First, I offer excellent references on helicopters: John Montgomery’s book "Sikorsky Helicopter Flight Theory for Pilots and Mechanics", originally issued in 1964. Get a copy if you can.

All helicopters have the same basic controls: a collective stick to enable power changes and vertical motion, a cyclic stick to enable turns and forward speed, and anti-torque (also known as rudder or tail rotor) pedals for anti-torque to the main rotor (conventional helicopters), turn coordination, and turns about a vertical axis while in a hover or performing a ground reference maneuver. More advanced aircraft may offer hydraulic controls, automatic flight control systems for stabilization and/or autopilot and/or trim, or fly-by-wire controls, and collective-yaw coupling. Regardless of what you are flying, the controls will be laid out in a similar fashion (with some slight exceptions, which will be discussed for general interest) and accomplish the same aircraft tasks.

As mentioned in the chapter on rotor systems, the rotor is controlled by three inputs from a swashplate. The primary job of the flight control system is to take the pilot’s control stick inputs (which should resemble what is desired of the aircraft) and transform them into something that is meaningful to the rotor system via these three inputs.

Helicopters are usually flown from the right seat, although some aircraft (MD-500 and Enstrom come immediately to mind) are flown from the left seat. The reason for this is lost in legend, although I think the best reason is that the pilot can fly with the right hand on the cyclic control (by far more sensitive) while attending to the instrument stack with the left. The collective control is typically more stable, and the aircraft can tolerate hands off here for a while. Some early helicopters (the Sikorsky S-51, for instance) had a single collective stick, placed between the two pilots. The trainee or co-pilot actually had to learn to fly the cyclic with the left hand.

The cyclic stick provides control for the pitch and roll attitudes of the aircraft, allowing fore-aft and turning motion. The roll channel also allows sideward translation of the aircraft while in a hover. The fore-aft channel is often referred to as ‘longitudinal’ and the left-right channel is often referred to as ‘lateral’.

The stick is usually mounted on the floor in front of the pilot, although the Robinson R-22 has a single, centrally mounted stick with a teetering ‘Y’ shaped bar to place the grip in one or the other pilots’ lap. The grip usually contains some switches to place important or commonly used functions at the pilot’s hand. These functions commonly include radio and intercom ‘press to talk’ switches, trim, and external cargo release. More advanced aircraft include over-water flotation firing, autopilot controls, radio frequency selection, and weapons selection and firing. Because it controls rotor inputs that cause blade changes with each revolution of the rotor, this control is called ‘cyclic’.

From the cockpit, the control inputs move to the ‘mixer’ through a series of linkages, where the collective input is added and the controls are ‘sorted out’ for input to the main rotor. Some aircraft include hydraulic boost actuators here to isolate any feedback forces from the rotor. This is also a convenient place for trim, stabilization, and autopilot inputs.

Remember the three inputs to the swashplate that we talked about in the Rotors section? Let’s name those inputs, related to where they are located. For practical purposes, there are four possible locations of these inputs: forward, rearward, left, and right. Although we can have any combination of three of these locations, let’s use forward, rearward, and right. The stationary scissors in this example occupies the site of the left input.

Suppose we want our helicopter to move forward. We move the cyclic stick forward; this motion goes through the mixer, and results in an upward movement of the rear input and a downward movement of the forward input to the swashplate. These inputs are approximately equal. Imagine that only rearward input moved with a forward cyclic input. Since there is no motion to the right input, the swashplate will most likely tilt around a line between it and the forward input (time to break out your three fingers and CD-ROM jewel box again). The aircraft would move in a forward and right direction because the maximum lift on the rotor disk will be applied opposite that position. Coordinating the motions of the opposite inputs to the swashplate is just one function of the mixer.

The collective control gets its name because it controls the pitch of all the rotor blades at the same time, regardless of where the blade is as the rotor head turns. Input is from the collective stick, on the left side of the pilot. The collective stick also often has a grip which contains mission related switches, but this is not important to our discussion.

On aircraft that you are likely to fly (and certainly to train in, particularly piston engine helicopters), the collective stick will include a throttle control. This is usually a twist type grip. As collective pitch is increased, there will be an increased demand on the engine, and this twist grip allows the pilot to simultaneously apply this input as he applies collective. With practice, this action becomes almost transparent to the pilot.

Depending on the aircraft you fly, some anticipation of throttle input may be required to keep the rotor turning at the proper speed. The Robinson R-22, for example, also includes a throttle correlator that changes the throttle directly with collective input, decreasing pilot workload. Some regular attention to rotor speed will still be required, but it is less of a task. Robinson has recently added a governor, which decreases pilot workload in this area to almost nothing.

Remember the discussion above about how the forward and rear inputs to the swashplate move in opposite directions with a fore-aft cyclic input? When collective is applied, the mixer is responsible for making these two inputs move in the same direction. Also, it has to make the right input (or lateral input) move in the same direction as well.

Adjustment of the rotor for collective is somewhat interesting. The maximum allowable collective pitch of the blades should be somewhere near that which allows the engine to keep the rotor turning at its normal speed. The minimum allowable collective pitch should be such that rotor speed will be in the allowable range during an autorotation. Some aircraft are set so that rotor speed will not go beyond the high end when the collective is fully down, and some are set so that some application of upward collective will be required to check rotor speed. Regardless, a qualified mechanic in accordance with the applicable maintenance manual should make this adjustment for the aircraft.

Just to clear the air and avoid any confusion, the lifting ability of a helicopter is adjusted by changing the pitch of the rotor blades with the collective stick, not by increasing the rotor speed.

The pedals are used for anti-torque control, turn coordination, and turns about a vertical axis while in a hover. Operation and design is identical to that in an airplane: there are two pedals on the floor, and as one is pushed forward, the other moves rearward. This pilot input is typically transmitted directly to the tail rotor, with an input in one direction causing an increase in tail rotor pitch, and input in the other direction decreasing tail rotor pitch.

Turn coordination in a helicopter is almost automatic, so as you enter or exit a turn there will be minimal adjustment of the pedals. Application of collective control is another matter. As you increase collective pitch, engine and rotor torque will also increase, so some pedal input will be required. On western helicopters (where the rotor turns counter-clockwise when viewed from above), left pedal will be required with increased collective input.

We’ve spoken a little about the mysterious device called ‘the mixer’. This is usually a collection of mechanical linkages that take lateral, longitudinal, and collective inputs from the cockpit and transform them to the three inputs to the swashplate and the rotor system.

As a simple example, consider the V-tail Bonanza. Here, the normal two elevators and single rudder are contained in two control surfaces arranged at 30 to 45 degrees to the horizontal plane. To accomplish control in two axes of the aircraft (up-down and left-right), some control mixing is required. This is performed by a mechanism that is essentially the cousin of the helicopter mixer, only much simpler. To create the effect of lift (elevators), both surfaces will deflect up. To create the effect of a right turn, both surfaces will turn right (however one will be ‘up’ and the other ‘down’). So what will happen in a right climbing turn? Let’s apply the control inputs one at a time. For the climb, both surfaces will move up, but when the turn input is applied, they will move in the direction of the turn, one up and the other down. One surface will have greater lifting force, and the other greater turning force. The net total of lifting force will be the same, as will the net total of turning force, just split between two surfaces instead of three.

This situation can be used to examine helicopter blades during a forward climb. The forward blade will have less pitch than the rear blade, but when collective is applied, they will both increase in pitch (as will the blades in other positions). All pilot inputs to the main rotor head will go through the mixer.

It is also possible to apply anti-torque input to the tail rotor as a function of collective input in the mixer, but this is typically encountered only on much larger helicopters.

Just to complicate matters a bit, the Sikorsky Black Hawk helicopter has a canted tail rotor. That is, the tail rotor does not rotate in a pure vertical plane, but rather at a slight angle to it. This provides some vertical lift, which enhances the CG range capability of the aircraft, but also provides coupling between the tail rotor and the pitch channel. Considering that this aircraft also provides collective to yaw (tail rotor) coupling, a collective input will also affect aircraft pitch. These inputs are ‘de-coupled’ in the mixer. I guess you can see that the mechanics of all this gets real messy real fast, so mixers are usually designed by engineers with clear, ordered minds.

These are often referred to as AFCS, as opposed to MFCS (mechanical) or PFCS (primary). Modern AFCS are digital computers, electrical actuators, and sensors that augment the mechanical system to accomplish trim, autopilot, and/or stability enhancement. Because pilot stick inputs somewhat mimic desired aircraft response, it is convenient to place AFCS inputs between the control sticks and the mixer. The input is summed with the pilot input, and can respond much faster than the pilot ever can. These systems typically also have a lower inherent reliability than the mechanical system, so they have limited authority of maybe 10%.

The RAH-66 Comanche, currently under flight test by Boeing and Sikorsky, has a full authority AFCS system. This offers great weight savings potential because there is no need for separate AFCS actuators, boost actuators, a mechanical mixer, or control linkages. The system is fully electronic, from the control sticks (similar to the joysticks that you may use to fly X-Plane) to the inputs to the servos below the swashplate. The system also relies on inputs from an air data computer, which measures pressure, air speed, and air temperature, and also uses rate and acceleration damping as required to fully tailor the controls to the situation at hand. Among other enhancements are different flight control ‘laws’ for hover and cruise flight, as well as when the aircraft is on the ground or touching down.

Another feature of the Comanche AFCS is that the three main rotor servos are placed 120 degrees apart. In a mechanical system, the inputs are placed 90 degrees apart because the effort to include the resulting pitch and roll mixing would be enormous. The computer can readily handle the sine and cosine functions required for determining correct servo motion based on their locations under the swashplate. Remember, ultimately it is the motion and position of the swashplate that determines the pitch angles of the blades as they rotate.