The Zen of Hovering Helicopters
By Patrick Bryant Aug 22, 2005, 18:14

The fact that anyone can ever learn to hover a helicopter is said to be a testament to human adaptability - in fact, at an intellectual level, I think it's a skill that's outright impossible to learn. I suppose that's one of the reasons I took it up.

I remember when I first started my helicopter training; my instructor asked how many hours I had as a pilot in fixed-wing aircraft (as helicopter pilots like to call "airplanes"). When I told him, he looked concerned and said: "That's a lot, but if you work really hard, I still might be able to teach you to fly helicopters." Airplanes are fairly linear things to operate, and are as stable as a broad-beamed boat - it takes a lot to make one capsize. Testament to that are the occasional incidents where an airplane will take off without its pilot and then drone around for hours before its tanks run dry.

When I started to study the aerodynamics of helicopters, I quickly realized that the blasted contraptions can't fly at all without constant remedial action. They are dynamically and statically unstable in all axes, the control movements in response to any command or external perturbation are non-linear (move a control X amount, and Y happens - move it 2X amount and 10Y happens), and every control input had an interactive affect on every other parameter requiring additional control input! In my mind's eye, the challenge seemed to be a lot like the old saying about herding cats. For example: Take just trying to hover in one place at a constant altitude. For simplicity we'll pretend that there is no wind blowing across the ground (the example would otherwise be impossibly complex). You adjust the main rotor so it suspends the helicopter at a constant height above the ground. Fine so far.

But the engine creates torque that wants to spin the helicopter around in the direction opposite that of the main rotor's rotation (action-reaction). So you depress the tail rotor pedal just enough to blow just enough air sideways to counterbalance the torque's rotation. Good. But now you are blowing air to one side from the tail rotor, and the helicopter starts to scoot sideways along the ground in the opposite direction (action-reaction, again). Also, increasing the pitch of the tail rotor so it bites air differently to move more or less air to the side - to counter the engine torque - puts more or less of a load on the engine, and the main rotor slows or speeds up just a bit, causing the helicopter to climb or descend. To counter the sideways movement from the air off the tail rotor, you tilt the main rotor just a bit in the opposite direction which moves the main rotor's "lift vector" away from vertical and causes the helicopter to descend, and you consequently have to increase the pitch of the main rotor and adjust the throttle to compensate for the main rotor speed changing due to your tail rotor pedal input - which changes the engine torque - forcing you to readjust the tail rotor pitch with more or less pedal, which again changes the load on the engine. You get the idea.

If you try to think your way through these problems, flying the thing becomes an endless and impossibly complex loop of effects and pilot reactions. Instead, you have to feel what the partnership of you and the machine are doing. In a sense, you have to meld with and become the machine. It's the only way to fly it with any degree of finesse. Then there are all those forces from all those spinning parts that are either non-existent or negligible to an airplane pilot: coriolis forces, gyroscopic precession, torque, transverse flow effects, and translational lift! All have to be correlated, anticipated and controlled first in the pilot's mind - and even when all pilot actions are perfectly performed - there are the gusty unexpected effects of wind and chaotic turbulent air flows all around the aircraft that seem designed to destroy any attempts at precession. The control inputs from the pilot are generally infinitesimal: the controls hardly move at all and mere pressures on the controls are usually all that is needed. So of course, if you put a neophyte airplane pilot like myself at the controls - one who is used to slamming the flight controls around in an airplanes, you end up with a helicopter performing wild "pilot-induced" oscillations, and generally flying like a madman is at the controls. What an unflyable machine! It's absolutely offensive the engineering mind!

There are also the nasty behaviors unique to helicopters that have to be intellectualized. The forces in a helicopter are so divergent that, left to its own accord, a helicopter would rapidly roll over on its back and screw itself into the ground. Land some helicopters a little hard and it initiates an under-damped regenerative oscillation in the main rotor that literally causes the helicopter to shake itself apart in a few seconds; usually leaving the occupants with massive contusions from being bounced around worse than going over Niagara Falls in a barrel. It's called "ground resonance," and the result is dynamic disassembly of the entire aircraft: a cascading process that the pilot has maybe 2 seconds to halt. When a helicopter flies, it imparts a rapid downward force to a column of air below it. Fail to accurately picture in your mind where that column of air is, and you can find yourself descending rapidly inside your own "dirty air." "Settling with power," it's called, but the name sounds much more benign than the actual effect: the experience is one of going down a drain in a whirlpool.

Hovering in a gusty wind is akin to balancing a broom by the end of its handle - while riding on a roller coaster. And if you try to hover while facing downwind, you are literally flying backward in the airmass while sticking the tail boom into the wind. The helicopter naturally wants to act like a weathervane and swing 180 degrees around. So to keep that from happening, the pilot has to perform a continual and delicate tap-dance on the rotor pedals. Given all that, I concluded that these things probably can't be flown by any but a few unique, trained-in-their-early-youth, steel-nerved specialists.

But I decided to humble myself a bit more anyway. My hovering. if you could call it that, looked terrible. I'd sit there and sweat gallons while I jerked the controls this way and that, always lagging the inputs that were needed and inducing wild oscillations in all three dimensions; while my instructor sat beside me with a calm smile that made me question his sanity. My 11 year old son, who was naturally oblivious to the difficulties, stood on the sidelines watching my practices and was pink from laughter whenever I returned and poured my limp body out of the cockpit. One morning I drove to the airport fully prepared for another humbling experience and further confirmation of the conventional wisdom that airplane pilots with fully-wired airplane-pilot brains just can't fly helicopters.

I lifted my tormentor into the air -- and discovered that neuronal servo loops had somehow become wired, the forces needed to hover correlated deep in some primitive level of my mind, and my hands and feet started moving autonomically in response to a three dimensional image that had formed in my brain. My hands and feet were moving of their own accord, steered by some unperceived mental process that interfaced what I wanted the helicopter to do with my control inputs, and correlating any deviations into effortless and progressively smaller corrections. I could hover! Coarsely at first, but in just a few days with no conscious effort at all. I sat there amazed at how my hands and feet would respond in all these non-linear ways to what I felt and saw, guided by something that was entirely beyond my conscious perception. I would hover there with the helicopter motionless, while my hands and feet would wriggle around doing things that were barely comprehensible even to me. Once I could hover, I started to discover the Zen of flying helicopters. Simply think "up," and the helicopter goes up. Think "yaw right," and the helicopter rotates as if by sheer will alone. Axons and synapses are truly wonderful things! I guess I still had some extra wire left to connect things in my airplane-pilot brain. I couldn't hover at 9.9 hours of utterly frustrating instruction, but I could hover at 10.0 - and I hadn't learned one wit more in the interval. Instead, the wires to the servo loops finally made their tentative connections in my brain. It really is a testament to human adaptability.