Light Twin (AMEL) Training - MEI Techniques & Technical Data Search
This post is aimed at MEIs, or folks with solid technical knowledge of light twins.

For training purposes, introducing ME students to engine out scenarios, below 3000 AGL is fairly straight forward, and often simulated by pulling a throttle below 50% Vmc and above 400 AGL.

When it comes to actually shutting down the engine, I am interested in technical information out there that might educate a budding ME instructor on how to manifest the engine shutdown and feather drill, short of any further restrictions that an AFM/POH may dictate.

Development of a scenario that drives the student to secure the engine, such as rapidly decreasing oil pressure with rapidly rising temp and increasing roughness could be used.  But, I also think there is value in failing an engine to surprise and help the student deliberately assess and identify the correct engine to feather.  Accident data shows mis-identification as a real and reoccurring problem.

Have searched quite a bit thru FSIMS, FAA, AOPA, and other credible web sights for a discussion on the use of the throttles, mixture and/or shutoff valve in hopes of building a reference list of techniques, risks, and benefits for various scenarios/methods.

Some techniques recommend blocking the student's vision of the mixtures and failing either with the mixture.  This is good to a point, but inhibits the student actually executing doing the generic "AIRSPEED, MIXTURES, PROPS, THROTTLES, FLAPS, GEAR, IDENTIFY, VERIFY, FEATHER" drill.  There seems to be a line of thought out there that using the fuel shutoff valve may cause cavitation of the fuel system, but I have not been able to find technical source (i.e. manufacturer, FAA, NTSB, etc) to support that this is a documented issue.

Any technical references would be greatly appreciated with regard to cavitation  or damage to fuel system, especially with regard to use of the fuel shutoff valve. 

Primarily concerned about training in light twins (Piper & Cessna).  Thanks.
5 Replies
1147 Posts
Keith Teister:
  I am searching for technical data to support your comment, ". . . the fuel selector should never be used to simulate failure . . ."


That wasn't exactly "my comment", but rather an FAA rule for checkrides in the cited documents.  I suspect it is based not on any "technical data", but rather on experience.  The problem with using the fuel selector is that it empties the fuel line from the valve to the engine, so it can take a lot longer or otherwise make it harder to restart the engine.  In some circumstances this could compromise safety if the applicant/trainee allows things to get out of hand.  OTOH, restart is generally nearly instantaneous even with the mixture.

Thank you Ronald. 

Yes, the FAA guidance below 3000AGL (or higher if directed by AFM) is clear.

And you bring to light the general discussion I have had numerous times regarding above 3K--and I don't necessarily disagree.  I am searching for technical data to support your comment, ". . . the fuel selector should never be used to simulate failure . . ."

I hadn't included the references you cite in my original post--thanks for highlighting them in the discussion.


1147 Posts
You said, "For training purposes, introducing ME students to engine out scenarios, below 3000 AGL is fairly straight forward, and often simulated by pulling a throttle below 50% Vmc and above 400 AGL."

I assume you mean doing the engine failure before 0.5 Vmc on the ground before liftoff,  For engine failure after liftoff, all guidance I know (8900.2, ACS, etc.) says not to fail the engine until Vyse (or Vsse) has been achieved as well as not below 400AGL, by which time the gear should already be up.  As for how to simulate the engine failure, the FAA's position is that below 3000 AGL, only the throttle should be used.  Mixture should be used only above 3000 AGL (or higher, if specified in the POH), and the fuel selector should never be used to simulate failure or turned off during a simulated engine failure drill.  And as Matt discussed, you don't turn a simulated emergency into a real emergency in training -- the engine should never be fully secured except to learn/demonstrate feathering above 3000 AGL or higher if so recommended in the POH.

Also, in this process, it's critical that the instructor guard the prop controls lest the trainee inadvertently pull the wrong prop back after identifying with the throttle.  Some MEI's avoid this problem by telling the trainee to just pull the prop only halfway back, or even to just tap the prop control.  The problem with this technique is the Laws of Exercise and Primacy -- if the trainee is taught this during initial ME training, they're too likely to do the same if an engine really fails later.  I teach my ME trainees to pull the prop back until it stops, either by my blocking it or reaching the stop.

Private Pilot ACS page A-18, "Multiengine Considerations"
FAA Order 8900.2, Examiner's Handbook, Section 7, paragraph 13c, "Simulated Engine Failure", page 7-26

Thank you--definitely not off topic and supports why I'm searching.  I agree in the use of simulators, and not rushing to failure.  Deliberate, thoughtful, and affirmative action is very important.  The E-11 crash is a very recent reminder, and one of the reasons I'm searching.  I (as many) have been blessed to have had rigorous training (and orchestrated training in sims as the instructor) for multiple aircraft, and wish they were more accessible for our budding airmen.

I also believe in understanding the risk as objectively as possible.  The cavitation conversation always comes up.  And, although it makes sense, and I don't necessarily disagree in principle, I really would like to point to evidence on why or why not to do something.

Sharing what I find is the best part of using this awesome forum.

Take Care!


This is probably not helpful or what you're looking for but if your goal is to reenforce the memory items and muscle memory reaction, I suggest a simulator or flight training device. You can fail an engine in any number of ways and repeatedly in a short amount of time. The students can complete the full checklist and work through whatever scenario you present. Once they have that down move to the airplane and let them experience the startle response, yaw/roll sensations and required rudder input.

In the actual airplane the primary goal is not to allow the simulated emergency to become an actual emergency. I think in light twins using the mixture to simulate the failure is the best option.

In my experience, teaching students to take thoughtful deliberate actions is most important. How you simulate the failure less so. Especially as we progress to larger more capable aircraft, doing it correctly is more important than doing it quickly. The US Air Force E-11A crash in Afghanistan is a recent tragic example of a crew responding too quickly and incorrectly to an abnormal situation, (an inflight engine failure at 43,000 ft) turning it into an emergency situation, (shutting down the operating engine and creating a dual engine out scenario) not reevaluating their poor decisions, (the engine they incorrectly secured would have restarted if they attempted it) and crashing an otherwise flyable airplane. (They took too long to turn towards an alternate airport and glided onto an open uneven field and did not survive.) 

I apologize if I strayed off the topic. It is always great to find new techniques and teaching points. If you come up with a good solution I would be interested in hearing about it.