Embraer Ejets fbw has a direct mode which does nothing other than replace the cables with electrical signals with no software interference. I think it's similar on the 777.

In normal mode this direct signal is mixed with software generated signals to dampen, limit, shape the control output.

That's correct - the 777 "normal mode" takes pilot input and mixes it with inputs from the Primary Flight Computer. In the "direct mode" the flight computer is skipped altogether and pilot inputs are sent directly to the flight controls. However, even in direct mode it is still considered FBW.

But even still it is not a "pure" FBW system as the controls are mechanically linked to one spoiler on each wing for redundancy

Concorde FBW
 

Salute!

Not sure how much "backup" the Concorde had, but it sure as hell had a lotta FBW concepts implemented.

Our Viper had ZERO mechancial connections of any kind to the flight control surfaces.

Pprune has a great thread about the Concorde, including engineers and at least two pilots contributing.

Thanks for calling me out, Tofeez.

old Viper pilot sends....

RetiredF4, thanks for the PM. Maybe I am confusing people by talking about attitude - I am referring to pitch and bank. Nothing to do with engine thrust.



U

In total, an A320 has 7 flight control computers that process pilot inputs according to 4 different modes of operation, called LAWS:
1) Normal Law
2) Alternate Law
3) Direct Law
4) Mechanical Back Up

In Normal and Alternate Law, pilot commands are processed from the computers and then converted to flight surface movement. In reality an Airbus pilot when moving the sidestick asks for “G’ load” in Pitch and “roll rate” in Roll. The computers will calculate the appropriate control surface deflection and actuate them. Furthermore, the aircraft doesn’t need the pilot to trim it to maintain a given attitude. The FBW will auto-trim the control surfaces to maintain both pitch and bank. (It is strange though that the B787 has trim controls on the control wheel…)

In Direct Law, the control surfaces movement is directly proportional to the sidestick movement.

All control surfaces are electrically controlled and hydraulically activated.
All primary control surfaces have at least two hydraulic actuators (for redundancy) which receive electrical signals (orders) from the flight control computers.

Only the Rudder and the Trimable Horizontal Stabilizer have a direct mechanical link to be used as a last resort, provided hydraulic power is available.

PFC's (Powered Flying Controls)
 

Sorry gums, but I have to trump both of those!

You are about two decades late!

An aircraft with 'ZERO mechancial connections of any kind to the flight control surfaces' first flew on 30th. August 1952. It was the VX770 prototype for what became the Avro Vulcan.

Avro_Vulcan

The term used for electrically controlled hydraulics with feedback was 'Powered Flying Controls'. The 'Controls' even had 'autostabilisation in the form of pitch and yaw dampers' and later an 'auto mach trimmer'.

As with FBW, loosing electrical power was critical - 'Because there was no manual reversion, a total electrical failure would result in a loss of control.'

Interestingly the Vulcan was powered by Olympus engines, which later were to power Concorde.

A Vulcan airframe was used as a testbed for the Concorde engines.

Re-reading the Wiki highlights that many safety features that are currently used were developed for the B2 variant of the Vulcan (RAT, AAPP->APU).

The Vulcan is of the same vintage as the De Havilland Comet - the first commercial jetliner, yet some of it's engineering developments took many decades to appear in commercial jetliners!

From a Software Engineer's perspective, your worry is pretty far from the reality. The relationship between the computer systems in aircraft flight control compared to the computers we interact with in a home or business sense is somewhat akin to the relationship between an old pick-up truck or marine diesel engine versus the engines used in Formula 1.

By which I mean that the former are designed to plod along doing their thing until doomsday without anything going wrong, whereas the latter are tuned for performance with the expectation that they will crash occasionally. There is no complex operating system involved in flight control logic, and the software has been exhaustively tested - it won't necessarily be perfect, but it won't do anything in and of itself to put the aircraft in danger.

Most of the previous posts are correct - digital FBW is implemented in the Airbus models of the A320, A330/A340 and A380 series, as well as the Boeing B777 and B787. Analogue FBW was implemented in Concorde.

Fundamentally, FBW means that the connection between the flight controls and the flight surfaces is electronic rather than electro-mechanical - that's about it.

So then the A320 is not "pure" FBW because it too has mechanical connections to the rudder and THS "for redundancy"... Methinks you need to redefine "pure".