In recent years the common wisdom on using full flight simulators (FFS) for UPRT was that they were not suitable platforms for training beyond the boundaries of routine flight. Chris Long examines the new processes for UPRT training at Airbus.

The envelope beyond stall had not been explored during the initial developmental flight testing which addressed the regulatory requirements for certification. The argument was that simply extrapolating that data was not merely uncertain and potentially inaccurate, but could actually result in negative training. The industry common standard for UPRT stopped short of the full stall and beyond.

As a result of several tragic mishandling accidents, the industry had to rethink this historic approach, and regulators stepped in to redefine the training requirements. During a presentation on the subject at APATS 2018 in Singapore, Xavier Lesceu, Airbus Test Pilot and TRI/E, detailed the process now developed by Airbus in response to this concern. Two key directives have been published - both for aircrew and for the devices used to train them:

- FAA 14 CFR Part 121-424 training to aircrew FAA Part 121-423 stall training
- EASA Part FCL: UPRT and stall training

- FAA PART 60 Change 2: Full stall capability required in the FSTD
- EASA CS-FSTD Issue 2: Full stall capability in the FSTD (ED 2018/006/R)

Airbus Solution

The start point for this evolution was to get the data of flight beyond the certification envelope. The datum aircraft was the A350, and flights were carried out up to and beyond the full stall to measure the performance and handling characteristics. Particular note was made of the aerodynamic warnings and the presence or otherwise of wing drop, together with noting the response to standard stall recovery procedures. As with all test flying, this was carried out progressively in a controlled manner, but all test points were successfully noted and recorded. That must have been an interesting exercise.

The next phase was to see how those characteristics could be replicated using a FFS. Among the challenges were to understand what range of movement was necessary (in fact the events were relatively benign and regular/repeatable) and how to give the pilot realistic motion cues. The tests revealed that the stall buffet was regular and consistent in amplitude as the airflow detached/re-attached to the wing as the angle of incidence oscillated. Of particular note was that there was no discernible wing drop (undemanded roll) observed throughout the test series, so the existing range of movement of the FFS is sufficient for realistic training.The key to give the same sensation of movement in the simulator as experienced in the aircraft was to tune the motion cues correctly for the buffet modelling and to induce yaw cues to simulate the wing drop at or near the stall in the simulator. The types and amplitude of the motion cues were measured by accelerometers in the FFS pilot seats, and adjusted until they matched the flight test data. Consequently it was determined that the conversion of existing FFS to the enhanced UPRT capability could be achieved by means of a software upload.


The first FFS to be modified was the A350 FFS at the Airbus Training Centre in Miami. All the other A350 FFSs will be modified, and the same process of test and evaluation will be applied to all the fly-by-wire Airbus types and engine variations.

The process of installing this package is based on three models:  AEROdynamic model, Roll-off (wing drop) and Buffet.  The Airbus stall training package also includes Validation Data: Qualification Test Guide (QTG); simulator validated envelopes; and Status of compliance including final subject matter expert (SME) assessment.

The annual verification of the FFS will check that to conform to the QTG run by the SME and the whole programme includes briefing and debriefing material for the pilots and instructors, and can be integrated seamlessly into the airline training process.

Published in CAT issue 5/2018