Autopilot — Page 363, Lesson 432

Let’s pick this up right where the autopilot’s engagement logic leaves off. I want to walk you through the automatic disengagement conditions first, because these are the exact situations where the A/P will drop out of control on its own — and as a professional pilot you need to know every single one of them cold. Here’s the full list, in the order the book gives it. The A/P automatically disengages when any of the following occur: Pressing either A/P disengage switch. That’s the dedicated disengage button on the control wheel — either pilot’s. Pressing either TOGA switch with a single A/P engaged in CWS or CMD below 2000 ft RA. TOGA stands for Take-Off/Go-Around. CWS is Control Wheel Steering, CMD is Command mode, and RA is Radio Altitude. So if you’ve got just one autopilot engaged, in either CWS or CMD, and you’re below 2000 feet radio altitude, pressing either TOGA switch disengages it. Pressing either TOGA switch after touchdown with both A/Ps engaged in CMD. Notice the contrast — here both autopilots are engaged, both in Command mode, and the trigger is after touchdown. That’s a different condition from the single-A/P case. Pressing a lighted A/P engage switch. If the engage switch is lit, it means that autopilot is active — pressing it again disengages. Pushing the A/P disengage bar down. That’s the mechanical bar on the glareshield. Activating either pilot’s control wheel trim switch. If either pilot moves the trim switch on the control wheel, the autopilot drops out. Moving the Stabilizer Trim Autopilot Cut-out Switch to CUT-OUT. That’s the switch that isolates the stabilizer trim from the autopilot — moving it to CUT-OUT disengages the A/P. Loss of respective hydraulic system pressure. Each autopilot depends on its own hydraulic system; if that pressure is lost, the A/P disengages. Repositioning the EFI transfer switch. EFI is the Electronic Flight Instrument transfer switch — moving it disengages the autopilot. Either left or right IRS system failure or FAULT light illuminated. IRS is the Inertial Reference System. If either side fails, or its FAULT light comes on, the autopilot disengages. Loss of electrical power or a sensor input which prevents proper operation of the engaged A/P and mode. This is the catch-all — if the autopilot can’t get the power or the sensor data it needs to run its engaged mode properly, it drops out. Now, the engagement logic. Only one A/P can be engaged at a given time unless the approach mode — APP — is engaged. So if you press the engage switch for a second autopilot while not in APP mode, the second autopilot engages as selected, and the first one disengages. The second A/P then operates in CWS or CMD without interrupting the CWS or command operation. In other words, the handover is seamless — the new autopilot takes over without a break in control. There’s one more subtle engagement case. If an A/P is engaged with the CMD engage switch during FD-only operation — that’s Flight Director only, no autopilot — while pitch or roll commands are more than half a scale from centred, then the A/P automatically engages in CWS for pitch and/or roll, and the FD command bars retract. So instead of grabbing command and snapping to a large command, it engages in Control Wheel Steering, and the flight director bars pull back out of the way. Now let’s talk about automatic synchronization, because this is the heart of a smooth engagement. Beyond the pre-engage requirement that the autopilot circuits are electrically complete, we also have to ensure that on engagement the ‘take-over’ is effected smoothly and without ‘snatching’ of the aircraft’s control system. Snatching means the controls jerking — you never want that. So the aircraft must be trimmed for the desired flight attitude before engagement, and the automatic control system must be synchronized to maintain that attitude on engagement. In the majority of autopilot systems, synchronization is effected by specifically designed synchronizing circuits. These circuits automatically sense any existing ‘standing signals’ in the pitch and roll channels, and automatically reduce or ‘wash out’ these signals to zero. A standing signal is a residual error signal sitting in the channel — if you engaged with that signal present, the servo would immediately drive to correct it, causing a snatch. The synchronizing circuit washes it out to zero. That stops the servo actuator in a position which is synchronized with the datum attitude detected by the sensing element. And that position is indicated by the return of the trim indicator pointer to its central position. So when you see the trim pointer centred, you know the autopilot is synchronized to the datum attitude and ready for a smooth take-over. Let me show you the actuator arrangement so you can picture how this fits into the control system. That’s the parallel actuator installation — the autopilot actuator works in parallel with the pilot’s controls, which is exactly why synchronization matters. And here’s the switching arrangement for engagement and disengagement. So to tie it together: you’ve got a precise list of automatic disengagement conditions, a rule that only one A/P engages unless you’re in APP mode, a special case where CMD engagement during FD-only operation with large commands drops you into CWS, and the synchronization circuits that wash out standing signals so the take-over is smooth and the trim pointer centres. That’s the complete engagement and disengagement picture.