It is common knowledge that a helicopter by itself does not wish to fly. According to Harry Reasoner, “it is maintained in the air by a variety of forces and controls working in opposition to each other, and if there is any disturbance in this delicate balance the helicopter stops flying; immediately and disastrously.” He goes on to add, “there is no such thing as a gliding helicopter“.

Actually there is. A gliding helicopter is in a condition of flight known as “autorotation”.

If the helicopter has to stay up in the air and achieve the miracles of vertical lift, it has to pretty much beat the air into submission. And for this, a constant supply of engine power commensurate with the flight condition is a must. If the helicopter is equipped with two or more engines and one of them fails, chances are, the other engine(s) will step up or be utilised to recover to safe one engine inoperative (OEI) flight. On a single-engined helicopter if the engine cuts, the rotor blades spinning at their nominal revolutions per minute (RPM) will instantly respond to physics and start feeding on its own kinetic energy.

An energy-exchange process will now be set into motion as the helicopter starts descending steeply through the air, trading potential energy for kinetic energy. The resultant upward flow of air through the rotors now drives the rotor blades into a state called ‘autorotation’. For this to happen, the pilot must intervene on the power lever, also known as “collective”, and reduce blade pitch (close to flat pitch) such that the descent airflow maintains main rotor speed within prescribed limits. A steep glide is established that can, with adequate training, be turned into a successful engine off landing (EOL).

There is thus such a thing as “gliding” in a helicopter, much as Harry Reasoner would like to believe otherwise. It requires skillful handling and all single-engined helicopters MUST demonstrate successful, repeatable EOLs as part of the certification process. I have narrowed down the focus of this article to autorotation on single-engined conventional helicopters. However, the fundamentals can be applied to conventional rotorcraft without much distortion in any EOL situation.

Why autorotate?

A helicopter may be required to autorotate for a number of reasons, primary among them being any situation where the power source driving the rotors is lost or deliberately shut down to respond to an emergency. Loss of tail rotor (TR), TR drive or failure of critical components in the anti-torque system may also force the pilot to remove torque by entering autorotation and shutting down the engine. Helicopters have a freewheel mechanism to decouple the engine from the rotors when the drive (engine RPM) lags behind the ‘driven’ (rotor RPM) thus allowing the rotor to “autorotate”.

In some cases, the pilot may choose to enter autorotation even with the powerplant working. Such a manoeuvre known as ‘power-on autorotation’ may be used to set up for an urgent force landing, steep descent, or for training purpose. In this case, the freewheeling transmission and engine power turbine section is re-coupled as the pilot raises the collective lever for landing.

Designing the helicopter such that a safe transition to autorotation / EOL is achievable within the capacity of an average pilot is thus an essential prerequisite for certification. Proficiency in autorotation is the ultimate ace up the pilot’s sleeve and comes early in basic training at any rotary flight school. Particularly in the case of light singles, key attributes of a healthy pilot reaction time to respond to engine failure, handling qualities in autorotation, audio and visual cues of power loss, glide ratio, rotor inertia and touchdown speed (as close to zero as possible) should be considered “essential”. For those interested in knowing the nature of flight tests conducted during the developmental phase of a new (conventional) rotorcraft, I recommend this excellent article on ‘Autorotation Flight Tests‘ by fellow professional, experimental test pilot Gokhan Virlan (via LinkedIn).

Phases of autorotation

Broadly, autorotation manoeuvre comprises three key phases, viz.

The entry into autorotation where the pilot is required to lower the collective lever to a setting that allows the helicopter to descend and reverse the airflow through the rotors such that the blades “autorotate”. This must be achieved without delay to ensure that rotor RPM does not droop to levels from which recovery will be impossible. Factors such as pilot reaction time, rotor inertia, condition of flight, delay time, etc, come into play here.

Steady state autorotation where the helicopter has fully stabilised in autorotative flight at a combination of forward and downward velocity that maintains the RPM within permissible range. The pilot manoeuvres the helicopter in glide to position for a landing which ensures maximum survivability. The pilot will have to plan for factors such as available height, terrain and landing field, speed for maximum endurance or range, wind direction and speed, etc during this phase.

The flare and recovery manouevre where the pilot “flares” (raises the attitude) to a point such that the rate of descent and forward airspeed are sacrificed to store kinetic energy into the rotors for the final landing manouevre. This is followed by a “check” (optional in some types), “level” and raising of collective to milk the kinetic energy stored in the auto-rotating rotors for cushioning the landing impact. If everything goes to plan, almost zero-speed touchdown within the limiting sink rate of landing gear can be achieved on a certified single-engine helicopter.

Design, predict, test, validate

Much of the background calculations to predict autorotative behaviour of the helicopter happens during preliminary design, concurrent with design imperatives for performance and handling qualities. Many competing demands and requirements will vie for attention. It would be unreasonable to expect the designer to design for optimum autorotative characteristics by giving up some or the other key deliverable. The final design is always the best achievable compromise of safety and capability. But there are boundaries between ‘undesirable’ and ‘unacceptable’ and these boundaries must be respected. It is essential that designers and test crew remain vigilant to ensure undesirable or unacceptable autorotative characteristics do not creep-in early into the design, or are glossed over only to be detected at a later stage.

For example, a plot of normalised rotor rate of descent versus normalised forward speed in stabilised autorotation should inform the designer of a practical boundary for autorotation without hitting the vortex ring zone for the rotor type. If the glide angle is very steep and/or rate of descent in stable auto is very high, the time available to reduce speed in the landing flare without hitting lower limit of rotor RPM will be too short. A landing with excessive roll-on speed (or more adverse outcomes) may in all likelihood follow. Such characteristics would invariably show up in well-executed flight tests and post-flight analyses. Needless to say, shortcomings in these critical areas, if any, must be addressed early in the program and not passed on to customers to discover through costly accidents or loss of life.

Personal experience

The EOL phase that precedes engine failure tests in the test pilot school curriculum is a mandatory check point to earn the coveted graduate patch. In Indian armed forces, EOLs are only permitted for training purposes during the flight test course (FTC). Like all rotary student TPs, I have personally practiced EOLs on Chetak during my FTC. As a rotary wing test flying instructor (TFI), I have also imparted EOL training and cleared student test pilots to undertake solo EOLs. I still recall vividly the run-on EOLs I completed as a student TP and the zero speed EOL touchdowns demonstrated by experienced STFIs (then) Wg Cdr NS Krishna, Gp Capt YS Rajora, Sqn Ldr Ajay Shukla et al. The confidence in the Chetak’s autorotative characteristics was reinforced through repeated EOLs, including one where Gp Capt YS Rajora cut the fuel lever from a free air hover at 1200 feet. The benign handling qualities of the Chetak, rotor inertia, control response and recurrent training from innumerable practice power-on autorotations came together in a blink. The high-risk transition from powered flight to a one-way-ticket to the parallel taxi track at HAL airport’s Runway 09 was an exercise we as test crew could confidently pass on to the average line pilot as “satisfactory and acceptable”.

Sim training for light twins; on-type for light singles

I have also conducted thousands of autorotations in the simulator, both as a student and instructor, starting from benign to the most complex (on light to medium twins). Juggling with multiple knowns and unknowns, wide spectrum of pilot experience, weight, altitude, temperature, nature of terrain etc., autorotations can be practised in a safe manner in the simulator. Coming from experience, I can understand why the jury is still out on “what constitutes realistic autorotation training“. Most twin-engine helicopters prohibit full touchdown autorotations in the Rotorcraft Flight Manual. However, for single-engine helicopter pilots, autorotation should come as second nature — possible only if such manoeuvres are frequently practised on type, ending with power-on recovery for trainees and full-down autos for trainers and test pilots engaged in the certification process. The FAA Practical Test Standards do not require the applicant to demonstrate proficiency in full touchdown autorotations during the practical test for a private, commercial or ATP certificate. However, the Flight Instructor Practical Test Standards require a CFI applicant to demonstrate proficiency in full touchdown autorotations. Read this linked US Helicopter Safety Team bulletin on autorotation training to know more.

Risk assessment in autorotation training 

Of course, such manoeuvres come with their own share of risks and are best practiced in a simulator (if one exists) or in power-on mode where at the last stage of the recovery the engine couples with the transmission as the pilot raises the collective (power) lever. As per Federal Aviation Administration document FAA P-8740-71 on ‘Planning Autorotations‘, “autorotation training is used to instill habit patterns in a student/pilot, which will, in an actual emergency, become an automatic response”. The document lays down broad guidelines for instructors conducting practice autorotations while drawing their attention to the possible mistakes and mitigating techniques thereof. Habit patterns cannot be built on a shaky foundation of inadequate or insufficient recurrent training, neither can it be a replacement for sound design.

The 2011 Compendium Report of US Joint Helicopter Safety Assessment Team (U.S.-JHSAT) analysed 523 U.S. registered helicopter accidents that occurred in 2000, 2001 and 2006. Per their findings, dual instruction/training accounted for 13.8% of all accidents under the accident demography of ‘Accidents by Activity’. Under the classification of ‘Accidents by Occurence’, autorotations, both practice and emergency, accounted for some of the most frequently observed accidents (32%).

The industry as such is divided on the issue of training for ‘practice autorotation with power-on recovery versus full-down, power-off autorotation’ given the risk-benefit payoffs of such exercise, reliability of modern aero engines, high-fidelity simulators etc (read a short research paper here). But no two helicopter pilots will possibly disagree on the need for light single-engine helicopters to have acceptable characteristics that allow full down autorotations with maximum survivability in the event of an actual engine failure or shut down. Except for a limited height-velocity envelope from where safe landing after an engine failure is precluded by the laws of physics, every single-engine helicopter should be designed to enable a survivable engine-off landing via autorotation with an average pilot at the controls. It is non-negotiable. How to realistically train for such an event can be debated but not the need for acceptable autorotative characteristics leading to a successful touchdown. Reaction time, rate of descent (versus forward speed), rotor inertia, rate of RPM decay, time in flare, min/max rotor RPM, landing speed obtainable, etc. are all crucial characteristics that must find itself in the “acceptable” range for handing over the aircraft to the line.

Brief case studies in autorotation landings

• On Mar 11, 2018, an AS350 B2 operated by FlyNYON for a “doors off” photo tour suddenly experienced engine failure after a passenger’s NYONAir-provided harness tether snagged with the fuel shut-off lever and rolled back the engine. The pilot reacted with alacrity, successfully autorotated and carried out a copybook zero-speed splashdown in New York City’s East River. Five passengers drowned as they could not free themselves of their supplemental harnesses from the overturned helicopter. The successful autorotation and ditching saved the pax but the impossible harnesses killed them, indicating the nature of variables in the environment that helicopter pilots have to contend with. At least one of the FlyNYON pilots had expressed strong reservations about the harness which the management ignored, leading to the needless loss of lives (NTSB report here).
• Year 2005. An Indian Navy Chetak (Alouette III) helicopter IN 483 on a routine night flying sortie in naval air station Visakhapatnam’s local flying area experienced a sudden engine failure. Against dire odds, the crew carried out a successful night autorotative landing to the unprepared ground 2000 feet below. Skillful handling coupled with the inherently benign autorotative characteristics of Chetak saved both lives and the helicopter. Landing off an autorotation after engine failure at night can be considered as the acme of all emergencies on a light single — possible only with the hardwired instincts of well-trained crew, on a helicopter with excellent autorotative characteristics.
• Another IN Chetak flying over the sea off Vizag in 2013 was however not as lucky. Following an engine seizure over sea, the helicopter was forced to ditch with inadequate height. The inauspicious combination of low height, hostile terrain below (sea) and high touchdown speed sent the helicopter cartwheeling on the sea surface, resulting in two fatalities.
• Closer home to the epicentre of all helicopter flight testing in India, a light-single Schweizer 300C (VT-HAV) from Hindustan Aeronautics Limited (HAL)’s Rotary Wing Academy suffered engine failure while on a training flight in Apr 2012. With hardly an open patch anywhere in sight, the highly experienced instructor pulled off an incredible autorotative landing on a rooftop. A few feet or knots more and the helicopter with crew would have tipped off the rooftop to certain death.

Two more recent examples of accidents during practice autorotation should serve to inform the debate around power-on versus power-off practice autorotations (obtained from the website of Federal Bureau of Aircraft Accident Investigation, Braunschweig, Germany). Both were training flights with full-down autorotations that led to serious aircraft damage not attributed to design of the helicopter or its autorotative characteristics in the accident report.

• On Mar 27, 2024, during a practice full-down autorotation, the main rotor of an AB 206B smashed through the tail boom of the helicopter. The aircraft was severely damaged but the crew escaped without injuries since the touchdown was over flat ground of the grass runway 07 at Landshut airfield (EDML, Germany). The pilots had planned the auto with engine to idle, but due to mismanagement of rotor RPM and delayed power recovery, the helicopter down touched hard and pitched forward, leading to main rotor strike on the tail boom (report in German here).
• Another training flight on Mar 6, 2024 on a Cabri G2 helicopter (light 2-seater with skid landing gear) led to a similar occurrence of unplanned hard ground contact with low rotor speed. The main rotor smashed through the tail boom. Here again, the crew of two escaped without injury as the landing area was a flat open field. The investigation concluded that the accident was a result of “non-optimal coordination between the instructor and the pilot” (report in German here).

Operational limitations & training opportunities

Helicopter pilots from this writer’s generation grew up with hardly any restriction on practice autorotations. The fulsome training in basic and advanced autorotations received at Helicopter Training School and ample opportunities to hone the skills in frontline Chetak/Cheetah units ensured high proficiency and survivability given the excellent autorotative characteristics of the Alouette and Lama. Over time, owing to an ageing fleet and repeated engine/transmission troubles (pointing back in part to repeated cyclical stresses of full flare power-on autorotations), restrictions were placed by the manufacturer HAL on the number and types of practice autos one could perform in a sortie.

While advent of simulators allowed recurrent training to continue unhindered for larger, twin-engine helicopters, pilots from the humble Chetak/Cheetah fleet — with no access to simulators — had to make do with lesser and lesser opportunities for practice autos. However, with ALH (and a Level D simulator) becoming the mainstay of helicopter fleet in the triservices, such training shortfalls on the Chetak/Cheetah fleet did not lead to a directly relatable spike in unfortunate consequences.

LUH induction & the road ahead

This is set to change with the proposed induction of HAL’s single-engine Light Utility Helicopter (LUH). The LUH in large numbers will form the mainstay of rotary fleet going forward. Most IAF/army pilots will cut their teeth on this machine for the foreseeable future. Excellent autorotative characteristics and unrestricted training opportunities for practice autorotations should be an essential prerequisite for accepting the LUH into military (or civil) service. At the moment, if anything, there is evidence to the contrary insofar as the LUH is concerned.

More on this in a subsequent report soon to follow after curtains fall on Aero India 25.

In the endgame of autorotation, there will invariably be the luck factor or “hand of God” behind every successful EOL. But such unquantifiables are not relevant to the discussion when designing for safety and survivability. The designer must work through very complex often contradictory requirements to provide excellent autorotative characteristics. The machine is then put through the paces by test crew in free air, full touchdown EOLs, progressively leading to ‘lever delay tests’ and experimental validation of the height-velocity diagram. The culmination of such trials ensures that the helicopter can be put down safely by an average pilot should the (only) engine ever fail (and fail it does, as the users and OEM know only too well). There really is no room for debate for this on manned light-singles where this is potentially a life and death matter for those onboard.

The irony should not be lost on anyone, let alone Union Ministers looking for photo opportunities. When you fly in a fighter, you are briefed comprehensively on the “ejection drill”. When you fly in a single-engine helicopter, you are at the mercy of the helicopter’s autorotation characteristics and crashworthiness. Can one replace the other? Do you know the difference? Were you briefed?

Or were you just winging it like an uninformed fanboy?

************

©KP Sanjeev Kumar, 2025. All rights reserved. Views are personal. I can be reached at realkaypius@gmail.com or on X (formerly Twitter) @realkaypius. Feedback is most welcome.

Disclaimer: If you are a flight crew, please consult national regulations, the Rotorcraft Flight Manual (RFM), Pilot’s Operating Handbook (POH) or Aircraft Flight Manual and your company’s Operations Manual as applicable to the type you fly.