Teaching Autorotations

by Philip Greenspun, ATP, CFII, March 2014

Site Home : Flying : One Article

We recently interviewed a fresh CFI/CFII graduate of an excellent Part 141 school. We asked how he would teach a student to perform an autorotation. We started by asking how long, given a starting altitude of 500' above the ground, the maneuver would take, from the time of initiation (or simulated engine failure) until the time when the skids were on the ground. He hadn't thought about this and could not come up with an answer (considering that the standard rate of descent in an auto is about 1500 feet per minute, 20-30 seconds seems like a good estimate and, plainly, is insufficient time for an extended sequence of actions).

After supplying him with the 20-30 second time period we asked "How many times should the student look inside at the gauges versus outside at the horizon during this maneuver?" The suggested answer was "More than ten" and that was a consequence of the fact that this CFI planned to teach students to conduct a scan: (1) outside, (2) yaw strings, (3) rotor RPM, (4) airspeed. As far as flight control inputs were concerned, the CFI stressed lowering the collective.

His ideas for teaching turned out to be completely opposite from mine. From my experience, autorotations are an attitude-flying maneuver like most others in any aircraft. If a reasonable attitude is maintained, everything else falls into place fairly easily. If the nose is allowed to drop upon entry, on the other hand, it is rare that the autorotation can be saved by the average beginner. I tell students that they should be looking outside at the attitude indicator (i.e., the horizon) and use cyclic and pedal to maintain whatever attitude they had prior to the engine failure or to lowering the collective. The collective should be lowered to the floor and, after a few seconds, cracked slightly up (about 1/2 inch on an R44). Only then would I suggest that the student take a quick look inside at the RPM and airspeed. If the airspeed is between 60 and 70 knots, I tell the student to go back outside and try to hold that attitude. If the low RPM warning horn sounds, the pilot should push the collective halfway down to the floor from wherever it is. I suggest that the student look back inside only once more during the maneuver, e.g., shortly before beginning the flare. In my world, the pilot is looking outside 95 percent of the time during an auto. In the new CFI's world, the pilot was looking outside 25 percent of the time (since the attitude was only 1 item in a 4-item scan). In my world the cyclic input had equal status to any collective lowering. In the new CFI's world it was all about the collective.

Who's right? The FAA's Rotorcraft Flying Handbook starts the autorotation technique section with the following:

From level flight at the manufacturer's recommended airspeed, between 500 to 700 feet AGL, and heading into the wind, smoothly, but firmly lower the collective pitch control to the full down position, maintaining r.p.m. in the green arc with throttle. Coordinate the collective movement with proper antitorque pedal for trim, and apply aft cyclic control to maintain proper airspeed. Once the collective is fully lowered, decrease throttle to ensure a clean split of the needles. After splitting the needles, readjust the throttle to keep engine r.p.m. above normal idling speed, but not high enough to cause rejoining of the needles. The manufacturer often recommends the proper r.p.m.
It then goes on to suggest "Continually cross-check attitude, trim, rotor r.p.m., and airspeed." So the FAA concurs with the young CFI that the pilot should conduct a scan in which attitude is just one of four items. It also stresses the primacy of the collective pitch input. I.e., the young CFI and his 141 school are right.

At Heli-Expo 2014, however, a two-hour seminar on autorotations revealed that I am not the only heretic. Experts from the NTSB, FAA, Airbus/Eurocopter, Robinson, and Bell all agreed that the Rotorcraft Flying Handbook was wrong and that cyclic pitch input was at least as important in the moments after an engine failure as lowering the collective. Two accidents were discussed, with an emphasis on one in which an Air Methods medevac A Star flamed out from fuel exhaustion in Mosby, Missouri in August 2011. The pilot should have been expecting an engine failure due to the "low fuel" caution light having been illuminated for some time (sadly the helicopter crashed just 1 mile from the airport of intended refueling). Yet somehow he did not successfully enter an autorotation following the engine quitting and, within 5 seconds, everyone on board was dead. The second crash was of a Marine Corps UH-1 (Huey) that suffered a drive shaft failure and, despite the generally amazing training of military helicopter pilots, did not manage to enter an auto.

Following the A Star crash, people from the NTSB and Eurocopter (now Airbus) worked in a simulator experimenting with flight control inputs following an engine flameout. It turned out that, starting from the rather high cruise speed of the Mosby crash, aft cyclic needed to be applied within 2 seconds or the aircraft would be unrecoverable.

The chief instructor for Airbus/Eurocopter's U.S. training center warmed my little CFI heart by saying that he tells people not to look inside during autos and that, after a quick verification of airspeed, pilots should look outside for the rest of the maneuver.


To prepare students for the possibility of a real-world engine failure, consider teaching that preserving aircraft attitude with cyclic and pedal is just as important as lowering collective and, in fact, may be more urgent. Also consider following the Airbus/Eurocopter school's approach of deemphasizing scanning gauges. If the aircraft attitude is held constant the autorotation will likely go pretty well.

About the Author

The author is a senior flight instructor for East Coast Aero Club's helicopter program.


Copyright 2014 Philip Greenspun.