Engine Failure After Takeoff & Turnback (SEL)

[Big Pistons Forever]

The "impossible" turn is completely possible and safe. I think it should be taught to students. If a student is able to perform a spin entry and recovery, why can't a student perform a turnback after a simulated engine failure? Both involve memorized procedures that need to be practiced. There is no difference other than it being a different flying maneuver.

The problem with teaching this manoever at altitude is that there is no way to simulate the ground rush you get when manoevering close to the ground. This has resulted in many stall/spin accidents when pilots instinctively pull back on the stick when they see the ground rushing up.

Therefore IMO you give the student a false sense of security if this manoever is taught at altitude. Teaching it close to the ground, that is actually pulling the throttle at say 500 ft AGL after takeoff, is IMO in the same league as actually pulling an engine on a light twin right after rotation. The chance of an accident far outweighs the training value.

[dr.aero]

BPF...

pulling the throttle at say 500 ft AGL after takeoff, is IMO in the same league as actually pulling an engine on a light twin right after rotation. The chance of an accident far outweighs the training value.

Maybe you're trying to illustrate that you don't approve of that one bit, but I completely disagree that those two maneuvers are even remotely related. First off, an engine failure in a single produces no control issues whatsoever - an engine failure in a twin produces significant control issues and is more dangerous, control wise. Secondly, one is happening at 500' AGL and the other is mere feet off the ground. Do you really think they're in the same league?

If you think a turnback is so dangerous, you should do some glider training - specifically show up on the day they're practicing rope breaks at 300' AGL (if I remember correctly that was the altitude used during my training). The glider has absolutely NO choice but to land back at the airport. When practicing in a powered aircraft you are a lot safer doing this maneuver as you can abort it at any time and try again - glider pilots don't get that luxury. Powered pilots also get the luxury of doing an overshoot when their approach isn't great - glider pilots don't get that luxury. If you have a properly trained instructor, there is no reason why you shouldn't practice the turnback maneuver. If an instructor is competent enough to prevent a plane from going into a spin while turning onto final, they should be competent to ensure that it doesn't happen during this maneuver either. Obviously there are other factors that the instructor needs to know but I'm not writing a post on how to teach the maneuver to a beginner student.

[North Shore]

^ Not sure I agree with the comparisons/analogies to gliders here...the whole point is that they are designed to *glide* - light trainers (C172) not so much.

I, like BPF, think that you are asking a lot of your 'average' PPL with a few hundred hours over 10 or 15 years to complete this safely.

[Chuck Ellsworth]

The glider has absolutely NO choice but to land back at the airport. When practicing in a powered aircraft you are a lot safer doing this maneuver as you can abort it at any time and try again - glider pilots don't get that luxury. Powered pilots also get the luxury of doing an overshoot when their approach isn't great - glider pilots don't get that luxury. If you have a properly trained instructor, there is no reason why you shouldn't practice the turnback maneuver.

dr.aero you are not going to change any minds here.

If glider pilots can learn the turn back early in their training why can't powered pilots.

Those who argue gliders are designed to glide are failing to take into consideration if a glider instructor screws up there is no way to go around.

The problem here is myopic thinking by some pilots.

[Big Pistons Forever]

BPF...

pulling the throttle at say 500 ft AGL after takeoff, is IMO in the same league as actually pulling an engine on a light twin right after rotation. The chance of an accident far outweighs the training value.

Maybe you're trying to illustrate that you don't approve of that one bit, but I completely disagree that those two maneuvers are even remotely related. First off, an engine failure in a single produces no control issues whatsoever - an engine failure in a twin produces significant control issues and is more dangerous, control wise. Secondly, one is happening at 500' AGL and the other is mere feet off the ground. Do you really think they're in the same league?

If you think a turnback is so dangerous, you should do some glider training - specifically show up on the day they're practicing rope breaks at 300' AGL (if I remember correctly that was the altitude used during my training). The glider has absolutely NO choice but to land back at the airport. When practicing in a powered aircraft you are a lot safer doing this maneuver as you can abort it at any time and try again - glider pilots don't get that luxury. Powered pilots also get the luxury of doing an overshoot when their approach isn't great - glider pilots don't get that luxury. If you have a properly trained instructor, there is no reason why you shouldn't practice the turnback maneuver. If an instructor is competent enough to prevent a plane from going into a spin while turning onto final, they should be competent to ensure that it doesn't happen during this maneuver either. Obviously there are other factors that the instructor needs to know but I'm not writing a post on how to teach the maneuver to a beginner student.

As North Shore pointed out gliders are designed to glide. Even an average performance glider will glide at least 4 times better then a typical SEP like a C 172. It will also do this gliding at a speed about 2/3's as fast as a C 172. So to replicate the experience of a 300 ft AGL rope break in terms of how it will take from the start of the glide to touch down the C 172 would have to be at least 1200 feet AGL before the start of the turn back. I, like pretty much every instructor I know, get students to do PFL's from that height all the time.

If you want to compare apples to apples then you would have to say that glider instructors should practice a rope break at 75 feet AGL and turn back to the runway. I don't see any glider instructors doing that for what I think are pretty much the same reasons not very many powered aircraft instructors think this is a good idea.

I have a question for Colonel Saunders. Do you ever close the throttle at 500 feet after takeoff on your PPL students and expect them to turn back to the runway ?

[iflyforpie]

Sorry dr aero, comparing glider rope break practice to an EFATO and return to the field in a light trainer is comparing apples and oranges.

Even a slug like the 2-33 has double the glide ratio of a light trainer, its best glide speed is slower than the dirty stall speed of a typical light trainer, its minimum sink is ten MPH slower than that, and stall is a paltry 30 and change miles an hour.

Plus, it landing off the runway is almost a non-event in the case of practice, not so with a light trainer doing EFATO and return to the field practice.

[white_knuckle_flyer]

In watching that 182 turnback video, I would not necessarily argue that this pilot should be 100% validated by his decision to turn back. Is he skilled ? Yeah. Did it work out for everyone involved ? Sure did. But you can't tell me that even had this pilot been less skilled, that he wouldn't have still chosen to turn back. Hard to tell from the vid, but there appears to be several open areas that could have got the passengers down safely but at the expense of the aircraft. Even if I am wrong, I'm sure you can think up a scenario exactly like I am describing.

A pilot who owns his own a/c may very well have his judgement clouded when trying to decide whether to turn back or not. Especially with pax, I think the pilot has to be 99% sure that they will make the field. Perhaps the skill of this particular pilot ( and that fact that it all worked out okay ) has obscured the fact that his decision may not have been completely objective. Had he been renting, would the decision have been different ? If not for him, how 'bout the next guy ?

[North Shore]

The problem here is myopic thinking by some pilots.

No, not at all! It's guys with thousands of hours of judgement/experience assuming that their level of comfort in an aeroplane (in all sorts of odd attitudes and wx conditions) is the same as that of a bare-ass PPL. "If I can do it with no problem, then why can't they?!"

Oh, and.. what BPF said about gliders, too!

[ettw]

So this reply has nothing specifically to do with the thread. Full disclosure.

What really gets under my skin (and maybe this is what CS is going for, getting under peoples skin) is that people like CS and CE are, apparently, VERY experienced pilots. Great! That's what we need, experience. But boys, your delivery SUCKS!

I think you lose an audience when your arrogance shows through as strongly and as frequently as yours does.

The delivery is as important as the message.

I read this thread with interest. I am always open to learning and, as importantly, re-learning.

Maybe just a bit less "look at me...this is easy...whats wrong with you people?" and "when I learned how to fly, I had to build my airplane first" and more "hey folks, think about this, its really important and here is a real world example."

Stop thumping your chests and try to convey the VERY important messages you both have to offer to our community in a manner that will be received by the audience successfully, not in a belittling and arrogant manner.

I have to go to bed and get some sleep. Have to work in the morning.

Cheers

ETTW

[Chuck Ellsworth]

Maybe C.S and I just look at these things differently.

As far as arrogance and look at me Im better than you, that is just the way you preceive us.

Quite frankly I Personally could give a rats ass less if you don't like my delivery or style.

Facts speak for themself, I finished a career of over fifty years flying most every conceivable aircraft all over the planet accident and violation free after flying in over siixty differemt countries.

Should I go with the herd and meekly agree with every idea they think is the right way to do things jut to be P.C. and warm and fuzzy?

[Colonel Sanders]

On the previous page, I said:

A fresh PPL, perhaps slightly on
the left hand side of the bell curve, that won't take on
a 10 knot crosswind, probably doesn't have the skills
to safely perform a descending steep turn (turnback),
trading off altitude to maintain airspeed.

However, a CPL with over 1000TT and a season (or two)
of jumper dumping or float flying or maybe a class 2 instructor
should be a little sharper. A descending steep turn should be
within his ability.

Not sure why that is so unreasonable.

I personally expect an awful lot from the few PPL's
that I do, including tailwheel, aerobatics and formation.
These PPL's are far to the right of the bell curve. I
admit that they are not average, nor (horrors) on
the left side of the curve.

The delivery is as important as the message

I suppose I could apologize for the factual content
of my posts upsetting you. Would you like your
money back, that you paid me?

[Colonel Sanders]

A little background: If you have never taken a
statistics course (and as such have never learned
not to buy a lottery ticket) ....

If one collects a lot of samples of stuff in nature -
such as people's IQ - you will notice that the data
forms a curve, with some data points with small
values, most data points clustered around the
average, and some data points with high values.

This is pretty neat:

http://en.wikipedia.org/wiki/Normal_distribution

Here's a picture:



Notice the symmetry of the curve. The average is
thusly in the middle, and the shape of the curve is
determined by something called the "standard deviation".

95% of the data points are within two times the
standard deviation of the average. 2.5% of the
points are above that range of 2 sigma, and 2.5%
are below that range of 2 signa that.

How on earth is this aviation-related? Don't teach
the turnback to PPL's on the left hand side of the
curve (i.e. with performance less than average).

A PPL with performance at least one sigma (certainly
two) on the right side of the curve is a candidate
for learning the skills to perform a descending steep
turn.

PS As previously mentioned, what can really help
the ability to perform the turnback, is practicing
a wingover (at altitude). The wingover teaches
you that you can fly very steep angles of bank
at very low G and very slow airspeed, which might
sound familiar.

A wingover is not terribly difficult to perform, nor
does it involve high G. It does not stress the
aircraft in the least. Here's a picture of one:



A wingover is a really nice way to turn around.
It is a gentle and elegant 180 degree turn. But
instead of doing it with high speed and high G,
what you do (in the first half) is a climbing turn,
to convert your airspeed into altitude. Power
is never touched. The slower speed allows you
to turn with a tight radius, using light G. The
second half of the wingover is symmetrical,
a descending turn.

I personally like to see around 90 degrees of
back at the apogee (highest point) of the wingover.
The G is light and positive, and the ball is in the
center. The aircraft is ballistic at that point.

IMHO a wingover at 3000 feet is definitely within
the capabilities of nearly any PPL on the right
side of the curve.

[PilotDAR]

Yeah but...

The curve proposes to depict a whole emotionless skill set. Let's say pilot's in this case. Overlaying that, but not depicted, are the other characteristics of daring, ego/macho, and carelessness, which are disconnected from the curve, but related to the skill in some way. Without the skill to fly at all, you can't go and do foolish things in planes.

So overlay these non skill characteristics, which for some pilots no amount of skill seems to overwhelm, and stupid things seem to happen in planes. Not so much because the action was wrong, but the judgement as to when to apply that action was wrong. The right thing at the wrong time is still the wrong thing.

As I had said, among an audience of who knows who, describing and promoting skills which are at the "skilled" end of the curve, as being applicable to the broad group of people described by the curve is setting out on an untoward ethical journey...

[Colonel Sanders]

I shall be fascinated to learn who is offended by
my gentle statistics & wingover lesson in the
previous posts.

I suppose what people find offensive is relative
to their previous experiences. Most of the young
people here probably have new-age sensitive
hippie baby boomer parents, and played soccer
without keeping score.

Does this make you nostalgic?

The curve proposes to depict a whole emotionless skill set

I suppose it could, but it might be more useful to measure
their performance as pilots which encompasses pretty
much everything. Not sure how "emotionless" snuck in there.

[PilotDAR]

Yeah, "emotion" is not the perfect word. What do we call it when a pilot performs a task exactly as trained, without falling short in performance because of distraction or other human failings, but also does not attempt to out perform because of bravado or stupidity?

[photofly]

A PPL with performance at least one sigma (certainly
two) on the right side of the curve is a candidate
for learning the skills to perform a descending steep
turn.

By your own statistics, that would exclude 9 out of 10 pilots.

For a manoeuvre to be of general use to the pilot population it has to be flyable by pilots at least one and preferably two s.d.s below average skill.

Is this such?

On a statistical quibble, the Normal distribution is great for things like adult height and shoe size but not perhaps the most appropriate to model pilot skills. One problem is that it's hard to come up with a metric for something as diverse as "pilot skill". Secondly, and empirically speaking, rather like the distribution of incomes, it's likely that any skill curve has a long tail to the right. There are a bunch of pilots with roughly the same ability, a few with a bit less (they did have to pass a check ride, after all, so they can at least takeoff and land in some fashion) and then a whole lot who vary from a bit better than average, to awesome, to exceptional. If this were a Normal distribution the probability of Bob Hoover existing, being that far from the mean, would be indistinguishable from zero. Instead you might want to look at the log-normal distribution and see how that fits: http://en.wikipedia.org/wiki/Log-normal_distribution

A corollary of pilot skill being an asymmetric distribution with a tail to the right, is that the mean is greater than the median, in other words, most of us are actually below average in skill.

[Colonel Sanders]

For a manoeuvre to be of general use to the pilot population it has to be flyable by pilots at least one and preferably two s.d.s below average skill

Two sigma below average is 97.5% of the pilot population -
I just don't see that happening any time soon. I'm honestly
only interested in the right hand side of the curve - top 50%.

On a statistical quibble, the Normal distribution is great for things like adult height and shoe size but not perhaps the most appropriate to model pilot skills

Not sure I totally agree with that. Take any class - say 30 students -
of any age, learning any subject. Take their marks on the course
and plot it on a curve. Betcha you get a bell!

Pilots are required to take written and flight tests which are graded
in detail. Take the average of their written and flight tests and there's
a simple scalar metric. Plot the data on a curve. I'll wager you get
a normal distribution.

[photofly]

Not sure I totally agree with that. Take any class - say 30 students -
of any age, learning any subject. Take their marks on the course
and plot it on a curve. Betcha you get a bell!

To some extent that's self-fulfilling. If you don't get a bell then people complain about the course being too difficult for the weaker students and require more easy test questions, or else insufficiently distinguishing of the best students and require more challenging material. And 30 students isn't really a big enough sample to see the difference between something that's roughly bell shaped, and a close fit to a Normal distribution with a good correlation to tail size out to two standard deviations.

Pilots are required to take written and flight tests which are graded
in detail. Take the average of their written and flight tests and there's
a simple scalar metric. Plot the data on a curve. I'll wager you get
a normal distribution.

Of course. If you design a test to distort any existing distribution into a bell curve (and don't think TC doesn't check) then it's no surprise when you get a bell curve.

It's even distorted at the examiner level, where the examiner looks for quibbles to knock a few points off a strong candidate that he knows is going to pass - just because he doesn't want the guy or girl to get too cocky - or gives away a few points to weaker candidates.

But are TC's flight and written tests a true measure of skill?

Colonel, You and Bob Hoover are both good pilots. Let's assume, hypothetically speaking, and for the sake of blushes that - as good as you both are, he's much more skillful than you. if you and Bob Hoover both sat the CPL written and flight tests today, would your scores reveal that difference? Since you'd both get close to full marks, it wouldn't. So the test scores aren't actually a good metric for skill. That applies particularly to the top echelons - the long tail of skills on the right is artificially compressed.

[Colonel Sanders]

the test scores aren't actually a good metric for skill

You are addressing a very tricky problem - the calculation of
the simple scalar metric, which I would consider the norm of
the vector of the appropriate dimensions, of which there are
many, of different weight.

I would run and hide from that rathole. You are far more
courageous than I, to grab that particular bull by the horns

It's fair to say that there is no one right way to calculate
the norm of that particular vector. The conclusion of that
is that the scalar metric is going to be a wee bit imprecise,
which is part of life.

[photofly]

You are addressing a very tricky problem - the calculation of
the simple scalar metric, which I would consider the norm of
the vector of the appropriate dimensions, of which there are
many, of different weight.

I note your assumption that the vector space actually has a norm.

Where's that hole...

[LousyFisherman]

C150 at a density altitude of 5000 feet.
By the time I'm at 500 feet AGL I'm probably 1.5 miles from the airstrip.
Doesn't matter how good a pilot I am "youse can't get there from here" (A little Ottawa Valley accent for the Colonel)

I also don't see how I can do small tight circuits if circuit height is 1000 ft AGL

LF

[Colonel Sanders]

By the time I'm at 500 feet AGL I'm probably 1.5 miles from the airstrip

Yes, the climb performance of a C150 on a hot day
is not very impressive. As I said before, many aircraft
have a glide which is far steeper than their climb angle
which makes the turnback not particularly useful, unless
the runway is 25,000 feet long, or you have 50 knots
of headwind on takeoff

I actually prefer to do circuits in a buck fifty at 500 feet
in the summer, even down here at sea level. Far too
much time wasted climbing - and as you point out, far
out of gliding distance of the runway.

vector space actually has a norm

I really doubt the Euclidean norm is going to cut it
for you ...

[Beefitarian]

Who is Norm?

[Colonel Sanders]

He is the Chief Vector. His middle name is Eigen.

I note your assumption that the vector space actually has a norm

Sounds like a litmus test to determine if someone
is a mathematician or an engineer

There is mathematics, which is elegant and pure.
You can spend a lifetime getting lost in it.

Then there is physics, which is applied mathematics.
We're getting a little grungy here, in the corners.

Then there is engineering, which is applied physics,
and things can get really horrible from a mathematician's
standpoint, in order to achieve something practical.

A very long time ago, I was an acolyte of Mathematics.
I've gone downhill a lot, down the slippery slope since
them. But I can still fly the snot out of an aircraft down
low, without hitting the ground. And if that ain't applied
physics, what is?

[Beefitarian]

I just now watched your youtube. Be honest, do you know Candice in the biblical sense, Airshow groupie?