Products page contents:

Report Airplane Control after Engine Failure

Paper The Effect of Bank Angle and Weight on V_MCA

Paper Staying Alive with a Dead Engine

Report Imperfections and Deficiencies in FAR and CS

Dutch products page

Products of AvioConsult

Prevention of Airplane Accidents after Engine Failure

Introduction.  Five years after the beginning of aviation, Orville Wright and passenger Lt. Thomas Selfridge took off to conduct a test flight in a new aircraft.  In a tight turn, at a height of approximately 150 ft, one of the propellers broke and Orville lost control; the plane plunged to the ground.  Lt Selfridge was killed and Orville Wright himself was seriously injured.  Aviation claimed its first victim, the cause being the loss of control due to asymmetrical thrust.  The Wright bothers' aircraft had one engine, driving two propellers and were therefore to be considered multi-engine airplanes.  Orville did not know yet how to control an airplane after the failure of one of the propellers. 

Today, 100 years later, engine failures of multi-engine airplanes continue to take their toll.  All around the globe, accidents with both small and big multi-engine airplanes continue to happen quite frequently following the failure of an engine during takeoff, go-around and during training, despite the fact that all airplane types are thoroughly flight-tested and operational limitations are published in the airplane's flight manual.  Pilots are obviously still not fully aware of the control limitations that apply after engine failure.  The question how accidents after engine failure can be prevented is therefore actual since the beginning of aviation.  

AvioConsult, trained at one of the few Experimental Test Pilot Schools, researched many accident investigation reports and concluded that these reports often present only 'loss of control' or sometimes even 'a stall' as the cause of an engine failure related accident.  AvioConsult noticed that the investigators never took a close look at the test methods and conditions that were used to flight-test the airplanes while an engine is inoperative and hence not at the conditions at which the airspeed limitations that are presented in the Airplane Flight Manual (AFM) are valid either.  In addition, investigators almost never evaluated the engine emergency procedures in AFM's to verify their correctness.  
Therefore, AvioConsult took the initiative to investigate many engine failure related accidents and write a report that should really prevent accidents after engine failure.

Tail design limitations.  The vertical tail and rudder of a multi-engine airplane are designed to counter the biggest asymmetrical yawing moment caused by the remaining engine(s) after engine failure.  The aerodynamic side force that can be generated by the vertical tail and rudder depends highly on the (square of the) airspeed.  As the airspeed decreases, the rudder deflection has to increase since the engine thrust remains the same.  But there are mechanical limits to the rudder deflection angle and the vertical tail is also limited in size.  So there must be a lower speed at which the vertical tail plus rudder generate just a high enough side force to counter the asymmetrical thrust and hence, to maintain the heading.  
The designers of the vertical tail know that the size of the tail and/ or rudder deflection can be reduced by banking a few degrees away from the inoperative engine.  FAR 23.149, 25.149 and equivalent allow a bank angle of max. 5 degrees.  So, in order to save manufacturing cost and weight, manufacturers design and build the vertical fin and rudder of their airplane to a size that generates just enough side force for maintaining straight flight after engine failure, while banking a few, up to 5, degrees away from the inoperative engine at an airspeed that is lower than or equal to 1.2 V_S.  

This is where the operational problems begin: 
Nobody ever told the airline pilots about these design or hardware limitations.  Pilots do not know that they should maintain both straight flight as well as a small bank angle away from the inoperative engine for the vertical tail to be able to maintain the equilibrium of side forces and yawing moments, because this is neither prescribed in the Airplane Flight Manual nor included in the engine emergency procedures.  This is, to the opinion of AvioConsult, why accidents after engine failure happen. 
The following warning would therefore apply to operations with multi-engine airplanes:

Flight crew procedures.  Because of these design considerations (hardware) and, hence, inherent limitations of the vertical tail for the lateral control of the airplane, it is required to design operational procedures (software) to ensure that an engine failure and the limited size of tail and rudder do not lead to the loss of control and to catastrophic accidents.  For this purpose, the airplane is flight-tested during which the lowest airspeed is determined at which straight flight can be maintained with the most critical engine made inoperative.  During the test, most variables that have influence on the engine-out equilibrium of forces and moments are kept at their worst case value, returning the highest  -- the worst case -- speed for maintaining lateral control.  This speed is called the minimum control speed in the air, V_MCA.  The bank angle used during measuring V_MCA is the same as was used to design the vertical tail: mostly 5 degrees away from the inoperative engine, but it is up to the manufacturer to select an angle, provided it is max. 5 degrees.  V_MCA is published as a limitation in the Flight Manual of all multi-engine airplanes.  The smaller (and cheaper) the tail, the higher V_MCA would be, but Regulations require the V_MCA to be less than 1.2 V_S.  This is therefore a factor that links the hard- and software.  Another factor for the tail design is that V_MCA should be as low as possible since the lift-off speed is then lower, enabling shorter runways or higher payload.  Refer to the Regulations and /or to the report presented below for other limits to control deflections and forces.

Many pilots consider V_MCA a useless speed, but it is still determined and indicated on the airspeed indicator of Part 23 airplanes by a red radial line – for a purpose, which is the limited size of the vertical tail!  
On big Part 25 airplanes, rotation speed V_R and takeoff safety speed V₂ are used instead of V_MCA, but V_MCA is used to calculate these procedural speeds.

V_MCA and/or  V₂ are determined to be used as operational limitations by all pilots to ensure a safe takeoff, go-around and flight, whether or not an engine fails or is inoperative.  Despite these safety ensuring speeds, catastrophic accidents continue to happen, both after engine failure and during flight while an engine is inoperative.  
After reading many accident investigation reports, airplane flight manuals and student pilot textbooks, AvioConsult noticed that (airline) pilots have a very different understanding of V_MCA than is intended by regulations, is used by manufacturers for designing the vertical tail and is used by experimental test pilots during flight-tests to determine V_MCA.  The mishap pilots did not maintain straight flight and did not maintain a small bank angle away from the inoperative engine while the airspeed was low and the (asymmetrical) engine thrust was high.  The aerodynamic force generated by tail and rudder became too small to counter the asymmetrical engine thrust. 

V_MCA is not just a constant.  Following additional research, it became clear that V_MCA is not just a constant number, but that V_MCA varies a lot with bank angle.  If the wings are kept level instead of 5 degrees away from the inoperative engine, V_MCA is at least 8 knots higher than the V_MCA that is listed in the Airplane Flight Manual (AFM).  For Boeing 707 and DC-8 type airplanes, the actual wings level V_MCA is 35 knots higher than the AFM listed V_MCA that was determined while banking 5 degrees away from the inoperative engine!  

An example: In the linitation section of the AFM of a multi-engine airplane, a V_MCA of 95 kt is listed. This V_MCA was determined while banking 3 degrees away from the inoperative engine, which was in this example the left engine, and at low gross weight (red line).  
The red line (dip at 5 degrees) presents the actual V_MCA while the airplane is at low weight (the test weight); the blue line shows the actual V_MCA at high weight, both while varying the bank angle and with max. thrust set on the opposite operating engine.  

The figure, that is calculated using the stability derivatives of the airplane (refer to the paper presented below), shows the effect of bank angle on actual V_MCA.  It in fact shows that a higher speed is required to maintain straight flight with an inoperative engine, for the given size of the vertical tail at different bank angles.  
The figure also shows that if a turn is initiated while the airspeed is close to the AFM-listed V_MCA, as is the case during takeoff, go-around and initial climb, then the actual V_MCA will increase above the indicated airspeed which leads to the loss of control; the vertical tail is simply not big enough to prevent the loss of control.  Turning at low speed and high power settings is therefore extremely dangerous.  A catastrophe following the failure of engine and while the other engine(s) are producing max. thrust can only be prevented by maintaining straight flight while keeping the bank angle 3 to 5 degrees (as opted by the manufacturer) away from the inoperative engine, because this is how the vertical tail was designed and dimensioned and how V_MCA was measured.  A lower thrust setting than max., or a forward cg of course reduces the asymmetrical thrust moment or increases rudder power and hence decreases the actual V_MCA.  This is why not all takeoff or go-around engine failures end in a disaster.

Deficiencies in manuals.  AvioConsult also concluded that the definition of V_MCA, that is used in textbooks, in Flight, Operating  and Training Manuals, was often copied straight out of Aviation Regulations (FAR/ EASA/CS 23.149 or 25.149 or equivalent).  From the manufacturers point of view, this might look safe and also acceptable to the authorities that have to approve the manuals.  However, FAR and EASA/CS Part 23 and 25 are for the certification of an airplane and definitely not for operations with the airplanes.  The inappropriate copying of the definition of V_MCA out of Aviation Regulations leads to incorrect and incomplete definitions of V_MCA in the manuals used by operational pilots and hence leads to inappropriate crew response to propulsion system malfunction.  
In most Flight Manuals of multi-engine airplanes (big and small), there are at least four (4!) deficiencies and imperfections in the definition of V_MCA and are two important conditions missing as a consequence of the inappropriate copywork.  These deficient and imperfect definitions have led and will lead again to a dangerous misunderstanding of air minimum control speed V_MCA and of takeoff safety speed V₂ by most, if not all, multi-engine rated pilots and will continue to lead to catastrophic accidents in case an engine fails during takeoff or go-around, or during maneuvering while an engine is inoperative.

Conclusion.  During the search for the real cause of engine failure related accidents, AvioConsult reviewed many accident investigation reports, flight manuals and training material and noticed that V_MCA was almost never considered a factor as cause of such an accident because it was unkown to the investigators (and to the mishap pilots) that the actual V_MCA changes considerably with bank angle and engine thrust.  Comments on a limited number of these accidents are presented on a separate page; please click here.  
AvioConsult decided to do something for the prevention of airplane accidents after engine failure and wrote papers and reports that are presented and downloadable below.  

One of the recommendations is to include a warning with V_MCA (and V₂) data in each and every Flight and/ or Operating Manual of multi-engine airplanes and in student pilot textbooks.  The warning could look like this:

 Home | News | Services | Products | Downloads | Accidents | Links | Contact