I was pondering today, as we wade through books and books of fine detail about the systems on a modern jet aircraft and how they interact, how simple ideas rapidly end up so damn complicated.
An aircraft is fundamentally different from any kind of ground transport in that it really has to work, and keep working in absolutely all foreseeable situations. It cannot ever be allowed to stop in mid-air, or for any important pieces to break or fall off. The consequences of a failure in flight are so grave that the thing just has to be reliable, as reliable as we can possibly make it. It is, in fact, reliable
by law. The chances of a catastrophic failure in a transport aircraft have to be less than 1 in 1000000000 per hour. How they test that one is anyone's guess.
The sensible approach to making something reliable is to make it as simple as possible, something I'm sure aircraft engineers would agree with. Yet modern aircraft are amazingly complex, positively dripping with computers, electronics, hydraulics, pneumatics, sensors and other devious mechanisms, mostly in duplicate or even triplicate in case of failures. How did this happen?
Back to basics
 |
Yes, I realise this is not a glider. Please ignore
the item labelled 'propeller'. |
An aircraft in its most basic form is a glider — is an amazingly simple machine that really needs only three moving parts, which relate to the three dimensions of space we live in. Easy.
It needs an elevator, which is usually a large hinged flap mounted horizontally somewhere on the tail. This enables you to move the nose of the aircraft up or down, known as
pitch. Or if you are relating it to the horizon,
attitude. Attitude in turn controls your speed through the air. It does not, despite the name control whether you go up or down.
The elevator has just demonstrated three of my golden rules of flying:
- Rule one: The plane is almost never going the way it is pointing
More of which later
- Rule two: Common sense, when it comes to flying, is wrong
Flying slowly not too far from the ground is safer than bombing along at 5000 feet, surely. The exact opposite. The stick makes it go up and down and the throttle makes it go faster and slower, right? Err... no. If the ground is rushing up to meet you, common sense would tempt you to avoid it by yanking back on the controls. And it would be very, very wrong. You get the idea.
- Rule three: Everything in aviation has at least two names
You would probably like to be able to change the direction of the aircraft, right? Given the lack of any solid object to yank on can only be done by tipping it over and letting some of the lift from the wings pull you around the corner. This is known as
banking or
rolling and is generally accomplished by
ailerons — a pair of largish hinged flaps, one attached to the back edge of each wing, usually towards the ends.
When you move the stick, one aileron will move down, increasing lift on that wing while the other moves up, reducing lift on the other. The plane will quite rapidly bank, and if you don't release the pressure on the stick it will keep right on banking until it is upside down, it doesn't know or care but you probably do.
If you are paying attention you have probably realised that I have already described three moving parts and I don't have any left for the third dimension. Well, fair enough, you got me.
But I can wriggle out of this one, as strictly speaking you don't absolutely
need a rudder, whose job is to swing the nose left or right, known as
yaw. A rudder, surprise surprise, is a large hinged flap only this time it is vertical and usually attached to the upright part of the tail (the fin).
You can actually fly an aircraft without a functioning rudder, albeit in a slightly 'drunken' manner that lacks style. I know of a glider that was successfully launched, flown and landed with the rudder cables
reversed and survived to tell the tale. In fact (after it was sorted out) I bought it. But that is another story.
On a boat, the purpose of a rudder is to steer. In you try this in an aeroplane, both rule one and rule two will gang up on you; firstly you will not change direction at all, and secondly you will be flying along sideways which is not usually what you want.
So what is the point of the rudder? It does two useful jobs — firstly it allows you to steer the aeroplane
on the ground. Secondly in the air it makes turns neater and more efficient by combating something called adverse yaw, which is where the ailerons tend to swing the nose the wrong way when you bank the wings. Using exactly the right amount of rudder will result in a
balanced or
coordinated turn (see rule three) — one that passengers will not even notice unless they look out of the window.
Using too little rudder is scruffy flying, but too much rudder is far worse as it can get you into a spin. This is very nasty and explains why training aircraft tend to have small rudders. Granted it will also help you get out of a spin, assuming you've not hit the ground already, but that argument is a bit like saying a fast car is safer because you can accelerate out of trouble.
Still with me?
I have tried to describe what the three basic flying controls are for as simply as I can and I am already running up against multiple ifs and buts. I have not even mentioned how or why the control surfaces work. I have quietly ignored the fact that the practical glider will also need air brakes, at least one wheel preferably with its own brake, possibly flaps, definitely instruments, a launch system and more. Each of these items could be the subject of a complete post.
 |
A simplified schematic of just the hydraulic
system of a Boeing 737. |
This, however, is
nothing compared to the systems needed on a modern jet. Wherever you start, you immediately run into complexity.
For example a jet transport aircraft will obviously need engines. Even if we consider the engines to be a 'black box', we can see they will need fuel, which will require carefully designed tanks, with pumps and heaters to get the fuel to the right place at the right temperature and pressure.
The fuel going into engines will be carefully controlled and monitored to give the correct amount of power and run at the correct speed and temperatures by complex computers called FADECs, that will of course be duplicated. The FADECs require a plethora of sensors to do their job properly, and so must communicate both ways with the flight deck computers (in fact they often communicate their parameters to the manufacturer in real time as well). The details of the fuel and engine control systems can, and do, take up whole books.
But the engines don't just provide thrust, they also generate electricity to power the many electrical systems, hot 'bleed' air for the air conditioning and anti-ice systems and hydraulic pressure for the flying controls, undercarriage, steering and brakes. Each of these sub systems will be duplicated at least once on each engine, so there can be at least four, possibly eight. They will be crammed with safety devices such as filters, pressure release valves, temperature sensors, fuses and so on. The systems and their associated safety devices will all require careful monitoring, meaning a plethora of sensors and associated wiring, which takes a computer (duplicated of course) to make sense of it and give meaningful information to the pilot.
The modern airliner is a genuine feat of engineering, it
stretches the limits of technology in so many disciplines —
aerodynamics, structures, electronics, communications, metallurgy,
materials. It is the most complicated machine mankind has ever produced.
The fact that a transport aeroplane that can do the things we want it to do — fly large distances quickly, economically and in reasonable comfort — just cannot be achieved in a simple way. It's pretty amazing that we can do it at all; to do it with outstanding reliability and safety really is incredible.
