The genesis of this article first appeared in the Holbrook Ultralight Club's "Flyer" early in the 1990's with an expanded version later appearing in the AOPA Australia magazine. The author has reviewed and updated the document for publication in Australian Ultralights but asks you to forgive any errors ...
When watching the efforts of someone at the Holbrook Ultralight Club attempting to start the Rotax on a Thruster I recalled similar frustrating attempts to start naval aircraft engines 50 years previously. The major procedural difference then was that the pilot fired a shotgun-like cartridge instead of pulling a cord. Less physically taxing but sometimes equally frustrating.
The piston engines which powered the Royal Navy aircraft of the day were the 12 cylinder, inclined vee, overhead valve, liquid cooled Rolls Royce Griffon in the Supermarine Seafires and Fairey Fireflies and the 18 cylinder, two row radial, sleeve valve, air cooled Bristol Centaurus in the Hawker Sea Fury and Blackburn Firebrand. The 12 cylinder Rolls Royce Merlin powered the twin engine, two seat, de Havilland Sea Hornet night fighter. The Centaurus engine capacity was about 54 litres, the Griffon about 36 and the Merlin about 27 litres. The rule of thumb then was that, in a high performance fighter aircraft, a well installed radial engine required about 1.5 times the capacity of an in-line engine to provide the same aircraft performance. Presumably because of the increased profile and cooling drag, although the in-lines did seem to produce more horsepower per cylinder – and per litre.
The Griffon 88 installed in the Seafire 47 (the ultimate model of the Spitfire in squadron service) developed about 2200 bhp and the aircraft had a top performance around 380 knots at 25 000 feet at the typical service load (full internal fuel and ammunition but no external stores) of 5000 kg. The Centaurus developed 2550 bhp and provided the same top speed with the Sea Fury service weight of 5700 kg. These engines culminated a line of Bristol sleeve-valve engines and a line of 36 litre Rolls Royce engines – the latter included the 2500 hp R engine which powered the Supermarine S6B seaplane – 1931 Schneider Trophy winner. After the trophy run the S6B broke the 1931 world speed record while hauling around a pair of floats! The twin 2000 hp Merlins also drove the Sea Hornet at much the same top speed but at a service weight of 8000 kg
The Griffon 88 was fitted with a two speed, two stage supercharger, gear driven off the crankshaft, with an intercooler between the last stage and the induction manifolds. The engine drove a 6 blade counter rotating propeller. The Centaurus 18 engine in the Sea Fury FB11 was fitted with a two speed single stage supercharger and drove a 5 blade, 3.9 metre diameter propeller. Both engines drove the propellers through gearing which reduced rpm by about 40%, and the blades had a pitch movement variable through 35 degrees.
As you know, in an unsupercharged engine, the pressure in the induction manifold is less than atmospheric until the engine is running at full throttle when the induction pressure is then close to atmospheric pressure – which decreases with altitude. The mechanical supercharger was a radial vane centrifugal impeller [driven from the engine crankshaft through a high speed gear train] which drew the fuel charge in at its centre and threw it off at high velocity into an expanding tube, or diffuser, which converted high velocity into pressure thus boosting the weight of the charge delivered to the cylinders at sea level, and maintaining it as altitude increased. The impeller speed was about 5.5 times engine speed in medium supercharging gear and about 8 times in full supercharging gear. The rotor diameters were about 25 – 35 cm, which means that the outer portion of the impeller would be exceeding Mach1 at maximum engine rpm. The intercooler, if fitted, reduced the high temperature of the compressed charge thus increasing density. The amount of boost is dictated by the throttle opening.
It is interesting to note that the Merlin's horsepower was more than doubled during its service period in operational squadrons. Nearly all this increase was provided by advances in mechanical supercharging. One of the problems with the mechanical superchargers was that they also consume considerable engine horsepower – as much as 300 bhp in FS [fast speed] gear. To overcome this American aircraft engines, like the R-2800, often used turbo supercharging with a notable example being the large turbocharger fitted in the rear fuselage of the P47D Thunderbolt – the engine exhaust, inlet and compressed air return ducts passed through the cockpit. Some installations used an engine exhaust turbine to supply intake air to a mechanical supercharger. The Wright Aeronautical Corporation took another tack by adding an additional system of three turbine power recovery units – each turbine utilising the energy of the exhaust gas flow from six cylinders. The turbines provided additional power – via fluid couplings – directly to the crankshaft of their mechanically supercharged Cyclone 18 cylinder R-3350 engine – of similar capacity to the Centaurus. This Turbo Compound version of the R-3350 increased hp by more than 20%, to about 3200, with only a minor increase in fuel consumption. Later variants of the R-3350 produced even more power but obviously they were – and still are – a very complex engine.
During the mid 1940s Rolls-Royce scaled up the two stage compressor of the Merlins and Griffons, joined a reduction box and prop to the front end and a two stage gas turbine to the back end and created the first Rolls-Royce Dart turbo-prop engine.
Pilot's engine power controls in the aircraft were quite simple. Mixture control was entirely automatic, although some early Firefly marks originally had manual control but were later modified to automatic. Unfortunately the modification did not require removal of the disconnected lever in the quadrant so I fancy a few aircraft were snagged after flight by puzzled pilots. The 2-position supercharger gear change control was usually switched to 'automatic' where an altitude switch effected the change between low and high gear when appropriate. The pilot could set the control at 'MS' [medium speed] in which case low gear was maintained at all altitudes. There was also a means provided for testing high gear operation during engine ground run-up.
The position of the rpm control lever adjusted the operating mechanism of the constant speed unit mounted on the engine. The CSU governed the engine rpm between about 1400 and 2700 in the Centaurus and 1800 and 3000 in the Griffon. The CSU functions by constantly adjusting the blade pitch to variations in power and in load imposed by flight functions. In Griffon engined aircraft the control lever could be pulled back through a gate which allowed rpm below 1800, but I think this might have been outside the governing range of the CSU. The Centaurus allowed an additional facility – pulling the lever back through the gate put rpm control into automatic mode and rpm are then controlled by the position of the interconnected throttle lever. When flying the Sea Fury the rpm lever was set to maximum rpm for take-off, pulled back to auto when airborne and usually left there until rejoining the circuit.
So, for some of these high performance aircraft, mixture, boost gear and rpm control were all automatic and the only engine control the pilot varied in flight was the throttle lever – just like most ultralights.
The throttle lever was connected to the automatic boost control which, in turn, controlled the opening of the throttle butterflies. Manifold pressure was boosted by the supercharger up to 18 lbs/sq.in above atmospheric pressure in the Griffon and 9 lbs in the Centaurus. [American practice is to measure boost as absolute pressure, expressed as inches of mercury, thus manifold pressure 30 in/hg in American engines equals zero boost in British]. The pilot sets the climbing power needed, for example, in the Seafire, maximum rate of climb 150 knots IAS, power +9 lbs and 2600 rpm , and the automatic boost control progressively opens the butterflies in the climb until the full throttle height, for that boost and rpm setting, is reached and boost begins to fall. Shortly after, the altitude switch should change the supercharger to high gear [which comes in with quite a 'clunk'], boost increases, throttle opening is consequently reduced and, as the climb continues, progressively re-opened until a second full throttle height is reached, then boost again falls. While this is going on the pilot sits back but needs to reduce IAS by about three knots per 1000 feet once past 25 000, thus maintaining a near constant TAS until at about 40 000 – 42 000 feet the engine starts to run out of grunt and the aircraft is approaching stall at 160 knots TAS. Airspeed also affects full throttle height, the ram effect of high airspeed helps the supercharger considerably, adding maybe 3000 feet at maximum airspeed.
Continual monitoring of engine temperatures and oil pressure during flight was vital. In a sleeve valve engine the sleeve is between the cylinder wall and the piston, its movement opens and closes the inlet and exhaust ports. Because of the sleeve valve design the Centaurus was prone to damage if the oil pressure dropped below 80 lbs/sq.inch but the normal operating pressure was only 20 lbs higher and the oil pressure in an overheating engine dropped rapidly. The Griffon had a greater range, normal 80 minimum 40. Most of the oil cooling and engine coolant systems were automatic or semi-automatic but the pilot did have to operate the Centaurus engine cooling shutters. The Centaurus was a rather oily engine and fouling among its 36 plugs was common.
The constant speed unit, coupled with propeller blades that had 35 degree variable pitch movement, ensured that the lower limit of 1800 rpm (about 1100 propeller rpm) in the Seafire 47 gave sufficiently coarse pitch to enable descent up to the maximum permissible diving speed of 450 knots IAS at lower levels and 240 knots IAS above 35 000 feet. The procedure for a fast descent – I don't think any pilots preferred a slow one – was switch to low gear, set boost about +4 lbs, rpm 1800, half roll, pull the nose down into the 30 degree dive, aileron turn upright and start gently easing on elevator and rudder trim as airspeed built up – rather rapidly as the Seafire was slippery indeed.
The aircraft was quite stable, once the 30 gallon rear fuselage fuel tank was empty, and it took a lot of stick force to recover from the dive. Winding back elevator trim as you pulled was useful but the trim was both very sensitive and very powerful, overuse could easily result in a high speed stall which was generally extremely punishing. One low cost, low weight instrument notable by its absence was an accelerometer – I suppose the powers that be didn't want to frighten the pilots.
A Griffon engined Seafire 17 has reached Mach 0.88 in trials without the wings twisting off – in a 45° dive from 40 000 feet – becoming the Royal Navy's fastest piston-engined aircraft. Pretty good for a wing that was designed in 1935 and remained unaltered aerodynamically – except for occasionally cropping or lengthening the wingtips. It had just the right combination of thinness and strength, the thickness/chord ratio at the wing root was 13% – very low in comparison to other fighter aircraft, the Mustang for instance was 16%. Also the total weight of external stores (bombs, rockets, long range fuel tanks etc.) that were eventually hung from that single spar1935 wing , plus the extra fuel tanks, four 20 mm cannon – with about 160 rounds of ammunition for each – stowed internally, plus the weight of the wing folding and strengthened undercarriage mechanisms, was astounding. Particularly so when you consider that each mainplane in the first model of the Spitfire weighed less than 200 kg.
The CSU in the Seafire directly adjusted only the front set of 3 blades, the rear 3 blades conforming through a device called a translator bearing. I've forgotten how it worked but on occasion it didn't and if that happened at high speed the rear blades locked in coarse pitch and once airspeed dropped below 140 knots it couldn't be recovered, even with maximum throttle as the rear blades were stalled. Climbing was out of the question, as was going round again. It is interesting to note that over a period of just 7 years the propellers fitted to Spitfires changed from a simple fixed pitch 2 blade wooden prop to that 6 blade contra-rotating metal monster. (The Seafire 47, which entered squadron service in 1948, had twice the power, twice the maximum take-off weight and was 100 knots faster than the Seafire1b [virtually the Spitfire Vb] which entered squadron service in 1942.)
The big propeller made the Seafire 47 very nose heavy on the ground and the pneumatic brakes on naval aircraft were very effective though somewhat grabby, so one taxied very carefully. [Braking pressure was controlled by a lever on the control column spade grip and differential braking was controlled by rudder pedal movement]. I once nosed a Sea Fury over, when taxiing out of dispersal whilst spreading the wings, and the propeller clipped the hard-standing. I was at a satellite airfield and had to fly the aircraft back to base maintenance with the blade tips bent backwards. Something which I guess CASA might frown upon these days. The subsequent landing was prudently fast, and took a lot of runway. I always trusted the marshallers when moving out of dispersal, after that embarrassment. Very recently we were visited at home by the then crew chief for that aircraft. She was responsible for maintenance, re-fuelling and re-arming and, like the other 'Wren' crew chiefs, took a great pride in the airworthiness and presentation of their responsibility. She remembered being annoyed with me over that incident but really fed-up with 'Shorty', who put her charge through a wire fence a few days later, really messing up the paintwork and the loving polish.
The propeller of the Merlin powered Spitfires and Seafires rotated clockwise, viewed from the cockpit. The rotating slipstream pushed against the left side of the fin and rudder, also the torque of the propeller was trying to rotate the aircraft anticlockwise around the propeller shaft and increasing the load on the left undercarriage wheel, consequently the aircraft wanted to swing left during take-off. Also the gyroscopic precession effect of the heavy spinning propeller, when the tail was raised on take-off, also wanted to make the aircraft swing left. [Any applied force which changes the axis position of a gyroscope causes the axis to move 90° to the applied force and in the direction of rotation]. The pre-take-off drill was to wind on full right rudder trim and start moving with right rudder applied. However the Griffon engines in the later models rotated in the opposite direction, for some unknown reason, and imparted an even more pronounced tendency to swing because of the larger propeller and greater torque, but to the right. Reputedly there were some rather amazing departures by bold pilots, flying a Griffon engined model for the first time, who didn't absorb the fine print in Pilots Notes, and who applied full right rudder trim and booted in plenty of right rudder! The Sea Hornet's Merlin engines were 'handed', one rotated clockwise the other anticlockwise.
The Seafire 47 did not have any problems with swing on take-off. The two sets of counter rotating blades produced a straight [non-revolving] slipstream and the gyroscopic effect and torque of one set was completely offset by the other. Thus the aircraft would just about take-off by itself even if you opened the engine up to full boost, +18 lbs. Also it was very directionally stable in a dive – there was no need to keep winding on rudder trim as the speed increased, also, due to the lack of gyroscopic effect, the nose didn't want to rise or fall when a level turn was initiated, as the other models did.
Looking back on those high performance aircraft something that comes to mind is how uncomfortable they were. The Seafires and Spitfires had little or no cockpit heating. Remarkable when you consider the engine coolant plumbing, running to the underwing radiators, which surely could have been utilised for cockpit heating.
The workplace layout would make a contemporary cockpit designer blanch, but I guess the many developments of the Spitfire each required something to be added to the layout so it just grew – and grew. One item that remained through all variants was a small crowbar clipped to the small drop-down cockpit door. Presumably to inspire the pilot's confidence in the emergency hood jettison system?
There were many levers added to the cockpit sides and floor – wing folding and locking, emergency undercarriage and flaps, RATOG carrier jettison, tail wheel locking, drop tank jettison and so on. All these levers seemed to incorporate sharp edges and finger traps, and be cunningly placed in close proximity to often used knobs and switches. Leather gloves were worn primarily to protect fingers from laceration, protection from the cold was secondary – at lower altitudes.
An example of poor ergonomics was the undercarriage control on the right hand cockpit wall. Standard procedure for carrier or airfield take-off was to raise the undercarriage on lift off then jink the aircraft to starboard, thus clearing wake vortices from the deck or runway and making life easier for the next aircraft. Thus during take-off, with left hand on throttle and right hand on control column, the pilot had to change the left hand to the stick and the right hand to operate the undercarriage lever, then perform the jinking manoeuvre, then reverse the hand change procedure to get flap up, power down and retrim. It was also advisable, sometime during the procedure, to squeeze the pneumatic brake lever on the control column to stop wheel rotation before stowage, as tyre clearance from the wheel well structure was sometimes lacking. I watched an aircraft put a wingtip into the ground whilst executing this manoeuvre. Pilot error was the verdict – nothing changes! – but how much did the hand swapping routine contribute?
Another example was placing the bubble canopy jettison 'golf ball' at the front bow of the canopy just above eye level. To jettison the pilot lowered his seat to the full extent (if he had time and forethought) tucking his head down as far as possible while reaching up and pulling down the ball which, via a cable, disengaged latching pins from the guide rails. The pilot then had to push the canopy outwards, off the rails, with his elbows. Hopefully the hood then flew off cleanly without taking any part of the pilot with it. This potentially dangerous design feature remained through all versions of the Spitfires.
Canopy jettisoning was improved somewhat in the Sea Fury. The jettison handle was placed near the pilot's right knee with no requirement to push the canopy off its guide rails. However the canopy wouldn't jettison cleanly at airspeeds below 210 knots and unless the pilot was not tall and had his seat right down the probability of being struck on the head was high, and 'bone domes' were not then in use. As one usually closed the canopy on take-off before 200 feet and then climbed out at 185 knots, canopy jettison, following engine failure, was perilous - as a friend found out the hard way.
The Sea Fury was much more comfortable than the Seafire, roomier, quieter and warmer. The cockpit layout was better planned with shelves on both sides carrying instruments, buttons, switches, etc. grouped in sub-panels. The undercarriage lever was placed on the left side, just below the throttle quadrant, in a position where the left hand naturally dropped from throttle to undercarriage lever.
The pilot wore a 'Mae West' (name derived from the amply endowed film star of the '30s) inflatable life jacket tied with tape across the chest and between the legs, and sat on an inflatable one-man dinghy pack attached by press studs to the parachute pack, the latter being positioned in the aluminium bucket seat. The dinghy pack included a small metal carbon dioxide inflation cylinder – with handwheel – which always seemed to manoeuvre itself into an uncomfortable position. The Mae West was equipped with a small dagger, amongst other strange devices, the prime purpose of which was to stab the dinghy should it come to life during flight.
The pilot climbed into the cockpit wearing the Mae West with the parachute and dinghy pack already in the bucket seat, sat on the dinghy and clipped the dinghy lanyard to a connector on the Mae West. The parachute harness was then tightly secured - four straps, the lower pair passing through a crotch loop, all terminating in a fairly large metal quick release box positioned on the chest. The seat harness was then secured, again four straps – two shoulder, two leg – terminating in another clip box positioned on the solar plexus. There weren't any inertia reels and the harness had to be very tight. The shoulder straps could be set in two positions - locked or free - which allowed reasonable movement.
Next step was to put on the cloth or leather helmet with attached goggles and oxygen mask. The helmet, providing absolutely no protection from injury, contained the headphones and the microphone was in the oxygen mask. The mask tube was then bayonet fitted to the aircraft's oxygen supply system and the microphone and headphone leads plugged into the communications system. If about to do a high altitude run in a Seafire 47 then electrically heated boots and gloves were also plugged in. It seems a lot of strapping and plugging but I can tell you, from personal experience, it can all be undone in a flash when the need arises.
The seat harness was designed to provide protection for deceleration forces up to 14 g. Crash landing or ditching forces are probably around 10g at which level the tightly secured harness, and the pilot's neck, stretch sufficiently for the pilot's head to strike the gyro gunsight mounted about 30 or 40 cm in front of the face. I still bear the scars.
Most aircraft were equipped with a pilot's 'relief tube' but the extreme difficulty of the extraction process made the effort of relief debatable. Immersion suits were provided to wear on flights over very cold waters. These were one piece with integral boots and helmet and made from material which became waterproof when wet. Normal flies/zips were replaced with a plug device which made the relief operation even more difficult. Perhaps it was the immersion suit that gave birth to the expression "in your boot!"?
End of part 1, see Part 2