By Rich Stowell
Pilots historically have been preoccupied with turning back to the runway following failure of the airplanes lone powerplant on takeoff. The maneuvers successful outcome generally depends on a variety of factors, not least of which include altitude, airspeed, runway length and pilot technique. Each year some pilots try the turnback maneuver after an engine failure and succeed while others dont.
Of course, we dont hear too much about the success stories, since they dont show up in the accident statistics-only the failures do.
But whats the rationale behind attempting a turnback after engine failure in a single? One popular answer: Returning to the airport improves the odds of a successful forced landing. If success is measured in terms of survivability, this line of thinking is demonstrably flawed.
As part of the General Aviation Crashworthiness Project conducted in the 1980s, the National Transportation Safety Board (NTSB) reviewed 535 light airplane accidents involving 1267 occupants. Of the accidents studied, 391 involved at least one fatality; the remainder involved no fatalities. The principal impact angle and impact speed was estimated for each case based on a detailed analysis of accident data.
Overlaying the graphical distributions of those occupants who survived the crashes (or who could have survived had they been wearing shoulder harnesses or sitting on energy-absorbing seats) and those occupants who did not survive revealed a survivable crash envelope bound by the following points: 0 degrees at 75 knots; 45 degrees at 60 knots; 90 degrees at 45 knots.
In addition, the cockpits of modern light airplanes are designed to withstand at least a 9.0 G forward deceleration. Thus, during a crash landing, the required stopping distance over which a livable cockpit volume will still be maintained can be surprisingly short.
For example, a Cessna 152 gliding per its POH does so at 60 knots along a glide slope of about six degrees. Even if the pilot fails to flare the airplane and impacts the ground at the full 60 knots under a 4.5 G deceleration, the stopping distance is just 35 feet-well within the survivability envelope. Short but survivable crash zones are not hard to find even in the most heavily congested environments surrounding some airports.
Impact speed and impact angle are the primary determinants of crash survivability. In other words, survivability does not depend on where the crash occurs, but rather how the crash occurs. Contacting obstacles and terrain-trees, water, buildings, parking lots, ball fields-while low and slow, in the landing attitude and with at least some distance over which the airplane can decelerate all significantly increase the chances of survival. On the other hand, stalling or spinning into the ground sharply decreases these odds.
Turnbacks Are Riskier
During a crash landing, the pilot must maintain control of the impacts angle and speed as far into the forced landing process as possible-never stop flying the airplane. But does that mean a turnback to the runway is inherently a higher-risk maneuver? Yes.
Compared to landing straight ahead, a Canadian study of stall/spin accidents over a 10-year period assessed the risk of death or serious injury as eight times greater when a turnback to the airport was attempted.
Some argue that such accident statistics overestimate the turnback risk, that a number of successful turnbacks may not be counted because no accident report was filed. But it is equally likely that a percentage of successful off-airport landings fail to appear in accident reports as well. The airplanes in these instances are retrieved, hangared, and repaired without fanfare. We can, however, use the results of a turnback-specific study as a predictor of the likely outcome.
The study looked at the feasibility of successfully executing a 180-degree turnaround following an engine failure at 500 feet agl. A variable-stability flight simulator emulating the characteristics of a light, single-engine airplane with fixed landing gear and a fixed-pitch propeller was used. The simulated airplane was capable of completing a turnback in 500 feet. Useable data was collected for 20 test subjects experiencing 147 engine failures spread over seven different sorties. The flight sequences typically lasted about two minutes.
The test subjects ranged from a 40-hour Student Pilot, to a Flight Instructor, to a former military pilot with more than 5000 hours. Nearly two-thirds of the subjects were Private Pilots; the rest were equally divided between Student and Commercial Pilots. Thirty-five percent of the pilots had logged fewer than 100 hours, 45 percent had between 100 and 200 hours, and 20 percent had more than 200 hours.
The study defined successful outcomes as follows: In all cases, the maximum rate of descent could not exceed 2500 fpm, rate of descent at touchdown could not exceed 500 fpm, and bank angle had to be within five degrees of wings-level below 100 feet agl. For turnbacks, the airplane had to complete at least 175 degrees of heading change without exceeding a 55 degrees of bank.
One hundred percent of the attempts to proceed straight ahead (35/35) resulted in successful outcomes-pilots maintained control of the airplane all the way down to the ground every time. The probability of survival in an actual emergency: high. The probability of crash damage to the airplane: It depends on several factors.
By contrast, only 62 percent of all of the attempted turnbacks were successful (69/112). Thus, nearly two out of every five attempts failed. And the majority of failed turnarounds culminated in stall/spin departures. The probability of survival from the failed turnbacks: low. The probability of significant crash damage to the airplane: high.
Real Numbers, Real Airplane
In his book, Performance of Light Aircraft, physicist-pilot John T. Lowry develops a methodology to predict a given airplanes V-speeds as well as its rates and angles of climb or descent at any airspeed, any weight, any bank angle, and any throttle setting. Using data for a fully-loaded Cessna 172 climbing out at Vx under typical airport conditions, Lowry performs a sample calculation for an optimized turnback following an engine failure from 500 feet agl.
This airplane under these conditions can successfully complete the turnaround; the bad news, however, is that it contacts the ground nearly 1200 feet short of the runway. If some of the conditions had been different-a longer runway, or lower takeoff weight, or stronger headwind-the airplane could have completed the turnaround and landed downwind on the runway starting from 500 feet agl. Interestingly, flap setting had a negligible effect on the outcome for this airplane.
As the turnback study and Cessna 172 simulation illustrate, successfully turning around after an engine failure on takeoff isnt impossible; its just unlikely to have the desired result. The climb angles of light airplanes are relatively shallow, even when climbing at Vx. Consequently, even though we may have achieved sufficient altitude to complete a turnaround, we may by then be too far away to glide back to the runway. The capability to turn around varies based on the usual suspects: wind and runway conditions, the airplane and its configuration, density altitude, local obstacles and piloting technique.
All other things being equal, an emergency turnaround that might work after taking off from a quiet, 6000-foot runway on a calm-wind day might be futile when departing from a busy, 1800-foot strip into a 20-knot headwind.
In spite of the warnings against turning back, pilots still want to know the best way to handle an engine failure after takeoff. But the answer depends on the definition of best. For example, best survival remains the straight-ahead. Simply lower the nose of the airplane to its best glide attitude and land somewhere within an area extending outward at an angle between 45 and 60 degrees from either side of the runway centerline.
The aerodynamic best turnaround, on the other hand, occurs in a coordinated turn, at a bank angle of 45 degrees, at an airspeed just barely above the corresponding stall speed. Although the least amount of altitude is lost during this turn, the margin for error is slim-to-none. One mistake and you either stall/spin, or lose control over the bank angle. True, glider pilots are required to demonstrate this very turn at low altitude following a simulated rope break. Sadly, that demonstration has little to do with how to survive the emergency and everything to do with saving the glider-is this really what we want to reinforce?
An alternative to the aerodynamic best turnaround is a coordinated turn at 30 degrees of bank, again at an airspeed just above the corresponding stall speed. The tradeoff in terms of increased margin for error versus the additional altitude lost compared to the aerodynamic best turnback is well worth it. Even so, a misstep still could easily culminate in an inadvertent stall/spin.
Pilots attempting either of the above turnback maneuvers must be committed to operating critically close to the stall in a bank close to the ground, while under the duress of an actual emergency. To reap the benefits of these types of turnbacks, the turns need to be executed with the stall warning horn activated throughout.A third alternative is to perform the turnback at whatever bank angle you are most comfortable with, while maintaining the airplanes wings-level best glide speed. Although this may wind up losing more altitude than the other two turnaround methods, it perhaps offers the greatest safety factor. It allows you to operate within your natural comfort zone at a time when stress levels are elevated.
To get a feel for the demands of the various turnback strategies, practice each of the three types mentioned above at altitude from both Vx and Vy climbs (take an instructor along if youre at all uncomfortable with the prospect of stalling in a turn). Perform left and right turnarounds, noting the altitude lost each time.
Also bear in mind that of the successful turnarounds in the turnback study, the bank angles used broke down thusly: 14 percent of the successful turnarounds occurred with banks in the 20-30 degree range; 57 percent occurred in the 30-40 degree range; and 29 percent occurred in the 40-50 degree range.
From your flight experiments, decide which turnaround method you might adopt. Add an appropriate safety factor to the average altitude lost for the method youve chosen. And, before each takeoff, mentally add that number to the altimeter setting-this is your threshold altitude for the turnback. Based on the ambient conditions, decide if you will turn back should the engine quit at or above your threshold, which way youll turn, what your target bank angle and airspeed will be, etc.
By doing this, you will have already made critical decisions about how you will handle an engine failure on takeoff. But always keep in mind that the primary objective is to maintain control over the forced landing; Saving the airplane shouldnt even be a consideration. If youre high enough to make it to a nearby crosswind runway under control, by all means head for it. If you can confidently execute a turnback without sacrificing survivability, then do it. But if any doubt exists at all about turning around, dont. Just land straight ahead.
Turning back doesnt improve our odds of surviving an engine failure after takeoff. For that matter, it may not result in us even reaching the departure end of the runway anyway. So why bother?
The only possible justification for attempting a turnaround is that it might minimize damage to the airplane. But you must decide just how much of your survivability (and that of your passengers) you are willing to gamble to try to reduce damage to the airplane-one percent? Ten percent? Forty percent?
In the case of an engine failure shortly after takeoff, sacrificing the airplane to improve the chances of survival always is a superior choice to sacrificing survivability in the hopes of saving the airplane. Granted, it takes a certain amount of intestinal fortitude to commit to a controlled crash landing off-airport, especially if the landscape ahead appears inhospitable.
But in the final analysis, a controlled crash that allows the occupants to survive is far more desirable than an uncontrolled crash that consumes both the occupants and the airplane.
Also With This Article
“Inside The Simulator-Based Turnback Study”
-Rich Stowell has been named 2006 Western-Pacific Region Flight Instructor of the Year. In addition to being a NAFI Master Instructor of Aerobatics, Rich has walked away unharmed from two in-flight engine failures, one of which occurred shortly after takeoff (he landed straight ahead).