By Jim Sparks
Oil change, replacement of ignition plugs and checking the pressure of tires are some of the many things often associated with routine aircraft maintenance. Most of us will automatically inspect the wheel and repack the bearings at a tire change, but unfortunately, similar considerations are often not applied to auto-flight systems. Experience and familiarity will often make the technician more suspicious of certain areas of the main wheel where cracks or other defects are most prevalent. This same familiarity with auto-flight systems can make routine maintenance practices almost second nature.
Autopilot systems in different aircraft will rarely work the same. In fact, some specific types of aircraft may have several different autopilot systems certified for use. Prior to conducting maintenance on any auto-flight system, it is important to have a good understanding of how the system should work.
By the most basic definition, an autopilot is nothing more than a labor saving device and can be categorized as a single, dual or three-axis. Essentially what this means is that the autopilot is the muscle used to operate the aircraft flight control systems. For the pilots to have a successful trip, they need information available, such as how to get to the destination, aircraft speed, altitude and navigation data. For the autopilot to achieve the same result, it needs to be able to utilize the same information. Flight Guidance or Flight Director systems will observe all aircraft flight parameters and based on priorities established by the flight crew, will provide the information required by the autopilot. A basic system will use a gyro reference and serve as a wing leveler, while more advanced technology will allow the autopilot to control the aircraft through all phases of flight including landing. Many high performance turbine aircraft will depend on automatic flight technology to provide protection against the hazards of high speed and high altitude flight. This makes it imperative for those involved in return to service to be fully aware of the possible consequence of any action on these machines.
Causes of autopilot deviation
With the number of aircraft in operation increasing exponentially every few years, many of the highways in the sky have reached the saturation point. The only reasonable solution is to decrease the space between the lanes. In some areas, Reduced Vertical Separation Minimums (RVSM) are already in place. This is where the typical 2,000 feet vertical spacing between aircraft is now decreased to 1,000 feet.
Altitude indicating, as well as auto-flight systems, are now under a much closer scrutiny and have to maintain a very small margin of error. One example to consider would be an aircraft with static ports located behind a removable nose cone. If the nose is not properly secured, or the aerodynamic fit to the aircraft is marginal, the airflow over the static ports can be affected inducing an altitude error. The result is that the autopilot will not hold the assigned altitude. Such things as paint damage or protruding rivets may also distort airflow. Some cases of autopilot deviation have been attributed to replacement of a nose compartment key lock with one possessing better theft-resistant tendencies. Unfortunately, these new and improved type locks protrude into the airstream about three millimeters further than the original and disrupt the airflow over the static port. Degradation of a static port heater may be another cause of autopilot ill health.
Horizontal stabilizer trim actuator with electric motor.
In many aircraft, flight control systems are common to both crew and autopilot systems. Electric motors are a common means of converting autopilot commands into aircraft response. That is, these motors are often connected in parallel with the flight deck controls. The autopilot is frequently operating the same controls as the flight crew. In most cases, these electric motors include a clutch assembly that gives the flight crew the ability to override the automatic systems, providing the flight crewmember's input force exceeds the autopilot servomotor's force. This override torque can often be adjusted.
Appropriate maintenance procedures should always be consulted prior to making any adjustments. In the event a flight control cable should encounter an ice accumulation, perhaps at a pressure seal, it may prevent the auto-flight system from making a minor flight control adjustment as the clutch in the motor could slip. The end result may be an automatic disconnect and a perceived failure of the autopilot. In some cases, the auto-flight system may have multiple means of introducing an aircraft axis change. In the case of a pitch system, the initial change may be an elevator command with a follow up of a horizontal stabilizer until the elevators are faired with the stabilizer. Should a stabilizer trim motor run too fast or too slow, the auto-flight computer may have to continue making flight control adjustments, which can result in a porpoising (vertical oscillation) condition. Should the trim system be equipped with a DC motor using brushes, the resulting dust may accumulate inside the housing resulting in sluggish operation or, if contamination should get in the motor braking system, overrun could occur. Aircraft using cables operated flight control systems and operating at high speed and high altitudes require additional attention be paid to rigging and cable tension adjustment. These aircraft will often have an automatic tension compensator that will make corrections as temperature dramatically decreases at high altitude. As flight control cables are most frequently made of steel, and aircraft structure of various aluminum alloys, the different rates of expansion can result in drastic changes in control cable tension. This may result in an involuntary flight control deflection or possibly desensitize the control system. Overcompensation of the autopilot may result in an oscillation of the aircraft in the axis where the cable discrepancy is located. Once again, this airframe deficiency is often interpreted as an autopilot failure. In fact, should this situation occur in the aileron system, the roll-oscillation may be countered by the rudder Yaw Damping system.
This rolling phenomenon is referred to as "Dutch Roll" and can also be induced by an uneven airflow over the wingtips. If uncorrected, this rolling action could result in an aircraft getting into an inverted situation. Significant wing leading edge damage, missing or inappropriately installed airflow enhancing devices such as boundary layer energizers or vortex generators, may also affect this lateral oscillation.
Yaw Dampers are considered by many manufacturers to be a critical piece of equipment and in some cases if the system is not functioning properly the aircraft may be grounded. In other cases, aircraft speed or altitude may be severely limited until normal operation of the Yaw Damping system is achieved. Frequently, turbine-powered aircraft have adjustable engine mounts that will allow the thrust angle to be adjusted both vertically and horizontally. There is, of course, a specification on verifying a correct setting, but all too often, the adjustment may inadvertently get modified. It will only take a minor difference in thrust angle to create a discrepancy in both the flight and performance characteristics of the aircraft. Some yaw damp systems may try to correct this asymmetric thrust condition by adjusting the rudder position. As the majority of these systems are designed to correct an adverse yaw condition and then return to neutral, a situation like this may cause the Yaw Damp system to disconnect or even indicate a failure. Improper rigging in flight control systems such as flaps, slats, spoilers/airbrakes and even ailerons and elevators or rudder can generate a similar condition that may show itself as a failed yaw damp system. A basic autopilot may also try to "correct" for problems in aircraft flight control rigging.
One of the primary inputs for an auto-flight system is a gyro or some other inertia-sensitive device such as an Inertial Reference System (IRS). In fact, many Yaw Damping systems have their own Yaw Gyro. Frequently, compensation for normal aircraft deck angle has to be made for these rate sensors and is accomplished by installing a tapered shim under the gyro. Direction of installation or orientation to the aircraft nose is also essential. Many airframe manufacturers will elect to operate this gyro anytime the electrical system is energized. Even though these devices are intended to remain in service for a significant number of flight hours, extended use on the ground during maintenance operations will significantly reduce a gyro's useful life. It is generally advantageous to deactivate this type device by pulling circuit breakers.
Flight control systems are often operated by hydraulic power and when an autopilot computer sends a command for an aileron deflection, it expects to see some type of response. This may be an electrical signal response from the specific flight control servo or may even be a modified output from a gyro or Inertial system. If the auto-flight computer waits for a response due to an aircraft movement and the aircraft is not in flight, excessive flight control deflection may occur. If unexpected this could be detrimental should personnel be in the area or if gust locks are installed.
Autopilot authority is the speed or range of flight control movement. This speed or range is often adjusted as a result of aircraft speed or even the position of flaps or slats. In other words, when the aircraft is in a slow flight condition, the flight controls need to move faster and possibly further than with the aircraft in a clean, high-speed configuration. Often, external switches are fitted to secondary flight controls such as flaps strictly for autopilot sensitivity biasing. These devices are frequently confronted with the brushes and spraying equipment of aircraft cleaners or even possibly the application of deicing fluids. Often, the only purpose of this device is autopilot control so an internal electrical failure may not be easily recognizable.
Virtually all automatic flight control systems have some means of disconnect so that the human pilot can take control. Some of these systems are automatic, while others require some specific action of the pilot in command. One example may be if the aircraft is flying with the autopilot holding altitude and the pilot were to introduce a change in elevator trim or horizontal stabilizer angle of attack. The autopilot may respond with a command to the elevator so the aircraft will continue to hold altitude. This is a "Mis-trim" condition and could also occur if the pilot were to apply a force on the control column without first advising the auto-flight computer. Once again, the computer may attempt to override the pilot and a severe pitch oscillation may result. On many aircraft, the autopilot is inhibited anytime the pilot is using the manual trim systems. Should the crew need to make a pitch change and not want to disconnect the automatic system, switches on the control wheel such as "Pitch Sync" (PS), "Touch Control Steering" (TCS), or "Control Wheel Steering" (CWS) may be actuated. Once the manual maneuver is completed, the switches are released and the autopilot regains control of the aircraft. "Go Around" (GA) switches may also serve as an autopilot disconnect while on some aircraft they serve as an input to the flight guidance system as an automatic response to a missed approach to landing.
Testing for overuse and underuse
It is very important to understand all the interlocks in a specific aircraft that will cause the autopilot to become disabled. In many situations, flight crews get used to using only one means of isolating the auto-flight system and none of the alternate methods get exercised. Most of us in the aircraft maintenance field realize that the lack of use of a device causes almost as many problems as overuse. If not already incorporated in the aircraft normal maintenance schedule, it would be worthwhile to include a designated time where all the autopilot functions can be tested. Most of the later technology digital auto-flight systems include a maintenance test where all the concerned switches are exercised and monitored.
When performing even the most basic maintenance on an aircraft, it is worth considering not only how it could affect the pilot but also how it could affect the autopilot.