The MoD funded trials objectives would be:
1. Demonstrate the feasibility of the Harrier STO (short take-off) from the ramp.
2. Measure STO performance.
3. Measure the undercarriage oleo closure to verify load predictions.
4. Identify any handling problems (pitch up) from the nose up pitch rate due to the ramp curvature. This was a concern as Harriers had experienced pitch ups after STOs off a flat deck where full nose down control had been needed. The ramp might make this worse. In the event it was found that the pitch down which occurred as the nose wheel fell off the end of the ramp cancelled the pitch up.
5. Achieve limiting performance launch conditions where the aircraft achieved horizontal flight after STO, neither sinking nor climbing.
6. Provide data to support an early decision to fit a ramp on HMS Invincible.
    The standard Harrier STO manoeuvre is:
1. Normal method for runway or ship take-offs.
2. Ground/deck roll at full throttle with nozzles aft.
3. Nozzle rotation to predetermined angle at scheduled speed or deck end.
4. Aircraft leaves ground or deck and is pitched up from 8 degrees angle of attack (AOA) to 12 degrees to safely optimise wing lift.
5. Aircraft weight equals wing lift plus jet lift. Off a ramp aircraft weight would be greater than wing lift plus jet lift until lift is gained when accelerating during the ballistic phase.
    The Ski Jump ramp to be built at RAE Bedford was designed by Kingston’s Ground Test Services. It was 180 ft long, of circular arc profile and built from some forty, 40 ft wide planks mounted on hinged beams supported by jacks and adjustable props allowing the ramp exit angle to be varied from 6 degrees to 20 degrees via 9, 12, 15 and 17.5 degrees. At 6 deg the Harrier pilot could see over the ramp end but not at 9 deg. At 20 deg the ramp end was a formidable 25 ft above the runway.
    Two test aircraft were used. The sixth Development Batch Harrier XV281 had a magnetic tape recording system covering a full range of aircraft handling and performance parameters. Production Harrier XZ136 had two A13 paper trace recorders. Both had head-up display video cameras, undercarriage oleo closure indicators and recorded engine bleed pressure which is related reaction control usage. Ground based instrumentation included ramp end speed cameras, kinetheodolites for trajectory recording, anemometers for wind speed measurement, air temperature thermometers for temperature correction, the Dunsfold F-47 take-off performance camera and a high speed cine camera to cover nose undercarriage behaviour.
    The trials programme, flown initially by John Farley. was cautious and progressive:
1. Measure hover performance to establish installed engine thrust for use in launch planning.
2. Perform RVTO (rolling vertical take-off) alongside ramp from STO start point to give the pilot confidence in the performance of the aircraft.
3. Perform RVTO from ramp for the pilot to become used to the feel of the ramp surface which was a series of short flats created by the planks.
4. Perform light weight STO from the ramp selecting nozzles down as the ramp end disappeared from the pilot’s view.
5. Perform a series of ramp STOs at a range of increasing weights, a range of centre of gravity positions and tailplane trim settings, in zero, head, tail and cross winds, with and without external stores, with the auto stabiliser on and off.
    Risk reduction measures were:
1. Employ the usual incremental approach in test conditions.
2. Ensure performance margins were available by flying with reduced thrust (97% fan rpm) and adjusting the weight to give the desired thrust-to-weight ratio, having water injection available and the option to jettison the water, carrying jettisonable stores, having an AOA margin from the 8 deg normal to 12 deg, having engine limiter override available. All these factors, not all of which were available on every flight, would make a rapid increase in lift possible in the event that a sink should develop.
3. The undercarriage oleo closure boundary was checked with instrumentation and an extra margin was used with non-instrumented aircraft. Closure extent was confirmed by putting grease on the oleo which indicated the maximum compression.
    Once the engine performance was established by the performance hovers, and using well known aircraft ground roll acceleration performance data, launch conditions could be planned to achieve a predicted rate of climb. Examination of the instrumentation data post-flight showed what had actually been achieved thus allowing planning the next launch. Undercarriage closure was also monitored as of course were handling qualities, pitch rate and control margins.
    The flight test programme began with XV281 flying off the ramp set at 6 degrees exit angle allowing AOA to reach 12 degrees then holding. The pilot noted a ‘clunk’ as the aircraft left the ramp. This was the undercarriage legs dropping as they became unloaded and was cured by setting the exit plank at a reduced angle so the leg was unloaded gradually. Otherwise there was nothing untoward so the ramp angle was increased to 9 degrees. The performance improvement continued and the handling qualities were unaltered. At 12 degrees the ‘natural angle’ was found; there were no control demands on the ramp and airborne the aircraft automatically achieved the desired 12 AOA without any pilot action, the sink in the trajectory being just enough. Again, there were no handling problems and the performance gains continued.
    The company two-seater, G-VTOL flew off the ramp at 15 degs which allowed John to demonstrate to other pilots that the ski jump was “the easiest way to get airborne in a Harrier”. The trials continued through 17.5 degrees to 20 degrees where the ramp end was 25 ft above the runway so all you could see, said Dick, who flew with John off most of the angles, was a grey wall with a black stripe down the middle, so the take-off was “a real act of faith that you were going to get up and over it”.
    An interesting unexpected phenomenon was that at higher ramp angles less nozzle deflection was required to prevent the aircraft stabilising at a low air speed and not accelerating to complete the transition. It was concluded that at least 4 knot/second acceleration was needed so nozzle deflections were adjusted accordingly.
    The proven benefits of the ski-jump launch were:
1. Much reduced deck run at a given weight.
2. Increased payload for a given deck run.
3. Need for lower wind-over-deck speeds to war-ship speeds easing ship handling.
4. Increased safety because for all launches the deck is inclined upwards away from the sea even when the ship is bow-down or pitching bow-down.
5. Guaranteed rate of climb.
    For example with a flat deck a 600 ft deck roll allows take-off with a 10,000 lb load at 120 kn end speed. With a 15 degree ramp and a 600 ft deck roll the load is 13,000 lb at 110 knots. With a 15 degree ramp the 10,000 lb load can be lifted from a 200 ft deck roll at 70 kn. All the above are for a 25 kn wind-over deck. A 10,000 lb load equates to full internal fuel plus five 1000 lb bombs.
    The first public demonstration was at the Farnborough Air Show in 19     . The army built a 15 degree ski jump using standard Medium Girder Bridge components. The ramp was assembled on the ground, the entry end was fixed in concrete, the exit end was hoisted by a crane then it was all propped up. This gave a catenary profile which resulted in a much smoother ride. The US Marines were so impressed by the ramp that they bought it; so another one had to be built for the Paris Air Show.
    After many questions Dick was thanked for this excellent talk with lots of illustrations and videos.