Tiger Moth Crash, Winchampton, Dorset Using the unique capabilities of the j2 Universal tool-kit to challenge the findings of an AAIB report cited as evidence in a trial for manslaughter in the UK. The j2 Universal Tool-kit was used to establish marked differences between the pilots reported experience and the AAIB Investigation conclusions that resulted in a loss of control of the aircraft.


On the 15th May 2011, a leisure flight of a 1941 Tiger Moth aircraft turned into a disaster when the aircraft crashed, killing the passenger and badly injuring the pilot, Mr Scott Hoyle.

In the aftermath of the accident and subsequent investigation by the AAIB, the report found that the incident was the result of pilot loss of control when attempting an aerobatic loop that was not completed. This conclusion was reached despite the pilot claiming to have experienced a stuck control surface (rudder) which remained fully deployed after exiting a left turn manoeuvre to return to the airfield.

This case history walks the reader through the work j2 AAI did in helping to establish what actually happened. It looks at how the proprietary j2 Universal Tool‐Kit was applied to demonstrate the data patterns produced from analysis of the available data sources and the flight physics analysis ‐ only possible within j2 software. The results supported the Pilot’s view of events in preference to the conclusions reached in the AAIB report and other prosecution evidence in a court of law resulting in the pilot being found ‘Not Guilty’ of all charges.


  • The j2 Universal Tool‐Kit is used to build a full non‐linear predictive solution.
  • The full physics‐based model of aircraft kinematics produced a physically feasible aircraft.
  • The flight construction process was able to reproduce the exact flight path from GPS data and a range of variants.
  • It was possible to build a range of flight manoeuvres and mass profiles to match available weight and balance data and compare data patterns produced from each.
  • 3‐D visualisation enabled views of the dynamic behaviour of the aircraft from all angles, a major aide in helping the jury to better understand.

Model Build and Sanity Check

The initial flight physics based model build of the 1941 Tiger Moth used data provided by De Havilland Support and a variety of other named sources. This enabled a high fidelity math model of the aircraft to be built. This model was further `sanity checked’ against data accepted as being representative of the DH82A Tiger Moth aircraft. To identify the characteristics of the aircraft, a series of angle of attack and sideslip sweeps were run, along with angular velocities. These numbers were compared to values from an aircraft of a similar type and class to ensure that there were no significant differences.

Once this model was further verified by reference to experienced Tiger Moth pilots, the analytical interrogation of the model could begin.

Model Interrogation Process

The first steps in the interrogation process examined the aircraft behaviour and characteristics resulting from an aerobatic loop. This allowed j2 to investigate the impact of a variety of weights and balances, aircraft speeds and pilot control inputs likely to have resulted in a successful loop being completed.

J2 then examined further the sensitivities associated with performing a loop (total aircraft weight, changing CG and speed of entry) and identifying from a flight physics standpoint the relative impact of each on performing a loop.

J2 was also able to examine how variations in each of these attributes might result in the aircraft stalling due to an unsuccessful loop.

All these findings where then presented in a graphical format, establishing the data patterns, and were available as interactive 3‐D playbacks J2 were also able to establish, through experienced pilot de‐briefings, the recovery procedures in order to regain control in the event of a failed loop ‐ assuming no restrictions in the control surface movements. This ‘evidence’ was also presented as graphed data and then ‘flown’ in the j2 software and each flight viewed and compared in the 3‐D playback facility.

Accident Scenario Recreation

J2 was able to recreate the Pilot’s description of events. These events are summarised below as the aircraft entered a left turn moving from SW to NE:

  • In deploying the rudder to the left, a small resistance had to be overcome and this was felt through the force in the pedal.
  • On exiting the turn the rudder remained deployed fully left.
  • Once in level flight, the influence of the deflected rudder was causing the aircraft to yaw and roll to the left relative to the direction of flight (NE).
  • The Pilot deployed right aileron to control the roll influence. This is known as ‘holding off’.
  • The Pilot then began a process of review as to what was happening and how he could free up the control surface. Because of the high drag (yaw) profile he was also losing air speed. The decision was made to centre the controls in the hope that the rudder might free up.
  • On centring the controls this did not free the rudder, the Tiger Moth then initiated a rapid, nose down pitch, and yaw and roll to the left.
  • To assert control the pilot moved the stick back and to the right. This initiated a rapid nose up attitude and loss of air speed which produced a wing stall. The aircraft then entered a spin to the left and the stuck rudder prevented the pilot regaining control.
  • The aircraft crashed to the ground rotating to the left.

Initially, the j2 Universal Tool‐Kit was used to reconstruct the aircraft states. It was additionally possible to reference radar and GPS data. Interrogation of the available GPS and radar data allowed j2 AAI to verify aircraft speeds, headings and patterns of interest over time.


  • The flight path reconstruction was verified by the GPS and radar data.
  • It is possible to identify the aero engineering data patterns associated with each set of aircraft manoeuvres.
  • The time‐step capability was able to identify the critical pilot inputs and their effect on aircraft behaviour. It was also possible to compare various recovery scenarios with and without a stuck rudder.
  • The final results from the model are robust and validated by reference to a documented series of steps and checks.
  • The model and analytical work were able to clearly support the Pilot’s view of events to the satisfaction of the court and the Jury.
  • Any aircraft model created in j2 can be used directly into a simulator, allowing further Pilot assessment of the manoeuvre and possible recovery scen