Otto Pilot, Mechanical Mike, or Gyro George — it's a lifesaver
The goal of every autopilot is to get back to being idle," says Bob Ferguson of Autopilots Central, an avionics shop in Tulsa. For simple autopilots this means the airplane's wings are level. In more sophisticated autopilots this means that the autopilot is steadily doing what the pilot has selected (climb, maintain altitude, descend, track, or intercept).
With the advent of navigation, weather data, chart, and flight-plan information displayed on panel-mounted multifunction displays (MFDs), pilots are spending more heads-down time in the cockpit — like it or not. An autopilot can take over the flying chores to allow the pilot to better study MFD information — and take a look outside more often.
In addition to giving pilots relief from the constant chore of hand-flying even the most deftly trimmed airplane, autopilots can be lifesavers, especially during single-pilot instrument flying.
According to the 2003 Nall Report, which is an annual report published by the AOPA Air Safety Foundation that looks back on the safety of general aviation flying during the past year, loss of airplane control was a factor in 20 percent of the fatal IFR accidents in the single-engine fixed-gear fleet and in 33 percent of the multiengine fatal accidents. Often, losing control follows spatial disorientation brought on by the rigors of attempting to juggle multiple tasks during instrument meteorological conditions (IMC).
Attitude-based autopilots
At the most basic there are two types of autopilots. They are attitude-based (also called position-based) and rate-based. Until recently both types detected deviations from the pilot-selected flight path by reference to the spinning mass of a pneumatically or electrically powered gyroscope. With the increased use of affordable solid-state attitude heading reference systems (AHRSs) the main liability of using pneumatically driven gyro systems — wear that causes precession, and unpredictable reliability and service life — will slowly fade away in importance. But that day is not here yet.
On September 24, 1929, then-Lt. Jimmy Doolittle made the first successful takeoff, flight, and landing using gyroscopic instruments for attitude and heading reference. Since that date pneumatically driven gyroscopic instruments have given pilots heading, roll, and pitch information related to the airplane's position in space.
When the autopilot senses a deviation, the second part of the system — the selecting/computing part crunches input from the attitude gyro and the heading bug in the simplest systems and the accelerometers, GPS receivers, pressure altitude sensors, and glideslope and VOR pointers in more complex systems to send correction signals to actuators known as servos. This control part of the system moves ailerons, elevators, and in some cases, rudders to get the airplane back on course and altitude. The new Chelton AP-3C (see " Chelton AP-3C," page 109) and typical Honeywell Bendix/King and Century autopilots (except for the Century I) are examples of position-based systems. Century autopilots were also recognizable under the Mitchell, Edo-Aire, and Piper nameplates.
The artificial horizon is stable in a plane that is perpendicular to the Earth's surface. Gimbals within the instrument case permit movement in two planes — roll and pitch. The gyro cannot move in yaw. Position-based autopilots use pickups within the gyro-stabilized artificial horizon to sense roll and pitch deviations.
This is advantageous in turbulence because the roll signals are not affected by deviations in yaw; therefore, recovery from turbulence-induced roll deviations is purer and quicker than the corrections of a rate-based system. Attitude-based, or position-based, autopilots typically correct airplane position by banking the airplane to a preset maximum bank angle, regardless of the airplane speed. The combination of quick and positive responses to roll deviations is one of the attitude-based systems' advantages over the rate-based systems.
Unfortunately the big advantage to the attitude-based systems — pure pitch and roll deviation sensing from the artificial horizon — can translate into a huge liability. If the vacuum pump that supplies pneumatic power for the gyros fails, the attitude reference instruments and autopilot are almost instantly converted into dangerous imposters. It is necessary to pound this point home. Pilots who fly IFR in airplanes with a single pneumatic source are gambling until they install a backup vacuum source — those who depend on their attitude-based autopilots to lighten their pilot chores in IMC should realize that the only thing between them and hand-flying in IMC on partial-panel instruments is a $350 vacuum pump (see " Airframe & Powerplant: Spinning Instruments," January 2003 Pilot).
An extra word of warning to owners with attitude-based autopilots. Precise Flight Inc., of Bend, Oregon, the manufacturer of a backup vacuum system that utilizes engine manifold pressure (it's called manifold pressure, but it's always less than atmospheric pressure and can be utilized to drive vacuum instruments under certain conditions) as a source of vacuum, prohibits the use of its STCed backup vacuum system in airplanes with attitude-based autopilots.
Rate-based autopilots
According to Jerry Walters of Brittain Industries Inc. in Tulsa, Alick Clarkson invented the first rate-based autopilot in 1954. This was followed by the invention of the turn-coordinator-mounted gyro. Rate-based autopilots depend on an electrically driven gyro that mounts in a turn coordinator case. Electric turn coordinators are very dependable. Brittain was founded in 1958 and sold its first two-axis lightplane autopilot in 1961.
The genius of rate-based systems is a canted, or inclined, gyro in which the axis of rotation is offset from vertical by approximately 33 degrees. This canting permits the gyro to sense deviations in both yaw and roll. Early models of Brittain rate-based autopilots supplemented the canted gyro sensors with an airspeed sensor (pitch detection), while S-Tec, the hugely successful modern-day rate-based autopilot manufacturer of Mineral Wells, Texas, uses accelerometers and barometric pressure sensors (altitude hold sensor) to further refine the sensing capabilities of its autopilots.
Unlike attitude-based systems, a turbulence-created roll-deviation signal is slightly muddied because of simultaneous yawing. This complication causes the rate-based computer to compute less roll than is actually taking place. This is the primary reason that attitude-based autopilots are somewhat quicker to respond to deviations than are rate-based autopilots. While this is a fact of operation, Ferguson, who started working on autopilots in 1972, says, "I can tell the difference but most pilots will never detect it." To compensate for yaw-confused signals, accelerometers are used to detect the rate of change of vertical acceleration. Not only do these accelerometers detect the vertical acceleration, which equates to pitch-up/pitch-down information, but also the rate of the return to steady-state flying is modulated to limit pitch maneuvering during normal operation.
The two advantages of the rate-based autopilot system are that a turn coordinator gyro will never tumble nor will loss of the pneumatic power cause a loss of autopilot function. Some pilots rationalize that if they had to choose between a backup vacuum system and a rate-based autopilot the better choice as far as IFR redundancy is concerned would be the rate-based autopilot (see " Airframe & Powerplant: Standby for Safety," April 2001 Pilot).
The electrically driven gyro in even the most basic wing leveler system will do an excellent job of keeping the aircraft under control after the loss of a single pneumatic source while the pilot takes a deep breath, covers up the pneumatically driven gyros, and forms a plan to fly toward VFR conditions. Even when an alternator or generator fails, a savvy pilot who understands electrical load shedding as a way to preserve the battery's electrical power should have at least 30 minutes of autopilot control before the voltage drops so much that the gyro motor can't maintain sufficient rpm to be dependable.
The first rate-based autopilot
The Brittain system was the autopilot of choice for Beechcraft in the 1960s (there was a Brittain B-4 in the 1966 V35 Bonanza that AOPA refurbished for the 2001 AOPA Bonanza Sweepstakes). Each flight-control surface in the Brittain system is controlled by a set of two pneumatic servos — four are required to control both ailerons and rudders.
When a deviation is sensed a rotary valve opens one servo to atmospheric pressure at the same time it directs vacuum to the matching servo. The rotary valve position is controlled by a single angularly mounted gyro that, when displaced, modulates servo action to return the airplane to straight-and-level flight. According to Walters, the B-4 autopilot that was installed in the AOPA Sweepstakes Bonanza was a three-axis autopilot — which could have been returned to service for an estimated $1,800. Walters says a good percentage of the Brittain autopilots in the field can be returned to serviceability. "One of the best things about the Brittain autopilots is the smoothness of their operation," says Walters.
He explained that contrary to what many people think, Brittain is actively in business and has 380 STCs (supplemental type certificates) for installation of its autopilots. "The pneumatic servos quit working after about 30 years, but we exchange servos for $80," says Walters.
It's probably a sign of how much light-aircraft electrical systems have changed since the days of generators and vibrating point regulators, because a sentence in a 1964 edition of Brittain's installation, operation, maintenance, and service manual touts that the Brittain basic autopilot system provides for basic stabilization of the aircraft on all three axes even in the event of electrical failure. Times have changed because vacuum sources are now widely regarded as the Achilles' heel of instrument power sources. Alternator-based electrical systems with backup batteries are more reliable than vacuum sources.
The future
Beginning in the late 1990s Cirrus started producing certified airplanes without vacuum systems. Other manufacturers have followed suit. Does this mean that pneumatically controlled autopilots are no longer viable? Not at all — the pure roll responses and the large number of good attitude-based autopilots in the fleet will keep the industry supporting these autopilots for a long time.
As has already been said, the weak point of both attitude- and rate-based autopilots is the less-than-sterling reliability in the sensor drive systems. This brings us to modern glass-panel airplanes. There's no spinning gyro in glass-panel-equipped airplanes unless little Johnny is playing with his yo-yo in the backseat. That's because modern solid-state AHRSs are providing roll, pitch, and heading data for autopilots. Crossbow Technology Inc., of San Jose, California, is one company that has developed stand-alone MEMS-based (micro-electromechanical systems) AHRSs that are STCed for installation in many GA airplanes. Each Crossbow AHRS utilizes a cluster of inertial sensors and is capable of updating an airplane's position, velocity, attitude, heading, and what Crossbows calls its inertial factors — which translates to G forces and rotational rates — 100 times a second. This information is exported in an RS-232 format and is available to the pilot on the glass-panel displays. A typical Crossbow AHRS measures 5 inches by 5 inches by 5 inches. Electrical power requirements are less than 4 watts or four-tenths of an ampere in a 12-volt system.
A pair of Crossbow AHRSs have been selected by Eclipse Aviation for use with the Avio Total Aircraft Integration system in its Eclipse 500 very light jet (VLJ). Chelton Flight Systems utilizes a Crossbow AHRS in its 3-D FlightLogic Synthetic Vision electronic flight in-formation system displays that will be installed in the AOPA 2005 Commander Countdown Sweepstakes along with the Chelton AP-3C autopilot — the AHRS brain, not a spinning gyro, will keep this airplane upright and on path.
E-mail the author at [email protected].
The Care and Feeding of Your Autopilot
Ron Hitchcock is the owner of Executive AutoPilots in Sacramento, California. When asked for hints on how to care for an autopilot Hitchcock practically shouts, "Use it for at least 20 minutes on every flight." Hitchcock explains that many autopilots utilize multicontact relays that need to be activated regularly to keep corrosion scrubbed off the contacts. Activating autopilots also warms up the electrical servos, which helps drive out moisture.
Hitchcock also says he believes that too many pilots have gotten into the habit of taking off without performing a pretakeoff autopilot functional test. Hitchcock labels modern autopilots as "fantastic" but caution that they require a lot of study to fully utilize all the functions.
Bob Ferguson of Autopilots Central in Tulsa, when asked the same question, chooses to debunk the old wives' tale that says, depending on an autopilot in turbulence is fraught with danger and it should always be turned off when flying in turbulence. "Turn off the altitude-hold function, but don't turn off the autopilot," says Ferguson. All certified autopilots must be able to maintain control in every corner of the flight envelope — from redline speeds down to stall speeds at full-forward and full-aft center of gravity — without exceeding the certification standards. "An autopilot can process changes 25 times faster than a pilot can, and can sense deviations down to one-tenth of a degree," adds Ferguson when pressing his point about autopilot capabilities. — SWE
Related Articles
