Speaking as an engineer...
 

I have always been curious as to at which point the test pilot gets involved with the project. How do you guys go about convincing yourselves (say) that the latest FBW creation of mad groups of engineers will convincingly do what it's told, with no uncommanded moments...

Mart

Theoretically, there isn't time to prove that will never happen. There are too many possible input combinations to guarantee that all have been tested.

To ensure that it probably won't fail, quality control throughout development is necessary, then flight test will try to cover all flight conditions within a specified envelope and a specified role.

To be fair, it really only has to be as or more reliable than a mechanical system.

All test programs for manufacturers involve a careful build-up of the knowledge. First comes analysis and design (which establishes the way the part/system/aircraft ought to behave) then comes subsystem tests (which attempt to see how the part/subsystem behaves with limited connectivity to the rest of the aircraft) then comes whole system tests (where the whole sub-system is benched and tested - like an iron-bird or tie-down test).

These ground tests and the designers data build up a "smart book" that describes what the aircraft should do in flight. The instrumentation is designed and installed to both check the fitness of the aircraft, and also to track how well each part/subsystem follows the designer's original understanding. Flight test is like exploring a mine field, where you get extra points by not just keeping from blowing up, but also by finding and defusing the mines through prediction. Nobody tests by "diving till the wings almost fall off" any more. In effect, flight test is a validation process, so the trends of the structural and performance data are matched against the actual measurements to assure that the aircraft behaves, but also that the designer knew what he was doing to begin with. Excellent agreement is desired, he loses points if the aircraft is very much better than he thought!

For the record, most accidents occur by surprise, when you least expect it, and during simple flight tests. The really hairy structural demos are usually exciting but uneventful. The damn radio check flight on a Thursday afternoon, or the ferry to a new test facility are the way you lose it, and the failure is by definition a complete surprise to the team.

Also, the typical TPS course is not a complete preparation for a person to perform initial flight tests of aircraft, because the schools do not understand nor teach the critical issues that face the initial flight test team. The TPS curricula are designed to allow military TP's to do acceptance and spec compliance evals, which are important, but not at all what a development test pilot faces in the shakedown and development flight test portion of a program. Few military test organizations do actual initial envelope expansion of new aircraft.

TPS does teach the way to think about a test, and the way to safely conduct yourself, so it is an excellent primer. IMHO, an engineer who is an experienced helo pilot and who has spent time in a manufacturer's design shop, or in a manufacturer's flight test organization is the right candidate for a manufacturer's TP.

As a plank Engineer who became a light-plank Test Pilot (at-least part time, sadly I'm still only paid for being an Engineer) I agree completely with Nick save one thing.

What is far from apparent to anybody outside of the test flying community is the monstrous amounts of paperwork - whether it's plannings, written briefings, test cards, flight test reports, analysis, summary and compliance reports - it's almost endless and a far greater demand upon a flight testers time than flying ever is.

So, paperwork is the worst bit of Test Pilot's job.

Dr.G

Crossover:
The entry for 'Special Qualifications' in my old RAF Logbook says (aircraft types) 'Various' (qualification) 'tp'. I did the RW course, but the qualification doesn't actually differentiate. (Although the ETPS graduation Scroll does!)
What matters is Role Relation. As an experienced light aeroplane pilot and instructor, I feel adequately equipped to comment expertly on such types, despite having earned my daily bread mostly on helos.
I know of several 'rotes' who have 'crossed' over. One, for example, is CTP of Marsahll Aerospace and now flies very large aeroplanes, having transitioned through smaller twin FW.