it would work for real. The other problem with practicing turn backs at altitude is the "ground rush" you will experience with the steeply banked very nose down attitude required for a successful turn back. This often results in inadvertently applying back stick and leads to a stall spin. The only way to get over this is to practice the manoever at actual low altitudes which is insanely risky.

Finally I think it is important to qualify the turnback. The aim of this manoever should not be to get back to the runway, it should be to reach any part of the flat, level and mostly unobstructed ground within the airport perimeter. If you can make the runway that is a total bonus.

But as I said in my opening post I strongly feel that for all but the most experienced pilots if a total engine failure occurs below 1000 feet AGL you are invariable going to better off gliding straight ahead and turning only to avoid major obstacles, and that is what I teach my PPL's.

[englishal]

It'll be different for different situations though so I don't see much value in practicing it. For noise abatement we try to turn ASAP once in the air...certainly below 500', and not far from the runway end. The result is that before 3-500' you have already turned 90 degrees ...but are heading away from the airfield . This could make it easier or harder to get back...Easier because you can turn less than 180 degrees and you are aiming at the airfield, but to line up you need to make another turn wasting more energy.

At these alts I'd either land ahead in one of the fields (depending on whether I have a head or tailwind now), or accept a landing anywhere on the airfield even if it is off runway.

[n5296s]

Well, at risk of attracting another set of nanny flames...

DON'T DO THIS AT HOME. NOR IN AN AEROPLANE.

OK, that out of the way... I decided to see how much difference airspeed and bank angle make. This time I did it at altitude and measured time for a 360 turn rather than the teardrop manouevre, since all I was interested in was relative altitude loss.

The result was quite a surprise. Holding speed at the hairy edge of a stall does make a BIG difference. I flew each combination at least twice and the results were pretty consistent.

Bank Speed Altitude Loss
45 55 500
45 60 300
45 70 500
60 70 500+

At 55 knots in a 45 degree bank, the wing is most definitely partially stalled. There is a kind of buffet but it comes more from the nose bobbing up and down than from turbulence over the horizontal stab. What really surprised me is the big difference between flying on the brink of a stall, and 10 knots faster. Also that flying in a partial stall is no worse than flying 10 knots faster. That's reassuring from a safety pov.

The 60 degree result is clearly worse. It's also MUCH harder to fly, in my plane anyway - it's so nose heavy that precise speed control gets harder. If I wanted to get really repeatable results I'd have to practise some more, but it's clearly inferior both in terms of altitude loss and certainly in terms of safety, so there's not much point.

My methodology was: climb to altitude (3500'), pull power and slow down to desired speed, enter bank, fly a 360 degree turn holding bank angle and airspeed, roll level anticipating 360 degrees, check altitude. This was all in my 1980 TR182.

The result suggests that even 400' might be doable in my plane, IF you can hold airspeed/pitch precisely. At airports where landing off runway is not too bad (e.g. Livermore where there is a field at the end of the runway), it might be better than the alternative (though not actually at Livermore because the field is long). I don't think I'll be trying this for real close to the ground though.

DON'T DO THIS AT HOME.

OK, let the flames commence...

[Genghis the Engineer]

Well i'll flame you if you insist N, but for my money that was a really nicely constructed little research project that certainly gives extra confidence to one of Rogers' assertions.

G

[BackPacker]

Yep. Nicely done.

So essentially you confirmed the Rogers (?) paper that the lowest loss of altitude, assuming certain stability conditions, is achieved by making a 45-degree banked turn at the edge of the stall. Good.

GtE, I'm still interested in your stab at doing a paper when you remove those stability conditions, and analyze a proper EFATO situation. Mark1234 found out that the best results in that situation are achieved by initially setting a 60 degree bank, pulling to the stall, and reducing the bank angle as you fly the turn and the nose drops.