Every well-trained crew, or any competent instructor, always makes it absolutely clear who has the controls. Even pilots who have flown together for hundreds or thousands of hours always announce aloud, “You have the controls,” and the other pilot responds, “I have the controls.”
These procedures come from the hard-learned lessons that somebody must always be in control, but two can never be in control effectively and safely at the same time.
I was reminded of this when my Baron was at Mayday Avionics in Grand Rapids this week to find and fix a problem with the automatic trim in my autopilot. The autotrim had been intermittent for some time, but had now finally quit functioning. A total failure is always so much easier to fix than an intermittent.
Autopilots fly an airplane exactly as we humans do. Both see an error in the desired flight path and move the controls to correct the error. Both the human and autopilot have a feedback loop. Sensors – eyes or electronic – see a deviation from the desired attitude, course, altitude, or whatever, and move the controls in a way that corrects the error. The airplane responds and the loop is complete when our sensors detect the error being corrected.
The big difference is that autopilots have only a single loop for each control and that loop cares about nothing else, while the human brain can maintain the flying feedback loop and also consider dozens of other factors. That versatility gives the human the edge when it comes to being a manager, but the 100 percent concentration on a single task makes the autopilot a more precise pilot. For a human to fly an ILS, for example, with the same routine precision as an autopilot, total concentration is required. That’s why in crew flying the pilot on the controls at critical times does nothing but fly while his crew mate handles radios, callouts, and overall situational monitoring.
The other way humans and autopilots are alike is that neither has the strength to control the airplane – at least not in every situation – without the aid of a trim system. When either type of pilot feels the controls pulling or pushing away from the desired flight path the solution is to adjust the trim until the pressure is gone.
The big difference – and potentially dangerous difference – between a human pilot trimming and an autopilot trimming is that the autopilot doesn’t know what is causing the force on the controls, while the human should. When the autopilot pitch servo senses force on the controls it adjusts the trim until the force is removed. Most autopilot pitch servos have an internal spring that allows for a certain amount of out of trim force. When the force becomes too high the servo twists on its spring and electrical contacts close powering the trim servo to move the pitch trim system. Simple, and effective.
The potentially dangerous aspect is that the pitch servo and its spring don’t know if the force is caused by air loads, or if the human pilot is pushing or pulling on the controls. If the autopilot is engaged and flying, and the human pulls back on the controls, for example, because he doesn’t like what the autopilot is doing, the pitch servo feels that as a need for nose-down trim. As long as the human pilot keeps pulling, the servo will twist on its spring and call for more nose-down trim.
Autopilot servos have a clutch that we humans can easily overpower so an autopilot can never take the airplane away from you. But the pitch trim system can overpower a human if the autotrim runs it to its limits. Too many pilots have pulled or pushed on the controls with the autopilot engaged until the trim system overpowered the human. Many pilots believe that the autopilot has great strength and that they are fighting the autopilot to the death, when it’s really their pushing or pulling that is causing the trim to run and possibly creating an unrecoverable situation.
Avoiding any potential problems with autopilot trim systems is as easy as remembering that only one pilot can fly at the same time. If you don’t like what the autopilot is doing, don’t “help” it, disengage it and take over. It’s exactly the same as transferring the controls from one human pilot to the other.
What was wrong with the KFC 200 autopilot in my airplane? After I described the way the system would not autotrim during the preflight test, or in flight, but the normal electric trim worked fine, Mayday’s Bob Weber knew that almost certainly the problem was in the pitch servo. Bob had it out of the tail in a few minutes and found the contact switches inside that power the trim servo had worn. In a short time he had them replaced and my autopilot is happily trimming itself as it must to function properly.