We all (presumably) know about stick shakers and audible stall warnings that tell when a stall has or is about to occur but what I am confused about is how a stall or potential stall is actually detected.
Is the actual physical stall somehow detected by sensors reading air pressures etc, or does some particularly complicated bit of computer wizardry determine that the aircraft should be stalling "about now" given it's airspeed, configuration and external climatic conditions?
There's no sensor on top of the wing of which I'm aware, but methods of actually detecting a stall are nearly as varied as there are airplanes...and rightfully so, because airflow around each type of airplane is different.
A stall is an increase in angle of attack to the point of airflow separation, and a big buildup of induced drag and a loss of lift. These things don't happen all at once, necessarily. The key to understanding them is monitoring the angle of attack, and stall warning systems using stick shakers, pushers, etc, do this by sampling the angle of attack. They also sample the position of flight controls such as flaps...because the angle of attack at which the wing stalls will change with the flaps setting as the wing shape and configuration changes. It also takes into account air data information received from an air data computer. How the airplane does this depends on the airplane and the systems in use. Some are very simple, some are not.
Some airplanes like the airbus simply won't let the airplane stall due to "laws" built into the system. Some airplanes such as a single engine Cessna may use a small tab or even a simple suction port on the leading edge of the wing to produce a noise, and nothing more...and yet other airplanes don't even do that...they're simply detectable by a buffeting in the airframe or controls.
An angle of attack vane may be a movable vane which acts like a weathervane of sorts, or it may simply be a probe with holes that sample the way air flows around it to derive angle information. This is then applied to the aircraft data to determine what the actual angle of attack is (local angle of attack isn't the same as the free airstream, and AoA changes with maneuvering or configuration changes may mean that the AoA isn't the same as what the probe or vane is seeing at any given time...computers correct this). The system in use then takes this information in, and outputs whatever signal is necessary...such as a stick shaker or pusher signal.
When the pilots are checking the control surfaces during taxi, why don't they check the flaps and spoilers ??
ie. Why not FULL flaps and then return them to TO configuration.
What's checked really depends on the airplane. The airplane will fly with or without flaps...but not with or without ailerons, rudder, and elevator.
We check all the flight controls after we've set the flaps to their takeoff setting. In cycling the flaps we also move a number of other parts, including the leading edge devices which cycle in stages. We can only test certain things with flaps up, some with flaps out...and we have specific set procedures for everything. In some airplanes, spoilers or aileron augmentation is tested at the same time...happens on our airplane. On others we test them and then test emergency stow systems...some thrust reversers are tested and stowed, and on others, they're not.
The primary reason for testing the flight controls is to make sure they're powered, and that they're not locked for any reason. For example, in our airplane, when the flaps are up, the outboard ailerons are locked out. We have to extend the flaps to get the ailerons to work. We're checking for freedom of flight control movement and that our indicators in the cockpit correspond. There's no reason to run the flaps down, especially during ground operations when there's a greater possibility of damage, hitting something, etc. More importantly, we certainly don't want any possibility that they're overextended and a takeoff roll started with them in that dangerous condition. Further, there's no guarantee that simply because they extended before flight, that they'll work when it comes time to land.
Especially in a big airplane, moving flaps is a big deal, especially on the ground. Even powering a system, such as electrical or hydraulics, can have severe consequences...or moving a flight control for that matter. What's moved, when, where, and how fast must be carefully considered at all times...especially on the ground. We have flight control surfaces longer than a bus. Our horizontal stabilizer and elevators are bigger than the wings on a DC-3, with a wider span. There's a lot of metal moving out there. The flaps extend down a considerable distance, and there's just no good reason to put them down and then up again.
As the engines spun down I could hear this god awful clunk with each revolution of the starboard engine. The ground crew didn't seem concerned, neither did the captain when he did a walkaround and gave the fan a spin with his hand (and the clunk was there as he did it).
Fan blades generally sit loose in the hub, and tend to rattle when they're not under a load. This is done for several reasons, including issues regarding thermal expansion. hearing a rattle or clicking sound isn't at all unusual, when the engine winds down or is windmilling on the ground. You may hear other sounds, depending on the type of engine, associated with other parts of the engine, or even accessories.