In the movie Blue Thunder, one of the plot devices used to build up the mystique of Roy Scheider's character is his claim to have looped a helicopter, which everyone else says is impossible. For years, I believed looping a helicopter must be impossible (movies are all true, aren't they?), but it's not. When I got involved in RC helicopters, I was amazed to see model helicopters doing loops, rolls, inverted flight, and all sorts of other "impossible" maneuvers.
For an example, see this video. If you've never seen something like this before, you'll think you're watching a special effects demo reel. Laws of physics are seemingly violated, but it's all real. Full-size helicopters are capable of these feats too, if they are rigged the right way. I suspect that when people report seeing aircraft doing 90-degree turns over Area 51, they're not seeing alien spacecraft: they're watching experimental helicopters. Helicopter aerobatics make the most sophisticated airplane aerobatics look boring.
Of course, I've wanted to do these maneuvers with my own RC helicopter. The first helicopter I bought, the Ikarus Piccolo, was incapable of such maneuvers. My current helicopter, the Hornet II, is capable. The difference is that the Hornet features collective pitch control. The thrust produced by a helicopter's rotor blades is dependent on two things: the speed at which the rotor is turning, known as the head speed and controlled by the throttle, and the angle of attack of the blades with the air, known as the blade pitch. These throttle and blade-pitch controls are known as the collective. As the head speed increases, the lift produced by the blades increases, just like an airplane's wing increases lift as its speed increases. Increasing the angle of attack also increases lift. The Piccolo has fixed-pitch rotor blades, meaning that the angle of attack never changes, so lift is determined solely by the throttle. However, the Hornet can control both the throttle and the pitch, providing for a greater degree of control over lift. Collective pitch also provides quicker response, as changing the pitch of blades can happen almost instantaneously, but speeding up or slowing down the head speed takes some time.
To fly inverted (upside down), a helicopter pilot adjusts the blade pitch so that it is opposite of normal blade pitch. This is called negative pitch. Most real-world helicopters provide some degree of negative pitch, which is needed for certain high-performance maneuvers and for landing in high winds. However, real-world helicopters do not provide enough negative pitch to hover or fly inverted. They could be rigged for more negative pitch, but manufacturers do not want the liability associated with certifying their helicopters for aerobatic flight. In contrast, RC helicopters with collective pitch usually are capable of inverted flight.
I first tried loops and rolls a few months ago, without much success. I was able to roll successfully a few times, but I also crashed a few times. My attempts to loop never worked out: I would end up with a messy flip and recovery, or I would end up walking across the field to pick up the pieces. The problem was that I didn't have enough cyclic power, which is what controls left-right rolling and forward-backward pitch. The helicopter just wouldn't turn over fast enough to get back to level flight before hitting the ground.
I read of two potential fixes to this problem. The first fix was a higher head speed. Higher head speeds increase the responsiveness of the helicopter, but also make it more jumpy and difficult to control. The second fix was addition of a Bell-Hiller mixer, which increases the power of the cyclic, but also makes the helicopter more jumpy and difficult to control. To smooth out the additional jumpiness, I added weighted rotor blades, which stabilize the helicopter by increasing the gyroscopic effect of the rotor.
I made these changes two months ago, but hadn't been able to try them out. We had a few hurricanes blow through, which really screwed up our weather for a few weeks. Then work ate up all my free time. Last week, I finally tried it out. I was rusty, so I just did some simple hovering and a few figure-eights. It was definitely a lot more responsive to the controls, so much so that I had difficulty keeping it under control. i did manage to keep it in the air for thirty minutes, and take it home intact.
During the week, a made a few adjustments to my transmitter. I made the cyclic control less sensitive and increased rudder sensitivity (I had noticed difficulty getting the heli to turn sharply). I also did a lot of practicing with the Reflex XTR simulator.
Today, I gave it another try. The air was calm when I packed up and got in my car, but by the time I arrived at the flying site, there was a 10 MPH wind. Undeterred, I lifted off and did some easy flying for a while. The wind was strong, but steady, so I figured I could try some crazy stuff. I flew into the wind, went into a climb, applied negative pitch as it went over the top of the loop, then leveled out. It wasn't pretty, but it was a loop. I did a few more. None of them were pretty, and the wind was too strong for me to know how they would look in steady wind, but it was a lot easier than I expected. To tempt fate, I also did a couple of rolls, and they were without incident.
So now I officially classify myself as an "intermediate" RC helicopter pilot. Whenever we have a calm day, I'll try some inverted hovering, which is what the Real Men do.
[UPDATE: Later in the day, I did my inverted hovering and a few more loops and rolls. Unfortunately, I crashed during a simple right-side-up hovering maneuver. I'm now waiting for $85 worth of replacement parts before I can try again.]