Engine failure in Vy climbout

[n5296s]

This post relates to a discussion a few weeks back, but I can't find the right one now. One of the things under discussion was what happens if the engine fails while in a Vy (or similar) climbout - i.e. will you stall before you have time to react.

I set up the following experiment: set full takeoff power, trim for Vy, let things stabilise, pull power.

As I expected, nothing bad happened. The airspeed went down to to about 80 knots (Vy in my plane is 90) but by then the nose was pointing downwards and I was picking up speed again.

I didn't try it at Vx - maybe next time. But I'd expect it work much the same.

Someone will no doubt say that this isn't realistic - the real-life case is the pilot hanging on desperately to the yoke trying to maintain a visual climb attitude. And they may be right. There's no need to flight test that one, there's no doubt about what will happen...

(Oh, and someone will criticise my grammar, and several people will accuse me of irresponsibility for daring to try something so risky, and others will go on to accuse me of trying to incite inexperienced pilots to kill themselves. And to all of you I say, a plague on all your houses, or less poetically <invective of choice> off).

[Pitts2112]

Personally, I think you've done well to practice a potential failure scenario. Well done.

Now go do it for Vx next time.

I did the same in the Pitts for the same reason - but more specifically because of the extreme nose-high attitude of a Pitts on climbout. Same result - it was a non-event and very easy to keep things perfectly together, in so far as a drill is just that.

[abgd]

Why would we call that risky? I assume you did the experiment at a reasonable altitude.

What aircraft were you flying, and how was it loaded?

[Big Pistons Forever]

Big difference between Vy and Vx as Vx is typically less than 10 knots above power off stall. In addition if you are at Vx you will be typically close to the ground and in a steep nose up attitude. When you try the test try it at altitude and assume the engine has failed at 100 AGL. I think you will be surprised at how far the nose has to go down and how fast the 100 feet disappears .....

[BackPacker]

It was me who did the test a few weeks ago. I was equally surprised to see that nothing bad would happen, regardless how hard I tried.

Whether to simply let the nose drop of its own accord, or to use a 1/2G pushover, or a 0G pushover all didn't matter. I was fully established in a Vbg glide *before* any altitude loss would occur.

The only thing that could conceivably considered "bad" was when I deliberately tried to maintain the attitude and let the aircraft stall. That lead to a loss of about 400 feet before being established in a stable Vbg glide. But it took three to four seconds from loss-of-power to the stall, so you should have plenty time to think "oh ****".

What did not work at all was simply releasing the stick and let the aircraft sort itself out. After having lost well over 1000 feet in a speed/attitude oscillation I called it quits.

[astir 8]

There must be strong similarities to winch lauch failures with gliders. e.g. you are climbing at a very steep angle, relatively close to stalling speed & then all motive power is suddenly removed.

The remedial action is then to lower the nose as quickly as possible (to about 30 degrees nose down is taught) and then WAIT until the required flying speed (normally approach speed with addition for for windspeed) comes back before returneing to a more normal attitude and then either a) landing straight ahead if possible or b) if landing ahead is not an option because you're too high, turning and re-assessing the options for a safe landing.

Just lowering the nose to a normal (or best glide attitude) means that it takes a much longer time to restore flying speed.

(Please note that option b is not necessarily a good idea in a powered aircraft)

It's educational watching how quickly airspeed is lost during the process of getting the nose down from over 45 degrees up to 30 down. It's also educational to see how long it takes to get back to a safe manoevring speed even when the nose is pointing quite steeply downwards.

The big no-no in the process is to start turning before the airspeed is restored as the glider is then being invited to spin at low level. Glider instructors get quite upset about that!:mad:

Also it's a very good reason for not rotating too quickly (i.e while still too low) into the full climb during a winch launch, because by the time you've got the nose down and airspeed restored, the ground has already intervened in the process. :ooh::ooh:

[peterh337]

You should not simply stall just because the engine has failed - in any flight regime - because the aircraft will continue flying at the trimmed speed.

It is the trim that sets the flying speed. The engine merely affects the rate of climb :)

[Mark1234]

Considerably less extreme than the glider example, winch launches are a bit special (and unnatural :E) Even the pitts doesn't climb as steeply as a winch launch, and the glider has a lot less inertia. As the OP says, it's a bit of a non event in a 'standard' piston single really.

[foxmoth]

With all this, some will question why instructors emphasise so much getting the nose down in the event of an EFATO if it does it itself. The truth is that in the event of a real failure there are many pilots who's brain does not function quite as they would like in a real emergency and having the training to lower it will stop them sitting there pulling the nose up and trying to hold the climb attitude.:ok:

[Pilot DAR]

During last fall's flight test program with a highly modified (and draggy) Cessna Caravan, it became necessary to reduce the published Vy, to obtain the minimum required climb rate to pass certification. Because of the added drag, Vy reduced a little = less drag, so better rate of climb.

Once I declared this, Transport Canada said fine, but that means that your target speed at 50 feet will now be the lesser speed to be declared Vy, and therefore I was required to demonstrate an EFATO from 50 feet, at that lower speed. This was not required for the lower yet Vx (thank goodness!).

At my new Vy, this was only just achievable, and was indeed to most scary thing I did during all of the test flying for that program. I had to admit that a few times, it was necessary to add power on the way down to make it work (and adding "a bit" of power in a Caravan is not as easy as it sounds).

The manuever was akin to autorotations, in that prompt and considerable "down" control was required, with the intent to accelerate, and then an agressive flare at the bottom. This required considerable practice, and was very stressful in a new plane, which was supposed to be landed nose low to prevent a groundstike of the payload.

We all agreed that the maneuver could be done, on the premise that the aircraft did not have be undamaged after landing, just no one hurt. That would have been doable.

The popularity of STOL kits also brings this to the forefront. Whether it is an EFATO, or simply an approach flown rather slowly, the very real hazard is there. In the early days I remember finding that my STOL C150 would very happily glide at 50 MPH, rather than 60+ stated in the flight manual. This was fine "up there" but very risky close to the ground. You could approach the ground power off at 50 MPH, but you had no inertia left with which to flare. Just slamming into the ground unarrested was a very real risk. Again, akin to the autorotation, where if you have not built up rotor RPM for the flare, you're just going to hit the ground.

Though flight training should teach the basics of this, the "fine points" are sometimes not understood by the instructor, and not taught well. In general, many aircraft will fly certain maneuvers at speeds less than the flight manual stated speed, but if the engine quits, you will no longer have the margin to maintain safe flight for the FAA's pilot who is flying without applying "unusual skill, attention and strength".

Though never to be published, for my Caravan program, I was required to fly a Vref -5 circuit. From liftoff to touchdown at my new speeds -5 knots. I did it, but the stall warning was sounding through most of it. That was the required margin between "normal skill and attention" and what I should be able to demeonstrate with my total of 40 flight test hours on type!

[UV]

Someone will no doubt say that this isn't realistic

It probably isnt!

What did you do during your simulation? I suspect you reduced the power to idle.

There is a BIG difference in a real engine failure. Has the prop siezed? Is it windmilling? Have you still got take off flap selected?
Trim will be affected by a stopped or seized engine (much more nose down... meaning lots of nose up trim..which may run out).


By throttling back to idle you still have a little residual power (put simply somewhat like feathering) which reduces drag. It can be nothing like the real thing!

Try stopping the engine at altitude with the mixture control, then switch off both mags. You may be quite surprised at the result, and thats probably without the prop actually stopped!

[maxred]

n5296s - the discussion you refer to revolved around TURN BACK, where in the red corner were the land straight ahead brigade, and in the blue corner, the advocates of a successful turn back. It looked closely at Vx, Vy, and at what altitude a successful turn back with safe landing could be achieved, or not as the case might be.

The situation of what happens at EFTAO, the pilots reaction, and importantly, the aeroplanes reaction, was discussed at great length. Several individuals, including you if I remember correctly, were going to conduct experiments, particular to type.

With respect, pulling the power, at stages of speed/climb, would be a fairly benign incident, height dependent, but it was the input of turn that created the full debate, and discussions as to whether this be a good idea.

[Big Pistons Forever]

I am posting a section of an article from the latest Transport Canada Aviation Safety Letter which discusses GA stall spin accidents

Quote


The occurrences broke down into three principal groups:

a. stall or spin accidents resulting from aircraft handling (27);

b. stalls or spins following engine failure (9); and

c. stalls or spins resulting from loss of control in IMC (3).

Handling Accidents

Twenty-seven accidents resulted from mishandling the aircraft into an aerodynamic stall. These accidents resulted in 26 fatalities and 16 serious injuries. In two cases, it appears that the engine was not producing full power but the aircraft was capable of controlled flight and the stall was avoidable. In all cases, the stall, which sometimes precipitated a spin or wing drop, occurred at low altitude and at low airspeed. The stalls and spins occurred at a height where recovery was very difficult and probably impossible. Sixteen stalls resulted from turning at low airspeed, 10 occurred in straight ahead flight, and one inverted spin developed when the pilot was practising aerobatics at about 1 500 ft AGL.

Most of the 27 handling accidents happened during the takeoff/initial climb-out or approach phase. There were 13 stalls during the climb-out after taking off and at least six of these occurred during a low speed, low altitude turn. Five stalls, all in turns, occurred during the approach/landing phase, most often on turning base to final. One practice overshoot ended in a stall when the instructor waited too long to take control and the airspeed fell too low.

Unquote

Lowering the nose after a low altitude engine failure, or even a inadvertant loss of airspeed, may sound obvious but in the heat of the moment the accident statistics say pilots are getting this wrong. I firmly believe this is an area where flight schools need to work harder to make sure students have an automatic condition response to lower the nose in the event of a loss of engine power or any stall indication.

Too many pilots are saying "well that will never happen to me", the problem is it IS happening to pilots just like you .......

[abgd]

You should not simply stall just because the engine has failed - in any flight regime - because the aircraft will continue flying at the trimmed speed.

I see that if you gradually reduce power over (say) a minute the aircraft will naturally tend to a trimmed glide without pilot input, but I don't see why there should be any law to the effect that if you suddenly cut power the aircraft will immediately tend to the glide attitude.

In fact, the drop in power could be considered instantaneous, but it will take a finite time for the aircraft to change attitude. During this time, it's airspeed must inevitably decrease and I would have thought that whether this decrease takes it below stalling speed would be very much type-dependent.

My reading is that Pilot Dar's experience tends to confirm this.

[BackPacker]

abgd, I have tried this, as I described earlier. Establish the aircraft in a stable, full power, Vy climb, properly trimmed and all. This is about a 20-25 degree nose up attitude. Release the stick altogether and chop the power.

The aircraft got into a wild speed/pitch oscillation and had not sorted itself out after 1000' of altitude loss so I called it quits. However, the aircraft did indeed NOT stall. (Although I don't remember if it got close enough to the stall for the stall warner to go off.)

Furthermore, even though Vbg is equal to Vy in this aircraft, the trim settings were completely different. With full power there is a lot of airflow over the all-flying tailplane, with zero power there is not.

[abgd]

I didn't mean to ignore your test - just to point out that IMO there's not a theoretical justification to think that the aircraft should simply sort itself out. Which I think your post also confirms.

[Pilot DAR]

there's not a theoretical justification to think that the aircraft should simply sort itself out.

This is key...

Most every plane I fly, I stall - numerous times, and in every configuration. I continue to be amazed at the variations in handling, and "sorting itself out" which I experience. Some are great, a few marginal, and one recent (Italian type certified) one did not enough pitch control available to prevent a stall in certain "normal" configurations. I like to find these little surprises with lots of altitude, and all other things being in hand. EFATO is not the time.

As obviously does the OP:ok:, if you commonly fly one or a few types of aircraft, it is excellent to cautiously experiment (with mentoring, if appropriate). The more you know about what your plane will do, the safer you'll be.

There have been remarks about stopped propellers, and the differences in drag. Yes, this is often the case, but I consider this risk to be well down the scale. For my experience, though failures of engines to develop useful power occur, seizures are rather rare. If your skills are good at handling the aircraft with an idled engine, the differences you could experience with a stopped prop will not be so different as to invalidate your skill.

Then there are the really odd situations.... Last week while flight testing a Piper Cheyenne II with an external survey modification, the left main gear decided it was very happy staying in the wing, despite my seletion for it to appear. After the methodical checking of systems, and repeated attempts, but before going to full blown emergency procedures, I decided that a few G's might coax it down.

Though I can happily find a bit more than 2G in a coordintated turn in most aircraft, and anger it closer to 3G by pulling once there - not it a Cheyenne! At speeds appropriate for the gear being extended, and with the flaps up, it gives an unmistakable stall buffet before you get to 2G. Two attempts at that were enough to convince me that fooling around accelerated stalls in a Cheyenne was not for me! Though perhaps it helped, as shortly afterward, the gear presented itself as desired, and it all worked out in the end.

It is easy to see that when something goes wrong, pilots can lose track of the primary role of mantaining controlled flight, and the problems compound. Engine failure during climb would be one of these opportunities. That probably factors well into the points with Big Pistons has presented, with which I completely agree!

[Sir George Cayley]

I'm told that an EASA proposal was that a single engined aircraft had to continue climbing after the engine failed:confused:


SGC

[bingofuel]

It will, briefly, until gravity overcomes inertia!

[Big Pistons Forever]

I'm told that an EASA proposal was that a single engined aircraft had to continue climbing after the engine failed:confused:


SGC

I would have thought a more likely response from EASA would be to create a committee tasked with creating a regulation that prohibited engine failures. :rolleyes:

[fwjc]

Big Pistons Forever -

"Like"

[Pilot DAR]

A stationary propeller produces less drag than a windmilling one

Well... I honestly am not sure about this - I do not know.... I have heard both sides of this presented very persuasively. For the few times I have experimented with this, my own results have been inconclusive.

Can anyone present data which will support a position on this either way, and if so, how much different? I agree that any flight manual I have seen has either said "windmilling" or "feathered". So, perhaps there is a difference, which manufacturer's know, but don't want to say....

[24Carrot]

Sorry, no data, but the "Windmilling Drag" or backwards thrust must surely depend on how fast the blades are turning, which must in turn depend on the friction provided by the broken engine.

Which might vary a lot, depending on what broke?

I suspect that zero-friction is the worst case, where the propeller does it best to resemble an auto-rotating helicopter.

[Mark1234]

If the prop is windmilling in a steady state at some RPM, it must be absorbing the right amount of energy from the airstream to balance what is needed to turn the engine over at that RPM against the internal friction, compression, pumping etc. If it were true zero friction, it would absorb no power.. if it can't overcome the friction, it will stop.

If the prop is stopped, you're looking at whatever is pretty close to the drag figure for a flat plate of the same projected area at that speed - that should be pretty straightforward to figure. My gut feel says less, in the same way that a stalled wing can lift less than an unstalled one (the prop is just a wing..), but honestly I don't know.

Having actually performed a glide approach with the engine shut down, I can't say the transition from idle power to windmilling, to stopped was noticable - I wasn't looking at the VSI etc., but in terms of seat of pants perception of glide angle, it wasn't evident. It did take a lot of effort to actually stop the prop however (well into stall buffet). I did find it a little disconcerting that the prop suddenly stops being transparent, and in this case (decathlon) with centreline seating, it chose to stop dead vertical, right in front of my nose, no missing it..

[n5296s]

Wow, lots of responses and discussion here - nice to see, thanks everyone.

@BPF - Big difference between Vy and Vx as Vx is typically less than 10 knots above power off stall.
I meant to try Vx too but for various reasons didn't. Next time. In my 182 though Vx (at gross) is 78, Vs1 is 51 - closer to 30 knots than 10.
@UV - What did you do during your simulation? I suspect you reduced the power to idle... By throttling back to idle you still have a little residual power (put simply somewhat like feathering) which reduces drag. It can be nothing like the real thing!
Conventional wisdom is that an idling prop is actually generating net negative thrust (obvious really since RPM at idle in flight is higher than on the ground). I suspect the effect on the overall situation is small.
@UV - Try stopping the engine at altitude with the mixture control, then switch off both mags. You may be quite surprised at the result, and thats probably without the prop actually stopped!
I'm all in favour of a bit of experimental flying, but I don't harbour a death wish, so no thanks. But STOPPING the prop according to most conventional wisdom actually REDUCES drag. Otoh unless the engine suddenly and immediately seizes (unlikely but of course possible), it's not something you're going to have time for in an EFATO. I don't see why a stopped prop would have any impact on trim/attitude but maybe you can explain.
@maxred - n5296s - the discussion you refer to revolved around TURN BACK
I know. I started it, or at least a segment of it. But at some point the question of what happens immediately following the engine failure led to a discussion and I said I'd try it. Just delivering on my commitments...

[Big Pistons Forever]

I meant to try Vx too but for various reasons didn't. Next time. In my 182 though Vx (at gross) is 78, Vs1 is 51 - closer to 30 knots than 10.
..

For the C 182 Q POH

Vx (best angle of climb) is 54 kts

Vy (best rate of climb) is 78 kts

Stall speed at flaps 20 is 47 knots

There is only a 7 kt spread between Vx and flap 20 stall speed

[Ex FSO GRIFFO]

Hey Mark,

At least you have the starter motor available to make the prop horizontal - if that's what you prefer....gets it out of your view, and eliminates the prospect of a ground strike....

I saw the results of a Cessna 310 with twin bladed props many years ago, who had a gear problem, so he arranged to get over the smooth grassy area, cut both engines, then use the starter motor to make the props horizontal before gliding it gently onto the grass at SY (AUS).

Minimum damage was the result.

Although I do believe the current thinking is to put it onto the bitumin instead...but that's another thread.
:ok:

[Mark1234]

Not really! The 'stopped' prop would 'creep' over the compression of the engine every few minutes, and complete a half revolution to stop precisely vertical again.. I'm thinking that with a 4 pot, 4 cycle engine the compression 'stops' are going to be 180degrees apart, so with a 2 blade prop it is a simple matter of where someone bolted it on. Much more scope with > 2 blades and/or > 4 cyls I guess.

n5296s, just on the idling prop; While an idling prop has net negative thrust, it has *less* net negative thrust than an idling prop without the engine running - the chemical energy from the fuel is helping to turn the engine over. Whether that has a significant effect upon glide profile I am not sure.

P.S. No deathwish was involved, engine was stopped at 5000agl directly overhead a large, but quiet airfield at the instructor's suggestion. I found it a worthwhile and interesting exercise, though I'm sure the practise is a little controversial.

[BackPacker]

P.S. No deathwish was involved, engine was stopped at 5000agl directly overhead a large, but quiet airfield at the instructor's suggestion. I found it a worthwhile and interesting exercise, though I'm sure the practise is a little controversial.

I have done the same thing, once. Very interesting exercise. But, like you, done at 5000' above a place with plenty landing opportunities. (In my case, lots of fields instead of an airport. And yes, I had my field selected beforehand.)

During my EFATO tests I tried the difference in effect between an idling prop (throttle closed) and a windmilling prop (mixture closed). The difference was in the order of 100-200 fpm. I have not (yet) established the V/S with a stopped prop.

But as others said, with a fixed pitch prop, actually getting the prop to stop is very hard in itself. So unless the engine has completely seized solid, you will have to do your forced landing with a windmilling prop. Maneuvering to get a windmilling prop to stop (which required a half G pushover at well below Vs in the R2160 when I tried it) is not something you are going to experiment with in a for-real situation.

[Pilot DAR]

On the other hand, if you do get the prop stopped in flight, it's going to be really hard to get it turning again without a starter. I used to fly a J-3 on skis, whose engine had no accelerator pump, and did not idle worth a darn. The result was that sloppy use of the throttle would make it quit, an I would have to dead stick onto the forzen lake (which was my intended landing site anyway). There, I would hand prop it.

I have windmill started my 150, having stopped the prop in flight. It's not easy, you've got to be going really fast.

[BackPacker]

It's not easy, you've got to be going really fast.

140 knots in the R2160. That required over 30 degrees nose down, and I lost about 2000 feet in the process.

[n5296s]

@BPF - For the C 182 Q POH...

Interesting. The figures I gave weren't quite right (shouldn't enter them without the book in front of me) but they were close... from the 1980 TR182 POH (which I DO have in front of me now) Vx=75 KIAS/KCAS at sea level, Vs1=54 KCAS at sea level (39 KIAS). So a broader spread than the fixed-gear version.

[Pilot DAR]

Vx=75 KIAS/KCAS at sea level, Vs1=54 KCAS at sea level (39 KIAS). So a broader spread than the fixed-gear version.

Just 'cause I'm curious (and I no longer have access to the flight manual for 182 RG I used to fly), is Vx flown gear up, or gear down? I'm presuming gear up, and therefore, the speed could be a little faster, as there is much less drag from no gear, to make a slower Vx the better performing speed.

That said, you are trying to keep Vx slower, so you don't make too much speed across the ground, to bring down your apparent angle. When you actually start doing the math, it's a real numbers game, and not always what it would appear!

[n5296s]

Good question. I'm at work now and don't have the book in front of me, but it seems pretty obvious that it should be gear up. I mean, here you are desperately trying to get enough altitude to get over the cliff face that is inconveniently just in front of the runway... you're not going to keep the gear down.

[Big Pistons Forever]

Interesting. The figures I gave weren't quite right (shouldn't enter them without the book in front of me) but they were close... from the 1980 TR182 POH (which I DO have in front of me now) Vx=75 KIAS/KCAS at sea level, Vs1=54 KCAS at sea level (39 KIAS). So a broader spread than the fixed-gear version.

It is interesting why there would such a difference in Vx between the two C 182 models.:confused: I did note however that the TR182 POH sprcifies that for a short field takeoff a speed of only 59 KIAS (60 KCAS), or 6 knots above stall speed, is recommended until all obstacles are cleared. I wonder if the Vx in the later models was raised to provide a larger margin above stall.

In any case I would not want to be 50 to 100 feet off the ground with a pitch attiude neccessary to maintain 59 kts, and have the engine fail :eek:

[n5296s]

That's odd, I agree. It led to quite a conversation between me and my instructor when I was doing my CPL. Aerodynamically, Vx is Vx - there can't be one Vx at takeoff and another at altitude (once you're out of ground effect anyway).

One possible explanation is that the PoH also says that a short field takeoff should be flown gear and flaps down until clear of the obstacle. This makes absolutely no sense to me, though of course I did it for the CPL. If I've got a tree looming up, the last thing I need is the huge extra drag of the gear. But maybe 58 knots is Vx gear/flaps down, which would make some sense.

[Big Pistons Forever]

Actually not retracting the gear until clear of obstacles does marks sense as when the gear retracts it turns sideways significantly increasing its drag over the profile it has when extended. Since the gear retraction is a somewhat leisurely processes the effect will persist for awhile. This extra drag is even more noticeable on the older style high wing retractable Cessna's fitted with main gear doors.

[n5296s]

Makes sense for the FAA-approved 50 foot obstacle, I guess, since you'd be clear of it before the gear had finished retracting anyway. Makes less sense for something significantly higher. I agree that there is more drag for a few seconds as the gear retracts, but no idea how much and hence how to figure out the obstacle height where retracting first works in your favour.

[Humaround]

I'm curious why, in the normal course if events, anyone would want to climb at Vx? Obviously if there are obstacles to clear. But otherwise? Surely better and much safer to climb at higher speed, giving more time to react if the donk stops. And better cooling for the engine, too.