EASA and FAA have significantly modified the stall-spin requirements for part 23 aircraft during the recent re-write of CS/FAR23. The “spin-recovery” requirement for non-aerobatic aircraft has been removed and instead a new “departure aversion score” has been introduced. The concept follows an initiative already undertaken by the FAA back in 1991 and makes room for innovation.
Airplanes performing a “spin” are a spectacular thing to look at. A spin is a complex interaction between aerodynamic and inertial forces that is characterized by an autorotation and rapid descent with angle of attack (AOA) beyond stall [1] [2]. In fact, it is a rather disorienting maneuver, and most airplanes are not approved for spins. Nevertheless, stall-spin accidents account for a significant part of general aviation accidents, mainly because pilots get into inadvertent spins at altitudes so low, that recovery is fundamentally impossible [3]. In such cases, the capability of the aircraft to recover from a spin is simply irrelevant – what matters is how “easily” the aircraft departs controlled flight. This is the fulcrum of the new regulations.
The phases of a spin
Industry practice is to divide the spin maneuver into four phases: Entry, incipient, steady and recovery phase [3]. As Figure 2 depicts, the vertical trajectory changes significantly during a spin and some time may pass, until the aircraft transitions from one phase to the next.
The flight dynamics behind a spin
To this day, spins remain somewhat mysterious, as there is no method that would reliably predict spin-behavior for new aircraft [2]. Every spin flight test or every spin for that matter is therefore associated with some degree of uncertainty.
“With the machine mowing forwards, the air flying backwards, the propellers turning sideways and nothing standing still, it seemed impossible to find a starting point from which to trace the various simultaneous reactions. Contemplation of it was confusing. After long arguments we often found ourselves in the ludicrous position of each having been converted to the other’s side with no more agreement than when the discussion began.”
- The Wright brothers
Mathematics involved in understanding spins are quite complex and would be beyond the scope of this article. There is one aspect that every pilot should memorize:
A spin can only occur, if the airplane is stalled on at least one wing [1][2][3].
The secret lies in aircraft stability, a topic that engineers tackle using “stability derivatives”. These are mathematical terms that describe the dependence of flight parameters on each other. They can have stable or unstable regions and obviously the aircraft design engineer will make them stable in the normal operating regime [2]. Figure 3 shows what a possible variation of these derivatives looks like, once the aircraft is stalled:
It is this transition from stable to unstable in a combination of these derivatives that permits the aircraft to depart controlled flight [1][2]. Note: Not all of these have to get unstable, there are various combinations that promote departure from controlled flight [2].
Certification requirements
CS23 pre-amdt 5 / FAR23 pre-amdt 64 (in the past)
So, what to do about this? Historically the certification authorities accepted designs (part 23) that were departing controlled flight when mishandled, as long as it was demonstrated that the aircraft could be recovered [4]. There were some rules, on when the recovery inputs were to be applied and how quickly the aircraft should recover. Figure 4 summarizes the FAA certification requirements before FAR23 amdt 64, similar requirements were stipulated in CS23 before amdt 5 [4]. Originally, there were two options: “Non spin-approved” aircraft had to demonstrate, that they would recover within one turn, after pro-spin control inputs were applied for 1 turn or three seconds, whichever took longer acc. FAR23.221(a) [3]. For “Spin-approved” aircraft, recovery was initiated after six turns, and the aircraft had to recover within one and a half additional turn [3].
“Spin-resistant”, an FAA initiative
Already in 1991, the FAA introduced a different option, called “spin-resistant design” [4]. The idea was based on an extensive NASA study and accident review and took into account the fact that most spins that led to accidents were actually entered inadvertently with insufficient altitude for recovery [4]. The idea was simple: Aircraft designers should be given the option to develop solutions that would make it very hard for pilots to inadvertently enter a spin and thus make the spin recovery demonstration obsolete [4]. The focus of that research was on Modified Outboard Leading Edge (MOLE) designs, one variant being the “cuffed wing” (see Figure 5). It could be demonstrated that such designs maintain good controllability as the airflow remains attached in the outer span regions (where the ailerons are located). The flow separation forming at the wing root does not propagate outboard, as the sharp discontinuity in the leading edge creates a vortex, that acts as an aerodynamic wing fence [4] [6].
There is a catch however: Experiments showed that wing designs that are spin-resistant, generally have poor spin-recovery properties once they do enter a spin [4]. Nevertheless, it is considered to be a safer option, as the level of safety is improved by making it very hard for the pilot to inadvertently enter a spin. The certification requirements for spin-resistance in 1991 were essentially a set of “abusive maneuvers” to verify that the aircraft remains controllable. Note: These requirements did not cover all possible combinations, especially not high-power and high pitch-rate maneuvers [4].
Typical representatives of the 1991 spin-resistant option are the Cirrus SR-20/22 (see Figure 5) or the Textron LC40-550FG. Neither of the two designs met all spin-resistance requirements and both were finally approved under an Equivalent Level of Safety (ELOS) finding [5].
CS23 post-amdt 5 / FAR23 post-amdt 64 (today)
The latest requirements are an evolution of the 1991 FAA spin-resistant option, aiming to create a more holistic approach. Credit will be granted for stall warning devices, departure resistance and further safety enhancing equipment [5]. This is important, as more and more general aviation aircraft are equipped with electronic envelope protection systems, such as the Electronic Stability and Protection (ESP) function provided by the latest Garmin autopilots [5]. Figure 6 summarizes the new certification scheme: The requirements for “spin-approved” aircraft remain essentially unchanged, departure characteristics are introduced for “non spin-approved” designs (ASTM F3180). It is worth noting, that there is a new requirement for light twins, not to “hazardously depart” controlled flight after a critical loss of thrust [5]. This is obviously an attempt to reduce the number of “roll-over” accidents in that category of aircraft.
Departure aversion score – the new concept
Depending on the aircraft category, now called “level”, the departure aversion score required is more or less stringent. The manufacturers are given the option to achieve that score using all areas, such as stall warning, departure resistance and safety enhancing features [5]. The total required score cannot be achieved by one area alone (e.g. stall warning only) and thus the manufacturers are forced to employ multiple layers of safety [5]. Figure 7 below presents the context of departure aversion score.
Depending on the level of the aircraft, the manufacturers have the choice between 3 alternatives to demonstrate the departure resistance of the aircraft, as depicted in Figure 8. Alternative 1 and 3 represent basically the spin-resistant maneuvers of 1991 and apply a pass/fail-criteria. Alternative 2 allows cumulative scoring, depending on the outcome of each maneuver and thus provides a more accurate representation of the aircraft capabilities, while also being more cost-effective [5].
Back to the future…spin-proof, the “ultimate” solution?
As early as 1937, aircraft designers came up with ideas to make an airplanes “non-spinnable”. The Engineering and Research Corporation (ERCO), located in Maryland, promoted the ERCO Ercoupe (later Aircoupe) aircraft as “incapable of spinning” [8]. The design was rather unusual, as the aircraft had no rudder pedals and was controlled by a simple yoke. There was an aileron-rudder linkage for turn coordination and a nose-wheel steering controlled by that same yoke [8]. Additionally, the elevator authority was limited, to prevent reaching high AOA values.
While this design solution was certainly generally very safe, it had significant drawbacks, such as its limited ability to adapt for different CG positions. Additionally, it shall be noted that gusts may still induce high AOA values, even beyond those achievable by the primary flight controls and thus possibly never investigated [1]. That is to say, the aircraft might still spin, even if this is not achievable by pilot inputs only. Later versions were equipped with rudder pedals, as the p-factor of more powerful engines needed to be accounted for.
It remains to be seen, if the new certification requirements will lead to safer designs and ultimately to less accidents. Initial signs are certainly very promising and it is good to see, that the new certification paradigm allows for flexible innovation.
Revision/20201202
References
[1] Stinton, Design of the aeroplane, Blackwell, 1983
[2] Hoff, The Aeroplane Spin Motion and an Investigation into Factors Affecting The Aeroplane Spin, PhD Thesis, Brunel University London, 2014
[3] CASA, AC 61-16v1.0, Spin avoidance and stall recovery training, CASA Australia, 2020
[4] EASA, Safety Aspects of Light Aircraft Spin Resistance Concept, Final Report EASA_REP_RESEA_2008_3, 2008
[5] Borer, Development of a New Departure Aversion Standard for Light Aircraft, AIAA, 2017
[6] Kroeger & Feistel, Reduction of Stall-Spin Entry tendences through wing aerodynamic design, Society of Automotive Engineers & NASA, 1976
[7] https://i.stack.imgur.com/F4Nal.jpg, 29/11/2020
[8] US centennial of flight commission, ERCO ercoupe, https://web.archive.org/web/20051230193424/http://www.centennialofflight.gov/essay/GENERAL_AVIATION/ERCO/GA12.htm, 24/11/2020
[9] GORDON W. HUBBARD COLLECTION No. 3545. Erco XPQ-13 (41-25196 c/n 110) USAAF Aeroplane Photo Supply (APS) Photo No. 4473