Multi-Engine Drill

[Hedley]

Following is an email I sent recently to a student,
which may be of interest:

-- cut --

There is a drill, or sequence of checks, that every
multi-engine pilot should have completely memorized,
in case of an engine failure right after takeoff.

An engine failure in cruise (or even easier,
in descent) is generally a non-event because of
all the energy (altitude, airspeed) that you have to
play with. All you have to worry about is your
single-engine ceiling, which is much less of a
problem if you have a turbocharger.

However, an engine failure after takeoff is the
nightmare that everyone trains for, because you
have no altitude or airspeed to spare, and the
correct actions must be taken promptly to reduce
drag and continue flying. I should mention that
if you're below 50 feet (or thereabouts) it might
be safer to simply lower the nose, pull the
throttles and land.

Anyways, here's the engine-out drill that you need
to memorize to fly multi-engine aircraft:

CONTROL
POWER
DRAG
IDENTIFY
VERIFY
FEATHER

When you wake up in the middle of the night, you should
blurt out, "Control-power-drag-identify-verify-feather!"

Let's talk about each of these in turn.

CONTROL

maintaining aircraft control is the most important thing
you can do. It's not just something you do at the start
of the engine-out drill, you must continue to do it during
the rest of the checks. You want to look at the attitude
indicator, get the wings level, establish a nose-up pitch
attitude, maintain heading, and above all, CHECK AIRSPEED.
Blue line (Vyse) is the best airspeed, if you slow down
below red line (Vmc) you roll upside down and spin in.
Common error is to maintain the same nose-up attitude
that you had with two engines pulling, which results in too
slow an airspeed, with an engine out, you MUST lower
the nose as required to maintain airspeed.

POWER

You want maximum power, and you do this as follows:
from right to left, push in both mixtures, both prop
controls and both throttles all the way in. You don't
know which engine has failed at this point.

DRAG

Now you want to eliminate drag. You do this by immediately
raising the landing gear and flaps. Visually check the
handles and indicators to ensure that you don't hurry and
skip over this important step.

IDENTIFY

Now you want to figure out which engine has failed. One
engine will be pulling forward strongly, and one engine will be
windmilling, and in fact pulling back on that wing. Let's
say the right-side engine failed. The aircraft will yaw
to the right, requiring you to hold a boot full of LEFT rudder
to maintain your heading (eg IFR departure). So,
the mantra here is "DEAD FOOT, DEAD ENGINE". To
maintain heading, your right foot will be flat on the floor,
and your left foot will be hard on the left rudder pedal. So,
your RIGHT foot is DEAD, and so is the right engine.

VERIFY

Since your right foot is DEAD, slowly move the RIGHT
throttle aft. If that makes no difference to the sound
or performance of the aircraft, it is indeed the DEAD
engine. I should mention that it is possible to have
a prop overspeed instead of an engine failure, but
that's a story for another time.

FEATHER

Once that you have verified that right engine is DEAD,
you want to feather the prop on the right engine to
eliminate the windmilling drag, which can make the
difference between climbing and descending. You
do this just like during the runup, by pulling the
RIGHT prop control all the way aft, past the detent.

Let's not forget about CONTROL.

Now that you have configured the aircraft for
maximum power and minimum drag, you need to
fly it to get the best climb rate. You do this by
setting a pitch attitude which maintains BLUE LINE
on the VSI, and by minutely banking into the
good engine - think of RAISING THE DEAD in
the movie "Night of the Living Dead". You don't
want 5 degrees, you want half that, or 2.5 degrees
of bank. Don't center the ball - if you do, you
are sideslipping (long story) - you want the ball
half out of the cage, to obtain the best climb rate.

The above is the important stuff. There is lots
of other stuff to talk about, but:

CONTROL-POWER-DRAG-IDENTIFY-VERIFY-FEATHER

is what you want to understand, and above all MEMORIZE.

If you're suffering from insomnia, here's an article
I wrote about 10 years ago. Multi-engine aircraft
haven't changed much since then:

http://www.avweb.com/news/airman/184438-1.html

FWIW.

[Airtids]

Good post, Hedley. Good article, too, BTW.

Here's the one I teach, very similar to yours:

Control
What you said
Attitude/Altitude
I throw this one in for IFR purposes to remind one of their required/minimum altitude
Power
What you said
Drag
What you said. Important to actually think about this procedure. Cowl flaps? Depends. Gear? Depends (337, forget about the gear if it's still down).
Identify
What you said, although once again, in the 337, you've gotta go to the gauges because there will be no yaw to help ID.
Verify
What you said. at this point I ask "Oil, smoke or fire?" If the answer to any of those is "Yes", then we skip the cause check and go immediately to feather.
Cause Check
Left to right flow check remembering the engione needs fuel, air, and a spark to run. Shame to cage an engine because I forgot to switch fuel back to the mains . No joy: Feather
Feather
What you said.
Shutdown/Secure
By the checklist/POH. Consider what phase of flight you're in. If in cruise, or on approach, you probably don't want full power on your good engine. Consider managing that good engine; increase cooling, reset mixture, etc.

When I wake in the middle of the night, I scream CAPDIV (captive) CFS (as in Canada Flight Supplement). OK, I don't really, but I expect my multi students to do so.

[Big Pistons Forever]

I used to teach the engine failure at takeoff the hedley way untill I took a twin cessna refresher course at SIMCOM. They in my opinion have a better way.

First before you talk about engine failure you have to deal with the question of decision speeds. Piston twins will not fly level below blue line airspeed, therefore with any engine failure below blue line the only option is to close both throttles and land straight ahead (hopefully on the remaining runway). It is only after you achieve blue line speed and with the gear coming up do you have a real option to continue the takeoff and then only if you immediately remove the drag of a windmilling prop.

the SIMCOM drill is as follows.

1) Whenever possible use 0 flap and no rolling takeoffs, all take offs begin with advancing the throttles to 20 in (normally aspirated) or 25 in (turbo charged). The engine is allowed to stabilize and a quick engine guage check is down prior to break release.
Power is then smoothly increased to full power (no reduced power takeoffs should ever be done)

2) Your hand should now be holding both throttles firmly forward and you should be looking down the runway. At seventy knots cross check the airspeeds and do a quick cross check of engine instruments. If the airplane will allow it, stay on the ground untill blue line (most twin cessna's and the PA 31 will do happily do this, particularly with no flap), if not lift off but fly level just above the runway untill blue line. In the event of an engine failure or other significant abnormality close both throttles (easy to do since your hands are allready on the levers).

3) At blueline with a positive rate of climb take your hands off the throttles and raise the gear. Now place your hands on the prop lever.
In the event of a strong yaw indicating a possible engine failure pitch the nose down to about 5 deg nose up, keep the airplane straight with rudder. Identify with the dead foot method, then feather the failed engine ( again easy to do because you fingers are allready on the prop levers)

4) Do nothing except fly the airplane straight and away from the ground (remember at this stage the gear has allready been selected up, the flaps are up and the prop is now feathering) untill you are 1000 ft AGL or so and then complete the appropriate engine failure checklist.

After flying a variety of transport propliners I realized that this proceedure mimics large aircraft SOP's. In effect we are saying blueline is V1/VR and your fingers are the auto feather system.


SIMCOM has a 1.5 million dollar twin cessna simulator with a very realisitic flight model. When the instructor told me about the SIMCOM method I thought it was stupid, so he challenge me to try the "standard EFATO drill and his method with an engine cut at blue line. The bottom line is I crashed with the standard method and flew away every time using his method. Since then that is how I fly every light twin and what I teach students.

A related pet peeve is when we teach for the multi rating the only failures are total ones. I would argue that most actual engine failure are either a partial failure or a surging/unstable engine. When the airplane is close to the ground you do not have time to troubleshoot and have to feather immediately. However if the airplane is cleaned up at climb airspeed and a couple of hundred feet in the air then you have the luxury of time to determine wether the yaw was from a partial engine failure or perhaps a prop overspeed. But to do that you have top know what to look for, so I go over the MP and RPM indications you would expect for various scenarios. Again this seems to be an area not well understood by many pilots.

[Cat Driver]

Excellent post BPF.

We have been flying twins like for decades before SIMCOM or Flight Safety were even thought of.

Nothing concerns me more than watching pilots take their hands off the throttles and fly with both hands on the control wheel before the thing is even airborne.

Good post....especially the part about partial engine failures on piston engine airplanes...when you have sufficient velocity you have the time to do the most important action...as you maintain control take the time to confirm " what the f.ck is going on here? "

Cat

[Donald]

The SIMCOM deal sounds good, and I agree with the similarities to big airplane ops. There is just one nagging thing in my head...

At blueline with a positive rate of climb take your hands off the throttles and raise the gear. Now place your hands on the prop lever.

Identify with the dead foot method, then feather the failed engine ( again easy to do because you fingers are allready on the prop levers)

However if the airplane is cleaned up at climb airspeed and a couple of hundred feet in the air then you have the luxury of time to determine wether the yaw was from a partial engine failure or perhaps a prop overspeed.

I just wonder, if the guy driving is one of the "quick hands" kind of guys, what are the odds of shutting down the good engine in a prop overspeed situation?

It's been a while since I flew the PA31, but isn't blue line around 104-107kts depending on mods? Seems to me it liked to rotate around 75-85kts.

Otherwise I think it's a good engine failure plan of action, especially the emphasis on landing ahead when the climbout is not a smart choice.

[Donald]

Oh yeah, and a caveat to my post is that I am not an instructor. I am just a driver

[Hedley]

Fascinating ... the Simcom method is faster - which is arguably important at low altitude and airspeed - because it appears to skip the "VERIFY" stage of throttle wiggling.

I admit, it is tempting, to have put your right hand on the props after takeoff so you're spring-loaded to feather the wind-milling engine.

However, here is the scenario .... you take off, and hear the engines go out of sync, and you yaw right, so you immediately feather the right engine.

But what happened is that the left engine had a governor failure, and overspeed. You have just shut down a functioning engine (on the right) and you are now flying only with the malfunctioning left engine

If the "VERIFY" step had been performed - ie wiggling the right engine throttle - then you would have noticed that the right engine was indeed producing power, which would have halted the drill.

Fascinating.

Bottom line: is there really time to do it the "right way" at 100 feet? What is the likelihood of an overspeed? fwiw, it happened to a friend of mine in a King Air, and he was damned glad he did the "verify" throttle wiggle.

P.S. I'm an instructor, but as we all know, that and a buck may get you a cup of coffee. I will freely admit to teaching aerobatics, which I know best - anyone want to learn how to tumble? - but as a multi-engine instructor, I'm frankly a low-rent substitute for the real thing

[scm]

Another very informative post that i've seen on avcanada lately

thanks for this



PS. I want to learn how to tumble. Where do you teach at?

[Wasn't Me]

Everything above is good but lets remember speed. If your climbing at blue line when an engine fails it means you have to lower the nose to maitain or accelerate to the blue line speed after you start to slow down because of the failure. Always find a good compromise speed that allows you to clear and climb safely. It is not a good thing to have to accelerate to your control speeds its always best to be able to slow down to them. By planning your climb speed to be above your blue line speed you have a better chance of success. In most twins this procedure is not much of a problem.

[Big Pistons Forever]

Hedley et al

Your point is quite valid and brings up the central cunundrum. Where should rote procedure give way to pilot judgement ? I had a prop on a PA 31 roll back to 2100 RPM just after takeoff going through about 50 feet. I knew something was wrong but I was also able to appreciate very quickly that the airplane was not in fact falling out of the sky and was still climbing so I could not have had a full engine failure. A glance at the instruments showed the low but stable RPM so I just pulled the appropriate throttle back to a low cruise Manifold pressure, declared a Pan flew around the circuit and landed. However my experience at SIMCOM was conclusive for me. If an engine fails right around blue line, close to the ground, the airpalne will not give you time to wiggle the throttle and play patycake with the gear and flap levers, it will crash every time. If you use the SIMCOM method you can fly away.

However your point about yaw away from an overspeed is valid. All I can say is with this method if you feather the wrong engine at least you will still be close to the ground and with sufficent airspeed to keep control of the airplane as it flopped back on the ground right side up. I would however suggest that at blue line speed the prop is close to the fine pitch stop so even a catastophic governor or prop dome failure will not produce a large overspeed and therefore only a relatively mild yaw.

I think the central issue is you cannot train experience you can only train procedures. I like teaching this procedure to new pilots for a few reasons

1) The airplane is optimally configured for an engine failure at every stage of the procedure and

2) It minimizes the number of decisions that has to be performed when the pressure is on.

Finally I think it is interesting to note that the 30000 hour 747 Captain is not expected to make any decisions between brakes release and 400 feet. He follows a series of rote procedures depending on what part of the takeoff he is at. If this is the case why would we expect anymore from a multi student ?

[Airtids]

BPF: You nailed it. You cannot teach experience. You have to give the student the tools to survive long enough to gain the experience required to be able to use judgement effectively. More often that not, unfortunately, those tools are procedures they learn how to apply.

Here's a basic t/o brief I use and teach to address the failure on departure problem:

"This will be a (standard/short/soft) left seat takeoff from Runway XX at Someplace Airport. In the event of any abnormality prior to rotation, the call will be 'reject', I will retard the throttles and stop straight ahead. In the event of an engine failure after rotation, and below blue line, the call will be 'reject' I will retard the throttles, land and stop straight ahead, turning to avoid obstacles. After blue line, with the gear up or in transit, we will continue to climb, identify, verify, and feather the dead engine. We will climb to our emergency safe altitude of X,XXX' at which point we will request a priority landing runway XX at Someplace."

FWIW, I also use the 20" MP static start procedure, unless operating from soft fields. After about 1800MPIC, I still use this brief, even giving it to myself when Single Pilot. Overkill? maybe, but it makes me fell more comfortable doing it. Repositioning hands to the prop control after the gear is an interesting twist. I'll give it a go.

I'm not sure about the blue-line climb. Blue line is Vyse, which is generally lower than Vy, which is what I teach the student to accelerate up to and climb at , or Vy+10. For most light piston twins, a lowering of the nose will certainly be required to avoid the energy bleed that will eventually drop the airspeed below Vyse. Because your climb speed is likely to be above Vyse, you've got a bit of time.

It used to be that we had to perform with SPEED in the event of a failure. It seems more and more with multi rides, and PPC's that the stress isn't so much on speed anymore, as it is in proficiency.

[Cat Driver]

" It used to be that we had to perform with SPEED in the event of a failure. "

Do you mean with speed as in act quickly?

[Airtids]

Exactly. Unfortunately, that sometimes meant faster than the noggin was working, which was a recipe for disaster. Speed is still of the essence, but not at the expense of getting ahead of yourself. THINK THEN ACT is the new mantra.

Sorry, just re-read that and had a thought. Were you taking a shot at the drug culture out here?

[Big Pistons Forever]

I now realize I was not clear in my post. I am not advocating climbing at blue line but rather using the attainment of blueline airspeed and the act of raising the gear to be the transistion from a situation when the only option is to close both throttles to now having the option to continue. IMO to successfully continue flying after an engine failure close to the ground the airplane has to be allready configured in a low drag state and the prop must be feathered. This procedure is IMO the safest and most expeditous way to do this. I think it also has to be understood that the procedure works in reverse. That is if while performing the drills you allow the airspeed to decay below blueline you are now back into the area where your only option is to close both throttles and take your lumps straight ahead.

But the normal takeoff I teach is when you pass the blue line go no go point you set the required nose up attitude to allow the aircraft to accelerate to initial climb speed (usually around 125 knots). Hope this clarifies things.

[Cat Driver]

No I was not refering to the drug situation here, I was refering to the stupidity of doing any action quickly and before you have a clue of why.

" THINK THEN ACT is the new mantra. "

Forgive me for being out of the loop in how flight schools teach, but I find that statement to be disturbing bacause it means there was a time when you taught pilots to react quickly before they actually identified exactly why they were reacting.

If that was the case who was responsible for such a stupid method of teaching flying?

Cat

[Wasn't Me]

Cat

Not all flying schools teach to react. Most that I am close to teach to cinfirm all actions. This means confirm throttle lever by moving it back a little to confirm the correct action then proced, the same with the other levers. It is never a good idea to just jump and pull a knob. Many accidents have occurred because of speed. Some one in this forum has a saying that if your doing it fast your doing it wrong. He's right.

[Airtids]

Forgive me for being out of the loop in how flight schools teach, but I find that statement to be disturbing bacause it means there was a time when you taught pilots to react quickly before they actually identified exactly why they were reacting.

Back when I did my Multi ('89ish), there was an emphasis placed on speed when conducting the drills. I too found this a little strange, and being more of a thinker, my drills were slower than my instructor was happy with. When I questioned her about this idea of speed, she indicated that TC wanted to see the drills conducted quickly, because it showed that the candidate had the procedure own pat. Slowness was apparently seen less as the result of thought about why we were about to take a certain action, and more about what that next action was. It does sort of play towards BPFs concept about not being able to teach experience to newbies and therefore being reduced to teaching procedures only.

By the time I actually started instructing in addition to my duties as a charter pilot (with the associated annual PPC rides), a conversation with a TC examiner indicated that they had received enough negative feedback on this preference for speed, that they were now more interested in proficiency, and systems knowledge (which would account for taking your time and thinking things through- especially the cause check) would be better and more thoroughly addressed in the ground briefing portion of a ride. Evolution?

[Cat Driver]

The truth is I shouldn't even get involved in the discussions on how to teach flying any aircraft.

All it does is frustrate me because I see so many goofy new ideas on how to do something relatively simple such as think ahead of an airplane.

I have watched the dumming down of this industry for decades, what was at one time taught with emphisis on the art of aircraft control and using common sense in tandem with proven airmanship has been gradually morphed into a culture of paranoia of missing some mantra or check either visual, oral or mental while you are teaching pilots the simple mechanics of flying these basic aircraft.

A typical example of this goofy thinking would be back some time ago when TC came out with the "light up and be seen " concept for turning on the lights on aircraft and decided that nav lights should be on at all times.

Give me a fu.kin break in bright sunlight you have to walk up to the airplane and put your hand over a nav light to see if it is on. Instead of adding to safety it only burns out nav lights so that some poor sob who needs them for a night flight finds it does not work.

I decided to not renew my Class 2 instructor rating some years ago when the guy in charge TC flight training in Vancouver told me it was going to be very difficult for me to renew because I have to many pre conceived ideas about how to fly.

What a f.ckin moron to have the audacity to sit behind his desk and pre judge my abilities on his perception of how I would teach.

Anyhow I guess I am kicking a dead horse here in Canada with my opinions on how to fly and or teach because I have not been annointed by the Gods of flight instruction in TC.


Fortunately I manage to earn a very good living teaching outside of their kingdom.

Cat

[goldeneagle]

Following is an email I sent recently to a student,
which may be of interest:

Interesting start to a thread, equally interesting discussion follows. One thing most folks tend to agree on, there's a lot of debate about the 'right' way to do some of this in a light twin. The bottom line tho, average light twin, failure of an engine in a takeoff scenario is catastrophic, the performance just isnt there on a single engine if the airplane doesn't have an energy store (altitude=potential airspeed=kinetic), and the remaining engine doesn't really have enough poop to do the job anymore.

The 421's you and I fly are a little different, but, for light twins, they aren't average either. With an engine out, we've still got 375 ponies to draw on, that's more than a lot of trainers have with both engines operating. That brings me to the SIMCOM concept of having hands on prop levers, and being ready to do an immediate feather. The folks at simcom have done the statistics, and come to the conclusion, propeller overspeed is not the type of failure that is prevalent, so, they teach a philosophy that will save you 85% of the time, and kill you 15% of the time (numbers picked out of the blue to make a point). It's easy to rationalize farther, propeller overspeed is the kind of incident where fethering the wrong one can actually be 'gotten away with' in some cases, the 'bad' engine is still producing power, even if it's not likely to do that for very long, it'll keep you going long enough to realize mistake, and correct it. But it's really dependant on the plane. With some planes, you DO have the time to analyze, and deal with things. With other planes, you DONT have the time, so, you roll the dice, go with the statistics, and reach for the prop handle on the 'dead foot' side. I'm sure there's a lot of folks that have 'seen it' in the simcom simulator, and are convinced that instant feather is the right course of action. What they may not realize, it's a statistical thing, they are taking the bet that the dead foot is the dead engine. In reality, it's a pretty good bet, but, like all bets, there is a probability of losing.

I think it's good to have some rote procedures drilled into folks, procedures that will 'do the job' most of the time. It's those rote memory procedures you draw on at 2 am, tired from a long day of flying, and the shit hits the fan. In a perfect world, we all have the perfect procedures memorized, and, we can execute them perfectly even after a 16 hour day, in the dark, with a blustery wind and rain falling. If anybody knows what planet has that kind of perfection, let me know, I wanna move there

The one gripe I have with the way most schools teach the rote procedures, everybody leaves out one of the most important parts. Right after 'identify', in big bold letters, the rote memory item should say LOOK AT THE FAILED ENGINE. It takes a small fraction of a second, and can provide invaluable information which can dramatically effect the rest of the decision making process. If you see flames coming out of the cowling, or buckets of oil coming out, you are going to make different decisions than if you see nothing out of the ordinary. It boils down to situational awareness, and that quick look can make you keenly aware of a very bad situation, especially if you see open flames, and realize they are only a few inches away from a locker tank that's just been filled with fuel.

[Doc]

We have our own little drill....as the power comes up on take off, untill we are out of five hundred feet, we chant..."Please God dont quit, please God dont quit....." Blue line, on our beast, is a speed we just visit, on the way to a higher number, leaving our exposure to a tough engine failure to a minimum. Always wondered why, on rides, I always get an "SB" for my speed after an engine failure? If I'm blue line PLUS ten knots, WHY would I SLOW the aircraft, if I already have a good rate of climb?? Slowing down to blue line requires me to retrim, and an increse in trim input = and increase in drag(Okay, not much, but some)and we dont like drag, do we? And VMC....you sould still be on the ground at VMC?
Now, on Cat's remark about doing something fast....he's correct...dont. I used to fly with an old guy (older than Cat) who said, regarding engine failures...."First, put down the newspaper.".......On a ride on the DC3, our inspector, Martin Braeman, pulled an engine on this guy, he lit his pipe, then remarked..."Hum. we seem to have lost an engine..." He didn't do anything until he had his pipe lit!