I took the trouble to read my way through the whole thread (which is very interesting b.t.w.!) and come to the same conclusion: A purely academical exercise.

Almost impossible to teach (at least to PPL students or low-hour private pilots) and entirely dependent on the nerve strength of both the instructor and the student. Doing 45+ degree banked turns below 500ft will scare the hell out most pilots (instructors included) even if a well researched paper predicts a favorable outcome of the maneuver. Training this procedure at a safe altitude (or in a simulator) will not help at all, because all the human factors that make it so difficult in real life will be taken away.

I would rather opt for a software solution (like an app for a smartphone or iPad) that will emit a signal once a speed/position/altitude has been reached, from which an averagely skilled and trained pilot should be able to fly a successful return maneuver within the normal operating envelope of his aeroplane (best glide speed +/- 10kt, turns with no more than 30 degrees of bank, ample reaction times). If your device hasn't beeped yet, you glide to a landing straight ahead, if you heard it beep, you are safe to turn back.

n5296s, the input was very useful. Though in real life I guess the prop might be windmilling?

Thanks also to Big Pistons Forever, I have seldom seen such concentrated common sense, especially "under fire".

It seems to me, (by no means a "Sky God"), that I wouldn't even consider the turn back under 500 ft in a C172 or similar, by which time I'm typically turning onto cross-wind, or even earlier for noise abatement etc, as somebody else posted.

So could we perhaps tone down the "Health and Safety" warnings a little if we re-cast the question as "what best to do with Engine Failure On Crosswind?" (EFOC?)

What we generally need to know is how much turning an average pilot can squeeze into any given height, especially if much draggier than usual because of a windmilling prop.

Just my 2c.

I suspect that may be possible, or (me being a bit of a technological luddite) possibly a graph for any given aeroplane which can be referred to before take-off and says for runway length X and headwind Y in this aeroplane, below height Z you will not get away with a turnback, but above that you have a fighting chance if flown well enough.

There are four main reasons to do the research in my opinion, which for me are:-

(1) It'll be fun and interesting

(2) It'll give people who might consider a turnback a clear indicator of when NOT to consider it.

(3) There are lots of runways without a land ahead option. Runway 23 at Lee on Solent, runway 19 at Oban, runway 29 at Sheffield City, Grand Canyon Airport in Arizona, plenty of small island airports or mountain airports around the world. Nobody's going to close these runways - so if they're going to get used, perhaps pilots flying there should have some material they can use for training and practicing turnbacks.

(4) See (1).

G

If you are on a crosswind leg and the engine fails you are probably at or above 500 feet AGL and a 90 degree turn with 30 degrees of bank to the nearest flat surface within the aerodrome boundaries would IMO be completely doable for the average pilot and is in fact what I tell my students to do for this particular situation.

True. But if you had the presence of mind it would be windmilling in coarse pitch (assuming a constant speed prop). I'd LIKE to compare the difference between a prop windmiling in coarse pitch and one under idle power in fine pitch, but that's beyond my nerve/stupidity. I always do power-off exercises in fine pitch hoping that the two will balance out, more or less.

Of course if you have a fixed prop this is all irrelevant.

Pulling the prop to full decrease makes a very significant impact on the glide performance and should be part of the shutdown check on every aircraft with a constant speed prop.

Interesting experiments.......I might go and try some of my own.

I read a very good article once about EFATO. It was written by some really experienced aerobatic pilot (I forget where I read it, US AOPA I think), and the best bit of advice he gave was "push until the ground fills a two thirds of your windscreen". This is very good advice, as we had an EFATO recently from our airfield (below 500') and as is common the aeroplane stalled. Luckily for the pilot he was not that high but he still managed to end up mushing down and cartwheeling along the airfield as the wing dropped and "landed" first, and he was trapped underneath in a fuel loaded smashed up aeroplane. Hurt but not too seriously but covered from head to toe in avgas.

My buddy was there and sent me some pics, very scary stuff. Luckily all his holes didn't line up that day.....a bit higher, a bit faster, and I expect there would have been a different result.

Whatever you do, keep flying :ok:

I suspect that may be possible, or (me being a bit of a technological luddite) possibly a graph for any given aeroplane which can be referred to before take-off and says for runway length X and headwind Y in this aeroplane, below height Z you will not get away with a turnback, but above that you have a fighting chance if flown well enough.

G - Yes a good theoretical experiment, that may yield some interesting and anecdotal information reference said aeroplane. But that is the point, it will be that particular aeroplane, with a particular weight, with a particular configuration. The concern Backpacker was alluding to, correct me if I am wrong BP, was that someone MAY, read this, and think, I am going to try that next time I take off in my PA28 or whatever, and guess what - the 45 degree bank angle et al might not work.

This is Bonanza specific. You fly landing approaches at 1.3 times stall speed (good compromise between low speed and safety margin above stall. Therefore in a 45 degree bank angle, multiply the stall speed by 1.3 and you get a turnback speed with same safety margin for stall - app Vy.

Also, which I do not think anyone has pointed out yet, high density altitude airfields make this almost impossible because the climb gradient is flatter, and high weight also reduces climb gradient which in turn raises your stall speed, and turn speed, which increases your turn radius and decrease your rate off turn.

Let your watchward be take care.

Well, I guess the process is no different than trying to create a graph for, let's say, landing distance. You first figure out what the proper and achievable technique is. You then figure out the baseline results for ISA conditions, MTOW, nil wind etc. And from there you start figuring out what sort of difference different conditions make.

Having seen, and worked with the landing distance graphs in the PA28 POH, I am convinced that it should not be beyond the average PPL to use a more or less similar multi-step graph to figure out the turnback height. But you are right: That graph will be aircraft-specific. Although the research may yield some generic results that would make it easy to adapt the graphs to different aircraft.

And I would not be surprised if the eventual graph contained a "here be dragons" area or something like that. Meaning: Do not go here or there's absolutely no way out. Don't single engine helicopters not have a similar graph for that part of the flight envelope where you cannot survive an engine out (because you're too low and slow for autorotation, but high enough to end up dead)? Pilot training should then simply be about risk mitigation and avoidance of those circumstances.

The easiest way to mitigate the risk is not to turn back below 1000 feet AGL a simple easy number to remember and conservative enough to work pretty much all the time.

Risk mitigation is fine but if you want to be holistics about mitigating the risk then you should mitigate the risks for the whole flight. Instead of obsessing about the Turn back between 400 feet and 1000 feet I think the time and energy should be spent on other areas which experience has shown is much more likely to present a risk to a flight like

-practicing the very late overshoot, or

-short field takeoff and landings including a practical stratagy for determining a go no go point, or

-practicing the 180 degree turn on instruments that you would use on an inadvertant IMC encounter, or

-determinimg exactly what your cruise fuel burn with a series of planned test flights etc etc..........

Well, I'm not sure I'd call 2 hours of flying, when I just felt like flying and din't have anything very specific to do, exactly OBSESSING. Just speaking for myself here. I do do all those other things too...
Indeed. It's generally called the Dead Man Zone, just in case the meaning isn't obvious. But, unlike taking off in an aeroplane, it is avoidable. You don't HAVE to hover at 200'. (Well, actually for quite a lot of heli missions, e.g. long-line work, you do, but that's too bad). Whereas you do HAVE to be at 200' and Vy in an aeroplane, if you want to fly it at all.

Different topic: over on the Cessna Pilots Association, there was some discussion about how trim plays into all this. As in, you're 15-20 degrees nose up at Vy, and the engine stops. Suppose you just let go of all the controls. What happens? Some time soon, the plane will be flying at Vy (assuming you were trimmed in the first place). It may have lost a bit of altitude getting there, but it won't be stalled. I'll give this a try next time I get a chance.

I still don't really see how you get a real developed stall doing this, except maybe trying to arrest the descent close to the runway. I was actually surprised at how benign things were pulling back below stall speed in a descending turn. But clearly people do :-(

Be sure to let us know the results.

I see two issues here though.

First is that people are taught to fly by attitude, and by looking out the window, not by chasing the needles. Especially early in their career, VFR. So they will not let go of the controls but intuitively try to keep the nose up, until they realize what happens. But by then it may be too late.:=

Second is that even if you let go of the controls, there will be a lot of inertia in the aircraft, and it will take a while to settle in a stable glide (again at Vy, assuming it was trimmed for Vy in the first place). During that period there may well be speed excursions below Vy, and even below Vs, but these, I would imagine, will be at a load factor lower than 1G. I would be very, very surprised if the aircraft were actually to stall in this scenario.

Anyway, the point being that the oscillations before reaching a stable Vy descent may take too long. If you fly the maneuver instead of letting go of the controls you reach that stable state much earlier, and have much more opportunity (read height) to figure out what to do next (whatever that may be).

Okay, I've got the plane reserved for Saturday morning. The weather looks good and I'm keen to explore some of the EFATO turnback issues discussed here.

Please critique my "test plan".

Key numbers from the POH:
Vs = 63 knots (at MTOW)
Vbg = 78 knots (at MTOW) - supposedly yields a 1 in 8.7 glide ratio, and a V/S of ~900 ft/min (to confirm)

Vs in a 45 degree stable descending turn is thus theoretically ~75 knots. Close enough to Vbg to not need to explore the difference in performance between pulling to the stall at 45 degrees, and flying Vbg at 45 degrees. In fact (but to be confirmed) I expect the stall warner to sound continuously, with noticeable buffet, in a 45 degree Vbg stable turn.


Preliminary actions:

HASELL

(i) From height, establish the aircraft in a Vbg stable, trimmed descent. Mark attitude with some sticky tape on the canopy. Write down/time V/S.
(ii) From height, establish the aircraft in a Vbg stable, trimmed descent at 45 degree bank left and right. Mark attitude again with some sticky tape. Write down/time V/S.
(iii) From height, establish the aircraft in a Vbg stable, trimmed descent with the mixture closed. Write down/time V/S.

(i) and (ii) are so that the proper attitude for the subsequent tests can be set quickly. (iii) is to provide data on the difference between throttle closed and mixture closed (prop windmilling), so that all subsequent data points can be corrected for that.

Test items:

All items start from a stable, properly trimmed Vy climb along a line feature, into the wind as far as possible. When passing through a certain safe height (3000' most likely, weather permitting) I'm going to pull the power and:

a. Hold the nose up to see how long it takes for the aircraft to stall. Then recover the stall and see how much time/altitude I lose before in a stable Vbg glide.
b. Hold the stick in the same position, furthermore as in (a).
c. Completely release the stick, let the aircraft sort itself out, furthermore as in (a).
d. Fly the aircraft in a half-G pushover, furthermore as in (a).
e. Fly the aircraft in a ballistic arc (zero G), furthermore as in (a).

(a)-(e) are essentially to see how much "oh !!!!" time I've got, and how important ingrained reflexes are. If, for instance, (a) delivers the same as (d)/(e) (which I don't expect), the "oh !!!!" factor isn't all that important.

Subsequent tests include one second of "oh !!!!" time after closing the throttle, then:

f. Establish Vbg according to the best technique from (a)-(e), then turn into the crosswind 225 degrees at 45 degree bank, turn the other way 45 degrees at 45 degree bank.
g. Same as (e) but with 60 degree bank
h. Same as (e) but with 30 degree bank

(g) and (h) essentially to confirm Rogers

i. Like (f) but start with a left turn 45 degrees away from the crosswind, then turn 225 degrees into the crosswind (teardrop thus reversed). Altitude lost should be the same as (f), but you never know.

j. Like (f) but this time roll into a 45 degree bank while pushing the nose down (at half or zero G as appropriate), ready to pull as soon as the Vbg attitude has been set.

k. Like (f) but use the Mark1234 technique of immediately rolling to 60 degrees bank, pull to the stall warner while the nose drops, play angle of bank against stall once the nose is at the Vbg attitude.

At the end of all items a-k, write down altitude lost.

Further things to note: QNH, OAT, actual weight, actual wind. I will also run an outdoor-type GPS which has track recording so that I can see how big the teardrop shapes are and whether the 225/45 turn without any pause inbetween gets you aligned with the runway centerline. (Any suggestions on how to make sure I'm later able to correlate the GPS track with the test results that I wrote down? Other than making sure my watch is synced with the GPS time, and write down the time of each individual test?)

Any tips on record keeping? Anyone has a handy form or something for this kind of stuff?

Other safety items:
I'll be doing this in the R2160 which is cleared for aerobatics, will be flown within the "A" limits, and I am very current in aeros/spin recovery etc. in this aircraft. I have in the past pulled the mixture to ICO in-flight and I know the engine will windmill above Vs and catch immediately as soon as mixture is restored. Everything will be done at altitude, over an area with plenty emergency landing opportunities. Airspace will be the Rotterdam TMA (class E) with permission/traffic service from ATC.

Any additions?

I think the GPS is a good idea, as you never quite know what you will want to analyse later.

Noting the QNH, ground temperature (and ideally the OAT at some relevant height) allows for some CAS/TAS conversions later.

Which is important because at altitude your TAS should be higher and your turn rate (at any given bank angle) should be slower, though the angle of descent should be the same.

Which, if I have this right, means that real turn backs are harder on low density days, with the height loss inversely proportional to density.

I don't have very high hopes for that GPS. Best it does is one track point per second. And of course the vertical accuracy can be as little as +/- 50 feet. But since I've got that GPS anyway, why not?

A fellow pilot has some professional equipment that's used to accurately track racecars, and he's trying to adapt that for aerobatics tracking. Maybe once that's ready we'll give it another go.

The GPS might be more useful than you think. During my PPL training I usually had a Garmin H eTrex in the back. It was much more useful for Nav than analysing circuits though. (I did an awful lot of circuits:ugh:).

The absolute height accuracy can be well off, eg I tracked my first helicopter solo, and found I picked up and set down at -200ft agl, and no I wasn't in the Netherlands! However the error can be quite stable, which means descent rates can be quite accurate. Headings and turn rates require some assumption about the wind, but can also work out well.

Assuming the unit had enough satellites to start with, the big problem was when it lost or acquired one. That was when you got the sudden jumps. I was pretty sure of the reason because in a circuit the same jumps always occurred when I turned through the same headings.

If you analyse it in a spreadsheet, you can work out the energy change per time step, omit the "impossible" jumps, and look at the trends on either side. So with some effort, even "hiker GPS" data can be useful.

------
Incidentally, in case any PPL students reading this might think GPS tracks are a good way to analyse circuits, my experience suggests not. I never learned anything useful from them. By contrast, GPS tracks were incredibly useful during Nav, not least for showing the FI you had actually gone where you said you did!

I got an excellent plot off my GPS from my recent real turnback with a partial loss of power. It helped a lot in my personal debrief.

Backpacker - I do a lot of assessment of flight test plans and flight test reports, but within a structured test environment. I would recommend if you are going to do this safely that you make sure you understand briefing and debriefing for flight test, get somebody else to listen to and critique your brief, understand the use of test cards, and probably fly with an experienced pilot as a safety observer. Also make sure that all of the aspects of flying you'll be demonstrating you are fully current in, before putting them together. The standard technique for assessing difficulty of flying a manoeuvre (usually referred to as "pilot compensation") is the "Cooper Harper Rating Scale" - there's a lot about it on the web, but you may find it useful to run your CH task construction past a test pilot who has done the formal training in its use.

G

GPS altitude is normally accurate to within 10-20ft - so long as you get good reception.

However some older products, notably those using the popular SIRF-2 chip, had a systematic altitude error of about 160ft. I can't remember which way it went. Basically they did not correct for the deviation from the ellipsoid, IIRC, which in the UK is a certain fairly constant figure.

Just a couple of thoughts.

IF you can safely set up a small video (eg a gopro) that can film the panel / altimeter, start it, film the GPS time, then let it run. That way you can sync the two up. Just a variation on the theme of photographing the GPS time so you can geotag photos after the fact. There are also some GPS's (e.g my cycle one) that have baro altitude.)

The other thought is that you don't seem to have a way of considering positioning. I'm not sure it is feasible, but for example e.g. if method 1) makes the turn in 300ft, but leaves you 1/2 a mile further away than method 2), which takes 400ft, which is favoured?

Otherwise, I look forward to the results. In retrospect, I think the my approach probably isn't/wasn't representative, as it pre-supposes you're making the turn.

Pretty irrelevant - between start and end of the manoeuvre, the total altitude change will be below 1000ft in probably any light aeroplane. In analysing something like this, it's just differences, not absolute values - and the GPS is in that case, giving a more exact reading than anything a pressure altimeter will give you anyhow since it's unaffected by PEC.

OAT and pressure altitude will give density altitude near enough, and those can just be recorded at stable conditions before manouevring.

Even if flying take-off to landing for a manoeuvre like this, the same applies for data analysis because GPS will flatline on the runway (and given that PEC varies with airspeed and may be affected by ground effect, a pressure altimeter may not, although it'll be close enough to allow analysis if you had airspeed as well).

G

If you do a 360 turn at constant speed and bank, you can calculate wind speed and direction at that height from the GPS points.

Chris N.

Chris, good one. I'll do that at the start.

Mark, unfortunately I do not have a decent camera, and the half-decent (actually, more like quarter-decent) camera I have will be hard to mount properly. Nevertheless, I'm going to see if I can make it work with that one.

I had a different experience. I was using a garmin gps60 from memory to log takeoff performance with a sample rate IIRC of either 1 sec or 1/2 sec. There appeared to be some interaction between the aircraft acceleration and the altitude. I started motionless at pretty much Airfield Elevation. As I accelerated, I dipped about 20 feet below the surface and then seemed to come up to the surface at about rotation, and then showed a normal climb out. It made it quite difficult to assess where the actual wheels off the ground event occurred. The deepest 'under ground' was at slightly over half my estimated ground run and there wasn't a really clear trajectory change anywhere. (As one would expect in a series of max gross takeoffs the initial climb was quite slow)

Been a bit busy the last couple of days so missed this reply to my post - but what a STUPID comment. NOBODY, least of all me said that a turn back is the ONLY way to deal with a EFATO - landing ahead should always be the prefered option if the area allows and AFAIK this whole discussion was about if it could be done IF NEEDED this is of course going to mean it will only be for a limited number of locations and NOT

Okay, trip report. This was, as planned in the Robin R2160. Actual weight was approx. 730kg, vs. an MTOW of 900kg. QNH 1027, OAT zero C @ 3000' which was my "close throttle" height for most tests. OAT on the ground 4 C, dewpoint 3.

From the POH: Vs 63 kt, Vbg 78 kt, Vy 78 kt.

V/S at Vbg, zero AoB, with throttle closed was 800-900 ft/min.
V/S at Vbg, 45 AoB with throttle closed was 1100-1200 ft/min. Stall warner was going off, but no noticeable pre-stall buffet was felt.
V/S at Vbg, zero AoB, with mixture closed was 1000-1100 ft/min.

The sticky tape on the canopy was really useful. It allowed me to set the proper attitude in the subsequent tests immediately.

The difference between throttle closed and mixture closed is between 10 and 20%. So we need to correct all subsequent measurements for that.

All V/S were from the VSI. I did not want to spend a lot of time descending through 2000 feet or more and timing the descent with a stopwatch. Maybe next time.

A couple of surprises here.
- The fuselage angle between Vy (full throttle) and Vbg (closed throttle) is actually only about 40 degrees or so (20 degrees nose up vs. down). Because of this, it really doesn't matter whether you do a zero or half G pushover, or simply let the nose drop all by its own. In all cases (except a and c) I was fully established in a Vbg glide while I dropped through the "close throttle" altitude.
- Holding the nose up as in (a), it took 4 seconds for the stall warner to sound, and 7 seconds for the aircraft to be properly stalled. But even then, I was able to recover to a stable Vbg glide 100' below the "close throttle" altitude.
- Completely releasing the stick (as in (c)) got the aircraft in a huge pitch/speed oscillation. I let the aircraft descend over 1000' but by then it had not yet recovered from the oscillation so I ended that test prematurely.
- Despite Vy being equal to Vbg for this aircraft, there was a *huge* difference in trim. With the R2160 the all-flying tailplane is in the propellor slipstream and that will be the primary reason. I expect this to be different in different aircraft.

I actually used two seconds of "oh !!!!" recovery time after closing the throttle.

With (f) I was able to consistently complete the maneuver in 450'. At the end I was in a stable Vbg descent, and all I would need to do is flare. That flare obviously takes a few feet. The (limited, see further down) GPS data I have suggested that the 225/45 degrees works quite well to regain the centerline in the no-wind situation. I was too high up to confirm this overhead a line feature though.

(g) and (h) gave me 800' and 700', respectively. I only did this once but the data suggests another confirmation of Rogers.

(i) was also done in 450'.

Both (f) and (i) felt really comfortable. Especially during the 225 degrees turn, there is time to do something else, like issuing a very quick mayday call to make sure the runway is cleared.

This did not make a significant difference. Dropping the nose from Vy climb to Vbg is a very quick process, and so is rolling to 45 bank. I did this test only once and lost 450', just like (f) and (i).

I tried this a number of times. It feels like a very aggressive maneuver which requires very precise aircraft control. I would not be happy doing this for real - it feels like it's too easy to lose control. But it did give better performance: The full maneuver was completed in 400' both times I tried.

The track recording of the GPS turned out to have an interval of approx. 5 seconds. It doesn't really help me with the analysis. I also had a camera installed in the cabin (which was indeed not even half-decent) but unfortunately it was not aligned properly and the instruments are not visible.

To summarize, I was able to complete the turnback consistently in 450', with only moderate technique. Obviously this was a very limited test and that number should be corrected for a number of things:
- I was not able to confirm that I would be able to make the runway. Just that the turnback would be completed in that altitude. Properly reaching the runway might require a further glide and thus further altitude. And obviously all this depends on climb performance anyway: If your climb performance (angle) is less than your best glide performance (angle) you'll never reach the runway even if you were able to turn back without altitude loss.
- I flew all maneuvers with the throttle closed. As my initial tests showed, we need to correct the numbers with approximately 10-20% to compensate for the residual idle thrust.
- All figures still need to be corrected for OAT, altitude, QNH and the fact that I was well under MTOW.
- All figures are specific for the R2160
- All figures are specific for *my* pilot technique. YMMV.

Personal lessons learned:
- You have about 2-3 seconds to recover from the "Oh !!!!" reaction. It doesn't matter all that much how fast you react and how aggressive you pushover as long as you don't stall. Whatever energy you lose on the way up, is recovered on the way down. (Obviously the longer you dally, the further you end up from the runway.)
- You have to fly the aircraft to the Vbg stable attitude. Just letting the plane sort itself out doesn't work, especially considering the huge trim change.
- Knowing the proper Vbg attitude is very important. The faster you can set this, the earlier you can start the turn and concentrate on other matters. And the Vbg attitude was less nose down than I thought.
- The turnback (225/45 or 45/225 as appropriate, at 45 AoB), with the engine windmilling, will take *at least* 500 feet, and possibly closer to 600-700 feet under the wrong circumstances. You then might need some additional altitude to reach the threshold. For me personally, this means I'm going to keep on using 1000' as minimum turnback altitude, assuming a straight-out departure. (Turning crosswind at 500' obviously greatly improves your chances to make the runway though.)

Note: All the usual caveats apply. Do not change the technique you've been taught just because some anonymous bloke dumped a few numbers plus some mumbo-jumbo in an anonymous internet forum. If you decide to deviate from the technique that's been taught to you by a qualified instructor, you do so at your own risk, and I suggest you first validate your prospective technique yourself. Yada yada.

April 12, 2012

A light aircraft that crashed into two houses in Peel Green in Salford, killing the pilot, probably suffered a fuel supply problem, an air accident report revealed.

No-one on the ground was hurt but the Piper PA38's pilot, Ian Daglish, 59, died later in hospital, the report from the Air Accidents Investigation Branch (AAIB) said.

Mr Daglish's 19-year-old passenger Joel McNicholls was seriously hurt in the crash on the morning of July 29 last year.

Leaving Manchester/Barton City Airport, the aircraft suffered an engine stoppage on take-off at about 200ft. It rolled to the left, with the extension roof of the first house most likely being struck by the aircraft's right wing.

The underside of the aircraft then hit the side wall of a neighbouring house, with the wrecked and on-fire aircraft coming to rest in a driveway between the two homes.

The AAIB report said: "The account of the passenger and the findings from the investigation support a fuel supply problem as being the most likely cause of the engine stoppage."

The AAIB went on: "Although other potential causes for the engine stoppage could not be eliminated from the investigation, the most likely cause, based on the available evidence, was that stiffness of the fuel selector valve and wear on the rod connecting it to the selector handle, may have resulted in the valve being in an intermediate position during the take-off.

"This would have reduced the fuel flow to a level too low to sustain continuous engine operation."

The report added: "The suddenness of the engine stopping and the limited time available to react to it probably resulted in the pilot omitting to lower the nose before the aircraft stalled.

"Once the aircraft stalled, it is highly unlikely that he could have recovered the aircraft in the height available." Father-of-two Mr Daglish, from Alderley Edge, Cheshire, was a military historian and wrote a series of books about Second World War battles.

He described himself on his website as a "battlefield mythbuster". Colin Maher, whose home was hit, said at the time that he had run into his garden and saw the plane alight. "I heard a man shout for help and just put a hosepipe on him," Mr Maher told the BBC.


The above was quoted from another post.

Sadly another fatal EFATO crash that started with a departure stall. Theorizing about turnbacks is all fine and dandy but too often the point is moot because the aircraft was allowed to stall when the aircraft lost power. If you are looking for something new to practice on your next local flight establish the aircraft in a Vy climb, in trim, and retard the throttle to idle. The object is to go right to the glide attitude. In other word after the aircraft has stabilized its speed should be right on best glide without you having to make any further adjustments to the attitude. It is harder than it sounds and a good way to build the automatic reactions that will save your life.

As the lead theorizer about turnbacks, nonetheless I agree with you totally.

If, for any reason, an aeroplane is permitted to stall anywhere between about 20ft and 400ft, almost certainly all bets are off, the height loss will probably take the aeroplane into the ground out of control.

This is irrespective of what you are going to try and do next. There just isn't time and height for a pilot who wasn't expecting a stall to recover.

G

What worries me about the PA38 accident above is that pilot (who I knew, incidentaly) did not lower the nose when the engine failed yet transmitted a mayday and may have (according to the report) also switched off the mags and fuel.

Having suffered a partial engine failure in a Chipmunk at almost the exact same spot as Ian's accident (off 09R at Barton), I know the importance of INSTANT forward stick cannot be over emphasised. All the other stuff (radios, mags, fuel) can wait (especially the radio).

It's very hard to put yourself in the head of somebody who has had an accident and sadly is unable to talk about it.

However, I know that at a few points in my flying career where I've returned to training I've been picked up for following "procedure" (checklists, radio...) in preference to the primacy of controlling the aeroplane. As an instructor I've seen this in other pilots as well.

That *may* perhaps be what happened.

G

Agreed we'll never know for sure.

But I think there's a lesson here for pilots and instructors regardless. ALWAYS FLY THE AEROPLANE! If you haven't got the stick and rudder stuff right, nothing else matters. Only when the aeroplane is stable and under control should anything else be even considered. Checklists have their place, but YOU MUST FLY THE AEROPLANE as the number one consideration, regardless of any checklist or procedure. And last of all comes the radio.

The old addage 'aviate, navigate, communicate in that order' was never more true.

Couple of questions from a low hours PPL:

A stall during an EFATO when properly trimmed has to be induced by the pilot pulling back, allowing speed to bleed off. From my limited experience flying C172/PA-28, the amount of pulling back to cause such a stall has to be fairly significant and not easily missed. Is it easier to miss this sign than I realise, or does panic simply cause pilots to miss this important warning sign?

As for turning back, what are peoples views on doing so from a crosswind position? Is it worth making a turn onto crosswind even if for you'd otherwise be departing straight out as it keeps you closer to the airfield and if the airfield has crossing runways then a 180 turn could have you lined up nicely on final for an xwind landing?

Says it all, really.


well, it depends on the exact type and configuration. Also, if you see the ground rushing up at you the _tendency_ is to pull back to avoid it. Sometimes called groundrush (or panic, if you prefer)

Control the crash. If you don't control the crash, you will probably die.

As a student (note my lack of experience in what I'm saying below), I've always been taught to push forward on the stick to get the plane into the glide attitude.

You have to remember that when the engine failure happens, you are probably going to be in the climb attitude and trimmed as such. I would be surprised if there are many aircraft which have the same trim settings for climb and glide!

When you practice basic stalls, you normally start from straight and level flight, then reduce the throttle and try to maintain the straight and level attitude. The speed will of course slowly wash off, in such a way you can feel the different stall characteristics of your particular plane. It might be useful to try stalling from the climb attitude (I would talk to an instructor before trying this!) - the stall is likely to be quite different- speed loss will be much quicker. Gravity isn't your friend...

Yes. And the other one you should be taught is stalling in a descending turn (mostly with the intention of getting you to avoid it), to try to reduce the number of people who kill themselves turning final with a tailwind.

If the aeroplane is climbing after take off at full power, it will be trimmed for that - no pulling back required.

When the engine fails the aeroplane, unless the pilot does something about it, will remain pitched nose high (not quite as high as before as removal of power will induce pitch reduction in itself), the speed will rapidly bleed off due the nose-up pitch and no power, and within a few seconds (3 in this case of a benign PA38) the stalling angle will be exceeded and the aeroplane will depart controlled flight. All this with no pulling back.

In very light aeroplanes flown to max climb performance the stall will occur pretty much as soon as the engine fails. At Barton about 30 years ago a highly experienced GA instructor, examiner, and B737 training captain (the guy who had meticulously checked me out on the Chipmunk) was climbing at max rate in an Issacs Fury. The engine failed and the Fury stalled and spun immediately. He was seriously injured.

When the engine fails in a climb it is essential to LOWER THE NOSE IMMEDIATELY, and perhaps quite violently, to prevent loss of control.

I, as always, find people's carefully planned actions with EFATO's genuinely interesting. There are those who have had one, and those that might, one day.

There is a lot of theory printed above; theory that in the circumstances just won't have the chance to be activated. EFATO's often happen (real ones) at 300 feet or less. There isn't the time to start calculating. By the time you have realised what's happened you have lost 33% of your height already.

A stationary propeller will give an exceptionally poor glide ratio and you can't imagine how poor it is. The silence will scare you. The sound of the yoke rubbing on the panel and the control cables squeaking on their pulleys behind you will make you feel lonely.

You will drop like a brick. Literally. Not like the "idle engined" situation that you practice in your exams or your biennial.

My point is, if you plan too hard, you might end up doing yourself out of time when you need it most. Time that requires you to think about what the air is doing over your airframe, and what you can do to reduce your chances of potentially injurious rapid deceleration on the ground.

Sorry to sound blunt, but when it happens it is completely different to how the theorists describe it to be.

When it happened to me it was a shock that it'd happened but I sure didn't have a think before lowering the nose. I lowered the nose while my brain was getting into EFATO gear. Pavlov's dog syndrome - 'X' happens, you do 'Y'. No thinking involved - there ain't time!

I completely agree SSD

My comments were aimed more at the theorists who were quoting reams of scientific forumlae. :bored:

Just a thought from a glider and power jock. In a glider you always trim for glide speed (I know it can't be done that precisely but about half forward on the trim is usually about right) before take off, thus assisiting your aim of not stalling in case of a cable break. This means of course having highish forces on the stick on climb out during a winch launch but they are quite acceptable.

What would be the problem in doing that in a powered a/c, say up to a thousand feet then trim for the climb? Wouldn't it be worth doing? Has anyone tried trimming for glide speed, leaving the trimmer alone then trying a take off?

One thing that gliders prep you for is a launch/take off failure, cable breaks happen far more often than an engine failure. I've just looked in my logbook and I had six launch failures out of the last one hundred, that includes two winch failures/poor winch technique. Unconsciously I'm always expecting a 'launch fail' even in a power a/c.