Multi ? Methods for engine failures

[Traumahawk]

I saw this question a while ago somewhere else, wondering what you guys would make of it.

Simulating engine failures in twins. When I did my training is was with the mixture below 3000agl....and usually fuel selector above.
I think these methods are the most valuable tools we can use to create an experience for a new student getting into the twin engine world.
Many people dissagree and support only using throttle retarding to simulate any engine failure anywhere, and, sometimes, with good reason. Many say if you were to crash due to the mixture cable breaking etc... you would be held under wreckless endangerment ect... also saying cutting these two ways is just too risky.
Now, personally, mixture/fuel selectors was the way I was taught. It is the way all the examiners in my area perform the checkride. In addition, lycoming even specifically states it is much easier on the engines to use mixture cutting rather than throttle reatarding due to 'a cushioning by air pressure of the pistons - providing the throttle is open.'

Another point to remember, how real is pulling a throttle on your student on the roll? You move his/her hand out of the way, yank one back and almost make the decision for them.
In the air, the same applies, but instead of 2 throttles sitting side-by-side, one is fully closed, the emergency procedure and line of questioning just got shorter and easier. Whereas a fuel selector cut may induce some coughing, yawing, Suprise, as can a mixture cut on the roll while you block them both with a cupped hand.

Too risky, possible lawsuit??
You could also argue whether it is 'worth the risk' actually stopping an engine in flight during training and the flight test, or whether it is worth doing the 'Vmc demo', just in case an engine fails at the wrong moment, or doesn't re-start. I believe that these risks are worth taking, and indeed nescessary to breed a competent ME pilot.
I have heard of pilots being sued for wrecking an aircraft while practicing crosswind landings in a taildrager, because it was "foolish and reckless to practice purposefully with a crosswind situation."
At this point, I feel that if we eliminated every risk and worried about every potential lawsuit(that you probably shouldn't usually be worrying about in the air) then we wouldn't get farther than the ramp boundary, and thats after a 4 hour preflight and mechanic's inspection. I do think we can increase the realism and experience level of a new twin student by using the 2 methods named, especially students coming from 'pilot factories' who often lack any form of new, supporting experience to fully and properly excercise the ratings held. I'm, in some cases, one of them.

What do you guys do?

Sorry for the novel folks.
Safe flying'
---T-hawk

 

[avbug]

I've never heard anybody comment on "cushioning pistons"; sounds like codswallop to me. Operating at reduced power settings at any time when the airstream is driving the prop, rather than the engine, is hard on the engine no matter how you cut it.

As far as mixture or fuel selector; why would you want to starve the fuel pump of fuel if you don't need to? That fuel pump is a fuel wetted component; most have life limits if run without fuel. Once in a blue moon for training purposes is one thing, but for training in general, the mixture will suffice. Generally, you're best off to retard power on the throttle for general single engine maneuvering work.

For simulating an actual unexpected failure, the mixture works just fine. After the engine is identified and verified, block the prop levers if you don't want it feathered (or gaurd the feather button, as the case may be). Then restore the mixture to the appropriate setting to get power back, test it with the throttle, and set zero thrust if the student indicates he or she has made an attempt to feather.

Try to avoid massive power changes, cool and warm slowly, and avoid sudden power changes where possible. Even though it's training, respect that airplane; your life still rests on it's good functioning.

Before you worry about a law suit for shutting off the selector during a takeoff (I can't wait until later in the paragraph to say it...it's a very, very, stupid thing to do; idiotic, mindless, inane, pointless), worry about surviving the aborted takeoff. Currently, the FAA prefers that engine cuts on the takeoff roll occur at or less than 40% or your rotation speed. At what point are you going to shut off the fuel selector and hope this occurs before 40% speed is obtained?

Suppose something goes wrong; your student elects to continue, or you go long, or there's an obstacle...don't you want the option of doing something with that now-useless engine? I wouldn't be in any big hurry to kill that engine; simulate a power loss if you must, but don't take away your options. Your butt is on the line, too.

Thought about blocking a rudder or pushing a rudder? Even dragging a brake a bit? All you need to do is simulate a loss of directional control; some key point that will force the student to decide weather to stay, or go. You block a rudder when doing Vmc work (you should); why not do it on the takeoff roll? Much safer than shutting off the fuel during a takeoff, and much more intelligent. There are more ways than one to skin a cat.

Before getting into a protracted discussion on how best to simulate an engine failure, let me ask you about your actual engine failure experiences. What were they like? When did most of them occur? Were they dramatic, or hardly noticable? Let's start there, then have a discussion about simulating those engine failures you've already had. You want to make the experience more realistic; you do know what it feels like to really lose an engine, right?

 

[Traumahawk]

just my .02

Couple things.....
Perhaps I didn't make it clear that I don't condone fuel selector shutoffs anywhere below 3000agl either, and when used, have the student go through the req'd procedure, and then quickly throw it back on and simulate feather.

I too, have no qualms about lawsuits in any of these situations, I've got more to worry about up there.

I was taught simulating engine problems on t/o have to be done 1/2 of VMC or below....not to be done with a fuel selector...mixture is what I mentioned there....I'm not that dumb

Just throwing out a topic that is raised often by pilots and local instructors here, all, like myself, engine failureless, and lucky to be in that group so far. I hope this does not prevent us from having an opinion as to the best procedure for our students to be most prepared for a situation we hope to simulate as best we can, with, of course, safety and airplane care in mind.

--T-hawk

 

[bobbysamd]

Simulating engine failures

On the ground or above 3000 feet AGL I generally would fail engines with mixture or with throttles while blocking their view. At altitude in Seminoles, every MEI I remember used the fuel selector to create incipient engine "failures." I simulated engine failures on the takeoff roll by pulling mixtures, and pulled throttle for engine failures after take off (that's the only way!).

I also would block rudder, drag brake, etc., just as Avbug mentioned, to vary the experience. I also blocked prop controls, etc. after the students went through the identify, verify, feather routine. At least once during the course, at altitude and always near the airport, we always shut down and secured one in flight and practiced maneuvering. Always got it started again, warming it up with MP at the bottom of the green and prop at 2000 rpm until the CHT was at the bottom of the green. Sometimes, it was easier to start than other times. I got my multi and MEI in a B55 Baron and instructed some in B58 Barons. I always appreciated the accumulators in those great aircraft.

 

[JediNein]

Actual Engine Failures

Let's talk about actual engine failures in flight.
M= Me
S= instrument student pilot

At night, circling to gain altitude for the climb over the long grade known as the Castatic Park, in a Piper Warrior, about 20ish miles northwest of Van Nuys Airport. The aircraft is running beautifully and the climb is slow due to the heat, our weight, and general downdrafts over the entire area. I look at my watch and realize it is the 30 minute point, time to change tanks. (30 min: 1 hour: 1 hour: 1 hour: land).

M: Why don't you reach over there and change tanks?
S: Fuel Pump on, changing from right to left.
Pause
S: Checking fuel pressure, Ok, Fuel Pump of. ..
Aircraft engine (E): COUGH COUGH COUGH SPUTTER
M: Change back to the other tank, my airplane.

I continue the circle towards the nearest landable airport, Van Nuys and hold altitude to get best glide speed.

S: I CAN'T CHANGE IT! IT'S STUCK!
E: SPUTTER SPUTTER SPUTTER (barely running, around 800 RPM)

Student pushs seat back, unbuckles belt, gets mostly out of seat and starts hammering on selector valve.

E: SPUTTER Purrr... SPUTTER

M: keying the mike, advises Socal that we are returning to Van Nuys and have a problem. SoCal transfers us to Tower. I advise tower that we will be landing VNY immediately. (Whiteman was the same distance away at that point) Tower agrees to keep the airport open just for us. I think that the tower really doesn't have a choice. The student, in between grunts as he hammers on the valve, begs me to not declare an emergency.

Valve: SNAP!

E: COUGH PUURRRRRR

Collective sighs of relief. We land at Van Nuys. Tower chews on us a bit for keeping the airport open (was 2 minutes from closing for construction when I called SoCal). I just thank 'em for being there. Very puzzled, the controller clears us to parking.

On teardown we found the student had snapped three metal screws which had worked their way out into the selector valve's path. We did a teardown of our other Warrior and found track marks where the metal screws in that valve were about to do the same thing.

After a long discussion about the price of emergencies that were not declared, and a few other incidents -these of the student's causing, the student no longer trains with us.

Fly SAFE!
Jedi Nein

 

[ipilot]

the best way to simulate engine faliure on ground is to use brakes. the student won't even see you moving for the controls and you can catch him off gaurd.
in the pattern i like to simulate engine faliure with throttles as its safe. you don't want to end with student feathering the wrong engine at low altitude and then you don't have any engine to play with.
simulating engine faliure at altitudes 2000 or above you can use mixture or fuel selector because you will have enough time to react and they can also go through the checklist for restart. going for the fuel selector is a very obvious one as well unless you ask the student to look out to the left for traffic
i hope this helps......



________________________
Check that its three greens.....

 

[avbug]

Again, using the fuel selector to fail an engine cavitates the fuel pump and robs it of necessary lubrication. It also means you may not be able to restart. Something to consider. Most all fuel pumps are life limited when run without fuel, and it's cumulative. Something to think about.

Jedi, I appreciate your comment on the fuel selector issue. I recently had some participation in a thread on another site with a group of pilots about running tanks dry. I pointed out several cases in which pilots have run a tank dry, and then been unable to select the other tank because of various problems, including selector failure. If you don't mind, I may mention in generalities your account.

 

[HMR]

Run a search on FAA accident reports for the popular twin trainers. It’s remarkable how many students, instructors and even FAA examiners have been killed or injured during training/checkrides. My MEI had me do this before our first lesson and it was an eye opener. Now, a year later with 400hrs or so of dual given in light twins I’m all the more aware of how quickly simulated emergencies can turn into real ones.
Personally, I never shutdown an engine below 3,000’ AGL. Any lower I just retard the throttles. Above 3,000’ I alternate between mixtures and fuel selectors depending on the situation (if the student is banking to the left trying to show me their house, I go for the selector).
On the takeoff role I always have my hand on the mixtures in case we have to abort and the student freezes on the throttles. The students are used to seeing my hand up there during the role and rarely notice my hand move if I pull one of the mixtures. If you do fail an engine on takeoff role with the mixture make sure you do it immediately after brake release and be ready to pull the other quickly if the student doesn’t respond (I once did a Dukes of Hazard 360 on the touch down markers with a student who simultaneously jammed full left rudder and brake with full power on the right engine before I could shut everything down). That brings up another point, don’t let the students lock their elbows when they apply full throttle on takeoff. Once takeoff power is set they should have their hands cocked on the throttles so they can quickly abort if something goes wrong. Every time we takeoff we’re anticipating an engine failure and pleasantly surprised when we don’t have one, right?
On a similar note, when doing single engine pattern work make it clear to the student that they can use the “failed” engine at anytime if necessary for safety. There are few things more frightening for an MEI than having the tower tell you to go-around on short final and the student jams in full power on the “good” engine and starts pitching up.
Always make it clear who is going to fly the plane in the event that a simulated emergency becomes a real one.

 

[bobbysamd]

Pattern work

In the pattern we always called "simulated single-engine" so the tower knew our intentions and understood why we were flying a wider pattern.

I was never wild about the engine-failure-after-takeoff drill. The only way is to pull throttle. I was always a little worried that my trainees would "verify" by pulling the wrong throttle, even though I had retarded the throttle on the "bad" engine. I remember how one time I was training an MEI student. I was taking the takeoff and told him to pull one on me. On upwind he pulls a mixture on me! Well, let's just say we had a fairly serious discussion about pulling mixtures as not being the safe way of simulating an engine failure in the pattern.

I don't recall ever having to go around while on a simulated single-engine final. But I always had my hand near the "bad" engine's throttle and I would have added the power very slowly for the go-around.

 

[Timebuilder]

I did both my multi and MEI work in the venerable Seneca, a teriffic trainer for a "cabin class twin" control feel.

Below 3000 AGL we NEVER used any method besides the throttle for engine failure training. This gave us the option of bringing the engine back on line if needed.

Above 3000 AGL, the fuel selector was used once during each lesson, and the feathering procedure and air restart checklist was used to provide a real world type of experience. The balance of 3000 plus AGL work used the mixture, hiding both mixture levers with the checklist to practice the multiengine mantra.

VMC demos used rudder blocking at 50% of travel to cause an early loss of directional control before stalling. That's particularly important at high density altitudes.

Takeoff losses were taught below 50% of Vmc, and demonstrated at 50% on the checkride.

Live to fly anther day...

 

[FlyingSig]

Jedi,

Maybe I'm just not reading it in your post, but how can a stuck fuel selector be the student's fault? Not ragging on you, just curious what someone could do on a Warrior to cause what you describe.

 

[JediNein]

Pilot Errors

I should have been a bit clearer. The fuel selector problem was not the student's fault. Three other minor incidents were caused by the student's lack of judgment and unwillingness to accept instruction.

The combination led us to refuse further instruction until he decided he was ready to train and passed an interview with the chief instructor.

For the rest of the students and instructors, we had a long discussion at our biweekly meetings about complacency. If something doesn't work as it used to, we need to know so we can fix it. Airplanes give warnings before they fail. We also switched mechanics.

All of a sudden, the little squawks appeared in the book and people started calling immediately after a flight with a problem appearing in flight. The increased attention has found a cracked primer line, an alternator causing a high frequency harmonic with about 1 hour before the spinner and prop departed the airplane, a hydraulic leak in a landing gear system, four failed flight instruments, and mostly unusuable shoulder harnesses. Both aircraft owners had large shop bills last month, neither are complaining too much as a wrecked airplane would cost more. This month all three aircraft are set to make their target of 60 hours each and 20 in the complex.

Fly SAFE!
Jedi Nein

 

[surplus1]

Pardon me for butting in, but I can't resist some observations and general comments. Caveat: It's been quite a while since I've given any instruction in light aircraft, but its also been a very long time since I started flying for a living. I'm still alive and that should be every pilot's ultimate goal.

Some things to consider:

1) More "accidents" occur in training than anywhere else. Particularly multi-engine training. Practicing emergencies is more dangerous than real emergencies in many cases.

2) Loss of an engine in a Part 23 twin is definitely an emergency. If it happens for real, declare it immediately without hesitation. If it happens to you, don't stop flying the aircraft until it has stopped moving (preferably on the airport, but anywhere).

3) Most if not all Part 23 aircraft (twins) have very poor single engine performance. They are not "required" by the certification rules to be able to actually climb on one engine and most can't (assuming GTW). This is seldom taught in the initial ME training phase. Learn it yourself and then pass it on to everyone you work with.

Note: flying around the airport (or trying to) on one engine when you have a sink rate of 50 ft/min. or a climb rate of 50 ft/min is not a productive exercise and probably will not increase your longevity.

4) Most (not all) "multi-engine instructors" have very little flying experience and even less multi-engine experience. Most (including the "experienced ones") have never had a real engine failure (thanks to strong engines) in a light twin. In general terms, its the classic "blind leading the blind" scenario. I'm not demeaning anyone or depreciating anyone, it's just a statement of fact. I was in the same category at one point.

5) The total and sudden loss of all power on one of your 2 engines, is a very rare event in the real world. Most engine failures give advance warning of some sort, which often goes unnoticed. This is seldom emphasized in initial multi-engine training programs. Teach situational awarness. It will do more to increase a pilots longevity by avoiding emergencies than you'll ever teach by practicing them.

6) Lot's of emphasis is often misplaced on the student's "quick response" to a "sudden failure", with a memorized sequence of reactions or routine, all coming under what I call the guise, of "immediate action". The truth is the best "immediate" response you can have, is no response at all. Take your time and fly the airplane. With the exception of a pre Vmc or pre V1 failure (which always mandates an immediate abort, there is time to deal with inflight failures. Slow down. There's little worse than shutting down and feathering the prop on an engine that can still develop 60% of rated power, just becuase you hastily completed a routine of unnecessary "immediate action". There's lots of evidence in twisted aluminium, ashes and body bags to substantiate the results of a feathered prop when the only thing that failed was a turbocharger.

7) In most cases, a total failure of one engine on takeoff (after airborne) in a light twin operating at max gross weight is going to require an off airport landing and preclude a "turn back" to the airport, just like it does in a single-engine aircraft. The failure to teach this reality has resulted in a lot of "loss of control" accidents in light ME aircraft following engine failure on takeoff (whether the failure was real or began as a simulated exercise gone bad). ME instructors in light twins need to teach student pilots how to survive the loss of power on takeoff, not how to attempt to make the airplane do what it cannot do. This includes "cabin class" twins like the Navajo and Cessnas 421 or similar types.

8) Teach your students (and yourself) to develop a plan of action, in the event of power loss, BEFORE takeoff power is applied. This is essential. There is very seldom anything "unexpected or sudden" that happens in an airplane, when it's pilot has planned ahead. If you know the failure is coming on EVERY takeoff, its a lot easier to deal with it when it does happen. The moral is a simple one ..... no surprises. That's how you get gray hair and the chance to retire.

9) Forget the so-called realism idea (refer to #5). It isn't going to be "sudden and unexpected" if you've complied with #8. Don't simulate loss of power by creating a real loss of power at low altitudes (below 3,000 agl). Don't cut mixtures and don't cut fuel selectors close to the ground. Its not healthy and there is no need for the sake of realism. Don't do it. If what you do and say to the student is correct, you can use the throttle for every simulation with all the realism required. Safety is number one. Practice that first and the realism will take care of itself.

How long is this initial ME training going to last anyway, 10 hours? By the time the new ME pilot has 100 hours of multi time he/she will have forgotten most of what you taught in those 10 hours anyway (unless they're trying to teach it to somebody else), particularly if you focus on the wrong stuff. How you simulate power loss is not important from a training standpoint, therefore you should use the safest possible method and that does NOT include fuel selectors OR mixture controls (at low altitudes) and should not include fuel selectors at higher altitudes (where mixture is fine). By the time you get to where you need more "realism" you'll be using good simulators, which is the only place this sort of russian roulette should occur.

10) At some point, you do have to go through a complete shutdown and restart in flight. Explain to the student that it will NOT be the same in every type and, perhaps more importantly, if you have a real engine failure you're not going to be restarting anything. This is an academic exercise and should be treated as such. Before you do it, know your particular airplane and know what it takes to unfeather that propeller. There's always a good possibility that you won't be able to do it (in some types). When you rehearse this, decide where you're going to make your single engine landing BEFORE you begin the exercise. Plan ahead.

In many, many years of doing this flying thing for a living, I've only had one (1) partial loss of power in a light twin (Queen Air), many engine shutdowns (precautionary due to partial failures) in large (round engine) recips and several engine failures (including cylinders exiting through the cowling) in large recip engines. A few precautionary shutdowns in turbine engines and only one (1) engine failure in a turbine powerplant. Many thousands of hours (more than shown in the profile) are included.

The bottom line is simple. Don't kill your student (or yourself) trying to be "realistic" in how an engine fails. Teach your students to think, to plan ahead and to use good judgement. That way you'll both grow old. This profession isn't about heroics or theatrics. Don't use them as training tools.

Finally, when you learn of examiners or FAA types that are into heroics, use somebody else. Examiners that need to "prove" anything, especially their expertise, should be avoided like the plague that they are.

Fly safe.

 

[Timebuilder]

Good advice.

It's nice to hear someone else preach about safety in multi training. All of our single engine work (with engine shutdown) was within easy range of the airport, and beginning at 3,000 to 4,000 AGL.

We also used two FAA reprints in our school:

Flying Light Twins Safely

Always Leave Yourself an Out


They're good reading that can save your life.

 

[bobbysamd]

Good points above

Us, too. We used the two FAA handouts.

Really,what ME instructors have to stress is an engine loss in a light twin might give you more time to determine alternatives than an engine loss in a single. In a single, your checklist is the ABC's; Airspeed (best glide), Best place to land, and Carb heat. In a light twin that loses an engine after takeoff, your procedure really isn't much different than a single. You may have to consider an off-airport landing. You may not have the performance (especially in Prescott on a hot day) to make it around. Your pattern may get too wide, especially with all the shallow turns and trying to climb.

Good discussion. Multi and instruments were my two favorite areas of instruction.

 

[Traumahawk]

'Always Leave Yourself An Out' is an awesome article. 5 stars.
Anyone know where you can get a copy of 'Flying light twins safely' ??

---T-hawk

 

[bobbysamd]

FAA Reprints

I'd start by going to FSDO. FSDOs usually stock all the reprints.