EFATO in Twins

[bookworm]

Into the lion's den... :)

I've just had a week of training for the multi-engine IR, terminating with a CAAFU test. I'm convinced that the way the EFATO is taught and tested will probably kill more pilots than it saves in the long run. I don't have statistical evidence to back this up, but I would like to put the safety case for doing it differently.

The simulation, as I was taught and tested, takes the following form. The examiner/instructor covers the throttle quadrant and retards one throttle to idle.

The student then:

1) Identifies the failed engine (after which the examiner removes the cover)
2) Checks (or requests a check) for fire
3) Applies full power
4) Checks gear and flap up
5) Completes touch drills
a) verifies the dead engine by closing the throttle
b) touches the prop lever to feather
c) touches the mixture to pull

Note that the ordering of items 1 and 3 are an artefact of the simulation. The student cannot apply full power until the examiner uncovers the throttle quadrant.

The entire procedure is expected to take 12 to 15 seconds. The identification of the failed engine needs to take place in perhaps 4 to 5 seconds in order for the remaining parts of the drill to be timely. Thus, from the first signs of yaw, the student is taught, in effect, to make the decision to shut down an engine in less than 5 seconds.

Like most students have, in one of the many practices, but thankfully not in the exam :), I called the wrong engine. We know that in real life, identifying a failed engine correctly is challenging -- although the Kegworth accident has many differences in circumstance, I think it illustrates the point.

Do we really need that haste?

My understanding is that it is prompted by the need to:
i) feather the prop while it can still be feathered
ii) avoid the extra drag of a windmilling prop for an extended period

I would contend that:

a) most engine failures are progressive
b) many engine failures are partial -- some thrust remains which may be better than zero thrust
c) there are few failures that would lead to the prop stopping between, say, 10 seconds and 30 seconds after the first evidence of failure, i.e. leading to a different outcome between a drill taking 10 seconds and one taking much longer with careful analysis
d) even if it did stop before it could be feathered because of oil loss, it would probably feather or get close to feathered by the nature of the CSU mechanism.
e) on many if not most occasions, light twins are operated at weights and in terrain where an adequate rate of climb (or even shallow descent) can be maintained on one engine with the other prop windmilling

In other words, in training pilots for the worst case scenario, are we instilling in them a habit that may get them killed in a scenario with much higher probability?

[Malcolm G O Payne]

The present routine, while not ideal, is the result of too many pilots getting killed in the days when the test involved a complete shut-down. One of my ex-RAF colleagues was killed that way when he misidentified a cockpit switch. When an engine fails it is a sudden and confusing situation, especially when it involves complete failure. As a former simulator instructor at one of the commercial schools I used to give students loss of oil pressure on a twin or a surging engine. An awful lot of them were confused and needed prompting out of the situation.

[Big Pistons Forever]

There is extensive discussion on EFATO in the " dangers of multi instruction" thread further down this forum section. I cut and pasted part of one of my posts that may be of interest to you.


One point nobody has addressed yet is the point that a mixture cut simulates a instant total engine failure. However I think this is the least likely failure scenario in a light twin. A slow run down in power and/or rough running caused by internal mechanical failure or a ignition problem , or a surging engine caused by a fuel problem is IMO a more likely real world event.
Yet from what I have seen many students complete the multi rating course with out any exposure to these situations.

When I teach the muti rating I always make sure to distract the student when the aircraft is in a turn and then slowly pull the inside throttle back. I also hide the quandrant on final approach and reduce power on one engine. These exercises provide a good example of ther fact that a engine failures can be subtle and the massive yaw experienced with a high power low airspeed rapid throttle cut is not the only way engines fail.
To simulate a surging engine I cover the throttle quadrant and then rapidly move one throttle back and forth from idle to cruise. This demonstrates that with a surging engine, It may not initially be obvious which engine is acting up. Careful observation of the engine quages is vital before any actions are taken.
Finally one of my concerns with the standard multi curiculum is that the prop overspeed scenario seemes to be at best glossed over. A prop overspeed could cause a yaw away from the malfunctioning engine, so instead of reducing power and airspeed and flying away on two engines the pilot could feather the good engine and crash. This execise is very difficult to simulate in flight. The best I have come up with is to cover both RPM guages . hide the quandrant with my clipboard and reduce the RPM on one engine to the minimum allowed. I expect the student to call " suspected prop over/underspeed", point to the RPM guages and verbalize his corrective actions. Its not a great exercise but at least it does provide the tactile cues of yaw combined with the sound of the engines going massively out of sync, and will hopefully mean the student will not consider every power abnormality only within the context of a engine failure.

I believe the instructor must adequately prepare their students for the real dangers of multi engine flight, but in a way that does not put the instructor or the student deliberately in danger.

Finally I bet that virtually all those instructors who do mixture cuts at takeoff are low time themselves and have probably
little or no time as a working line pilot. I tell all my students that the yellow stripe down my back has gotten a lot bigger since I started flying 27 years ago

[Oktas8]

Correct me if necessary, but it seems bookworm is commenting on the order of the drills as much as the mode of engine failure.

I have just been required by my boss to start teaching students to identify before maximising power, and I don't like it - it teaches them that identifying a failed engine is more important than maximising power to stay in the air longer. And yes, I think it is done purely to satisfy the examiner so s/he can remove the clipboard.

The way I'm accustomed to teaching the EFATO is to ensure that maximum power is applied and gear & flap is up before thinking about which engine has failed.

I believe it instils a more thoughtful response, and is less likely to result in the wrong engine being shut down (because you've forgotten the identify while doing the maximise / minimise part of the drill.)

O8

[BlueLine]

Presumably you completed you MEP training prior to the IR course as they are seperate courses!

EFATO always looks at the "worst case" take the single engine EFATO and the PFL, how many practice partial power failures? There is an assumption, that if you can deal with the worst case you should be able to contend with a lesser problem. A good instructor will demonstrate alternative scenarios.

In the case of the twin, the essential element is CONTROL, without SPEED you cannot control the aircraft; without POWER you will descend rapidly so you need as much POWER as you can handle. If you can't handle, it CLOSE both throttles and land ahead the asymmetric problem has gone away. You do not have TIME on your side.

There is no point identifying, if you are not under control! Then it is essential to ensure you tidy up the dead engine and minimise the drag to aid the PERFORMANCE because you are probably still descending. Identification of the correct engine (Use your feet) and identification of the correct levers is very important hence the deliberation brought inrto the drill. Two RAF C130 crashed during EFATO practice resulting in a major rethink of the way things are done.

The drill you are learning, is to teach a sequence, if it ever happens for real you will fly the aeroplane; move all the levers forward; by now you should know which leg is dead, Identify the levers - throttle first to verify- feather, feather feather; Gear Up Flap Up, bank to the live engine and see if it will fly away! If not, you may have to resort to plan B!

Many light twins will not climb away from an engine failure close to the ground at low speed, be prepared to kill the asymmetric problem by closing the throttles and land ahead, it requires thinking about before you go, there isn't time when you are struggling to control the direction.

[Stan Evil]

Just a picky technical point:

d) even if it did stop before it could be feathered because of oil loss, it would probably feather or get close to feathered by the nature of the CSU mechanism.


The problem with piston twins is that, unless positively feathered reasonably early on in the drills there is a good chance that the prop will never feather. This is because the 'feathering latches' drop in at, typically, 600 rpm and stop the props from moving to the feathered position. The reason for these latches is to stop the props feathering when shutting down normally on the ground. You would not be able to ground start an engine with a feathered prop because of the propeller drag.

Some turbo-props, like the PT-6 have a free turbine and so feathering is not a problem - you can even do it while the engine is running!

[fireflybob]

Simplicity is, I believe, evidence of the most advanced form of teaching.

Rule 1 - Fly the a/c - keep straight with rudder...etc

I have seen a few different systems over the years. One I really liked was favoured by Austrian Airlines many years ago:-

Power, Performance, Action

Power - Full power on the live

Performance - Reduce Drag - Gear, Flaps, Cowl Flaps etc

Action - Feather/Shut Down

Most of the other guff can wait till later - Crossfeed etc etc.

[bookworm]

Thanks for the responses, which are all helpful.

Oktas8 correctly summarises one of the points, which I made rather poorly in the original post. There were really two:

1) The timescale over which the decision to shutdown an engine is made should not be as compressed as it is in the EFATO drill.

2) Forcing the identification before application of power is permitted is unrealistic, and further compresses the timescale for decision to be made.

Blueline makes the good point that drills always concentrate on the worst case. I think this is an extreme though. If I have a partial engine failure in an SEP, I would adjust speed for minimum drag and assess the situation. I wouldn't immediately click into "land the aeroplane without power in a field" mode.

But the MEP equivalent appears to require a decision to shut down the engine very rapidly (5 seconds), or the test is "failed". While the application of rudder and the control of pitch attitude must, without doubt, be immediate, the rest can follow after a careful assessment based on all the available information.

FWIW, I have spent three years flying MEPs before doing this upgrade, and it's perfectly reasonable to suggest that partial failures should be covered in basic MEP training, leaving only the worst-case drill to cover in the IR. But more than any other drill I've come across in aviation, I get the impression that the speed required in this one teaches bad habits.

This is certainly not the fault of the FTO -- five hours to train for the upgrade is a very short timescale, and presumably this is the sames timescale as used in the full-course IR for EFATO and SE work. And I'm not blaming the examination system as such. I'm just making a case that a re-evaluation of the nature of the failure simulated in the test.

An anecdote: Some weeks ago I was practising ILSs in twin. A few seconds into the go around, after cleaning up. I experienced a sudden and unexpected roll and yaw to the left. This is the situation in which, within 5 seconds, I should be taking the decision to shut down the left engine. I'm glad I didn't -- the
problem was that the right flap had stuck down. Returning the flap to its extended position fixed the problem. Shutting down the left engine would almost certainly have killed me and my safety pilot.

Finally, I'm indebted to Stan Evil for setting me straight on a technical point that I should have known. I do, however, think that the drag offered by a stationary propeller even in a partially feathered state is rather less than the windmilling fully-fine prop that we're used to thinking about.

[Oktas8]

As far as I'm aware, centrifugal feathering latches engage at the 900 rpm mark (±100 rpm) in most light twins. Ref. Robson from the Aviation Theory Centre series. I wonder at what CAS the propeller stops windmilling on a typical light twin, barring engine seizure? Could be close to VMC...

Also, a stationary propeller will give less drag than a windmilling propeller for piston powerd aircraft, as the energy to turn the engine is not being removed from the airflow. But a feathered propeller gives oh so much less drag again, which is why it is emphasised strongly in the EF drills.

O8

[BlueLine]

"I experienced a sudden and unexpected roll and yaw to the left. This is the situation in which, within 5 seconds, I should be taking the decision to shut down the left engine."

Think about it!

Control
Power (Check Speed)
Identify
Confirm! when you retard the throttle the aircraft will swing, so you don't shut the engine down - you investigate further. Thats why you start with the throttle.

[Big Pistons Forever]

With respect to the EFATO case ( ie slow and close to the ground )Why confirm with the throttle ? If I think an engine has totally failed ( based on all the proper cues )I do the dead foot dead engine check and then go right for the prop control. As you pull it back you are verifying because if you pull the wrong prop lever back you will know it :uhoh: The trick is to feather the right engine. Verifying with throttle wastes time and still will not prevent you from pulling the wrong prop lever back. The only time
I would adjust the throttle is if the engine was surgng or I had reason to believe that a reduced throttle setting would result in the engine being able to produce usefull thrust. When I teach new multi students I always review the accident stats. They are pretty unequivical. A lot of fatal accidents are caused by pilots trying to fly after an EFATO. Based on the performance of your typical light twin I teach my students to pull both throttles back and land straight ahead untill the aircraft has a positive rate of climb, is at or above blueline, and the landing gear is retracting. Then, and only then, should they try to fly away from an engine failure.:ok:

[ROB-x38]

For interest's sake the feathering locks engage below 950rpm in PA-44. I'd be interested to know what airspeed would give this RPM for a typical flat four.

[Oktas8]

I fly PA44's regularly.

From memory, at 88kts the dead engine windmills at about 1400 rpm, give or take a bit. It makes a small difference between mixture ICO and engine idling, but not much.

Have done several VMC demos in the last few days - and I think the dead engine was windmilling at a little over 1000. But that was idling, not mixture ICO.

O8

[bookworm]

Think about it!

Control
Power (Check Speed)
Identify
Confirm! when you retard the throttle the aircraft will swing, so you don't shut the engine down - you investigate further. Thats why you start with the throttle.

Again, it's a fair point, Blueline. But for me the issue is as much that it's confirm or verify rather than investigate. I think, in the light of my training experiences, the mindset developed is that the engine will be shut down.

In the event of a genuine engine problem, verification is a touch more problematic than in the drill. What would be your criterion for the decision to shut down immediately rather than investigate further? Would you shut down only if there were absolutely no yaw on closing the throttle? That would be a very unusual event, as most realistic failure modes leave some power being developed, or at least a fluctuating power situation. How long do you take to make the decision?

[FJJP]

I used to fly 4-jets. Each time I had a new co-pilot in the sim, I used to brief that in the event of an engine failure he was to carry out the drills as quickly as possible, without saying anything, to make the situation safe. Without exception, I ended up with the situation where the wrong LP cock was shut, thus turning a single engine failure into a double. And the sim instructor timed the event from start to finish. Then we froze the sim and put it back to the start of the runway. The second t/o brief was for him to identify the failed engine positively, touch each control and say what he was about to do before doing it, and then do it. Again the process was timed.

The difference was measured in seconds; but the main point was that we were now flying on 3 instead of 2 engines. It does not pay to rush the checks. There is no circumstance that I know of where you MUST get the engine shutdown drills done in record time. Rather you should take that extra second or 2 to make sure you get it right...

[DFC]

Big Pistons Forever has hit the nail on the head whan saying that a forced landing must be considered as a serious option in many light twins when EFATO occurs.

The CAA have published appropriate guidance in LASORS 2004.

"On most MEP aeroplanes there is usually no
provision for single engine climb performance
until the aeroplane is configured correctly (e.g.
landing gear and flap retracted, full power on
the live engine, propeller feathered on the
failed engine, single engine best rate of climb
speed Vyse). The act of raising the gear or
retracting the flap may, in some aeroplanes,
cause a temporary increase in drag, loss of lift
or even reduction in control margin. Thus, from
the time of engine failure on take off to
achieving the single-engine climb criteria, a
forced landing must be considered as a likely
outcome."

The CAA's advice which should form part of MEP training is that the options should be considdered in the following order:

1. Rejected Take-off
2. Land back on
3. Forced Landing
4. Continue Climb

Note that options 1, 2 and 3 involve closing BOTH throttles!

Regards,

DFC