Auto Pilots – Different Types & How They Work!

Michael Kaufman

by Michael Kaufman, CFII

When learning to fly in the mid 1960s in a Champ, the word “autopilot” was something you read about in the aviation magazines of that time. My first personal experience was flying a Mooney, which had what was then referred to as “PC” (Positive Control). In fact, it was an autopilot that was on all of the time except when you pushed the button. If John F. Kennedy, Jr. would have had this device and would have let go of the button, he would probably not have experienced his fatal crash.

Today, autopilots are an important part of most aircraft that come off the production lines and may even be available on some light sport aircraft. When traveling by light aircraft, the autopilot is famous for allowing the pilot to be more relaxed when arriving at his destination. On the other hand, aside from the safety and fatigue factors, there is a drawback to autopilots. We as pilots have become so dependent on autopilots that our flying skills, especially instrument skills, have deteriorated to the point that we are an accident ready to happen should the autopilot fail in IMC conditions. An interesting question appeared on the application from Flight Safety where I will be doing some recurrent training next month. It asks: “how much non-autopilot IFR have you flown in the last six months.” More comments will be made on this in future issues of Midwest Flyer Magazine.

There is so much to say about autopilots, it would be an entire book. First, let us mention the level of sophistication on the autopilot.

The basic autopilot is a wings leveler similar to the Positive Control system I mentioned previously about the Mooney. The next ascending level is to add a tracking device allowing the autopilot to fly a VOR or, perhaps, now a GPS signal. We can add a heading bug and currently describe it as a typical “single-axis autopilot.” Most of the single-axis autopilots may be better than nothing at all, but they do a poor job of holding a course in a crosswind. If we add altitude hold, now we are getting somewhere. This can now be considered a “two-axis” autopilot.

In an IFR flight, ATC always gets excited about your altitude…because that is how they separate traffic. I have had many pilots tell me that they did not need altitude hold. They would just trim the aircraft, and it would hold the assigned altitude for them. Hmmm! I would sure like to find an airplane like theirs that was so stable they did not need any pilot input to hold altitude without an autopilot.

The last level of sophistication is a “yaw damper.” This controls the rudder and makes for a smooth flying aircraft. When we were learning to fly, remember the instructor telling us to use the rudder to keep the ball in the center? This is what the yaw damper does. Some aircraft are so stable with a two-axis autopilot that they don’t need a yaw damper – a Cessna 182, for example. On the other side is my V-tail Bonanza where my wife tells me to turn on the yaw damper if we encounter even the slightest bit of turbulence.

So, how well do yaw dampers work? To answer that question, a veteran airline passenger sitting near the back of the airplane would surely become nauseated in even the slightest amount of turbulence without the yaw damper. Most airliners have two yaw dampers in the event that the first one would fail. In light planes, my V-tail Bonanza with a yaw damper will ride smoother in rough air than a Cessna 182 without one. We have now covered the basics of autopilots in design without adding a lot of sophistication.

How do autopilots work?

Many of us may think of an autopilot being a super device that would make Albert Einstein scratch his head, but the concepts are relatively simple. Bill Hale, one of the flight instructors in the Beechcraft Pilot Proficiency Program with me, designed and built his own autopilot for a certified aircraft! Bill did the research and paperwork necessary to get it certified by the FAA in his Beechcraft Bonanza. I have to mention that Bill is a retired Hewlett Packard electrical engineer.

There are two basic concepts in autopilot design, which are used widely in our aircraft today. They are “ATTITUDE BASED” and “RATE BASED,” and I will describe both concepts with the pros and cons of each.

The attitude-based autopilot is located on the attitude indicator on your instrument panel. The attitude-based design is a smoother flying autopilot than is the rate-based design, but is more prone to failure. Research shows that an attitude gyro failure rate is about 2,600 hours, compared to 8,000 hours for the turn coordinator. That is one of the main resources in the rate-based system. Another negative factor with an attitude-based autopilot is that the attitude gyros are typically driven by a dry vacuum pump that has a failure rate of about 650 hours. King Avionics manufactures mainly attitude-based systems, and Century manufactures both types depending on model. I must say that in my exposure to autopilots, the King KFC-200 has shown to be a very reliable and maintenance-free unit. S-Tec manufactures mostly rate-based autopilots that we will look at briefly.

Many of the attitude-based autopilots – including almost all two and three-axis autopilots – are coupled to both the attitude and heading indicators. (Most simple wing-levelers, on the other hand, are coupled only to a turn coordinator and are rate-based systems.) An attitude indicator in an autopilot-equipped airplane has two pickup coils, one to sense the bank angle and the other senses the pitch angle. There is also a third pickup coil in the heading indicator or HSI, attached to the heading bug that is used to drive the autopilot’s heading-hold function.

The rate-based system, though not as smooth in my opinion as the attitude-based system, obtains its flight information from the turn coordinator and “accelerometers.” Accelerometers are small switches that are placed in different locations on the aircraft and open and close with movement or turbulence. Rate- based systems, using turn rate rather than bank angle, form the attitude indicator that will cause the aircraft to turn at the same rate regardless of aircraft speed. The attitude-based system using bank angle will cause the aircraft to turn faster as airspeed increases. In both instances of attitude and rate-based systems, the information from the sensors are fed into a computer that controls the servos which move the aircraft controls at the autopilot’s commands.

Different aircraft flight characteris-tics cause the autopilot manufacturer to write different software for different aircraft. This is why the same autopilot removed from a Cessna Centurion will not work in a Beechcraft Bonanza and vice versa; the software must be rewritten! Some autopilots have their directional control connected to the rudder, rather than the ailerons, which is why the factory must do the install and flight test on the first of a particular aircraft series. I once picked up a new Maule amphibian at the S-Tec autopilot factory for delivery to its new owner for training. The flight characteristics required factory installation, and the directional servo was controlling the rudder. I spent several days in Texas as they rewrote and retested the aircraft numerous times before releasing it back into the field.

So much for “Autopilot Basics 101!” In future issues of Midwest Flyer Magazine, I will disclose “My Favorite Autopilot,” autopilot maintenance and failures. We will examine autopilot-related accidents and some autopilot glitches that could save your life. We will also discuss GPS-steering and altitude preselect.

Fly safe and always have a tailwind!

EDITOR’S NOTE: Michael “Mick” Kaufman is the program manager for the Beechcraft Pilot Proficiency Program and a flight instructor operating out of Lone Rock (LNR) and Eagle River (EGV), Wisconsin. Kaufman was named “FAA’s Safety Team Representative of the Year for Wisconsin” in 2008.