Autoland — Page 388, Lesson 472

Let’s start with the big picture, because autoland is one of the most precise things an aircraft will ever do. I’m going to walk you through the sequence of an automatic landing, and I want you to think of it as a series of gates — each one defined by a height above the ground, measured by the radio altimeter. First, the key instrument here is the Radio Altimeter, which we abbreviate RA. It measures the aircraft’s actual height above the terrain directly below, not above sea level. Every height I’m about to give you is an RA value unless I say otherwise. So, the sequence begins at 1500 feet RA. Shortly after the aircraft has captured both the LOC and the G/S — that’s the Localizer, which gives lateral guidance, and the Glideslope, which gives vertical guidance — and as it descends below 1500 feet, the second autopilot couples with the flight controls. At that moment, FLARE mode armed is annunciated — that’s a cockpit indication telling you the flare is ready. The A/P go-around mode also arms at this point, but note: it arms but is not annunciated. And if the aircraft has it, ROLL OUT mode will also arm now. The autoland status is then annunciated as either “LAND 2” or “LAND 3” — LAND 3 being for fail-operational aircraft, which I’ll explain in a moment. Here’s a critical limitation: from this point on, the pitch and roll axes cannot be manually overridden into CWS — that’s Control Wheel Steering. If you try, the autopilot will disengage. So the crew must not touch the controls in that way. Next gate: 800 feet RA. The second autopilot must have been engaged by this height to execute a dual autopilot approach. If it hasn’t been engaged by the time you descend through 800 feet, engagement of the second autopilot is inhibited — it simply won’t couple anymore. So there’s a hard deadline. Now, between 400 and 330 feet RA, something subtle but vital happens. The stabilizer is automatically trimmed an additional amount nose-up, while the elevators neutralize and hold that pitching-up moment. Why? This is called biasing. It aids the flare. And here’s the safety logic: if the autopilots subsequently disengage, you may need forward control column force to hold the desired pitch attitude. But in the event of a subsequent fail-passive — meaning the autopilot fails in a way that just disengages without commanding anything — the aircraft will tend to pitch nose-up. That nose-up tendency prevents a hard contact with the ground and aids the initiation of a go-around. So this biasing is a built-in safety feature. There’s another hard rule here: if FLARE is not armed by approximately 350 feet RA, both autopilots automatically disengage. So the system is self-checking — if the flare isn’t ready, it won’t continue. Now we move to 45 feet Gear Altitude, which is 50 feet RA. Let me introduce a second height reference: Gear Altitude, abbreviated GA. This is the height of the landing gear above the ground, and it’s different from radio altitude because the gear is lower than the radio altimeter antenna. I’ll come back to how it’s calculated. At about 50 feet RA, the A/P flare manoeuvre starts and is completed at touchdown. FLARE engaged is annunciated, and the FD command bars retract — that’s the Flight Director command bars disappearing from the display. The stabilizer trim is again automatically trimmed an additional amount nose-up. The FLARE mode is automatically engaged, replacing the glideslope, and its job is to reduce the vertical speed from around 10 to 12 feet per second at 50 feet down to 1 to 2 feet per second at the point of touchdown. The controlling signal for this is the rate of reduction of radio height — in other words, how quickly the aircraft is descending toward the ground. Also at this stage, the A/T — that’s the Autothrottle — begins retarding thrust at approximately 27 feet RA, so that it reaches idle at touchdown. So the throttles are already spooling back before you touch down. Now, about that gear altitude calculation. It’s pre-programmed into the computer, and it’s based on three things: radio altitude, pitch attitude, and the known distance between the landing gear, the fuselage, and the radio altimeter antenna. Because the antenna is somewhere on the fuselage and the gear is below it, the gear altitude is always less than the radio altitude — and the difference changes with pitch attitude. That’s why the calculation needs all three inputs. Next gate: about 5 feet GA. Here the flare mode is disengaged and there’s a transition to touchdown. The LOC disengages — lateral guidance is no longer needed. And the roll-out mode, if available, engages. At about 1 foot GA, the pitch attitude of the aircraft is decreased to 2 degrees — that’s the nose being lowered. And finally, at touchdown, a command signal is supplied to the elevators to lower the aircraft’s nose, bringing the nose landing gear wheels in contact with the runway and holding them there during the roll-out. Let me tie the two autopilot concepts together, because they matter for the LAND 2 versus LAND 3 distinction. A fail-passive system, as I mentioned, simply disengages on failure — it doesn’t command anything dangerous, but it leaves the crew to take over. A fail-operational system, which is what LAND 3 indicates, can continue the landing with the remaining autopilot after a failure. That’s the difference in capability that the annunciation is telling you about. So the whole sequence is a cascade of height gates: 1500 feet for arming, 800 feet as the engagement deadline, 400 to 330 feet for the stabilizer biasing, 50 feet for flare start, 27 feet for throttle retard, 5 feet for flare disengage and roll-out, 1 foot for the pitch reduction, and touchdown for the nose gear contact. Each gate has its own annunciation, its own logic, and its own failure mode. That’s the autoland sequence.