
I want to walk you through the concept of control laws in the autopilot, and then we'll look at a very specific example: the Boeing 737-400's command speed limiting and reversion modes.
Let's start with the big idea. The autopilot has to fly the aircraft the way a good human pilot would — logically and safely. That means it must never break the aircraft's limitations: speed limits, load factor limits, pitch limits, bank limits. But at the same time, it shouldn't be so timid that every manoeuvre takes forever. It needs to use a satisfactory amount of the aircraft's performance.
Here's the example the book gives. Imagine a light aircraft on a VFR navigational exercise. Overhead the next waypoint, the pilot needs an 80° left turn. The bank angle chosen determines how that turn feels and how long it takes. If the pilot used only 5° of bank, the turn would take an inordinate amount of time. If the pilot used 90° of bank, that would be excessively hard — it could break load factor limitations and probably cause a loss of height. The sensible choice is 30° to 45° of bank. But if the turn required was only about 10° instead of 80°, then banking to 30° would generally be considered too harsh.
So the autopilot needs to apply the same logic to all its actions. Here's the key principle: whenever the autopilot is required to make a correction — either for stability or when referring to a particular flight path — the control response will always be in proportion to the deviation or the corrective action required. But — and this is crucial — only up to a limit. That limit prevents the autopilot from attempting a manoeuvre that would cause an excursion from the safe flight envelope.
Now let's move to the Boeing 737-400 specifically. To prevent the AFS — that's the Autopilot Flight System — from causing a flight envelope excursion, there's a system of command speed limiting and reversion modes. And note this important point: AFS command limiting and reversion is independent of the stall warning and the airspeed/Mach warning systems. So this is a separate layer of protection.
Let's look at command speed limiting. The AFS provides speed, pitch, and thrust commands to avoid exceeding these limit speeds: VMO/MMO — that's maximum operating speed and maximum operating Mach number; wing flap limiting speeds; landing gear speeds; and minimum speeds. The commanded speed can be equal to a limit speed, but it will never exceed a limit speed.
Here's where it gets interesting. Speeds greater than VMO/MMO cannot be selected from the MCP — that's the Mode Control Panel, where the pilot sets the desired speed. However, speeds can be selected that exceed flap and gear limiting speeds, or that are less than the minimum flight speed. So the system allows the pilot to select an unsafe speed, but it won't let the aircraft actually reach it.
Now, what is minimum speed? It's based on an angle of attack, and it's approximately 1.3 VS for the current flap configuration. VS is the stall speed, so 1.3 times stall speed. It's sensed by the angle of attack vanes — one on either side of the forward fuselage.
Here's the limiting behaviour. If a speed greater than a placard speed — that's a published limit speed — or less than the minimum speed is selected, the AFS allows acceleration or deceleration to slightly short of the limit, then commands the limit speed. So it lets the aircraft get close to the limit, but not past it. And there's an indication: the over-speed or under-speed limiting symbol appears in the MCP IAS/Mach display when the commanded speed cannot be reached. IAS is indicated airspeed.
So to tie it together: the autopilot's control laws make it behave proportionally and sensibly, and the command speed limiting system is the specific mechanism on the 737-400 that keeps the aircraft inside its safe speed envelope, independent of the stall warning systems.
This is one saved preview. Continue from this exact book or paper with BlueFlash voice AI.
Continue in BlueFlash