Main Menu

Fairey Jet Gyrodyne

 

faireyjetgyrodyne


Towards the end of 1953 the surviving second prototype Gyrodyne reappeared in a very different form. A research contract had been received from the Ministry of Supply to try the principles of the tip-jet rotor-driving system.


This was necessary to test the tip-jet, and to develop handling and other procedures for the compound, or convertible, helicopter. By early January 1954 this aircraft, named the Jet Gyrodyne, and carrying the duplicated Ministry serial XD759, which was later to be changed to XJ389, was making tethered flights at White Waltham in the hands of John N. Dennis, who had joined Fairey as helicopter test pilot in June 1949. The first free flight was made in January 1950.


The Jet Gyrodyne retained the fuselage, tail unit, stub wings and tricycle undercarriage of the Gyrodyne, and was also powered by a specially modified Alvis Leonides nine-cylinder radial. The engine was used to drive, through a gearbox and shafts in the stub-wings, two variable-pitch pusher propellers of Fairey design. These propellers provided propulsion in cruising flight and slow-speed directional control through rudder pedals by means of a differential pitch-change which was superimposed on the collective-pitch action. A third drive from the Leonides was taken from the main gearbox, via a friction clutch, to two Rolls-Royce Merlin centrifugal compressors mounted face-to-face under the rotor pylon. Air from these compressors was fed to the fuel-burning pressure-jet units at the tips of the 18.29m two-blade rotor. Unlike that of the Gyrodyne, the rotor had a conventional helicopter control system - with collective-pitch change for varying the overall lift, and cyclic alterations of the blade-angle, by means of a conventional column, to provide fore-and-aft and lateral, or rolling, control.


In the Jet Gyrodyne tip burners, compressed-air from the blowers passed through the rotor blades, while centrifugal force fed the metered fuel through the blades to the jets. The compound, or convertible, helicopter principle is lift for take-off, slow flight and landing was provided by the jet-driven rotor. For transition to cruising flight the compressed air to the rotor tips was progressively reduced and available engine power transferred to the propellers, leaving the rotor as an auto-rotating lift unit which was supplemented (in a very minor proportion for the Jet Gyrodyne) by the fixed wing. The procedure was reversed for a return to the helicopter regime.


The problem for the Jet Gyrodyne was accentuated by, at a gross weight of 2720kg, under-powered for the work it had to perform and the Leonides was normally operated at maximum boost. The Jet Gyrodyne could not quite maintain level flight in the cruising, autorotative, mode. It was not until 1 March, 1955, that a transition was completed by John Dennis. Within four months of this first full transition, the techniques had become well established and transition was no longer being accompanied by a considerable loss of height. The cycle was demonstrated during the SBAC Display at Farnborough in September 1955; while practicing for this demonstration, 65 successful in-flight tip-jet re-lights were accomplished during eight days.


Even during the earlier period of testing, the transition from helicopter to autogyro flight was relatively easy. The basis of the operation was the transfer of engine power from the jet-feeding compressors to the propellers. The pitch of the propellers was progressively coarsened, thus absorbing more power and reducing the air delivery to the jets, which eventually flamed out, and the compressors were declutched. It was found that transition could be made at widely varying speeds, but 128km/h was found to be the most convenient, with a rotor speed of about 210 rpm.

 

Fairey-JGyro

 

The difficult operation was the transition back to helicopter flight. A great deal of flying was required before the best propeller-pitch and tip-jet re-lighting sequence could be established. A major part of the difficulty was that, with the compressors being driven, there was no reserve of engine power for the propellers during the re-lighting sequence, and the aircraft descended rapidly in autorotation until the jets had been re-lit. So long as the re-lighting was being done over or near an aerodrome there was no particular danger in this situation; a controlled landing could be made, and often was made, in autorotation.


As finally established, the drill was to throttle back momentarily, engage the compressor clutch, switch on the tip-jet ignition and fuel supply, and progressively fine-off the propeller pitch. This automatically opened the intake-valves for the compressors. The tip-jets re-lit when a certain head pressure had been reached and collective-pitch was increased to keep the rotor speed down. Propeller pitch was then slowly reduced to zero so that maximum power was available for the blowers to give full tip-jet thrust.


Because of the need to keep the gross weight down with the limited power available, the Jet Gyrodyne normally carried only sufficient fuel for 15 minutes or so of safe tip-jet burning endurance. Extra tanks were occasionally carried under the wings to provide 30 minutes' endurance, but, as already noted, there was no particular danger in fuel exhaustion for an aircraft which could make powerless autorotative landings from a best indicated gliding speed of only 72km/h. The high inertia of the big rotor allowed for a few moments of hovering before touch-down.


By September 1956 the Jet Gyrodyne had made 190 transitions and 140 autorotative landings. The techniques were by then familiar and reasonably well understood, so that there was already a sound basis for the procedures required for the Rotodyne, which was to make its first flight a year later.

 

Fair-JetGyro

 

Six Ministry of Supply pilots each flew the Jet Gyrodyne successfully after about an hour's instruction and practice. All the early test flying with the Jet Gyrodyne, including transition from helicopter to autogyro flight and vice versa in March 1955, was done by John N. Dennis.


At the time when the Jet Gyrodyne was making its earlier test flights the design team was led by Dr G. S. Hislop, chief designer (helicopters), and Capt A. G. Forsyth, chief helicopter engineer, who was responsible, among other features, for the tip-jets, and shared with Dr J. A. J. Bennett (who had by then left the company) the 1949 British Patent on which the Jet Gyrodyne and Rotodyne concepts were based. Later Jet Gyrodyne flight development was carried out by Sqn Ldr W. R. Gellatly and Lt Cdr J. G. P. Morton, who were to take the Rotodyne through its four years of testing.


Although scheduled for scrapping in 1961, the Jet Gyrodyne was rescued and eventually preserved.

 

Fair-JetGyro-ld

 

 

 


Copyright © 2017 all-aero. All Rights Reserved.
Joomla! is Free Software released under the GNU General Public License.