SOPWITH CAMEL: A REALISTIC JSBSIM FLIGHT MODEL - NOTES Version 2.0 Brent Hugh, 19 March 2017 brent@brenthugh.com PURPOSE In flying the existing Sopwith Camel flight models in Flightgear with Bombable in simulated combat situations, the shortcomings of these flight models became apparent. Flightgear flight models are typically designed and optimized for flying aircraft well within their design envelopes, and work very well in those situations. However, combat aircraft are often operated near the limits--or even over the limits--of their documented capabilities. Particularly for an aircraft like the Camel, working with and against the quirks of its design was an integral part of operating the aircraft in both everyday operations and in combat. These operational quirks were openly exploited by combat pilots. As one writer said: I enjoyed flying the Camel, but its vices of control instability, extreme control sensitivity and pronounced gyroscopic effects all combined to create the impression of balancing an egg on the point of a needle rather than flying an aircraft . . . It was never forced into manoeuvres - they were executed by light pressure on the controls and subsequently relaxing, or sometimes reversing, the pressure once the desired rate of response was attained. I set out to create a flight model of the Camel that would incorporate as many of the documented combat flight characteristics of the Camel as possible, and in as realistic a way as possible. Technical and operations manuals of the Camel are available that give a baseline for its capabilities and operation, and the JSBSim flight model now very closely matches those public specifications. However, extensive wind tunnel testing data, which would give far more detailed data about performance of the aircraft in different situations--and which is often available for more modern aircraft--is not available for the Camel as far as I know. For that reason, I have relied on detailed reports by pilots of historical and modern Sopwith Camels, which give an account of characteristics, quirks, and typical solutions used by pilots, and used these first-hand sources to guide the development of the flight model. CAMEL FLIGHT CHARACTERISTICS IMPLEMENTED IN THE FDM Camel flight characteristics are documented below and in the included files. Below is a list of Camel flight characteristics implemented in this JSBSim flight model, with a rating of the effectiveness of the current implementation on a scale of 1-5: 1 - Not implemented 2 - Partially implemented 3 - Well implemented 4 - As well implemented as possible, as far as I (a non-pilot) can determine 5 - As well implemented as possible, according to an expert Camel pilot 4 Weight, fuel capacity & weight, hp, dimensions and other technical details implemented as described in attached article, 'The Sopwith "Camel"', The Automobile, Nov 7, 1918. 4.2 Best available calculations for moment of inertia of engine/propellor, prop thrust, moment of inertia of aircraft wings and body. See included file Camel-JSBSim-calcs-2011-10D.xls. Note that value for gyroscopic moment as measured in FG matches quite well with values as measured in an actual Camel in Calculated Sopwith Camel article (see Docs folder). Flight testing in FG shows that changes from that point even at +50-100% are more in the line of tweaks with a noticable affect on flight performance than complete game-changers. Just installing a different engine (heavier, lighter, different distribution of weight) could have changed gyroscopic effects 10-40%. NOTE: These results could/should be verified and improved by actual measurement and testing of an aircraft, by wind tunnel data, or by extensive testing and feedback of the model by an experienced Camel pilot. 4.2 Although the gyroscopic affects are noticeable, under normal flight conditions and sufficient speed, they are well within the capability of the rudder and other flight surfaces to overcome. (See Calculated Camel article.) 4.8 Speed, RPM, horsepower of climbing and horizontal flight closely matches published number. See for example the speed/climb rate graphs comparing the JSBSim Camel and a historical 1917 Camel in the Docs subdirectory, file Camel-JSBSim-calcs-2011-10D.xls Real Sopwith Camel: 115 mph @ 6500 ft 113 mph @ 10000 ft 106.5 mph @ 15000 ft (All True Airspeed (TAS) Climb to 6500 ft = 6 min 0 sec 10000 ft = 10 min 35 sec 15000 ft = 20 min 40 sec JSBSim Camel 2.0: 100.5 kts @ 500 ft - 115.7 mph 102.1 kts @ 6500 ft - 117.5 mph 99.6 kts @ 10000 ft - 114.6 mph 94.4 kts @ 15000 ft - 108.5 mph So, each of these values is within 2.5 mph of the historically measured value. Climb to 6500 ft = 6 min 47 sec 10000 ft = 10 min 19 sec 15000 ft = 20 min 19 sec The climbing times, as well as the speeds at various altitudes, are a pretty reasonable match given that each individual aircraft varied in its precise specifications, and we are not certain of the precise protocol the test pilots followed in measuring these values. Values vary 1-2 mph in either direction in any given test, even in JSBSim. So the numbers we see here are well within the measurement error bars of the historical data. I'm assuming that airspeed numbers given in historical documents above are "true airspeed": http://wiki.flightgear.org/Aircraft_speed The JSBSim speeds given are also true airspeed ("vtrue"--on the property tree at /fdm/jsbsim/velocities/vtrue-kts). So don't expect those exact values to show up on your airspeed indicator, unless you are at sea level. 4.0 Max sustained turn rate about 76 deg/sec, ie 360 degrees in 4.7 seconds (near stall speed, level flight, sea level). NOTE: Current best time in FG for 360 degrees is about 8-10 seconds; this matches reported times by Camel pilots of the time and the 4.7 sec/76 deg/sec might not be the most reliable number, or possible only for short bursts, not for a complete 360 degree level turn? In 2.0 we're getting 52 deg/sec, as calculated during a level turn just before stall, presumably similar to the test discussed here. The limiting factor is the drop of wing during stall & a faster max turn rate could be obtained by weaking that behavior. However, that conflicts with known data about max alpha degrees of the wing profile. 4.2 Able to turn through 360 degrees (including roll in/out) in 8-10 seconds as reported by WWI era Camel pilots. 2.0 is very close to this--about 10.0 seconds for right & left 360 degree turns. This could potentially be brought more to 8 seconds by tweaking the wing-drop on stall behavior a little. 4 Flown basically with rudder and elevator; rudder initiates turns etc and ailerons only used as assist; many pilots didn't use ailerons much. 4.2 Quick/light response to controls 4.3 The gyroscopic effect causes the nose to rise in a left-hand turn (moreso that most other similar aircraft) 4.3 The gyroscopic effect causes nose to drop in a right-hand turn 4.3 Fairly large amounts of left rudder were needed in both L & R turns to correct 4 Sluggish in left turns. Quicker through 270 degrees to the right than 90 degrees to the left. See Calculated Sopwith Camel article. It is debatable what this means--quicker to roll? Or to turn once rolled? Flat, level turn or something else? Are there little idiosyncratic ways to turn one way or the other that that pilots often used? In 2.0 a split-S right to turn 180 degrees or even a bit more is noticeably quicker to the right than the left, for instance. 4.1 Known for stalling/spinning on take-off if inattentive piloting let nose rise during initial LH turn. 4.2 Sensitive in a turn, if the turn were tightened just a little, it was likely to whip into a tight spin. 4 All the weight of the engine, guns, and pilot was concentrated in the first seven feet of a short 18-foot fuselage. 4.2 Spins: Period of 1-2 seconds. 4.2 Loses 150-200 feet per revolution while spinning. In 2.0, one trial had 2.0 seconds per rotation and 418 feet of altitude lost per revolution. Another trail had 1.6 seconds per rotation and 175 feet of altitude lost per rotation. These are pretty clearly in the ballpark of the actual numbers. 4.2 Exit spin by gentle forward pressure in the stick, then *gradually* pull out. AC will *sometimes* self-recover from a spin, but many times not. 4.2 Flat spins - prone to entering these if the pilot pulls too hard during a turn. 3.8 General spin characteristics similar to Camel; ability to do spin-related maneuvers documented for the Camel--snap rolls, very quick RH Split-S turns, etc. Note that 'snap rolls' as often described in relation to the Camel appear to not be true aerobatic snap rolls as we would perform today (too vigorous for the Camel's light airframe) but a kind of roll coordinated between rudder & aileron. 4.1 Possible to fly inverted. Stalls at about 65 mph/56 knots when flying inverted. In ver. 2.0, inverted stall is almost exactly 56 knots. 4.0 In inverted flight, possible to stall and get into an inverted spin. In ver. 4.0 Camel's elevator was powerful and sensitive. 4.0 Rudder was too small and relatively ineffective. Per Calculate Camel, it is however very sufficient to overcome gyroscopic forces given normal flight and sufficient speed. 4.2 Straight/forward flight requires slight right rudder at all times. 4.1 Roll for turns initiated by rudder (due to large adverse yaw/drag of ailerons); ailerons use to control roll in turns rather than initiate. 4.1 ~2.5 seconds to roll from level to 60 degrees either left or right (data from Calculated Camel article). Ver. 2.0 takes 2.6 seconds left & 2.1 seconds right. 4.4 Accelerating the aircraft caused the nose to climb, and swing to the left. 4.1 At 110 miles per hour, there was about 15 to 20 pounds of back force on the control column, which could not be trimmed out because the aircraft had no pilot-adjustable, moveable horizontal stabilizer. (Though in FG you can trim it out--and to avoid arm cramps, it a probably a good idea to do so.) 4.2 Moving the stick forward to enter a dive caused the Camel to yaw left because of gyroscopic force, and correcting this yaw with right rudder caused the nose to pitch down sharply. Note that this is more noticeable if near stall speed. 4.1 Steerable tail skid of the Camel helped to overcome a slight swing to the left as I opened the throttle. (The steerable tail skid seems to be only a feature of replica Camels, not ever used during WWI. However it is implemented in FG as otherwise there is no way to maneuver the A/C on the ground. In WWI ground crew would have been used instead.) 4.2 The tail came up at about 20 knots. As of 2.0 is it more like 35 kts and it is very difficult to handle once the tail is up, even at 35 kts. (TODO) 4 Lifts off about 35-40 knots 4.2 Able to take off & land in a very short distance, especially on a field and with a headwind--see for instance: http://www.youtube.com/watch?v=VT9wtDNiKaI With a 15 mph headwind, JSBSim Camel now matches this takeoff almost exactly. The rollout matches pretty well, too--though obviously much depends on the exact surface and other details. 4.0 Right rudder needed to keep AC on course early in roll-out. 4.0 Left rudder needed later in roll-out as speed increases. NOTE: The JSBSim current requires R rudder on takeoff, not left, and not quite full rudder. Not certain of the reason for this. Both of the above effects are dependent on wind direction and other details of the takeoff. But in windless conditions, both are quite noticeable. The steerable tail drag in the JSBSim Camel changes some of the characteristics of early roll-out. For example, some Camel pilots would start roll-out with full rudder applied, gradually releasing it as speed and control authority increased. This isn't possible in JSBSim simply because the steerable tail drag exerts too much control. 3 Rudder loses effectiveness as soon as it drops on landing. 4.5 In a wind of 10 to 15 knots you are airborne in a couple of plane lengths at 35 mph. 4 Rate of climb of almost 1,000 feet a minute climbing out of take-off. 4.5 Capable of looping, 110 mph required. 4 Being slightly tail-heavy it goes up and over in an incredibly small circle in the sky, and faster than any other WWI aircraft I have flown. 4.2 Tail heavy with full fuel 4 Slightly nose heavy with empty fuel. 4 Ineffective ailerons: Slow roll/ailerons only with level flight requires 23 seconds to complete--if it can be done at all (it requires a special technique). This is reported in a radial engine replica; it is duplicated in Ver. 1.4-2.0 almost exactly, with the added feature that roll to the L is noticeable faster/easier than roll to the R, due to large torque of Clerget rotary engine. 4 Ailerons induce 'awe-inspiring' opposite yaw (ie, yaw to the R on L roll) and much drag. 4 Starting at level flight, adding full left aileron just moves the nose to the right and it stays there. You don't start rolling right until you coordinate with the rudder as well. (And ditto for full R aileron.) 4.1 In level flight at 100 mph indicated, just a hint of rudder is required for straight flight. 4 Thin airfoil shape affects lift, drag, AOA of stall, etc., and the effects are quite different than for later aircraft with thicker wings. Stall speed: Wikipedia and several other sources report 48 mph/41.7 knots. However, this does not match up with the Camel's reputed quick turn rate. Lower stall speed directly equates with quicker turn rate. A replica Camel reports a stall speed under 40 MPH (35 knots) and Frank Tallman reports stall speed of 35-40 mph (30.4-34.8 knots). A stall speed in this range seems to fit better with the Camel's reported turn rate. The JSBSim 2.0 Camel has a stall speed in the 35-40 mph range (30-35 kts). 4 Current (Ver. 1.4+) drag/lift uses tweaked historical lift/drag curves for the Camel's wing shape and also the known historical zero-lift drag coefficient for the Camel. 3 P factor - evidently the shape of the prop creates a somewhat higher P factor than for later similar aircraft, but the strength of this effect is an open question. 4.2 As you slow down over the top of the loop you must feed in rudder against the torque. 4.1 On strafing runs, as soon as the airspeed reaches 130 to 140 mph the nose begins to hunt up and down, and the elevator becomes extremely sensitive. I feel this action is due largely to the square windshield between the two Vickers guns, causing a substantial burble over the tail surfaces. 3.9 "In stalls at 35 to 40 mph the nose drops frighteningly fast and hard to the right, but you also get control back quickly, although a surprising amount of altitude has been lost. I have had the pleasure of limited dog fighting with other WWI fighters, and there are none that can stay with a Camel in a turn." JSBSim model stalls almost exactly as described, but nose tends to drop down and to the left, rather than to the right. This is in large measure because of the gyroscopic effect of the rotating engine. Particularly when the engine is off or engine RPMs are low, it can drop to the right OR left. This is a detail first-hand report, but it is difficult to know the exact configuration--and particularly the engine--of the aircraft flown. Some configurations or engines may tend to drop the nose to the right in a stall. 4 "Very quick half-roll (a touch of left rudder for a half roll, then a quick half-loop was a often-used maneuver to reverse direction)." In Ver. 2.0 the quick half-loop is possible in either direction, though both have certain quirks the pilot must deal with. 4 Twin Vickers machine guns, 400 round capacity (historically 250-400 seems to have been the range), best known realistic mass, ballistic, and aerodynamic characteristics for rounds, firing rate, characteristics of tracer rounds, etc. Historically realistic re-loading scenario (requires landing, full stop, engine off). See camel-alternative-submodels.xml for information and documentation. NOTE: Aerodynamic characteristics of machine gun rounds could be tweaked if better data were available; tracer visuals could be made more realistic. 2.5 Mixture needed constant manual adjustment; first adjustment about 250 ft above ground on liftoff. For highest realism, switch on manual mixture using the Camel menu. TODO: Mixture and throttle adjustments are not as tweaky as described by Camel pilots. Mixture lever or dial isn't implemented; a simple Flightgear menu displays current mixture & RPM when manual mixture is enabled. So we have a mixture control but it is rudimentary and doesn't seem to mirror the way mixture in a real Camel would work. 4.3 Control of engine thrust via blip switch and magneto rather than throttle. NOTE: Pilots can simulate the realistic historical way of controlling the Camel's engine using this method: Generally manipulate engine power through use of the magnetos and blip switch; use throttle only for a bit of fine tuning on top of that. In Ver. 2.0, engine power with one and with two magnetos engaged is very close to that described in technical specifications: Operating with the left magneto creates about 4/9 max RPM. Right magneto gives 5/9 RPM. This gives 3 basic power settings--each can be fine tuned further via the throttle. This is similar to the way (some models) of Camel engines operated, via a combination of throttle, mixture, magnetos, and blip switches. Other, more limited, models had only magneto and blip power control options with no throttle at all. 4 Engine power with both magnetos or one magneto matches historical engines. We now have three distinct settings--some research & testing may be required to determine whether the power output of each setting matches the historically accurate amount. 4 Engine start. Camel engines were notoriously difficult to start. Given the manual engine starting methods of the day, they must have started either with a quick crank or not at all. NOTE: This is modeled by an approx. 0.2 second cranking period, initiated by the 's' key, which starts the engine about 1/4 of attempts. Engine can also be started by clicking the propeller with the mouse. 4.2 Engine runs out of fuel/stops when inverted or pulling negative Gs. On slow roll (23 seconds), engine stops just after wings vertical and starts again soon after regaining erect flight. Fully implemented in Ver. 2.0. 4 Service ceiling 19,600 ft. Tested in ver. 1.7, by the standard of highest altitude able to sustain 100 fpm climb (1.66 fps). 3.9 The 130HP Clerget Camel was generally lower performance/underperforming for altitudes greater than 10,000-12,000 feet. This is implemented, but could be tested and tweaked further to match exact historical performance. SPECIFICATIONS Cruising speed for Camel with 130 HP Clerget, per SopwithCamelSpecs-1955.pdf - Flight, "Sopwith Camel", 22 April 1955, is 115 mph @ 6500 ft 113 mph @ 10000 ft 106.5 mph @ 15000 ft Climb to 6500 ft = 6 min 0 sec 10000 ft = 10 min 35 sec 15000 ft = 20 min 40 sec Sopwith Camel Vs Fokker Dr I: Western Front 1917-18 By Jon Guttman, p. 25, gives these values: 112.5 mph @ 10000 ft 106 mph at 15000 6500 ft = 6 min 0 sec 10000 ft = 10 min 35 sec 15000 ft = 21 min 5 sec We're assuming that all airspeed numbers above are "true airspeed": http://wiki.flightgear.org/Aircraft_speed - so don't expect so see them show up on your airspeed indicator (except at sea level). DOCUMENTS A number of relevant documents and notes, including facsimiles, articles, and calculations, are included in the Docs folder of this distribution. Below is a compilation of notes and comments by pilots with experience flying the Camel. --- Sopwith F.1 Camel; Clerget 9B; Wg 1 482 We 962 b 28.0 18.8 S 231 Wg/S 6.4 Wg/Po11.4 Vmax 105 (at 10,000 ft) Vs 48 Cd0 0.0378 f 8.73 A 4.11 L/Dmax 7.7 http://www.hq.nasa.gov/office/pao/History/SP-468/app-a.htm --- Based on the information contained in appendix II of reference 100 for the later Sopwith Snipe, the gyroscopic action of the engine caused a nose-up moment in a left turn and a nose-down moment in a right turn. Accordingly, left stick, a large amount of left rudder, and moderate back stick were required in a steep left turn; too much back stick caused the aircraft to stall and spin. Right stick, a moderate amount of left rudder, and full back stick were required in a steep right turn. There seems little doubt that these odd control techniques could cause confusion and indecision on the part of an inexperienced pilot. 100. Penrose, Harold: British Aviation. The Great War and Armistice - 1915-1919 (London: Putnam & Co., 1969). http://history.nasa.gov/SP-468/ch2-2.htm --- R&M 592 Report on Accidents to Certain Aeroplanes with Special Reference to Spinning - Aeronautical Research Committee *Note: This report refers to the Camel as "A"* II. BEHAVIOUR OF "A" 4. During the course of the investigation the Committee had the advantage of evidence on the behaviour of various aeroplanes from fourteen pilots with wide flying experience. 5. On many of the most important points their opinion was unanimous. Where differences occurred the contradiction was more apparent than real, and further evidence showed that such difference did not affect the actual manoeuvre. For instance, some pilots make use of the lateral control, but an agree that this does not materially affect the spinning of an aeroplane. 6. All the pilots agreed that the machine A had certain peculiarities which may be summarised as follows :* GENERAL CHARACTERISICS. 7. (a) Stalling.-A as usually rigged is tail heavy and longitudinally unstable. This is accentuated when a light Le Rhone engine is substituted for a heavier Clerget and/or where guns and ammunition are removed. During normal horizontal flight, it is necessary to keep a continuous forward pressure of about 14 Ibs. on the control column. Any relaxation of the pilot's effort is therefore likely to end in a stall. Because of the longitudinal instability of the aeroplane engine failure will be followed by a stall unless the pilot dives the machine. 8. (b) Turning.--Most of the pilots, in the course of their evidence drew attention to the fact that a steeply-banked right hand turn requires full left rudder. This is accounted for by gyroscopic effect. 9. The unsymmetrical forces and couples on an aeroplane arise from engine torque, slipstream and the gyroscopic couple. The first two, torque and slipstream, may produce an unsymmetric setting of the ailerons and rudder respectively. They do not produce effects which depend on the rate of turning to any appreciable extent. The gyroscopic couple alone is capable of explaining the effect mentioned above. 10. The propeller, viewed from the pilot's seat, turns clockwise, and when the aeroplane is turning to the right it tends to put its nose down as a consequence of the gyroscopic couple. The tendency is countered by left rudder and back (or top) elevator. Turning to the left, the nose of the aeroplane tends to go up, and this effect is countered by left rudder and forward (or bottom) elevator. From the evidence of the witnesses it is apparent that practically all flyers feel strongly the need of a larger rudder surface. 14. (d) Looping.-In the case of A all violent use of the controls must be avoided. With the stick pulled too hard back, the machine will fail to complete the loop. Further, if a straight loop is required, left rudder must be used to counteract gyroscopic effect, and it was stated by one witness that the rudder was barely sufficient for this manoeuvre. 15 (e) Spinning.-The general method adopted when getting into a tight hand spin is :* (i) Cut off engine. (ii) Pull back control stick. (iii) Put on right rudder. This describes the action usually taken, but the essential feature is that the machine is slowed down until it stalls, when the normal sequence is a spin. 16. There is general agreement as to the position of the controls when in a spin. The essential feature is that the stick is kept hard back; the position of rudder and ailerons is relatively unimportant. 17. The normal method of coming out of a spin is with the control stick slightly forward and other controls neutral. In this position the spin will soon stop, out the nose of the machine must be held down until flying speed is attained, when the stick should be pulled gently back, and the machine is flattened out. In the case of an emergency, the spin can be stopped more rapidly by putting the stick forward and reversing the rudder, but in this case a violent manoeuvre is induced, and want of skill may lead to the machine dropping back into the reverse spin. 18. It appears to be established that the rotation cannot be reversed without an intermediate period, during which the machine is re-stalled. This does not require the fuselage to be brought to a horizontal position. 19. As usually rigged, the aeroplane A will not come out of the spin with hands off, and will only come out slowly, if at all, with all the controls held strictly central. 21. There is no aeroplane at present in use which spins faster than A; its time period is 1 to 2 seconds, and the loss of height per turn 150 to 200 feet. 23. Accidents due to spinning.-The Committee have considered 41 accidents to machine A due to stalling and spinning. These accidents can be divided into two groups :* Group (i) includes accidents due to stalls at low level. Group (ii) includes accidents due to spins continued to the ground from a high level. A classified list will be found in Appendix IV. 24. On all aeroplanes loss of control occurs when stalled. At a low level this results in a nose dive or spin, and an accident ensues as lack of head room precludes the completion of the manoeuvre before the ground is reached. Accidents in group (0 belong to this category. The common feature is that the stall was unintentional, the primary cause of the stall varying (e.g., engine failure, faulty turn, general inexperience). 38. In cases where previous dual instruction on A is impossible the movements necessary in a spin must be impressed so forcibly on the pupil that they will return to his mind subconsciously when the difficulty arises. The routine method is : (i) Stick forward and central and rudder central. (ii) Wait for the machine to acquire flying speed. (iii) Pull out gently." Source: http://www.theaerodrome.com/forum/aircraft/56275-question-ways-get-killed-sopwith-camel-4.html --- Max sustained turn rate is about 76 deg/sec. That is near stall speed, level flight, at sea level. p. 7 A graph of speed vs climb rate is given for the Camel and four other aircraft from the same era (p. 7). http://home.comcast.net/~clipper-108/AIAAPaper2005-119.pdf Not su --- "I lived its quick response - it was so remarkably swift and sensitive that it demanded constant attention. The torque caused the nose to rise in a left-hand turn and drop in a right-hand turn. Fairly large amounts of left rudder were needed in both turns to correct for these idiosyncrasis. The tiny biplane was so sensitive in a turn, however, that if the turn were tightened just a little, it was likely to whip into a tight spin - quickly and without warning. Oddly this very ease for spinning was used in combat by many pilots to shake a persisten German from their tails. Sopwith Camel vs Fokker Dr 1: Western Front 1917-18 by Jon Guttman, Harry Dempsey, quoting from mark Curtis Kinney's memoir, I Flew a Camel --- Here is a description about the Camel's traits from BARKER VC, (via author Wayne Ralph at http://www.theaerodrome.com/forum/2001/11306-sopwith-camel-flight-characteristics.html ): Quote: All the weight of the engine, guns, and pilot was concentrated in the first seven feet of a short 18-foot fuselage. The various models of rotary engine fitted to the Camel, from a 110-hp Le Rhone, to a 130-hp Clerget, to a 150-hp Bentley, had the propeller attached to the engine crankcase and the crankshaft to the aircraft so that the propeller and engine revolved together at more than 1250 rpm. This heavy whirling engine was lubricated continually by castor oil which was not recycled, but rather was burned in the combustion process and vented overboard, soaking the aircraft, the pilot's flying suit, helmet and goggles. The gyroscopic forces that were generated varied in intensity based on a complex interaction of engine rpm, aircraft speed, and control input. The scout had a different character turning left or right, as did all the rotary-engined aeroplanes, but the Camel to an extreme degree. It was sluggish in left turns and the nose always pitched up, while right turns were very quick, with the nose dropping sharply. Without plenty of top (ie, left) rudder to correct the pitch down, it would spin. A Camel pilot had to apply left rudder turning left or right, the amount varying with aircraft speed and engine rpm. Since the engine rotation countered the input of left stick and rudder, some pilots made all turns to the right because the aircraft was quicker through 270 degrees, than 90 degrees to the left. The Camel's elevator was powerful and sensitive, while the rudder was too small and relatively ineffective. Coordinated turns demanded a fine touch. Correcting sideslip, the tendency of the aeroplane to slide to the inside of a banked turn, by applying opposite rudder, caused the Camel to tighten up in the turn. In steep turns over 45 degrees of bank, the aircraft tended to pitch up its nose, and tighten up its bank angle as speed reduced, requiring continual adjustments in the amount of forward pressure on the stick with engine rpm changes. In the words of test pilot W/C Paul A. Hartman, RCAF, the Camel has no 'dynamic longitudinal stability' in steep turns. Accelerating the aircraft caused the nose to climb, and swing to the left. At 110 miles per hour, there was about 15 to 20 pounds of back force on the control column, which could not be trimmed out because the aircraft had no pilot-adjustable, moveable horizontal stabilizer. Therefore, Camel pilots had to fly two-to-three hour missions, continually applying forward pressure just to maintain level flight. Letting go resulted in the aircraft pitching up, and rolling inverted to the right. Moving the stick forward to enter a dive caused the Camel to yaw left because of gyroscopic force, and correcting this yaw with right rudder caused the nose to pitch down sharply. By today's standards the Camel was completely unacceptable, and yet many pilots of 1917-8, most with under 100 hours of flying time used it as a lethal weapon. --- http://www.theaerodrome.com/forum/2001/11306-sopwith-camel-flight-characteristics-3.html In those notes are a description of when and why the mixture had to be adjusted. If memory serves, it was perilously low to the ground, under 250 feet. The shift to left rudder would have to have taken place either just before or just after this, because the whole point of changing the mixture was that the engine would have been on the verge of choking, which means that you'd be trying to maintain flying speed. My guess is that the shift would have to happen within 10-20 seconds of getting airborne under normal circumstances. --- "...all flights of these early machines were conducted from the grass surface of the airfield, not the paved runways, and the steerable tail skid of the Camel helped to overcome a slight swing to the left as I opened the throttle. The tail came up at about 20 knots. As the speed reached 40 knots I eased the stick back slightly and the Camel became airborne. It accelerated rapidly to 55 knots and I held it in a climb at that speed until I reached 500 feet. I leveled off, leaned the mixture, and started a left hand turn. The gyroscopic effect from the rotary engine quickly became apparent... ...I enjoyed flying the Camel, but its vices of control instability, extreme control sensitivity and pronounced gyroscopic effects all combined to create the impression of balancing an egg on the point of a needle rather than flying an aircraft...it was never forced into manoeuvres - they were executed by light pressure on the controls and subsequently relaxing, or sometimes reversing, the pressure once the desired rate of response was attained. By modern standards of stability and control, the Camel would be totally unacceptable as a military aircraft... (quoted from Canada's National Aviation Museum - Its History and Collection, by K.M. Molson) http://www.theaerodrome.com/forum/2001/11306-sopwith-camel-flight-characteristics.html --- "As I opened the throttle I simultaneously applied full left rudder and as I became airborn I found that I had full left rudder on. (This was the answer always, as I afterwards found, and I was never in trouble again from this cause)" August 98 issue of Flight journal has the flying of the Camel by Rich King. 160 Hp Gnome engine (yes, rotary), quote: Right rudder is needed to keep the Camel tracking straight ahead while it is on the runway, and as the tail comes up, I find that my rudder correction was right on. I m facing straight down the center of the runway.........A little back pressure on the control stick and the bumpy ground falls away behind me, and the Camels speed continues to increase at an exhilarating rate. A little forward pressure on the "closh handle" hold the aeroplane a few feet above the runway while its speed continues to increase. .... Left Rudder is now needed to keep the racing Camel straight....." end quote http://www.theaerodrome.com/forum/2001/11306-sopwith-camel-flight-characteristics-2.html --- From the Frank Tallman book "Flying the old planes"... My Sopwith Camel is, as far as I'm aware, the only original World War 1 Camel ever brought back to flying condition. It was originally owned by Colonel Jarrett of the Jarrett War Museum located on the old Steel Pier in Atlantic City; who in the 1930's had the best museum of WWI equipment ever assembled, including the Belgian War Museum in Brussels. The Jarrett Museum fell on hard times following WWII and, with time and money on my side after my service period, and a lifelong ambition of owning a WWI aircraft, I purchased for a small sum (by today's standards) several antique aircraft including the Camel, a Nieuport 28, a Pfalz D.XII, a Fokker D.VII and a SPAD VII. The Camel was the first WWI aircraft I brought back to flying condition and required some major rebuilding, which took several years, many thousands of dollars and a whole host of experts including Paul Poberzney of the Experimental Aircraft Association, the gifted master craftsman Ned Kensinger, the Hawker Siddeley Group and a number of very dedicated volunteer's. (NB: There is quite a bit on the rebuilding process in the Chapter but I have left it out of this extract for the sake of space). When the day finally came to fly the air was filled with great anticipation. On arrival at the airport though I was dismayed to hear from my team that they had been trying to get the temperamental 110 h.p. Le Rhone started since 8.00am that morning, without success. The lack of knowledge amongst us regarding the Le Rhone was appalling. Did we have spark? Yes. Was the mag set? Yes. Had the commutator ring been wiped off? Yes. Had we primed it? Only every other cylinder. With only a vague notion of what I was doing I clambered into the cockpit (a very tight fit) and reviewed the cord-wrapped Spade stick, the Block tube, carburettors next to one's knees, the flexible air intake to the outside air scoops, the wood wire brace longerons, the instrument panel with it's clutter and the duel control cables to the wooden rudder bar. At my request, the crew forced open the intake valves as the engine was pushed through (switch off) and shot a charge of fuel in each cylinder, as the cylinder came in front of the hole in the cowling. By accident, rather than by knowledge, I advanced the long lever controlling the air, and in pushing the manet (a small wheel knob on the miniature control quadrant) forward and then returning it, I had hit on the correct starting procedure. Wonder of wonders, as I flipped the porcelain-mounted switch up and called for contact, the Le Rhone started with a full-throated bellow, scaring both me and the crew! By shoving the fuel-controlling lever forward and using the coupe (cut-out) button on the stick, I was able to keep the engine running. Soon the never-to-be-forgotten smell of castor oil infused our area, and the sight of oil splattering the leading edge of the low wings indicated that the engine was lubricating properly. Taxiing practice ended ignominiously a hundred feet from the starting point, when my newfound knowledge wasn't equal to the delicate adjustment of fuel and air, and the Le Rhone quit. The revitalised ground crew hauled the 900 pound airplane over the grass and faced me into the wind. For safety sake we changed the plugs, and the Le Rhone started first try. I headed down the field with the throttle wide open. The tail came up almost instantly, and visibility was good, except for the Aldis sight and the twin Vickers. Not having planned on flight it came as something of a shock to find the Camel airborne at about 35 mph after a ground run of just 150 feet. Being afraid of jockeying with my ticklish fuel and air controls I stayed low and just got used to the Camel's sensitive ailerons, elevators and rudder. I circled the field once, got into position for landing, shut fuel air and switch off, and made a light forward slip, touching down gently on three points. Total landing couldn't have been much longer than the initial take-off run. So much for my first (unintentional) flight in the Camel. Since then I've spent more time flying the Camel than any of the other historical aircraft in our collection. I've also had more forced landings in it than all the rest of the WWI aircraft combined. It's that temperamental Le Rhone. Cylinders have blown, magneto's have failed, even fouled spark plugs have brought me down unceremoniously, with sweating hands and my heart in my mouth, desperately seeking a patch of open ground on which to land. Yet for all that it's the one I turn to first for any show or exhibition, as the Camel gets my blood going like no other. This is an aircraft that is a joy to fly. With the Le Rhone 9J, you cannot adjust either the fuel or air intake without running the risk of a dead-stick landing. You must leave them alone and use you Coupe (cut-out) button for all fight handling. The take-off run is easy. In a wind of 10 to 15 knots you are airborne in a couple of plane lengths at 35 mph and climbing out at 60 mph, with a rate of climb of almost 1,000 feet a minute. The elevators are sensitive, as is the rudder. Consequently, when fling for any distance I often put the heels of my shoes on the floor tie wires, because the vibration of the Le Rhone through the rudder bar exaggerates the rudder movements. In level flight at 100 mph indicated, the Camel is delightful, with just a hint of rudder being required for straight flight. The structure is rugged enough to feel comfortable in loops, and being slightly tail-heavy it goes up and over in an incredibly small circle in the sky, and faster than any other WWI aircraft I have flown. Sneeze and your halfway through a loop before your aware of what's happened. 110 mph is enough to carry you through, and as you slow down over the top you must feed in rudder against the torque. In military shows I have ground strafed, and as soon as the airspeed reaches 130 to 140 mph the nose begins to hunt up and down, and the elevator becomes extremely sensitive. I feel this action is due largely to the square windshield between the two Vickers guns, causing a substantial burble over the tail surfaces. Turns are what the Camel is all about. Turning to the right with the torque requires the top rudder to hold the nose up, and the speed with which you can complete a 360-degree turn is breathtaking. Left turns are slower, with the nose wanting to rise during the turn. But small rudder input easily keeps the nose level with the horizon. In stalls at 35 to 40 mph the nose drops frighteningly fast and hard to the right, but you also get control back quickly, although a surprising amount of altitude has been lost. I have had the pleasure of limited dog fighting with other WWI fighters, and there are none that can stay with a Camel in a turn. With the Le Rhone being temperamental as it is, flying the Camel is best done at times when there are few other aircraft in the sky, leaving easy access to the airport in cases of emergency. The Camel touches down easily but runs out of rudder control almost instantly, and if you bounce your landing at all, you are likely to find yourself in a hairy ground loop looking at a rapidly bending aileron dragging in the grass. For a wide variety of reasons, the Camel is a fascinating airplane, flight-wise as well as historically. But don't think I ever got out of the Camel after being airborne even in the coldest weather without buckets of perspiration and considerable gratitude that I had gotten the little girl home again without breaking her into splinters! http://www.theaerodrome.com/forum/aircraft/45869-sopwith-camel-myth-6.html --- Additional comments on Camel handling characteristics by Victor Yeates, author of "Winged Victory" (yes, it's a fictional novel, but his experiences of flying a camel were realistic, and are a valid, honest description)... Re training: "Camels were wonderful fliers when you had got used to them, which took about three months of hard flying. At the end of that time you were either dead, a nervous wreck, or the hell of a pilot and a terror to Huns . . ." Re turns: "And in the more legitimate matter of vertical turns, nothing in the skies could follow in so tight a circle..." Re the half-roll (Split S): "The same with the half-roll. Nothing would half-roll like a Camel. A twitch of the stick and flick of the rudder and you were on your back. The nose dropped at once and you pulled out having made a complete reversal of direction in the least possible time. Thomson, the squadron stunt expert told him that it (half-roll) was just the first half of a roll followed by the second half of a loop; the only stunt useful in fighting. If you were going the wrong way, it was the quickest known method of returning in your slipstream." Re the loop (he didn't like looping a Camel): "But a Camel had to be flown carefully round with exactly the right amount of left rudder, or else it would rear and buck and hang upside down and flop and spin." Re general flight: " . . . a Camel had to be held in flying position all the time, and was out of it in a flash. It was nose light, having a rotary engine weighing next to nothing per horse power, and was rigged tail heavy so that you had to be holding her down all the time. Take your hand off the stick and it would rear right up with a terrific jerk and stand on its tail." Re ground strafing (which he hated due to ground anti-ac machine gun fire): "Unfortunately, they were good machines for ground-strafing. They could dive straight down on anything, and when a few feet off the ground, go straight up again." Re speed:"...a Camel was a wonderful machine in a scrap. If only it had been fifty per cent faster! There was the rub. A Camel could neither catch anything except by surprise, nor hurry away from an awkward situation, and seldom had the option of accepting or declining combat...You couldn't have everything." http://www.theaerodrome.com/forum/aircraft/45869-sopwith-camel-myth-6.html --- Australian ace Edgar McCloughry: "I at once turned but they did not wait, one of the horrible characteristics of a camel being, as I will describe later, that it is unable to catch any other machine with the exception of the Fokker Triplane on the level." "One word on the 'Camel': There is not one pilot in the squadron who would not argue to the end for a Camel. Although slow, she could get around anything, also one could not run away from anything, which rather aimed for success." Arthur Cobby, another Australian ace: "In this manner we accounted for a few of the enemy, but they could dive faster than our Camels. Unless we got close to them early in their dive, they would just keep on diving and so get away. ... If we were only able to encourage the enemy to get in a dogfight, things were easy, as a Camel could out maneuver anything." Comments above are by Australian pilots who used the 130 HP Clerget model of the Camel; other squadrons used 140 HP engines; the 150 HP model was used by English squadrons at this time. http://www.kuro5hin.org/story/2002/12/1/02631/3666 --- Like many other World War I airplanes, the Camel's vertical tail was too small. Any stall automatically became a snap roll spin entry, even without intentional rudder deflection. Finally, once spinning, the Camel required vigorous rudder defliction against the spin to stop the motion. A well-behaved airplane, on theother hand, has to be held in a spin; letting the controls go free should result in automatic recovery. http://assets.cambridge.org/97805218/09924/sample/9780521809924ws.pdf --- Indeed you are correct about the Rudder - it's area was a paltry 4.9 sq. ft. (and fin at 3 sq. ft.) http://www.theaerodrome.com/forum/aircraft/47951-sopwith-camel-myth-2-a.html ---- The pathetically small fixed area on the fin/rudder made the problems in yaw even worse. A pilot generally doesn't hold the rudder pedals stiffly enough for the rudder to contribute very much to yaw stability. http://www.djaerotech.com/dj_askjd/dj_questions/short_coupled.html --- Replica Camel with rotary engine: The first flights on the plane proved to be very impressive with performance figures very close to the original aircraft, stall speed is below 40 mph, cruise comes in at 80 mph and climb is a conservative 1000 feet per minute. http://www.lightsportaircraftpilot.com/airventure2010/2.html http://en.wikipedia.org/wiki/Airdrome_Sopwith_Camel --- Arthur Cobby: We had not come into contact with it [Fokker DVII] to any extent as most of our patrol work was being done at lower altitude, but our fellow Australians in No.2 [No.2 Sqn Australian Flying Corps] were continually meeting them. Their SE5a's could get to greater heights than our Camels, which were at their best up to about 12,000 ft. We could get much higher of course, but the performance fell off rapidly above this level, and against the new Fokker, would put up an indifferent show. Later on we did meet them up higher and managed by sheer hard flying to hold our own, but unless one was an exceptional good pilot the odds were definitely not good. --- Speed: The trials for the Sopwith Camel was done with an 150 hp Bentley of which only the Royal Naval Air Service squadrons were equipped. The RAF, AFC and USAS squadrons equipped with the Sopwith Camel had 130 or 140 hp Clergets. The Bentley trials gave the Camel a top speed of 114 mph at 15,000 feet while the Snipe at the same height had a speed of 113 mph. However this statistic is meaningless as the Sopwith Camel was obsolete in comparison to the Fokker DVII above 12,000 feet. http://www.southsearepublic.org/article/1501/read/ --- Gyroscopic effect: In the case of the Camel, with approximately 400 pounds of gyroscope hanging on the nose, a turn to the left would cause a gyroscopic precession moment that would try to raise the nose. A turn to the right would shove the nose down. In a right turn this acts like down elevator. If you roll into a steep bank and shove in enough down, you will find yourself in knife-edge flight with no turn rate. OTOH, for a steep turn you normally need a pretty fair dose of "up", and the precession in a left turn would provide essentially the same thing. Pulling the nose up would cause a yaw to the right, and pushing the nose down would yaw the airplane to the left. If you stall, the rapid pitch-down at the stall break would cause a strong and rapid yaw to the left, just as if you had just stomped on the left rudder. And we all know what happens when you boot the rudder right at the stall break. http://www.djaerotech.com/dj_askjd/dj_questions/short_coupled.html --- Cause of take-off crashes: You take off, start a normal climb (nose well up, airspeed low, power at max), and just as you start into that standard 90 degree left turn after takeoff, the engine begins to sputter, instantly drawing your attention away from the horizon and into the cockpit. Meanwhile, unbeknownst to the pilot (whose head is in the cockpit, and who has never experienced a plane that automatically tries to change its pitch attitude dramatically whenever it turns) the left turn plus the gyroscopic effects starts inexorably pulling the nose up. The airplane's nearly non-existent pitch stability (due to the aft C/G) doesn't put up much resistance to this. The more nose-up attitude adds to the already decaying airspeed (caused by the loss of power from the sputtering engine), and sure enough, the airflow lets go from those thin wings and the plane stalls violently. The pilot, suddenly attentive but completely disoriented, looks up (far too late to do anything about the situation) just in time to see the nose yaw violently to the left as it falls through in the stall break. In this era long before the relationship between stall, angle of attack and airspeed was properly understood, our neophite pilot reacts (naturally) by pulling the stick back the rest of the way, and probably adding full right aileron as well. With utterly no chance for a recovery at this point, the airplane does about a quarter turn before smiting itself against terra-firma. The rotation is slight enough that his buddies on the ground don't recognize that it was indeed a spin, perpetuating the mystery and the myth, and delaying any meaningful insight into solving the problem. Today, however, we can recognize from the photos that it was indeed a spin that added our hapless hero to the Camel's list of statistics. Clearly visible in the photo is that fact that one set of wings are wrapped around the nose, and the other set are wrapped around the tail, indicating that the plane was rotating when it hit. http://www.djaerotech.com/dj_askjd/dj_questions/short_coupled.html --- Camel Spinning, Camel Gyroscopic couple & rudder effectiveness compared to the SE5a From: British Aviation, The Great War and Armistice 1915-1918 by Harold Penrose. page 376 Even more worrying were Camel crashes through spinning, for the S.E.5a was comparatively immune when flown under similar circumstances. Camels were usually rigged tail-heavy with full tank and were then longi­tudinally unstable like the D.H.6. This was accentuated when the lighter Le Rhone was substituted for a Clerget, or guns and ammunition were removed. Usually a continuous forward pressure of some 14 lb was re­quired on the control column when in normal horizontal flight; any relaxation ended in a stall, and so did engine failure unless the pilot instantly dived the machine. In the course of evidence on the machine's behaviour, most pilots drew attention to the necessity of applying full left rudder in a steeply banked right-hand turn. Torque and slipstream pro­duced unsymmetrical settings of aileron and rudder, but not effects dependent on rate of turn to any great extent, the gyroscopic couple alone being capable of that. Propeller and engine revolved clockwise, viewed from the pilot's seat, so when the aeroplane was turning right the gyro­scopic couple put the nose down, countered by top, that is left, rudder and back elevator. Analysis showed that the effective couple produced by a given setting of the rudder on a Camel was 40 per cent less than for the S.E.5a, whereas the gyroscopic couple to be counteracted was more than twice as great. To make the two machines equally effective would necessi­tate doubling the Camel's rudder area. It was immediately evident with a Camel that violent use of elevator must be avoided, for when the stick was pulled back too hard the machine would not complete a loop. By contrast, full left rudder could scarcely keep a loop straight, but surprisingly when it came to spinning, the rudder was relatively unimportant. As soon as the stick was pulled hard back and the machine stalled fully it would flick into a spin, but quickly stopped if the column was eased slightly forward with other controls neutral, and could be stopped almost instantly by reversing the rudder as well as giving forward stick -though this induced a violent manoeuvre, and lack of skill could result in the machine entering a reverse spin. If the pilot became dizzy the Camel would not come out of a spin with controls free, and only slowly ifall controls were held central. That forward push on the stick was essential. --- Flight International 2 May 1968, By RONALD SYKES, DFC "Pull up into a stall and apply the usual encouragement from the rudder; the Camel will then cartwheel over and then flick into a spin (which, with the stick held right back, will be a fast one). Centralize the controls and after about four more turns the machine will come out of the spin: it can be forced out more quickly, by applying opposite rudder and pushing the stick forward briskly, though this does not always have the desired result." --- JSBSIM & THE DIRECTION OF THE GYROSCOPIC EFFECT JSBSim seems to have the direction of force for the moment of inertia of the gyroscopic effect of the engine reversed. This creates a difficult problem for modelling the Camel, because the gyroscopic effect of the Camel is so pronounced. Below are my notes on the problem and now it was resolved in the Camel JSBSim flight modle: The docs say that sense =1 for CW propeller as seen from the cockpit. The camel has a CW propeller by every example I've been able to find. This source definitely says, "The direction of rotation was counter-clockwise as seen from the propeller-end of the engine." The photos also show a CCW (from propeller-end) rotating prop. That would be clockwise as seen from the pilot's seat and sense=1. http://www.gracesguide.co.uk/Clerget But sense=1 makes the Camel climb on RH turn and dive on LH turn. Lots of sources say the opposite, especially including this experienced pilot who is clearly flying a craft with a CW propeller (as seen from cockpit) and yet he says the Camel dives on RH turn and climbs on LH. http://www.youtube.com/watch?v=j6PnKUEFX8g Wikipedia says, "To easily ascertain the direction of gyro effect, simply remember that a rolling wheel tends, when it leans to the side, to turn in the direction of the lean." http://en.wikipedia.org/wiki/Gyroscope#Properties That means, for a CW (from the rear) propeller, a turn to the right causes nose to drop, exactly what is reported for the Camel. My conclusion is that JSBSim has it backwards and sense=1 corresponds to CCW rotation of the prop, as seen from the rear, as far as gyroscopic moment is concerned. However, as far as p_factor & torque is concerned, JSBSim seems to have it right. So what we're going to do is put in sense=1 here to get correct p_factor and torque effects, but then put in a correction down in the aero section to correct the moments, which JSBSim seems to have reversed. THICK VS THIN AIRFOILS; DRAG AND LIFT Thick airfoils had a great advantage besides the structural one. Compared with thin wings, thick wings could produce more lift, by about 25 percent, because the gentle roundness of the leading edges helped air follow the curvature of the airfoil and not break away. The added lift did not affect climb rate, but it improved maneuverability, because the space within which an airplane can turn is determined by its maximum lift. Triplanes could also fly in a very nose-high attitude, because the thick wing kept producing lift at angles at which the sharp-edged wings of Allied fighters had already given up. A U.S. pilot, James Hall, wrote of the Fokkers’ “trick of standing on their tails beneath one” with guns firing upward. http://www.airspacemag.com/history-of-flight/red_baron.html?c=y&page=2 --- ROTARY ENGINE CONTROLS It is often asserted that rotary engines had no carburettor and hence power could only be reduced by intermittently cutting the ignition using a "blip" switch, which grounded the magneto when pressed, shutting off power to the spark plugs and stopping ignition. However, rotaries did have a simple carburettor which combined a gasoline jet and a flap valve for throtting the air supply. Unlike modern carburettors, it could not keep the fuel/air ratio constant over a range of throttle openings; in use, a pilot would set the throttle to the desired setting (usually full open) then adjust the fuel/air mixture to suit using a separate "fine adjustment" lever that controlled the fuel valve. Due to the Gnôme's large inertia, it was possible to adjust the appropriate fuel/air mixture by trial and error without stalling it. After starting the engine with a known setting that allowed it to idle, the air valve was opened until maximum engine speed was obtained. Since the reverse process was difficult, "throttling" was accomplished by temporarily cutting the ignition using the blip switch. By the middle stages of World War I some throttling capability was found necessary to allow pilots to fly in formation, and the improved carburettors which entered use allowed a power reduction of up to 25%. The pilot would close off the air valve to the required position, then re-adjust the fuel/air mixture to suit. Experienced pilots would gently back off the fuel lever at frequent intervals to make sure that the mixture was not too rich: a too-lean mixture was preferable, since power recovery would be instant when the fuel supply was increased, whereas a too-rich mixture could take up to 7 seconds to recover and could also cause fouling of spark plugs and the cylinders to cut out. The Gnôme Monosoupape was an exception to this, since most of its air supply was taken in through the exhaust valve, and so could not be controlled via the crankcase intake. Monosoupapes therefore had a single petrol regulating control used for a limited degree of speed regulation. Early models also featured variable valve timing to give greater control, but this caused the valves to burn and therefore it was abandoned. Later rotaries still used blipping the ignition for landing, and some engines were equipped with a switch that cut out only some rather than all of the cylinders to ensure that the engine kept running and did not oil up. A few 9 cylinder rotaries had this capability, typically allowing 1, 3, or 6 cylinders to be kept running. Some 9 cylinder Monosoupapes had a selector switch which allowed the pilot to cut out six cylinders so that each cylinder fired only once per three engine revolutions but the engine remained in perfect balance. Some documentation regarding the Fokker Eindecker shows a rotary selector switch to cut out a selected number of cylinders suggesting that German rotaries did as well. By 1918 a Clerget handbook advised that all necessary control was to be effected using the throttle, and the engine was to be stopped and started by turning the fuel on and off. Pilots were advised to avoid use of the cut out switch as it would eventually damage the engine. The blip switch is, however, still recommended for use during landing rotary-engined aircraft in modern times as it allows pilots a more reliable, quick source of power that lends itself to modern airfields. The landing procedure using a blip switch involved shutting off the fuel using the fuel lever, while leaving the blip switch on. The windmilling propeller allowed the engine to continue to spin without delivering any power as the aircraft descended. It was important to leave the blip switch on while the fuel was shut off to allow the spark plugs to continue to spark and keep them from oiling up, while the engine could easily be restarted simply by re-opening the fuel valve. If a pilot shut the engine off by holding the blip switch down without cutting off the fuel, fuel would continue to pass through the engine without combusting and raw fuel/air mix would collect in the cowling. This could cause a serious fire when the switch was released, or alternatively could cause the spark plugs to oil up and prevent the engine restarting. Source: http://www.sim-outhouse.com/sohforums/showthread.php?6993-Rotary-Engines --- The early Gnome 50-80 hp 7 cylinder rotary aero engines introduced from 1908 onwards had no throttle control, and the fuel/air mixture could only be adjusted for fine tuning on the ground by a mechanic. The engine was therefore either on or off, although there was a blip-switch for the pilot to temporarily cut ignition. The development of the 100hp Gnome Monosoupape (or 'Mono') in 1913 introduced pilot-control of the fuel-air mixture for the first time, with a lever that regulated petrol flow to the engine. This permitted the pilot to make small adjustments to the RPM, or lean the fuel mixture at higher altitudes to maintain engine efficiency. Lt. R.T. Leighton provides a very good description of this in his pilot notes published by the Shuttleworth Collection: "The engine should give 1,150-1,200 rpm, as height is gained, then petrol should be cut down until engine is giving 1,050-1,100 rpm, when machine [Avro 505K] flies level at 65 mph. The machine at full revs flies level at 85 mph...[to descend] shut petrol off...glide down at 55 mph...do not 'buzz' engine...taxi in by buzzing the engine with petrol about 1" on adjustment". Pilots were now discouraged from using the blip-switch, as over-use could stress the engine and (in the mono) cause an engine fire. In the 100hp Monosoupape the engine rpm could only be reduced by about 10% or 20% without risking engine cut-out. In 1916 the 160 hp 14 cylinder version of the Gnome Monosoupape introduced another refinement in the form of a second magneto switch that could be used to cut ignition temporarily to two or more of the 14 cylinders in any one cycle to give reduced power rather than no power at all: a switch that could be used to reduce engine output by approximately 1/8, 1/4 or 1/2. The Le Rhone and Clerget engines introduced a second mixture control, to control the air added to the mixture (often referred to as the 'throttle' by WWI pilots) although the petrol adjustment lever was also retained (referred to under a number of different names, but most usualy as the 'fine adjustment' by RFC pilots). There has been quite a bit written on the Le Rhone, which was much liked by pilots. Cecil Lewis, in 'Sagittarius rising' comments "The rotary was an 80hp Le Rhone. It was a beauty, the sweetest running rotary ever built. It throttled down and ticked over like a water-cooled stationary, and was as smooth as silk over its whole range". The Le Rhone was also ahead of its time in linking the throttle to the needle valve which regulated the petrol supply: Lt. Leighton comments in his pilots notes that "Theoretically, the position of the fine adjustment can be found once and for all for every position of the throttle, so that having set fine adjustment once, it need not be moved again. The throttle lever then being worked as on a stationary engine". He adds, though, that "Practically, the engine will run if worked this way, but better results are obtained by varying the position of the fine adjustment with varying positions of the throttle lever". The two levers were positioned together normally on the left of the cockpit, on a quadrant marked from 1-10. The Clerget had the same arrangement of throttle and fine adjustment levers, but dispensed with the unreliable linkage between the two - it was much liked by ack emmas, as it was easier than the Le Rhone to maintain in the field, and by the War Department as it was slightly cheaper than the Le Rhone. In both the Le Rhone and the Clerget (and the later Bentley) the combination of a throttle and a fine adjustment did take some getting used to, and RPM could not be changed rapidly in flight even by experienced pilots. I do not believe the 'throttle' was used much, if at all, in combat - it was used mainly to ensure optimum endurance at altitude, to make formation flying easier, and to reduce power on landing. Engine rpm could not, in practice, be reduced below 50% or so of engine power, however, so a powered landing would still require use of the blip switch unless the pilot was confident enough to bring the airoplane in by gliding down with the petrol switched off. Robert W. Bradford (An Associate Director of the [Canadian?] National Aviation Museum) is quoted as saying "the 110/120 hp Le Rhone rotary has the characteristics of all early rotary engines - they have a high idling speed in proportion to the full power rpm. They simply do not 'tick over' as a radial or inline engine would do - in fact with the fixed pitch wooden propeller, they idle at about 45% of full engine speed (500 rpm as against 1150 rpm for take-off at full power)" [billybishop.net/bishopF.html]. I have seen a variety of other figures for the safe idling speed of Le Rhones and Clergets, from 600 rpm up to 800 rpm, and there is some evidence that the Bentley rotaries might have idled at a lower rpm [Blakemore,1986] - in practice, I suspect each engine would be slightly different depending on the make, length of service or time between overhauls, the skill of the fitter, maintenance standards in the field, etc. It does seem clear, however, that most pilots would be reluctant to risk an engine cut-out (particularly in the final approach to landing) by cutting rpm back further than 50% of engine speed. So these rotary engines could be 'throttled' up or down, but it was certainly not a simple procedure and pilots report that it took considerable practice before it became second nature (engine failure on take-off was often caused by the pilot getting the mixture wrong and then choking the engine). Also, these engines could not be 'throttled' up or down quickly - it took about 7 seconds before the change in the mixture setting had any effect on engine power [Nahum, 1987]. Using such a 'throttle' in combat was therefore unlikely to be practicable. Most pilots would continue to use the blip-switch to produce sudden changes in engine power, mostly on the final approach to landing (although they were officially discouraged from doing so, particularloy at full engine power, as it damaged the engine). German rotaries were similar to the early Gnome and Le Rhone engines, although German pilots had both throttle and mixture control for even those engines based on the early Gnome design (so the Fokker E.III/E.IV had both a throttle and a mixture lever). I think that the most interesting of the German rotaries is the late war contra-rotating 160hp SH III fitted to the Siemens Schuckert DIII and IV. This was "fitted with twin magnetos and speed was governed by a proper throttle control, sensitive down to about 350 rpm" [Profile 86]. Blakemore, L N. Bentley BR2 World War 1 rotary aero engine: building the one quarter scale working replica. Yalanga, 1986. Leighton, R T. Pilots' notes for the handling of World War I warplanes and their rotary engines. Shuttleworth Collection. [Pamphlet. Notes originally written in 1917 by an RFC pilot. Covers the Monosoupape, Clerget and Le Rhone rotaries, with notes on flying Avro, 1 1/2 Strutter and Pup]. Morse, William. Rotary engines of World War One. Nelson and Saunders, 1987 Nahum, Andrew. The rotary aero engine. HMSO, 1987. Profile Publication 86: The Siemens Schuckert DIII and IV. 1966. Lewis, Cecil. Sagittarius rising. Peter Davies, 1966. http://www.theaerodrome.com/forum/ai...-1-allied.html http://www.theaerodrome.com/forum/ai...-2-german.html Source: http://www.sim-outhouse.com/sohforums/showthread.php?6993-Rotary-Engines ------------------- Peter Garrison has written several articles about the Camel, some with info about a project to quantify the Camel's flight characteristics: http://www.flyingmag.com/sopwith/technicalities-scoring-sopwith http://www.flyingmag.com/pilots-places/pilots-adventures-more/nice-little-book http://www.airspacemag.com/history-of-flight/what-the-red-baron-never-knew-22968921 http://www.flyingmag.com/photo-gallery/photos/fokker-aircraft-and-sopwith-camel-spruce-creek-fly --- ACKNOWLEDGEMENTS The enclosed Sopwith Camel JSBSim flight model started as an Aeromatic-generated model and was then heavily modified based on ideas from the FlightGear and JSBSim forums, wiki, documentation, and other JSBSim models. In particular, the P-51D JSBSim FDM by Aeromatic, Jon Berndt, DATCOM, and modified extensively by Hal V. Engel provided a model of a JSBSim FDM that successful modeled asymmetric stall, snap rolls, induction of a spin during stall, and other characteristics found in the Camel FDM. Other than the JSBSim FDM, adaption of weapons models for the Camel, and other minor adaption as needed for these updated systems, the Camel models, sounds, systems, and XML files are all the versions distributed with FlightGear, which credits AJ MacLeod, Vivian Meazza. LICENSE, COPYING FlightGear is released under the GNU GENERAL PUBLIC LICENSE and this JSBSim FDM and all associated files are released under the same license.