Taming the autogyro: will the autogyro ever be truly domesticated?

Robinson, Sophie
Taming the autogyro: will the autogyro ever be truly domesticated? PhD thesis, University of Liverpool.

[img] Text
RobinsonSop_Jul2014_2006919.pdf - Unspecified
Available under License Creative Commons Attribution.

Download (13MB)


Aviation is a pastime enjoyed by many amateur pilots. Of the 21 000 aircraft registered in the UK [1], 96% are engaged in general aviation activities (non-commercial flying). The UK Civil Aviation Authority (UK CAA) classifies microlights, gliders and autogyros as recreational sports aircraft. Of the 21 000 UK-registered general aviation aircraft, only 306 are autogyros, compared to over 4300 microlights and almost 2600 gliders. Despite this fact, the autogyro has been seen to exhibit a fatal accident rate up to 100 times higher than those of the microlight or glider. In response to the identification of this high accident rate amongst autogyro type vehicles, the CAA commissioned a programme of research intended to understand the cause. This research, undertaken in the UK by the University of Glasgow, consisted of analytical, wind tunnel and flight test activities. These studies concluded that the autogyro displayed “conventional lateral and directional dynamic stability characteristics”, and that both the static and dynamic stability (in particular, a lightly damped phugoid mode) was highly sensitive to the vertical position of the centre of gravity (c.g.) relative to the propeller thrust line. The lack of provision within the autogyro community to collect meaningful data in relation to the airworthiness requirements was also highlighted. Outside of the work performed as part of the research programme that generated this report, there is still considered to be “little indication that rigorous scientific or engineering investigation of airworthiness has occurred”. Therefore, there remains significant scope for further research into just what makes autogyros so unsafe to fly and how to improve their airworthiness. Prior research into the autogyro and its aerodynamic characteristics can be broadly divided into two phases, the first being from its inception in 1923 to the beginning of World War II and the period between 1996 and the present day, when a resurgence of interest in the autogyro began to occur. Much of the early works concentrate on characterising the autogyro’s aerodynamic characteristics, relying heavily upon wind tunnel testing, flight testing and analytical investigation to establish an understanding of the fundamental flight dynamics of the autogyro. With the first flight of the first functional helicopter, the outbreak of World War II and the death of the inventor of the autogyro, research interest in this aircraft type was critically diminished. Only three papers on the subject of autogyros were published between 1939 and 1996. The Air Accident Investigation Branch (AAIB) review of the airworthiness of the grounded Air Command autogyro, conducted after the occurrence of 5 fatal accidents between 1989 and 1991, recommended the commissioning of a programme of research into both the airworthiness and aerodynamic characteristics of light autogyros. As a result of this recommendation, the autogyro experienced a resurgence in research interest, culminating in the publication of a CAA Paper which presented 4 recommendations intended to improve the airworthiness of the autogyro: 1. It is recommended that the vertical location of the centre of gravity (c.g.) should lie within a ± 2 inch envelope of the propeller thrust line. 2. Horizontal tailplanes are largely ineffective in improving the long term pitch dynamic stability (phugoid mode). 3. Extreme manoeuvring can lead to excessive rotor teeter angles during certain phases of flight, potentially resulting in the rotor blades striking the prop or empennage. 4. The chordwise centre of gravity of the rotor blades should always lie at or ahead of the 25% chord position to prevent rotor blade instability. One of the primary aims of this Thesis was to assess the validity and applicability of these recommendations; in order to do so, a simulation model of an autogyro was created. The model was based on the G-UNIV research autogyro owned by Glasgow University and validated against flight test data in order to ensure the required level of fidelity was achieved. Upon re-assessing the recommendations, in some cases, different conclusions were drawn. The first recommendation, while accepted as a sensible design aim, was found to be overly restrictive. BCAR Section T, the airworthiness specification for autogyros, specifies requirements on the period and time to half amplitude of any longitudinal oscillations present in the aircraft. Limiting the vertical position of the centre of gravity to within ±2 inches of the propeller thrustline resulted in forcing a design which is compliant with the requirements of BCAR Section T outside the specified range, to become non-compliant when the centre of gravity lies within the range. Recommendation 2 suggests that the removal of the tailplane of the aircraft has little impact on the longitudinal trim control positions and the characteristics of the phugoid mode. It was found that the results from the simulation model disagreed with this conclusion; the removal of the tailplane changed the characteristics of the phugoid model and the trimmed control positions significantly. The third recommendation highlighted the potential for a rotor blade to strike the propeller or the empennage under extreme manoeuvring. The simulation environment provided a safe environment in which to test this recommendation; it was found under the loading of an extreme manoeuvre it was possible for the main rotor to strike the tail, supporting the conclusion drawn in CAA Report 2009/02. It was not possible to assess the fourth and final recommendation, relating the positioning of the chord-wise position of the blade centre of gravity, due to the limitations of the simulation model developed. Another focus of the recent work surrounding the autogyro has been on quantifying and assessing the handling qualities of such vehicles. This presents many challenges, including the fact that no autogyro-specific handling qualities specifications currently exist. One of the main themes of this Thesis was to progress towards the creation of such a specification, either through development of a new methodology or development of existing specifications, such as ADS-33E-PRF. The first steps in this field have been taken by Glasgow University using ADS-33; the primary specification used in the assessment of military rotorcraft. Assessment of the autogyro was previously carried out in a real-world flight trial using existing Mission Task Elements (MTEs) taken from ADS-33, the Slalom and the Acceleration-Deceleration. The results from this trial were then used to derive proposed Level 1, 2 and 3 boundaries for quickness and pilot attack. This Thesis replicated this trial using real-time piloted simulation, and the method described in the work carried out by Glasgow University was also utilised to derive a set of predicted handling qualities Levels for both quickness and pilot attack. It was found that the predicted Level boundaries generated from the simulation trial did not agree well with those predicted in the original flight trial. There were several reasons for this; in the original flight trial the pilot used a non-standard technique to fly the Slalom manoeuvre, using sideslip to complete the test course. Additionally, both the original flight trial and the simulated flight trial used data from several different course geometries. This resulted in the ordering of the Level boundaries being reversed for the Levels predicted by the simulated flight trial, as those test points carried out on the more aggressive course geometries received lower handling qualities ratings, whilst using the most aggressive control inputs, compared to those on easier courses which used lower magnitude and aggression inputs, and thus a lower quickness, while receiving better handling qualities. Recommendations were made to address these issues in future iterations of this work. This Thesis also sought to establish whether the MTEs specified in ADS-33 highlighted the deficiencies within the autogyro in the same manner as they do in the helicopter, as well as identifying the fundamental differences between the autogyro and helicopter. For the most part, the MTEs chosen did highlight the deficiencies in the same manner for both autogyro and helicopter, with the exception of the Acceleration-Deceleration. When assessing the helicopter, the Acceleration-Deceleration is intended to highlight the presence of any undesirable roll due to pitch cross couplings present in the aircraft. As the autogyro uses throttle setting to accelerate, and not pitch attitude, the Acceleration-Deceleration cannot be used to assess the impact of this cross coupling in the autogyro. The Acceleration-Deceleration also revealed the presence of a roll due to throttle coupling in the autogyro which had not previously been reported. Alongside the repetition of this original flight trial, new MTEs were analysed. The Heave Hop and the Roll Step are MTEs originally intended to assess tilt rotor type aircraft; however, they have shown some utility in assessing the autogyro. The Heave Hop in particular highlighted another of the fundamental differences between the autogyro and the helicopter. The Heave Hop is intended to test the ability of the aircraft to sustain a positive load factor, before transitioning to a negative load factor. Much of the effect of this changing load factor is mitigated for the helicopter by the presence of a rotor speed governor, which maintains the rotor speed at a constant value for irrespective of the rotor loading. However, the autogyro rotor is unpowered, meaning that changes in the rotor loading also change the rotor speed. This results in potentially problematic changes in the autogyro handling qualities; reducing or increasing the available control power or quickness available to the pilot. The completion of the work described herein has raised many possible avenues for further work; although it has been shown that ADS-33 style predicted handling qualities represent a good baseline for development of autogyro specific handling qualities, there remains scope for redefinition of the Level 1/2 and Level 2/3 boundaries for the predicted handling qualities parameters, such as quickness and control power, as well as for the re-definition of MTEs to make them autogyro-specific. In order to draw absolute conclusions, these manoeuvres must also be reassessed using different autogyro types and configurations – often the work presented herein is only the second time such an assessment has taken place. The development of the easily reconfigurable autogyro model developed as part of this Thesis presents an ideal tool to achieve this goal. In summary, through development of existing work and introduction of new ideas, some progress has been made in the progression towards an autogyro-specific handling qualities specification. Whilst there is still a long way to go in thoroughly domesticating the autogyro, this Thesis represents a step in the right direction.

Item Type: Thesis (PhD)
Additional Information: Date: 2014-07 (completed)
Uncontrolled Keywords: Autogyro, Autogiro, FLIGHTLAB, simulation, modelling, handling qualities, rotary wing, ADS-33E-PRF
Subjects: ?? TA ??
Depositing User: Symplectic Admin
Date Deposited: 09 Sep 2015 15:35
Last Modified: 17 Dec 2022 01:18
DOI: 10.17638/02006919
URI: https://livrepository.liverpool.ac.uk/id/eprint/2006919