Jump to content


Photo
- - - - -

Data Topic for Airplanes Performance.


  • Please log in to reply
136 replies to this topic

#41 piecost

piecost
  • Posts: 1318

Posted 16 November 2010 - 17:45

Rudder effectiveness versus incidence for Bristol Fighter, B.A.T. Bantam & SE5a
Elevator effectiveness versus incidence for Bristol Fighter &B.A.T. Bantam


Taken from: Proceedings, Fifth Meeting, Second Half, 62nd Session, THE JOURNAL OF THE ROYAL AERONAUTICAL SOCIETY: THE SPINNING OF AEROPLANES

"The drop in control efficiency of rudder and elevators due to a simple increase in incidence is well shown in Figs 22 and 23, which are drawn without any special reference to the size of the control.

The rudder control diagram shows n_zeta referred to the chord axis. In general the decrease in control between alpha=20º and 40º is at leaste 50 percent…. In may be mentioned that at a given incidence n’zeta is roughly proportional to zeta up to zeta=25º in the incidence range shown. The elevator control diagram shows a smaller, but still serious, efficiency loss both for upward and downward elevator movement the loss is of the order of 30 percent between alpha=10º and 30º. At a given incidence m_eta is linear with eta between eta=-25º and +25º, … The practical control range is usually not more than +/-25º"

Attached Files


  • 0

#42 piecost

piecost
  • Posts: 1318

Posted 18 November 2010 - 22:07

Original Nieuport 28C-l (clipped wing?)

Extract from:

Flying the Old Planes
By Frank Tallman
Doubleday & Company, New York
1973

Engine: 160-hp Gnome Mono

CERTAINLY THE LOVELIEST race horse lines of all the World War I aircraft belonged to the French Nieuport 28 and like the beautiful thoroughbred it resembles it also had real speed and agility. I share the feeling of others lucky enough to fly the “28" that it certainly is the epitome of World War I flying. After you have flown this beauty you are ready to throw old rusty water-cooled engines at other much better publicized Allied and German aircraft.

So much unutterable garbage has been written about rotary engines that many of our current pilots ap­proach them with severe palpitations and vibrations in the area of the back­bone. Actually, the characteristics of the 160-hp Gnome as I know them are delightful. Unlike most rotaries, this engine has dual ignition. Few of our pilots of today have become acquainted with that so-well-remembered silence that occurs when the lone ignition system folds its tent and steals silently away. A rather common occurrence with magnetos on rotary engines aging now for half a century is that when they are actually turned up and flown, the very old winding on the armature can often swell, then drag, and finally shear the rotor shaft key, which promptly cancels your current flight plan. A rather competent aviator of the Old School said, "Two magnetos can be as comforting as the difference between a fifth and a quart."

We are shy some instruments that usually were carried by the 28, but presently we mount an airspeed indicator, oil pressure gauge, and altimeter. Missing also is the infamous cut-out switch, with which you could ground out three, six, or nine cylinders. According to legend, if you grounded out three or six, the distinctive sound was the signal for the pyrene-equipped World War I fire crews to run out on the field to greet the 28 as it landed in flames.

We face the little bird into the wind with a clear area ahead, and chock it. The single level controlling fuel air mixture is retarded while the gas is on and the propeller is pulled through. The dual ignition switch is off, and the coupe (cut-out) button on the top of the stick is depressed for added safety. There is pretty fair compression, necessitating a mechanic selected primarily for fine muscle tone and body structure to pull the prop through.

The Gnome starts without priming (unlike other rotaries), and with a real bark. As the little lady tugs at the chocks you are sure you really have a tiger in the cowling, even if the spots don't match the camouflage. With the appearance of spattered castor oil on the leading edge of the wing we know we are oiling and consequently ready for takeoff. The temperature is 63 degrees, the altitude is 53 feet, and the wind on our nose of 11 knots. With the chocks pulled and the Gnome winding up, it's like a Clydesdale or Percheron pulling, for there is no doubt as to the old engine's raw horsepower.

I know of no other aircraft except the Boeing F4BI (P-12) that has such complete and instant response. The tail is up in 15 feet of forward roll, and rudder is needed to overcome torque, though the actual torque is nothing like the hair-raising fables of the World War I-type pulp magazines.

With its short wingspan, it bounces off over several Jack Rabbit burrows in about 220 feet. Wow! The climb is spectacular, and the steep-climbing turn gives not the tiniest evidence of payoff. Unfortunately, as I level off in the pattern with essentially no throttle control, I am going by the modern civilian aircraft like they were at anchor.

In flight, the most noticeable fact of the 28 is that the big Gnome gives off heat like a cast-iron stove in a New England country store in winter. It's so d—hot you're sure the ship is afire.
Climb is on the order of 1300 feet a minute, and stalls are straight­forward unless one uses full back elevator.

In our present state of aircraft design, it's easy to forget that many aircraft of earlier vintage had enormous control throws on the order of two feet (elevators) instead of several inches. This excess of control is desirable in the hands of a professional but often deadly dangerous in the hands of an amateur. Certainly many of our World War I pilots who arrived at the front with great gallantry and noble purpose could only qualify today as beginners.

In flight, apart from the stove-like heat and the more than adequate con­trol throw, the most noticeable sensation is the overwhelming castor oil smell. Having flown many rotary­powered aircraft, I rather look for­ward as something of an aficionado to the castor oil. Not so in the 28, however, for the comfort of the driver, one needs a nice tight pressure-type oxygen mask.

Not having to outdive some leather­coated Prussian preparing to venti­late my coat tails with a pair of Spandaus, most of my aerobatics are done with "G" loads on the order of 2 1/2, and with some timidity on my part. The 28 loops beautifully, and with the exception of offset rudder at the top (to counteract torque), it might as well be on rails. I start the loop at 120 mph, and there is an immense sense of thrust from the big Gnome as it pulls you up and over.

Slow rolls are smooth, and you don't need forward stick to hold the nose up, just unshakable faith, for the engine quits and doesn't come back again for about 15 seconds. You would be surprised how much land­scape and real estate you can assess for a forced landing in this time. No other aerobatics were tried, but the plane picks up speed like a pig on a greased slide, and you can comfortably get 200 mph without too much of a dive.

Like almost all World War I planes, a nose-high forward slip with this plane can kill off speed and yet give you some idea of clearance in your landing area. In the 28 this is even more necessary, because with the Vickers guns in the staggered position (because of ammunition boxes inside the fuselage), your vision is restricted in a standard field pattern in a left turn. With its short wings the Nieuport pays off fast, and you better be close to the ground in any three-point attitude. But thanks to the graceful and generous rudder, you don't have to make the rudder correction on landing that you do with both the Spad and the Fokker DVIII. The apparent touchdown speed is between 48 and 50 mph. The skid takes hold quickly, and you drag to a stop in about 300 feet.

As I taxi in, with bursts of power from the big Gnome, 1feel like 1could use a brandy and milk, which seems to have been World War l's favorite stomach settler after an hour of concentrated atomized castor oil.

[photo caption] Flying formation with a Pfalz DXII during a filming session. The 28 could-and did this day-lick the DXII hands down.
  • 0

#43 =IRFC=AirBiscuit

=IRFC=AirBiscuit
  • Posts: 2455
  • LocationNaples, FL USA

Posted 24 November 2010 - 22:09

Awesome! More anecdotal evidence that the N.28 we have in ROF is rather messed up, too. Tail off the ground in 15 feet? Try 15 seconds.. And "agile" hardly fits the bill for our N.28, straining to perform anything resembling a combat turn, heh.
  • 0

=IRFC=Air Biscuit

http://quetoo.org


#44 J2_Adam

J2_Adam
  • Posts: 2453
  • LocationVancouver, BC

Posted 25 November 2010 - 01:44

Have you all seen "Wings"? There are N28's in there. Mostly taxiing and starting up but there is a shot of one taking off and climbing out. The tail comes up not in 15 feet but quicker than RoF. And before you say anything about the speed of old films, no, this didn't look sped up.

On a side note (and nothing to do with his thread, sorry) the sound of the N28 in wings is incredibly nasty and not like the higher pitch (note produced as in frequency) sounding 160HP engine we have in the RoF N28. Again I don't want to hear about sound recording and mic placement and microphone type. I have well trained ears. I don't care if you work around these engines or not. I can differentiate 2 different basic sounds :D

Redirect to engine sound post #76:
Fokker E.III Wing Warping Demo Video
  • 0

#45 piecost

piecost
  • Posts: 1318

Posted 07 January 2011 - 12:48

Behaviour in Dives, Terminal velocity of SE5a

taken from: British Aviation, The Great War and Armistice 1915-1918 by Harold Penrose. page 287

There was particular concern with behaviour in dives, not only of the [Bristol] Fighter but all other operational aircraft, for this so often proved the avenue of escape from German fighters. It was found that the S.E.5 could attain 250 mph in the first 1,500 ft drop with engine on, and the terminal velocity was 265 mph at an angle-they had believed vertical, but was nearer 45 to 60 degrees. The margin of strength was considered high, despite several occasions when wings had broken off although modified following Goodden's crash. Farnborough technicians calculated that in a vertical dive about half the weight was supported by the wing structure as drag, but the wings were also under torsion through a moment due to air forces even though the resultant of lift at right angles to the wind was zero. This put a down-load on the front spar and an equal up-load on the rear spar equal to 2·3 times the weight, but the factor of safety under the combined drag and twist was still 2 on the front spar and 3 on the rear, and stresses on the rest of the aeroplane were inconsiderable. 'One of the chief objections to diving at very high speeds is that intense vibration is often set up,' stated R & M No. 494 of 1917. 'This naturally gives the pilot cause to think that the machine is intrinsically weak, and in any case continuous vibration is very bad for any structure. Much of this vibration is undoubtedly due to high engine speed. Probably a great deal more is due to the wires, which are also responsible for much of the noise. But it is important to see if any is due to what appears as the rapid change of centre of pressure near no lift. Although the position does change very rapidly the moment on the wings changes quite slowly, in fact more slowly than at ordinary speeds, and it appears probable that very little of the vibration is due to this cause.'
  • 0

#46 piecost

piecost
  • Posts: 1318

Posted 07 January 2011 - 12:53

Camel Spinning, Camel Gyroscopic couple & rudder effectiveness compared to the SE5a

taken 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.

Some aeroplanes spun even faster than the Camel, with a period of l to 2 seconds a turn, and 150 to 200 ft lost.
  • 0

#47 Baal2

Baal2
  • Posts: 314

Posted 07 January 2011 - 13:06

Very nice find Piercost, the N28 test flight is very informative! Don't forget to send a link to Han and Jason
  • 0

#48 piecost

piecost
  • Posts: 1318

Posted 10 January 2011 - 16:41

Flying the SE5a

F1y1ng Scale M0de1s, Nov 2002

Andy Sephton, Chief Pilot of the Shuttleworth Museum Collection

In the hangar, the S.E.5a looks just great. A squat biplane with ailerons on each wing, she's Tiger Moth size with 200 HP up front and she's still cleared for aerobatic flight even after all those years. One can't wait to get her into the air, but lifting the tail onto a trolley to get her out of the hangar rapidly brings the budding SE pilot down to earth - she is heavy! It's no wonder there's four 'lift here' points marked on the fuselage. Help must be summoned and the tail eventually coaxed onto the hand-steering rig. At last, the machine can be moved from the hangar to the airfield.

Shuttleworth pilots must learn how to operate old aeroplanes from the start of their careers at the Shuttleworth Collection. The best way to learn is by practice, so we start by pushing aircraft into and out of the Hangar. As with all aircraft of this vintage, it's easier to push the S.E.5a backwards. Further, the pilot soon learns that hand pressure must only be placed on the ends of struts, not in the middle, and only on a rib joint when pushing on the leading edge of the wing. Some air­craft, the Hawker Tomtit for example, require an extra man to pre­vent the tail from jumping out of the box of the hand-steering gear, but not so the S.E.5a…

Let's have a look at the layout of the machine. Starting at the sharp, or rather blunt end in the case of the Collection's S.E.5a, we have a 200 HP Wolseley Viper, eight cylinder 'V', liquid cooled engine. The coolant system consists of a liquid filled (70/30 water/glycol) radiator at the front of the machine, with pilot operated shutters to control the temperature between a nominal 75 to 85 degrees centigrade. The control lever is located on the port cockpit wall; it comes easily to hand and operates in the natural sense - forward opens the shutters and allows increased cooling airflow through the radiator. The system is replenished through the tap on the top of the radiator, it holds seven and a half imperial gallons. A small condenser and expansion reservoir is fitted to the right side of the upper wing centre section and an overflow pipe exits at the right trailing edge - more of this later.

The engine oil system is self-contained and, therefore, is of little interest to the pilot. However, some oil fumes emit from various parts of the engine in flight, the location of which must be noted and, of course, oil pressure must be monitored during all times that the engine is running. In some respects it is a relief (no pun intended) not to have to monitor oil flow by the amount of caster oil thrown back at the pilot in flight, as with the rotary.

Fuel is carried in two tanks, one of four imperial gallons capacity fed by gravity and located in the left part of the upper wing centre section. The other, of twenty eight imperial gallons capacity, is located in the forward fuselage and is fed to the carburettor by air pressure from an engine driven pump or pilot operated hand pump. The fuselage tank is referred to as the 'service' tank, and the centre section tank as the 'emergency'. As with the water system overflow, that of the emergency tank is at the trailing edge of the wing centre section, but this time, it is on the left.

With four gallons of fuel, the emergency tank will only keep the engine running for about 20 minutes at cruise power, therefore, as far as possible, it is kept full. Replenishment is carried out by pressurising the service tank and opening the line from the latter to the gravity tank by turning the fuel selector to the appropriate mark. The fuel selector is a rotary switch mounted on the left side of the instrument panel which either allows the fuel tanks to be isolated, or it opens taps to connect either the service tank to the emergency tank, or, alternatively, both or separately, the service and emergency tanks to the carburettor. Replenishment is complete when the fuel overflows from the vent pipe.
Now, with two overflow pipes, both having the capability of dispensing scalding and/or noxious fluids, located just a few inches from the pilot's head, the pilot must beware. In normal flight, should the fluids overflow, most of the liquid will be taken straight back in the slipstream. But taxying in, with winds from all points of the compass, the direction of overflow can be anywhere. The desire to rid oneself of ones goggles immediately after landing must be resisted and the said items of protective clothing left firmly fixed in front of the eyes. It's also prudent to select the emergency tank for taxying in. It takes 'the top' off the fuel in the tank and thus, helps prevent some of the leakage.

As well as for transfer to the emergency tank, air pressure is used to transfer fuel from the service tank to the carburettor and it is provided by either, or both of, the engine driven and pilot operated hand pump. A second rotary controller fitted next to that for the fuel operates valves that either isolate the airflow, or open lines from either or both of the air pumps to the service tank. The cockpit hand pump has a further tap at its base that either isolates the pump, connects it to its air line, or allows air pressure to blow off to atmosphere. A pressure relief valve is fitted adjacent to the rotary switch that is reserved for ground crew adjustment of blow-off pressure. However, pilots with the relevant knowledge may use it to let off excess pressure in emergency by pulling up on the central valve spindle. A pressure gauge, red lined at two and a half pounds per square inch, also located on the pilots instrument panel, completes the air system.

Only the service fuel tank is gauged. But, as the gauge is fitted to the rear of the tank, it is almost completely out of the field of view of the pilot, being situated about two feet behind (in front of?) the main instrument panel.

To complete the description of the fuel system, a 1.8 pint priming fuel tank is fitted behind the engine firewall to allow priming of the engine via a ki-gass primer pump which is located low down on the port cockpit wall.

Walk-round of the aircraft prior to flight follows the normal pattern. Special note must be taken of the security of the aileron control circuit as, in the event of a wing drop during a previous take off or landing, the lower aileron horn would have contact­ed the ground first and if damaged, could lead to loss of aileron control on a future flight - close inspection of the Collection's S.E.5a will show repairs in this area from previous mishaps.

Well meaning ground crew always remind me to watch my head when mounting the aircraft, but at some stage in the entry/egress process, I always manage to 'nut' the Lewis gun in some way or other. Contemporary literature and pilot's reports of the period all criticise the Lewis gun mounting. I can understand why and not just from the 'nutting' viewpoint. With a Vickers gun mounted in the fuselage and arranged to fire through the propeller arc, why not mount another similar gun next to it. It would have provided a better concen­tration of bullets, a gun that didn't require reloading in flight and a sight line close to both gun axes. In contrast, the Lewis mounting allows for a poor harmonisation of shot and therefore a low bullet density per unit area at the target and a large par­allax between gun and sight. Further, and more importantly from a practical point of view, it requires reloading by pulling it towards the cockpit and changing a drum of ammunition which is now flat plate to the slipstream. Also, the drum must have been stored in some way, and believe me, the cockpit is small. Then, after reloading, the pilot must push the gun back into position against its dead weight and also against the Slipstream. How this could be achieved during combat manoeuvring beats me; it probably had them beat too … and in more ways than one.

Anyway, we're now seated in the cockpit and preparing for flight. Again, the ground crew will often aid the pilot by pumping the service tank to pressure during their own pre-flight inspection. If not, about five minutes of slow pumping will be required to achieve the magic 2.5 psi tank pressure. Owing to the small pipe size, fast pumping will only serve to exercise the blow-off valve and increase the job time by a further five minutes -patience is required.

With the tank pressurised, one can now strap in - the tasks are impossible the other way round as the hand pump valve is located just outside the pilot's reach when the pilot's harness is fully secured. When set­tled, fuel is selected from service to carb, tank pressure is selected from hand and engine pump, and the ground crew is informed that all is ready.

Consultation provides the number of priming strokes required and the propeller is turned as the priming fuel is injected. The magneto switches, left, right and starter, all located on the starboard cockpit wall, are set to 'ON' and the throttle is checked closed.

Ignition for starting is by starter magneto, the handle of which is located on the outside of the starboard cockpit wall. For starting, it is turned rapidly by a third ground crew member as two others pull the propeller over compression. Co-ordination is required as the magneto must not be turned until the propeller is moving, other­wise a kickback may result with it's inherent dangers to the swinging crew. The normal method employed at Shuttleworth is for the pilot to shout: 'Three -Two-One­GO!', at which point the propeller crew pull the propeller and as soon as the third member notes the propeller's movement, and not before, he rapidly turns the magneto handle. Normally, the engine starts at the first attempt, but the magneto must be kept turning even if the propeller apparently stops as a start can often be achieved even several seconds after the propeller has lost it's momentum. As the engine fires up, the throttle is 'cracked' slightly to catch it, the engine, that is, and the oil pressure is checked. A rise to at least 50 psi must be noted within thirty seconds, but it is equally important that the maximum of 120 psi is not exceeded as this could lead to failure of the oil pump drive. Engine speed is kept below 1,000 rpm until smooth running is achieved, revs are then set to 1,000 for warm up. The 1,000 figure must not be exceeded until the radiator temperature exceeds 60 degrees centigrade, but to prevent the coolant boiling due to temperature inertia, the radiator shutters are opened as the needle on the temperature gauge pass­es 50 degrees.
Power checks are carried out on the chocks, with ground crew holding the tail down. I'm not sure why we do this, given the excess weight of the SE, but it's probably because we do it to all other aircraft and it's best not to change SOP's (Standard Operating Procedures).

Full power must provide over 1,700 rpm, but is not normally confirmed until take off The magneto check is carried out at 1,600 rpm and should provide a limiting single magneto drop of 100 rpm. Slow running should be between 600 and 750 rpm; a higher figure leads to overrun on landing, a lower figure, a propensity for the engine to stop in flight.

Given the weight of the aircraft, the heavy steerable tailskid and an engine that can be controlled down to a useful idle, taxying of the S.E.5a is a relative delight. However, the field of view forward is poor and there are no brakes fitted, so appropriate care must be taken. In anything other than light wind conditions, it's also prudent to enlist wing walkers for taxy.

Normal pre-take off checks work for the SE, and they commence with tail trim, checked at neutral. Tail trim is achieved by an all-moving tailplane controlled by a wheel fixed to the port cockpit wall. The wheel moves in a number of detents, either thirteen or fifteen, I can't remember exactly. However, as the tailplane actuator is part of the tail skid mechanism, the aircraft weight acts against movement of the wheel and setting the trim on the ground with any weight in the cockpit requires Herculean strength -perhaps the same strength as that required to relocate the Lewis gun after reload? Thus, to limit the required trim wheel movement, suffice to say that, seven clicks nose down from full up puts the aircraft in trim at take off with my 210 lbs (95.5 kg.) in the cockpit.

The coolant temperature is confirmed within limits and the radiators set as appropriate - on an average UK summer day, they would normally be about 80 degrees and full open respectively.
Lined up, it is almost impossible to see directly forward, but there are adequate references on which to check the nose attitude for landing and the relative position of the horizon can be noted. The throttle is smoothly opened, 1,700 plus is achieved and oil pressure is checked. Rudder is required to prevent the swing, but the main sensation is that of positive acceleration ­ the 200 HP really makes itself felt, even given the airframe weight. The tail is raised and within a few seconds, a flying speed of around 50 to 55 mph is reached. The aircraft flies off the ground with a will and settles into a natural climb at about 60 to 65 mph.

Both control power and harmonisation are good. The S.E.5a is very similar to the Tiger Moth, except that roll rate is higher and the controls are slightly heavier. There is also the absence of the inertia caused by the Tiger's heavy centre section fuel tank. The climb is rapid and as briefed, slight oil mist/smoke is seen to emit from the radiator area and the starboard side of the engine cowl. The revs are reduced to 1,600 to save the engine, and coolant temperature and radiator flaps are monitored and regulated respectively.

An appropriate height for stalling is soon reached and the slow end of the flight envelope can be essayed. First, engine manage­ment and the radiator shutters are closed as the throttle is retard­ed. The latter must be done slowly as fast engine modulation brings unwanted stress on the engine and may lead to a damaging back­fire. The stall break occurs at about 45 mph with more than ade­quate buffet as a warning and a wings-level nose drop as an indicator. In flight, she's a benign aeroplane, but as we'll see later, the pilot-friendly stall does not carry over into ground effect. Further handling assessment shows a more-than-expected amount of adverse yaw, but no other significant anomalies. The adverse yaw leads to steady heading sideslip on aileron application alone, but with an appropriate amount of rudder deflection, balanced turns can easily be achieved.
The cockpit field of view is poor forward and down, but good in other respects. The upper wing gives some restriction to the view, but it is not excessive. Cockpit noise is high and as the engine is effectively silenced by the long exhaust pipes, it is difficult to sepa­rate engine noise from that of the slipstream. As called for in the Pilot's Notes, specific attention must be given to 'feeling' the engine's operation, as a miss-fire is difficult to perceive by ear alone.

Currently, the maximum airspeed is limited to 150 mph although, anecdotal evidence suggests that the machine was dived to in excess of 300 mph in combat. Aerobatics can be performed with ease at 120 to 130 mph, so the 150 mph never exceed is all that is required. Aerobatically, the machine handles like an overpowered Tiger Moth -aerobatic speeds can, after all, be achieved in level flight. However, engine rpm must be constantly monitored as an overspeed cannot be supported. Further, one eye must never be far from the coolant temperature gauge and the radiator shutters must be modulated as appropriate. In practice, a radiator setting of open for take off, half for display, closed for approach and open after landing meets most of the requirements of an average collection flight. Aerobatics are now limited to loops and barrel rolls only. The machine is more than capable of a crisp stall turn and a precision slow roll, but both manoeuvres reduce oil pressure to zero so as a mark of respect for the longevity of the engine, they are no longer flown.

So, we find ourselves in a delightfully powerful, manoeuvrable and easily flown machine. The only vice so far seen is an excess of adverse aileron yaw, but it's easily controlled with judicious use of
rudder. Surely the SE must have some vices….Now, after every flight must come a landing.

[Caption to photograph] Cockpit of the S.E.5a is a tight fit indeed. Shuttleworth Collection’s chief pilot Andy Sephton is by no means very wide shouldered, but to operate the aircraft, he needs to sit with his shoulders scewed and to operate the throttle control on the left side of the cockpit, he needs to use his right hand, crossed over; holding the control column with his left hand.

With no brakes and only a steerable tailskid for ground control, the SE must always be landed into wind. Those who have tried otherwise have suffered the inevitable ground loop with attendant wing drop and associated destruction of the lower aileron horn.


Also, it's worth noting that following any wing drop on landing, it is imperative that the machine is not flown again until the ailerons have been checked - thus, a go-around is not an option on a particularly bad landing.

The approach is set up into wind at about 60 to 65 mph, aiming to over-fly the airfield boundary at about 55 mph. A normal round out is attempted and then the fun starts. The stall in ground effect can lead to a nasty wing drop -beware aileron control - or if the stall is not coincident with touchdown and the machine comes down on the main wheels first, a rather unnerving bucketing motion occurs. The machine rocks aggressively between the main wheels and tailskid, giving the pilot a most uncomfortable ride. SE folklore has it that a go-around from this condition would only lead to a damaged propeller, so the best course of action is to hold the stick fully back and hope for the best The only consolation is that it feels and looks far worse from within the cockpit than it does from outside.

Landing over, the pilot must first resist the temptation to raise his goggles to wipe his sweating brow and then, to give himself the best chance of avoiding a petrol eyewash, turn the fuel cock to: 'FROM GRAVITY TO CARB'. Oh! and also open the radiator shutters.

After flight, the engine is allowed to temperature stabilise at fast idle for about 3 to 5 minutes, then it is shut down from slow idle by turning off the magnetos in turn. The fuel and air are then isolated and tank air pressure is blown off using the hand pump tap. All that remains is the traditional 'nutting' of the Lewis gun as the cockpit is vacated.

It's so easy to criticise an aircraft of the 1914 -18 era when com­pared to modem specifications written nearly a century later. But in the case of the SE, I believe we have an exceptional aircraft. True, the systems require more than normal monitoring, but they do provide several methods of getting fuel to the engine in the event of failure or battle damage. The coolant system also requires time and effort, but otherwise the operation of the machine is simple. In any event, the systems are easily learned and understood. Any pilot who is competent on the Tiger Moth would have no difficulty converting to the aircraft in very few hours indeed.

As a combat aircraft it was, and under the same circumstances, still would be an effective machine. Perhaps the only let-down is the characteristic Lewis gun over the top wing, but it just wouldn't look the same without it, would it?
  • 0

#49 hq_Reflected

hq_Reflected
  • Posts: 4711

Posted 10 January 2011 - 16:47

Excellent find, piecost!

Footnote: the Shuttleworth SE5a is the only airworthy original SE5a. Originally it was a Hisso, but it got re-engined with a Viper due to the unrileability of the Hispano Suiza. It also got some metal parts replaced with new ones made from heavier material so it's actually some 100lbs heavier than the service aircraft. (Source: my correspondance with the SW clollection)
  • 0

#50 piecost

piecost
  • Posts: 1318

Posted 10 January 2011 - 16:58

Glad you enjoyed it. I will post a couple of articles by other pilots of the same aeroplane. It is interesting to compare how modern test Pilots assess the same aeroplane.

I also have an article about the restoration and modification of the original Hisso engine. Perhaps I'll dig it out.
  • 0

#51 Kwiatek

Kwiatek
  • Posts: 680

Posted 10 January 2011 - 17:58

"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. "

Well i think Camel ROF had rather opposite left spin recovery procedure.
  • 0

#52 Chill31

Chill31
  • Posts: 1891

Posted 10 January 2011 - 21:01

What is your source for this quote?
  • 0

#53 Sensenmann

Sensenmann
  • Posts: 381

Posted 10 January 2011 - 21:27

What is your source for this quote?

Camel Spinning, Camel Gyroscopic couple & rudder effectiveness compared to the SE5a

taken 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.

Some aeroplanes spun even faster than the Camel, with a period of l to 2 seconds a turn, and 150 to 200 ft lost.

  • 0

#54 piecost

piecost
  • Posts: 1318

Posted 10 January 2011 - 23:51

From an earlier posts on the Camel spinning behavior: Current ROF Airplanes Flight Model Discussion Topic. post #73

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."

I suspect that Harold Penrose based his account partly on that of RONALD SYKES, he may even have discussed it with him (and other Camel pilots). In writing his definitive works on the early British Aviation industry he was well acquainted with former great war pilots and luminaries such as T.O.M. Sopwith. However, he did not have first hand experience of the Camel but had flown SE5, Bristol Fighter and Morane Parasol.
  • 0

#55 piecost

piecost
  • Posts: 1318

Posted 11 January 2011 - 18:14

Flying the SE5

from W1ngspan, Nov 1995

The flying characteristics that endeared the SE5a to the aces are equally appealing to those, like Mike Brooke, among the small coterie of experienced pilots entrusted with displaying the Shuttleworth Collection's example.

"The first time I flew it I was taken by its handling characteristics," he explains. "The very skittish, lively, low­stability fighters, like the Camel, Snipe and other rotary-engined aeroplanes of that era, were difficult to handle, both from an engine handling and control point of view. They were very agile, which is a good thing in a fighter, but it took a lot of the pilot's time and effort just looking after the engine and keep­ing the aeroplane going where he wanted to go.

"The SE5 is just the opposite. It handles like an aeroplane of the 'thirties or 'forties even. One of the aeroplane's technical advances - and I'm not sure if Folland, the SE5's designer, invented it, but he certainly incorporated it as a feature - was the variable incidence tailplane for trimming. It gives you very good control and stability. If they'd even bothered to have tailplane trimming before that it was probably done on a spring bias or movable tab on the control surface.

"You can almost fly the SE5 hands­-off because it's got what is called a good dihedral effect - if you put on a little rudder the aircraft yaws first then rolls in the same direction -so you can control any roll with your feet. I once wondered how long I could fly hands off, so on a transit flight from Farnborough to Old Warden I got it all trimmed out, took my hands off the stick and put them on a little coaming in the cockpit and kept my feet on the rudder. After 20 minutes I gave up, it got boring!"

… the eight hours flying time Mike has accumulated on the SE5a (out of his 7,000 hours total) does not sound much, but in relative terms it is quite high. ''There are aircraft at Shuttleworth that I still haven't flown for one hour but I've displayed four times. You end up not getting many hours, but flying a lot of trips"

So we are agreed that the SE5a is a stable aeroplane -a big surprise for something designed in 1916. And according to Mike Brooke it is not diffi­cult to manoeuvre despite being a little short of roll power. "You'd like it to roll quicker. At combat or cruising speeds it's alright, but at low speeds it's very poor. You actually run out of lateral control at a speed slightly higher than touchdown speed. So if a wing drops just before touchdown it's no good using the stick to try and pick it up. You can use rudder but it's not that powerful at that speed. So it's very important to land the SE5a into wind.

"They built the SE5a for safety. If you try to pick up the back-end of one you actually need two pairs of hands for sure, three preferably and four to make it easy. It's very solidly built. In the early days of the SE5 before it became the SE5a a couple of test pilots at Farnborough were lost through structural problems."

"You may have watched the SE5a return to its parking slot after a display and noticed fluid dripping from one or both of a pair of pipes ending on the top wing's trailing edge either side of the pilot's head." Mike Brooke explains: "It's a water cooled engine and you have to keep your eye on the water temperature and keep it round about 60 to 80 degrees. You have control over the radiator grills. If you let the water boil it's an embarrassment and I don't think it does the engine much good. The aircraft also gets its own back on the pilot. "The coolant tank is in the wing above the pilot's head and below the gun. The overflow from that comes in a little tube down to the trailing edge just above the pilot's right eye. If you let the engine overheat it tends to spit you don't notice it when you're flying, but when you land you get a face full off hot water. Oh, and on the other side is the gravity feed petrol tank overflow and that sometimes spits and throws petrol in your face. The worst bit is when you've stopped and petrol drips onto the exhaust pipe!"

None of which sounds particularly pilot friendly. But when flying the Shuttleworth Collection's aircraft you have to be prepared for the occasional eccentricity!
  • 0

#56 piecost

piecost
  • Posts: 1318

Posted 11 January 2011 - 18:34

Test Flying the RAF SE5A

by Roger 'Dodge' Bailey, The Shuttleworth Collection

Taken from World War 1 Aer0 No. 174

It is summer. Someone has painted the sky a uniform light blue except where an unseen hand has blotted off the colour to form scattered cumulus clouds. The wind is blowing from the northeast at the canter. I am standing in the open doors of a blister hangar looking at a small olive­ green biplane. In 2 days there is to be a big show; it is important that the machine be ready for it. Recent rectification requires that it must be air tested before the day. I am to fly the air test.

The engineers brief me on the repairs made and what to look out for and when. I try desperately to remember it all but know I shall forget something. The aircraft is pushed out of the hangar, rocking from side to side on its narrow track undercarriage. When it is arranged tastefully on the aerodrome, head to wind, the engineers busy themselves with last minute details. The chocks just so, the tanks all full and primed. I remove the pitot cover because I fear one day I will walk right past it and go flying with an unwanted extra streamer on the interplane strut. Now it is my turn to walk clockwise around the aircraft: I pretend to know what I am doing, waggling this and checking that. Are these ailerons differential? The radiator shutters are open (I must remember to check that the lever in the cockpit agrees and then close them for the warm-up). I double check that the pitot cover is off and having reached the tail check that the control surfaces are free to move.

It is time to mount the machine. Left foot in the 'kick-in' stirrup just behind the wing trailing edge and then, with one hamstring straining, kick, swing the straight right leg over the fuselage and over the headrest before bending it at the knee to let the foot fall into the cockpit and onto the seat. The left leg joins the right inside and I fidget my feet down into the negative G hoops on the rudder bar as I settle down into the seat. The cockpit is small and there is no room for a parachute.

Like many pilots nowadays the first thing I want to do is to fasten my harness, but in this aircraft there is one job to do first. A small brass tap residing somewhere close to my left ankle must be set in line with the adjacent pipe now, for once I am strapped in tightly it will be quite beyond my reach. Now, with the harness secure I am ready for the preliminaries to engine start. First I must pressurise the fuel tank to ensure a good feed to the engine. A mahogany-handled pump is used and always requires more pumps than I can comfortably provide without an aching left arm. Next, the confusing 5-position fuel selector is set to, let me see now, oh, yes, SERVICE TO CARB, and the air selector can go to FROM ENGINE AND HAND PUMPS.

I announce to the 2-man starting crew that I am ready to prime the engine. A typical start-up litany follows: "Switches Off, Throttle Set, Sucking In." As the propeller is turned one compression at a time, I operate another small pump adjacent to the first to squirt a mix of petrol and oil into the cylinders. I pump in 4 shots as the propeller is turned and then 2 more for luck. A voice from beyond the nose calls, "Ready for starting."

Now we must all be careful as there is danger for the prop swinger if things are not done just so. I can see the prop swinger with one hand on the propeller which he has positioned ready for the start; his other arm is held slightly unnaturally behind his back to keep it clear of the prop when it starts. Another engineer is immediately to my right ready to turn the starter magneto crank, although as yet his hand remains well clear. I call "Contact" to let everyone know that the mags are now live; and then call "3 -2 -1 -GO." On the "GO" the propeller is swung; when it moves, and never before, the starter mag is cranked vigorously. Nothing happens.

"Switches Off' comes from the swinger. I select the mags off and report "Switches OFF." The prop is repositioned and another call of "Ready for Starting" comes from up forward. The switches go on, "Contact" is announced, and then"3 -2-1-GO." This time we are rewarded by a distinct "Chuff"

We repeat the procedure twice more; we re-prime and try again. At last the motor fires and settles into a crackily fast idle, I adjust it to 1000rpm and check that the oil pressure is good and turn off the starter magneto.

There follows a longish wait for the coolant temperature to rise to the "opening up" value. Just before the radiator temperature reaches 60°C I open the rad shutters and once 60° is reached I circle a finger vertically indicating my desire to test the motor. One of the engineers holds the tail down while I hold the stick fully back (all this to prevent the aircraft from nosing over and smashing its prop) and open the throttle slowly to 1640rpm. The rpm needle oscillates slightly, averaging the required value. Now holding full back stick with my right forearm while guarding the throttle with my right hand, I reach over with my left hand to select the magneto switch from left to right.

On selecting L a small rpm drop is apparent but the engine note remains steady (that'll be OK). I move the switch to R and the rpm falls 150 and there is no mistaking the resultant misfiring. I throttle back, we have a brief discussion and 1ry once more with the same result. This won't do. I will have to switch off It is customary to open the throttle as the mags are switched off but this can be fraught with danger for the unwary. The ignition switches are not really on/off switches in the same way a light switch is. To ensure the motor will continue to run even with a faulty circuit it has been so designed that when the switch is ON the circuit is broken and the engine can run.

When the switch is turned OFF the circuit is made, earthing the magnetos and stopping the engine -but only if the circuit is complete. On one occasion I switched off and opened the throttle only to find the motor still running and the machine ready to nose over onto its prop. Not this time - with the stick held fully back, I switch off and when I hear the motor die I open the throttle. It stops cleanly.

Reluctantly I climb out and leave the machine with the engineers. Sometime later, after a plug set change and a clean-up of the suspect magneto, we are ready to try again. I have spent the time wandering among the other fighting machines here. There are scouts and 2-seaters in the various sheds, even a captured LVG.

We are getting old hands at starting this engine by now, and freed of its earlier ailments, it decides to give us no more trouble and settles into a good idle. The mag check goes off without further incident and I wave the chocks away. The aircraft is fitted with a new tailskid 'knife' giving better braking on the rock-hard grass than I would expect with a flat skid shoe. Directional control is no problem in these wind conditions even when taxiing downwind. I do the usual before-takeoff vital actions plus the specials for this aircraft. Seeing no other traffic approaching to land I let the aircraft turn into wind and roll forward slightly to straighten the tailskid. At this stage I take a few moments to have a last check around the cockpit. The oil pressure is 80psi, the rad temperature is 80°C, and the fuel pressure is 2-1/2 psi.

I open the throttle steadily - a slam will cause a distinct falter and cannot be good for the engine. The aircraft accelerates briskly, and before I have fully opened the throttle I have levelled the machine with forward stick. Now, with the tail up, I can see ahead. The aircraft is light on its wheels, skipping from bump to bump; I know it will fly now although I have not referred to the airspeed indicator. I select a climbing attitude and, when clear of· the trees, throttle back to 1800, the airspeed is 70mph; too fast; the best climb speed is 60, so I adjust the attitude and wait for the speed to settle.

As this is a test flight I climb straight out at the best climb speed monitoring the fuel pressure, the oil pressure and the rad temperature until reaching 1000'.

All is well. I look for other aircraft but the sky, as usual, seems completely empty. I have started a stop-watch on takeoff and I will time the climb to 3000'-it will be a good indication of the power output from the motor. During the remainder of the climb I monitor the engine instruments, look out for aircraft, particularly from the southeast -into sun -and make a series of gentle left turns to keep the aerodrome insight under the port trailing edge.

As height is gained I can relax a little, as height is like having savings in the bank, it provides a breathing space should things go wrong, in this case should the motor choose to stop. I can use height to try to diagnose the problem -Fuel? Ignition? Icing? I might attempt to restart if possible-or, if not, I can plan a leisurely glide landing back onto the aerodrome. The pilot that flies the machine next may not be able to climb as high as I can today, nor spend as much time studying the engine instruments. He may be flying in formation, with all his attention on his leader, or he may be manoeuvring aggressively concentra­ting on the task in hand but, if the engineers have done their job well and I do mine then, with luck, he should have no unpleasant surprises.

The scattered cumulus cloud is at 3000' and it has taken 4 minutes from the start of the takeoff roll to reach it. Now staying within about one mile of the airfield I throttle back gently to make sure that the motor continues to run at the minimum throttle setting. I slow down until the machine stalls at 4[sic] mph and note that I can keep the wings level into the stall; and although there is no appreciable warning the stall, when it occurs, is a classic wings level nose-drop. After 2 more I am satisfied that the stall handling is as it should be, the airspeed indicator is telling something close to the truth and the motor will not stop if the throttle is closed at low speed.

The next pilot may have to dive to high speed and fly combat manoeuvres against an enemy aircraft and so I must make sure the aircraft and engine are ready. I dive to 120mph keeping the rpm under 2000 by throttling back. Then pull up briskly, keeping the wings level until the horizon is lost under the nose, then by looking left at the upper wingtip I keep the orientation of the loop correct with small rudder and aileron inputs before looking back for the inverted horizon. When this arrives it is clear that I have not kept the wings completely level; but there is adequate control to put matters right before completing the loop, watching the airspeed and the rpm in the recovery dive to keep them well within the limits.

I wing-over to the right to stay close to the field, and from the wing-over I start a barrel roll to the left getting the nose very high to make sure that the second half is not over steep. From the barrel roll I pull up for a stall turn to the left. In a classic stall turn the aircraft is pulled into vertical flight and as it stops rudder is used to yaw the aircraft through 180° so that it is pointing vertically down; recovery to level flight is thereafter straightforward. It is a good way of reversing direction. But at the point of yawing the airspeed is near zero; and as the effectiveness of the flying controls is proportional to the square of the airspeed (and the square of zero is very small!) it is quite easy to put the aircraft into a position from which the pilot cannot prevent a tailslide or flop-over which could damage the aircraft or its engine. It is not my job today to see if I can break the machine, so I cheat on the stall turn by starting the yaw with airspeed to spare. With some opposite aileron to overcome the dihedral effect the machine rotates through the first 90° under control; the next 90° is considerably aided by gravity acting on the motor and this pulls the nose down cleanly. I try one the opposite way; the only difference seems to be less aileron control required to stay in the plane of the manoeuvre but this is due to the effect of the propeller.

Confident now that the machine can climb and manoeuvre, I dive to the maximum speed checking the engine rpm, temperatures and pressures - everything is fine and the motor sounds smooth and untroubled. I carry out a series of simulated strafing runs, tracking a point with the gun sight, and once again the aircraft responds normally. A few low-level evasion manoeuvres between the trees complete the test, and now I pull up and position for the landing. The fuel pressure and oil pressure are both good and the rad temperature is 85°. I slow the aircraft on the downwind leg to 70 mph and when abeam my landing point turn in, throttling back and easing the speed to 65. As the speed reduces, with little slipstream from the prop over the fin and rudder the machine feels very loose directionally: it is quite happy to yaw this way and that, and my attempts to control its head accurately meet with only limited success.

Now, lined up into wind I bring the speed back to 60 and wind the tail trim fully nose up for the landing. As I get closer to the ground I flatten out, surprised initially with the sensitive pitch control; but this is a scout after all. I memorised the 3-point attitude before takeoff - it is when the lower edge of the windscreen mount is on the horizon. I start a scan between the attitude the height as judged by looking at the grass just forward of the port wing. The ideal landing will be achieved if I put the aircraft in the 3-point attitude as the height reaches zero the wheels skid and the tail skid will all touch at the same moment. So much for the theory.

The aircraft touches down slightly before reaching the 3-point attitude - this aircraft exhibits little float once the throttle is closed and with the speed below 55 it stops flying pretty quickly. This is what has happened, and it has caught me by surprise - the aircraft has dropped the last foot onto its main wheels -I should know better by now but I must be a slow learner. The aircraft responds out of all proportion to the magnitude of my misdemeanor. It buckets from wheels to skid and back again 4 or 5 times but never more than a foot or so from the ground.

When things settle down I add power smoothly to take off again. I fly a tight circuit, not above 400' or more or less continuous 360° turn until I am flattening out again 3 or 4' above the grass. This time I am ready, and steeling myself not to over control, I hold off a little longer and this time things work well and the machine sets down without drama on 3 points. It starts to swing gently to the left, but this is easily corrected and I am reminded once again to pay attention to directional control right down to walking pace. I turn to the left through 90° and fully open the rad shutters. A leisurely taxi back to the sheds allows the heart to slow down and subside gratefully to its proper resting pace. Now it is possible to wear a confident smile and assume the appropriate aviator air.

I am met by the engineers who are eager to know if the machine needs more work. At their request I repeat the magneto check and then I allow the engine to idle for a while noting that the oil pressure is now down at 30psi. The idle rpm is lower than the book figure and really should be adjusted up; but there was no tendency for the motor to stop, and if set too high it makes for over-long landings.

After allowing the motor to idle for a couple of minutes I switch off and climb out of the cockpit answering the engineers questions as I go. We agree that the machine is ready for service. I walk towards the flight offices with a certain lightness of step because it is always a pleasure to fly a Scout Experimental No 5.

Where did this test flight take place and when? The depot at St Omer in August 1918 just before the big push at Biaches? No, at Old Warden Aerodrome, August 1999.
  • 0

#57 gavagai

gavagai
  • Posts: 15541

Posted 21 February 2011 - 04:53

Dr1 airspeeds, from http://www.theaerodr...imum-speed.html" onclick="window.open(this.href);return false;">http://www.theaerodr...com/forum/aircr … speed.html

Attached File  FokkerDrI.jpg   64.56KB   870 downloads

This is from Profile Publications:

Attached File  Dr1profilepublications.PNG   1.34MB   868 downloads
  • 0

#58 piecost

piecost
  • Posts: 1318

Posted 21 February 2011 - 16:33

Measured Turn Radius Compared to Theoretical Turn Radius
Measurement and Assessment of Roll Rate


To all those of us who like to calculate and discuss the maneuvering performance of Aeroplanes read on:

"February 1919 THE AERONAUTICAL JOURNAL page 58

CONTROLLABILlTY

Controllability of an aeroplane is mixed up to a great extent with stability; and it has proved even more difficult to measure. One of the simplest measurements that can be made is that rate of turn. It can easily be shown, that there is a smallest circle in which a machine can be turned. If the controls are powerful enough the diameter of this circle does not depend on the size or speed of the Machine. This is supported by experiment, though when we come to aeroplanes ,so large that the diameter of the smallest theoretical turning circle is approximately the same as a dimension of the machine, it will probably cease to be true.

The wings of the aeroplane will give the most lift at stalling angle and the lift will be the weight of the aeroplane multiplied by the ratio of the square of the actual speed and the stalling speed. If, then, the aeroplane is banked vertically this force will supply the necessary central acceleration, i.e., the mass of the aeroplane multiplied by the square of the actual velocity and divided by the radius. Therefore the minimum radius is independent of the mass or the actual speed, and is given, in fact by (stalling speed in ft/sec)^2 divided by the acceleration due to gravity. For a stalling speed of 40 m.p.h., say 60 ft/sec., the: radius is about 110 feet!

Many measurements of the smallest radius of turn which could be obtained have been made either by taking the time of a turn at a known speed, or by camera obscura observation; The maximum acceleration has also been measured directly. It was found that for a number of different aeroplanes the minimum radius obtainable was above the theoretical minimum but by an approximately constant percentage (20% to 30%).

This measurement is comparatively simple because the motion is fairly steady. When we come to lateral and directional control the motion cannot be kept up indefinitely, and not only does measurement become difficult but the actual motion ,of the aeroplane depends on the pilot and how rapidly he moves the controls to predetermined positions. For instance, the time to bank to 45° was measured for several aeroplanes and the time recorded was found to vary in apparently identical conditions on the same aeroplane with the same, pilot by as much as 1 sec in about 3 secs. By great care a very valuable set of results was eventually obtained, which did clearly put the machines in an order of excellence, but that order was not probably the order of preference of the average pilot of these machines.

I am not sure that we shall ever be able to put down everything about this very debated question in black and White -personal preferences will always have a good ideal of weight. But we shall gradually clear up a number of uncertainties which now trouble us, and the time will come when we shall be able to say to a man, “that machine will do what you want, and it is better for that purpose than any other machine," and if he says he cannot make it do what he wants, we shall be able to say that the fault is his and not the machine's. At present we are inclined to hang our heads and go away and try to satisfy what are in many cases purely personal prejudices."
  • 0

#59 piecost

piecost
  • Posts: 1318

Posted 21 February 2011 - 16:36

Propeller Slipstream at the tail of a BE2C

THE AERONAUTICAL JOURNAL February 1919 page 34

FULL SCALE AEROPLANE EXPERIMENTS.

Air Speed Exploration.

The air speed and direction at various points near the aeroplane have an important bearing on many parts of the design; particularly in connection with the tail surfaces, and with the propeller. This work is quite straight forward. I am unable give details of results, as I could not devote enough time, to the subject to do justice to it. Fig., 9 shows a very, striking result obtained by this method. The peculiar and irregular shape of the lines of equal speed give an Indication of the difficulties which face the designer when attempting to estimate the mean speed of the air over the tailplane; for example. This applies to many of the experiments which I shall mention briefly, and I hope the deficiency will be made up later on by others with more experience of the details of the individual experiments.

Attached Files


  • 0

#60 piecost

piecost
  • Posts: 1318

Posted 21 February 2011 - 16:40

Normal Accelerations on an SE5a and Bristol Fighter During Mock Combat

February 1919 THE AERONAUTICAL JOURNAL page 59

By permission of the Controller of the Technical Department of the Department of Aircraft Production I am able to show three typical records, two of accelerations on a scout and a two-seater during a mock fight (Fig. 13) and one of acceleration during various stunts (Fig. 14). Many other experiments have been made, loops, rolls, spins, etc., being shown very beautifully.

This instrument not only gives us, in any circumstances, the resultant air force on the aeroplane - information which is essential for fixing a proper standard of strength - but it also measures the time of rapid manoeuvres. It is no exaggeration to say that it is one of the most valuable instruments for aeroplane experiments - and, indeed, for any other experiments where rapidly changing accelerations have to be measured that has been devised.

Attached Files


  • 0

#61 piecost

piecost
  • Posts: 1318

Posted 23 February 2011 - 22:10

Drag of Stationary Propeller

THE JOURNAL OF THE ROYAL AERONAUTICAL SOCIETY, PROCEEDINGS, ELEVENTH MEETING, [piecost: 1924 IIRC]
THE WORK OF THE AERONAUTICAL RESEARCH COMMITTEE'S PANEL ON SCALE EFFECT

W. S. Farren M.B.E.

(3) The drag coefficient of a similar aeroplane, when gliding with the air­screw stopped, does not differ from that of the model by more than ±0.002. This is about ±7 per cent. of the minimum drag of the whole aeroplane and airscrew, or about ± 10 per cent: of the whole aeroplane without the airscrew.

[piecost note: Old British Drag Coefficient KD = 0.5 x CD ]
  • 0

#62 piecost

piecost
  • Posts: 1318

Posted 23 February 2011 - 22:36

Flight Tests & Wind Tunnel Lift & Drag Measurements of SE5a, SE5b & SE5c

THE JOURNAL OF THE ROYAL AERONAUTICAL SOCIETY, PROCEEDINGS, ELEVENTH MEETING, [piecost: 1924 IIRC]
THE WORK OF THE AERONAUTICAL RESEARCH COMMITTEE'S PANEL ON SCALE EFFECT

W. S. Farren M.B.E.

The next diagram, Fig. 8, shows results of experiments on S.E.5. with two shapes of body and two arrangements of Wing structure. The standard machine agrees completely with the model. These results are in fact Fig. 8. of Mr Wood's paper. On the other hand, with a different body and the same wings, there is a large discrepancy, the reduction of resistance predicted by the model not being realised on the full scale. The third set of results representing the remaining combination of wings and body suggests that the discrepancy is not due to the wings.

It may be noted that full scale experiments have also been made on these aeroplanes with the airscrew running. It is not easy to draw clear conclusions from a comparison of the results with those shown in Fig 8 since the radiator associated with the streamline body exposed, whereas it was withdrawn on the glides, but the broad result that there was hardly any difference between the performance of the two full scale aeroplanes, with the airscrew running, is not without interest.

[piecost note: Old British Drag Coefficient KD = 0.5 x CD, old British Lift Coefficient KL = 0.5 x CL ]

[piecost note: curves represent wind tunnel, dots are flight test measurements]

Attached Files


  • 0

#63 piecost

piecost
  • Posts: 1318

Posted 15 March 2011 - 19:38

Flight Test Measured Performance of Fokker EIII, Sopwith Triplane

NOTES ON THE PERFORMANCE OF AEROPLANES, BASED ON A REDUCTION OF THE OBSERVATIONS MADE AT THE CENTRAL FLYING SCHOOL DURING THE ACCEPTANCE TESTS OF AEROPLANES.

By L. BAIRSTOW, A.R.C.Sc., F.R.S., E. F. RELF, A.R.C.Sc., and C. H. POWELL, B.Sc.

Reports and Memoranda, No. 268. February, 1917.

SUMMARY.-The data were supplied by the War Office through the Aeronautical Inspection Department. Records of speed, both at ground and at height, and of the climb of aeroplanes, were given with a curve showing the engine output at given revolutions. Drawings of the machine were also supplied with details of weights of parts; the latter are not made use of in this report. From the original observations, curves have been deduced which show the horse-power necessary for horizontal flight at different speeds and the way in which speed and climb depend on the ratio of available horse-power to weight of aeroplane. The differences due to change in aerodynamic qualities are far less than those arising from the variation in h.p. per unit weight. Further data are being collected, but a summary of the records to date is given in Figs. 9 and 10, from which the performance of a well-designed aeroplane can be pre­dicted with moderate certainty.

Records of speed attained and rate of climb possible in a number of aeroplanes, together with a number of details as to weight, size and strength, have been provided by the Air Department of the War Office. The present abstract deals with all machines for which results are available, and has reference mainly to their performance, i.e., to the maximum speed and rate of climb reached by a given aeroplane with given engine.

The range of b.h.p. in the eight aeroplanes under considera­tion is from 100 to 350, and the weight variation from 1,400 to 8,000 lbs., the larger engines being in the heavier aeroplanes. The b.h.p. per 1,000 Ibs. weight of aeroplane ranges from 40 to 90.

The conclusion of greatest importance in the comparison is that, although machines differ in their aerodynamical qualities, the b.h.p. per 1,000 1bs. is the factor on which differences in relative performance largely depend.

The original records were not obtained specifically for the present purpose, and it ,vas necessary in the first place to decide on the best method of reduction. It was found that the data reduced satisfactorily if two main assumptions were made :_

(a) that the engine horse-power at a given number of r.p.m. was proportional to the density of the air, and

(b) that the horse-power absorbed in climbing could be estimated from the observed rate of climb, the known ­weight' of machine and a propeller efficiency of 70 per cent.

Assumption (a) may be wrong if the carburetor is not working satisfactorily at high altitudes, the error always being an over­estimation of the b.h.p. A check on the assumption is provided by the observation of the rate of climb at various altitudes, the observations giving a group of points (which is usually very well defined) at a low flying speed.

The original observations consist of :_
(I) Weight of aeroplane, usually by direct measurement.
(2) Speed at ground and at some considerable height such as 8,000 or 10,000 ft.., the speed being obtained by measured times of flight to and fro over fixed camera­-obscuras. In some instances speeds at many heights are recorded; observations of this character are most valuable in estimating aerodynamical performance. In. a few instances experiments were made with the engine throttled and, generally speaking, such observa­tions are useless for present purposes. Where there is a considerable difference between the propeller revolutions at top speed and at maximum climb there
is a possibility of deducing results of moderate accu­racy from observations with the engine throttled.

(3) Readings of time and aneroid durinq a climb.-These, after reduction, fix fairly definitely the maximum rate of climb and the least horse power for flight. The speed readings are taken on an air-speed meter and have to be corrected for density. The correction is sometimes made at the Central Flying School and sometimes at the National Physical Laboratory; in the latter case the fact is specially indicated, and little doubt exists as to the necessity for such correction when it has been made. In obtaining the rate of climb smooth curves through observed points have been used.

(4) Curves showing b.h.p. of engine when" all out."-These are almost always taken from the results of tests on the type and not from a test on the actual engine used in the aeroplane.

The observations mentioned above are given in Tables 1-8, with the exception of the b.h.p. against revs. This is plotted in the figure relating to each aeroplane. A drawing of the machine, to a uniform scale, is given in each case.

The results of the reduction of the original observations are finally presented in a form which shows, graphically for each aeroplane, the time of climb to a given height, the rate of climb at a given height, and the b.h.p. necessary for hori­zontal flight at various speeds near the ground. The curve showing b.h.p. for a given speed of horizontal flight near the ground is obtained by the methods described in R. & M. 216 (Section i), and is as follows :- Having deduced the speed, rate of climb, and h.p. for flight at any height where the relative density is rho, the corresponding experiment at ground level has been obtained by multiplying each of these quantities by rho^0.5 The details of the calculation are indicated in the table relating to the Sopwith Triplane. Values of rho have been taken from the Meteorological Glossary, assuming a value unity at the ground in all cases. Corrections for the variation of barometer and temperature at the ground have not been made, as the addition to the accuracy of the present analysis which would have resulted appeared to be negligible.

[TABLE]

It is understood that the data relating to weights and factors of safety as given in the original record need careful revision before acceptance, and for this reason no deductions are based on the results available.

In some cases it has been necessary to make estimates of various items of importance; where such has been the case a note has been given. It is believed that no estimate has been made where the possible error would vitally affect the conclusions arrived at.

Figs. 9 and 10 contain the result of all reductions on the basis of b.h.p. per 1000 lbs. Curves have been drawn which indicate roughly the limits of performance of modern aeroplanes, and these may e of use in drawing up new specifications.

Attached Files


  • 0

#64 piecost

piecost
  • Posts: 1318

Posted 31 May 2011 - 17:27

Windage Experiments with a Model of Rotary Engine B.R. 1

Reports & Memoranda No 448, May 1918

[power loss due to Windage on Bentley BR1 Powered Sopwith Camel]

Attached Files


  • 0

#65 piecost

piecost
  • Posts: 1318

Posted 31 May 2011 - 17:30

Windage Experiments with a Model of Rotary Engine B.R. 1

Reports & Memoranda No 448, May 1918

[power loss due to Windage on Bentley BR1 Powered Sopwith Camel]

…continued

Attached Files


  • 0

#66 JG14_Josf

JG14_Josf
  • Posts: 7

Posted 24 June 2011 - 17:34

Hi,

In response to a specific quote:

+++++++++++++
“that machine will do what you want, and it is better for that purpose than any other machine”
+++++++++++++

Those who choose the numbers to be programmed into the software know which plane is modeled with a higher number relative to another number, and depending upon how the program processes those variables, the scale of best to worst on any variable may simulate airplane performance variables well enough for simulating WWI air combat.

One number can be increased, and that flight model changes relative to the performance that flight model had before the change was made, and relative to all the other flight models of all the other planes that were not changed.

I am introducing my forum post in this manner so as to be very clear about my viewpoint from the start, as I offer my viewpoint for consideration, or moderation, whichever the case may be.

The code is indisputable, it is what it is, (unless the code can be altered without the knowledge of the programmers or the other simulating pilots), but the simulating pilots may not be able to manipulate their controls as well as another simulating pilot can, and in this way reality is well simulated - so long as no one is able to adjust the code independently.

Every plane in reality, as well as every plane in simulation, is capable of maximum performance, and then there is the human variable, since not every pilot is capable of reaching the maximum performance of the plane.

How can the measure of that difference between what the plane can do and what the pilots can do be documented, so as to avoid having too much focus and attention misdirected toward accusations of cheating, and or, inaccurate flight modeling?

I think that there are easy ways to measure that difference, and they are well worth doing.

One person already posted a very well written explanation of the problem, and an explanation of the solution to the problem.

I’ll quote some of that:

++++++++++++++++
Many measurements of the smallest radius of turn which could be obtained have been made either by taking the time of a turn at a known speed, or by camera obscura observation; The maximum acceleration has also been measured directly. It was found that for a number of different aeroplanes the minimum radius obtainable was above the theoretical minimum but by an approximately constant percentage (20% to 30%)
+++++++++++++++++

And:

++++++++++++++++
By great care a very valuable set of results was eventually obtained, which did clearly put the machines in an order of excellence, but that order was not probably the order of preference of the average pilot of these machines.

I am not sure that we shall ever be able to put down everything about this very debated question in black and White -personal preferences will always have a good ideal of weight. But we shall gradually clear up a number of uncertainties which now trouble us, and the time will come when we shall be able to say to a man, “that machine will do what you want, and it is better for that purpose than any other machine," and if he says he cannot make it do what he wants, we shall be able to say that the fault is his and not the machine's. At present we are inclined to hang our heads and go away and try to satisfy what are in many cases purely personal prejudices."
++++++++++++++++

An actual flight test in the real world that varies from the mathematical calculated result by 20% to 30% is a knowable measure of the difference between math and reality.

An actual in game flight test will also measure the difference between programmed performance numbers and actual simulated performance numbers; as far as know.

As far as I know the programmers do not plug in one number that enables a plane to turn a 5 g turn for 4 seconds before the plane slows down below a velocity sufficient to generate enough lift to maintain a 5 g turn, and then, as far as I know, the programmer can’t just increase that one number to cause that plane to turn on second longer in that 5 g turn before it stalls out at 5 g, in a downward spiraling energy burning turn from level flight top speed initially.

And that last sentence is my introduction to Energy Maneuverability; something not known until John Boyd developed the theory after World War II, and still used up to today, as far as know, in the modern military.

The programmers have a set of variables that intend to simulate flight, and from my experience with actual flying, the program in this simulation is very good, and good enough to begin quantifying just how good, by simulating the actual tests done by modern military pilots as those tests are done to quantify which “machine will do what you want, and it is better for that purpose than any other machine”

If there is one measure to pick as the best measure of relative combat performance then that measure is acceleration, and I can explain how that measure may solve a whole lot of problems for those who are interested in simulating air combat.

One forum member has already expressed this variable.

I’ll quote it:

++++++++++++++++
gavagai - think of two cars drag racing a 1/4 mile racetrack.

CAR A - crosses finish line at 4 seconds @ 200mph
CAR B - crosses finish line at 5 seconds @ 210mph

…which car is "slower"?

You can have a faster top speed - but take longer getting there. I suspect that is the Pup
++++++++++++++++

In modern performance testing there are a few very handy tests done so as to quantify Energy Maneuverability and one test is the level flight acceleration test. Please consider using that test, if no other test is worked out well enough to be truly useful.

If only one pilot tests any test, there will be no check on that result, and therefore the possible error may be well over 30%.

If only top speed is tested, the information needed to quantify relative combat performance is not being tested, as the drag race illustration above proves, beyond any doubt.

If the most important test is picked, by one person, what would be that one most important test? Many people may be inclined to pick a level flight sustained turn rate test, and if so, that test is about as useless as a top speed test, since it only measures maximum performance at minimum sustained speed, it does not quality any Energy Maneuverability, other than the bottom of the performance flight envelope.

Sustained level flight turn tests measure the bottom, and top speed measures the top, but what about all the needed information in the middle? A good start is level flight acceleration tests.

The level flight acceleration test is much more informative than either level flight sustained turn tests and/or top speed tests.

If multiple simulated level flight acceleration tests are done, by many people doing the tests, an average result will be possible, as well as a high end result, whereby one exceptional person will record that absolute best result until someone else can show a higher result, and the average may be more useful than the extreme, and explaining why the extreme result was extreme may uncover some useful information too.

If multiple simulated level flight acceleration tests are done, and plotted on graphs, there will be multiple acceleration curves to compare one against the other.

Each test should be done at the same altitude, and done the same way, and what will be measured is valuable information, because acceleration is the essence of Energy Maneuverability, or the ability to out maneuver an opponent.

On the graph there will be a vertical axis from zero acceleration on the bottom to maximum acceleration going up the left, vertical, scale; while airspeed is plotted on the horizontal axis, zero airspeed in the lower left and maximum airspeed at the far right.

Each plane flown by each pilot will reach a maximum rate of acceleration somewhere in the middle of the speed range, and each flight test will plot out as an arch, down to zero acceleration on the left at the level flight stall speed, and up to maximum acceleration in the middle, and back down to zero acceleration on the right at top speed in level flight.

Induced drag will be high at slow speed, and lessening as speed increase in level flight, and form drag (or parasite drag) will increase square with velocity as the plane accelerates, and drag will be minimum in the middle, where Corner Speed is found.

Corner Speed, a very useful bit of information, is roughly the same speed as maximum acceleration in level flight – not at stall, and not at top speed, somewhere in the middle the plane accelerates fastest, and all planes are not the same, and all pilots are not the same.

A plane having better Energy Maneuverability will be easy to see once it is plotted over a worse Energy Maneuverability plane, as the better curve on the graph will rise over the worse plane in the middle where the better plane accelerates better, and much better will be much better, and much worse will be much worse for many reasons, and one plane may accelerate much better at a higher speed compared to an almost as good plane that accelerates almost as good but at a much lower speed.

The flight tests will prove the facts, and the superimposed diagrams will show the pilot which “machine will do what you want, and it is better for that purpose than any other machine”.

Having a higher rate of acceleration in the middle is, in essence, a higher Specific Excess Power measure, which is the essence of John Boyd’s Energy Maneuverability Theory, which led to such things as The F-16, the F-18, and the A-10 Warthog.

Try it, you may like it.

A higher Specific Excess Power measure means that the plane will have more power and less drag compared to the other plane at the speed where it accelerates better, and then it can be assumed (but not confirmed without other tests that test specific things) that wing efficiency in turns is similar, and therefore the higher Specific Excess Power plane, at that speed, can maintain higher G force, tighter turns, and do so for a longer period of time, until speed drops down to a speed where acceleration is equal, or drops down to a speed where one plane can keep turning at a slower speed, such as a sustained level turn speed that is slower; which will also show up on the level speed acceleration tests, since level flight is equal to a 1 g turn, and a 2 g turn is akin to a level flight on a planet with 2 gs, the plane flying slower still flies slower.

Once the graphs are made, by enough people making them, it can be known which plane accelerates faster at which speeds, graphically illustrated, each plane superimposed over another, and there is no way that a slower accelerating plane can out maneuver a faster accelerating plane, by definition. What the pilot can do with angles, slips, geometry, or teamwork with other wingmen, is another story.

Level flight acceleration tests, plotted on graphs, each new test adding to the measure of reality (simulated reality), consistently improving the average measure, and consistently proving the maximum achievable limit of that range, can begin to document the question asked, and do so with a valid answer.

This:

Which “machine will do what you want, and it is better for that purpose than any other machine”?

How the programmers get the information they need to improve the game can be a useful feedback loop, or it can be something less valuable.
  • 0

#67 piecost

piecost
  • Posts: 1318

Posted 24 June 2011 - 21:51

JG14_Josf,

Interesting read
  • 0

#68 Chill31

Chill31
  • Posts: 1891

Posted 25 June 2011 - 02:53

I have some questions Josf

What are you meaning by corner speed? Corner speed is independent of acceleration and even independent of excess power. It is the lowest speed where maximum Gs can be applied.

Why would you say that acceleration is a better test than max level speed or max sustained turn rate?
I submit that max level speed is one of the most important test we can do in this sim due to the fact that max level speed is one of the few quantifiable facts we have on aircraft performance.

As for max sustained turn rate, that is based on excess power and wingloading, one of which is, again, one of the few quantifiable pieces of data that we have for these old planes.

I'm not denying that acceleration is an important and informative test! I'm only contending that acceleration is not the most important test we can perform when working on FMs/data for RoF.
  • 0

#69 WWBrian

WWBrian
  • Posts: 2418

Posted 25 June 2011 - 05:50

Why would you say that acceleration is a better test than max level speed or max sustained turn rate?

Hey Chill,

I'm sure you're quite sick of me by now, but since I am the one he quoted about CAR A and B, it sounds like you may not be giving enough emphasis on acceleration.

Now I certainly do not want to start another discussion about torque and horsepower, but just look at the old ten-speed bicycle you rode when you were a kid:

In 1st gear, you had great torque and acceleration…..yet your top speed might have only been, let's just say 5 miles an hour for sake of argument.

….but put that same bike in 10th gear, and you had less torque, thus slower acceleration…but a higher top end speed, of, let's say 20 miles an hour.

Start from a stop in 10th gear and see how long it takes you ( in minutes) to get to the same 5MPH, as you would in first gear. Higher top speed, but lower acceleration ( due to less torque - and same horsepower).

Furthermore, better acceleration can help sustain a better turn by "pulling" you through it.

Give it some thought.
  • 0

#70 Chill31

Chill31
  • Posts: 1891

Posted 25 June 2011 - 13:47

Again, not saying acceleration isnt an important performance parameter for a fighter.

All I am saying is that its not the most important test for working ROF FMs. We are discussing this in the FM data topic. Not the general aircraft performance topic. So in the interest of working the FMs, acceleration doesnt help us match any quantifiable data.

So I only want to point out here, before people say Chill31, why didnt you include an acceleration test when working on the FMs?? Well the reason is very simple, it doesnt produce any useful data point for working on these FMs.

Sustained turn is based on wingloading and excess power (we've already had the power vs thrust convo). Your wing has to be able to support higher G loading (lift) and you have to have enough extra power to support the increased drag. While acceleration might also be based on excess power, acceleration itself is not what is helping the turn.
  • 0

#71 JG14_Josf

JG14_Josf
  • Posts: 7

Posted 25 June 2011 - 23:10

What are you meaning by corner speed? Corner speed is independent of acceleration and even independent of excess power. It is the lowest speed where maximum Gs can be applied.


Chill31,

Corner speed is the slowest speed at maximum g, yes, that is what I mean, and if that isn’t exactly the same speed as the maximum rate of acceleration in level flight, then it isn’t, but it will be close to the same speed – it turns out that that is the essence of my point.


Why would you say that acceleration is a better test than max level speed or max sustained turn rate?


Please leave me out of it, as much as possible. It does not matter what I think, or why I would say anything, what matters is reality. So the question then is weather or not an acceleration test is a better test than a max level speed test, and another question is weather or not a level flight acceleration test is better than a level flight sustained turn test at minimum speed?

What is the goal? What is the point? Leave the opinions and subjectivity out, please, and if the goal is known, then competitive ways to get closer to the goal can be known.

Before I answer with what I know, I will quote your submission to the discussion on that point:


I submit that max level speed is one of the most important test we can do in this sim due to the fact that max level speed is one of the few quantifiable facts we have on aircraft performance.


If the idea is to measure the game relative to the known documented performance measurements recorded in history only, and that is all, then that is all, but that viewpoint has little to do with my stated viewpoint. If you only want to test the top speed tests and/or the level flight sustained turn tests at the slowest speed, then I can agree with your decision to do what you want to do, and I have no desire to argue with you about your decision.

I can also quote from a previous post in this tread if the only thing of concern is the known documented performance measurements recorded in history (which is too limited for me, as I’ve tried to get out in front from the start – already):

+++++++++++++++
The most remarkable of all the tests carried out at Orfordness is described by Sir Vernon Brown: 'One of the things we did not understand was that as the aircraft got faster and faster, so we experienced some rather extraordinary effects. Ifwe flew fast and then did a tight turn, or, for instance, dived an aircraft and then pulled it over in too tight a loop, a sort of haziness crept up over one's eyes, rather like a mist; and there were occasions if you held the tight turn for too long where you almost passed out altogether. In order to find out what was happening Lt Jones asked me to fly at as high a speed as possible over a hut in which was a camera obscura, projecting an image on to a table. As the aeroplane, Sopwith N 5430, flew over the top its image was shown through the lens on to the paper, and, by means of a metronome, he was able to point it every second as I made circles. Mterwards it was a simple calculation to find out what "g" had been applied. The triplane was one of the best of all aircraft to do it in, because if you did black-out and lose control, it always recovered itself automatically. It was really very safe indeed.

'On one occasion the aeroplane went into the most peculiar manceuvre at one thousand feet, and it was because I had passed out altogether that I came to at only two or three hundred feet in a shallow dive. I was very lucky it hadn't dropped into a spin. During these experiments once, I heard an awful bang; I didn't know what it was, and I came down very gently. When I landed the whole centre-section was found to have been wobbling about, as one of the strainers on the centre-section bracing wires had broken. But otherwise the triplane was fortunately rather free from these vices. 'Lt Jones found by repeated experiment that I could hold 4'5 g for 10 seconds, and that I could hold 6 g for 4-5 seconds. The results of these experiments were sent up to the War Office.'
++++++++++++++++

With all that up front, above, I can answer the questions stated earlier (from my point of view, perhaps not anyone else’s point of view):

Is a level flight acceleration test better than a top speed test when the goal is “to say to a man, “that machine will do what you want, and it is better for that purpose than any other machine," and if he says he cannot make it do what he wants, we shall be able to say that the fault is his and not the machine's. At present we are inclined to hang our heads and go away and try to satisfy what are in many cases purely personal prejudices."”?

A level flight acceleration test can be a rough, but well enough, measure of corner speed, or the maximum g force, at the minimum speed, a plane, and pilot, can manage, to the limit of the plane, and to the limit of the pilot.

Obviously there is a problem in a game as a game may or may not model pilot g tolerance, or it may fudge that measure in ways that change the planes performance due to a change in the way the game changes pilot performance one way in one plane, and another way in another plane.

A level flight sustained turn at minimum speed is not the same thing as a maximum performance turn, and as the quote above records, in history, a WWI plane was able to turn to the limit of the pilot in a maximum performance turn at 4’5 g for 10 seconds, and that plane, flown by that pilot, according to the report, could hold 6 g for 4-5 seconds.

The point about measuring level flight acceleration is such that it begins to quantify Energy Maneuverability which is what was being done in that historical record quoted above even if those pilots did not call what there were doing by the name Energy Maneuverability measuring.

My point is such that reality is what it is, and the game is what it is, and even if the game is absolutely wrong relative to history it remains to be absolutely what it is when simulating WWI Air Combat with that game, so why not, my point, why not measure what it is, so as to do a better job of recording things, as they are, than the pilots did in history, since Energy Maneuverability is now known, and tests that can quantify Energy Maneuverability are also known now, and Energy Maneuverability may not have been fully understood back in WWI, or WWII, or the Korean War, until John Boyd’s work, at least not by academics, and perhaps only known by the best pilots who knew Energy Maneuverability intuitively.

I will link a chart that helps illustrate the value of Energy Maneuverability, and offer further commentary on that chart if there is a demand for it.

Image

Back to the response:

As for max sustained turn rate, that is based on excess power and wingloading, one of which is, again, one of the few quantifiable pieces of data that we have for these old planes.

There are ways to calculate, on paper, how well a plane will perform once the plane is built, and having only excess power and wingloading is missing the measure of drag, unless your words are not precise on purpose, but I can't read between the lines.

What do you mean by a max sustained turn rate? My guess is that you mean a level flight turn at stall, where altitude is kept constant, which is not the same thing as a max sustained turn rate such as a turn at corner speed in a downward spiral, which could also be called a max sustained turn rate, where the idea is to sustain maximum g force at the slowest speed, or if the idea is to sustain the minimum turn radius possible, or if the idea is to sustain the maximum turn rate possible, if the idea isn't to maintain altitude.

So if you are going to say "max sustained turn rate", then there is a missing factor, that factor is sustaining altitude, if that is what you mean. When sustaining altitude the maximum turn rate is certainly not a maximum turn rate, it is only a maximum turn rate while maintaining a altitude, and even then it is only constant if the air speed is held constant, at stall.

A pilot can dive into a level flight turn and then the plane can turn a level flight turn at a much higher turn rate initially, and at a much tighter turn radius, initially, and only as speed drops down to stall, will the turn rate decrease to a higher turn rate, and only as speed drops down to a stall, will the turn radius increase to a larger turn radius, and only as speed drops down to a stall, will the g force decrease to a lower g force, since all the energy bleeds off, and the airspeed is then down to the stall speed, and all the energy is bled off and the engine power, and wing lift, is all that is left to keep the mass of the plane held level with gravity, as all the energy is bled from the plane.

The energy of speed is used to tighten the level flight turn, increase g, and decrease turn radius when a plane is dove down to an altitude and begin a level flight sustained turn test.

There may be a common misconception concerning at which speed a plane will turn the smallest turn rate possible, and at which speed a plane will turn the highest turn rate possible, and that speed is well above the level flight sustained (sustaining air speed and altitude) turn speed.

The use of the word "max" may be misleading to some people.

As for max sustained turn rate, that is based on excess power and wingloading, one of which is, again, one of the few quantifiable pieces of data that we have for these old planes.

A max sustained turn rate means what in the above context?

I'm not denying that acceleration is an important and informative test! I'm only contending that acceleration is not the most important test we can perform when working on FMs/data for RoF.

And to be clear, my point is to point out how a feedback loop between developers and simulators can include simple tests, if done with care, so as to help quantify, more accurately, the Energy Maneuverability of the planes being modeled, such as was done according to the data quoted in this thread already.

Example:

'Lt Jones found by repeated experiment that I could hold 4'5 g for 10 seconds, and that I could hold 6 g for 4-5 seconds. The results of these experiments were sent up to the War Office.'

  • 0

#72 piecost

piecost
  • Posts: 1318

Posted 26 June 2011 - 02:35

Josf,

E-M theory must be the best framework for comparing combat aircraft, but, without detailed historic E-M information the resulting discussion is subjective.

Can you provide any guidance of how to derive E-M information based on scant historic data? The most common parameters are top speed and time to height (best rate of climb - often lacking best climb speed). I understand that best rate of climb is analogous to best sustained sustained turn (at constant height).

I have found normal accelerations during mock combat for a few British aeroplanes but the data lacks other parameters to enable further conclusion. As an aside - I believe that flight simmers routinely pull many times more the normal acceleration experienced in reality (post#60 shows peaks of 3g for an SE5a in mock combat with an F2B). Non-flight model related limitations of sims compared to reality effect this.

I also wonder about the relationship between optimism/conservatism of flight models relative to fidelity of available data. Could it be "unfair" to model one aeroplane to higher fidelity than another? This may introduce advantages or disadvantages relative to the competition.
  • 0

#73 Chill31

Chill31
  • Posts: 1891

Posted 26 June 2011 - 03:32

Josf, Piecost,

Reading your posts, we are all on the same page. My only intent in asking questions about Josf original post was to clarify that for the purpose of FM development and refinement, E-M is a secondary and subjective method for comparing ROF FMs.

Not that it matters at this point, I will clarify sustained turn rate. Sustained turn rate implies level flight. If you are diving to maintain corner speed or you pull Gs and bleed speed, you are not sustaining your turn as you will either hit the ground or stall.

Good discussion and information by all. Thanks
  • 0

#74 JG14_Josf

JG14_Josf
  • Posts: 7

Posted 26 June 2011 - 13:47

E-M theory must be the best framework for comparing combat aircraft, but, without detailed historic E-M information the resulting discussion is subjective.


piecost,

If you can find a subjective thing I’ve written, then please point it out.

Energy Maneuverability is the opposite of subjectivity, so your quoted words above make no sense in the context of what I wrote, unless I missed something.

Can you provide any guidance of how to derive E-M information based on scant historic data?

I don’t have all the historical data. I’ve read some of it in this thread, and I began to quote those quotes that are relative to Energy Maneuverability, so as to do exactly what you are asking. If you don’t read what I write, or if you don’t understand what I write, then I can explain what I write, or you can read what I write. What else can I do?

Example:

Can you provide any guidance of how to derive E-M information based on scant historic data?

OK, you ask that question after I’ve already quoted the following:

'Lt Jones found by repeated experiment that I could hold 4'5 g for 10 seconds, and that I could hold 6 g for 4-5 seconds. The results of these experiments were sent up to the War Office.'


I am not claiming that the quote above is the absolute, documented, official, un-subjective, un-“anecdotal”, accurate, unerring, facts, concerning a specific WWI plane modeled in the game exactly as it should be, or any other thing other than what I am writing.

That is supposedly something documented by someone who measured the stuff of Energy Maneuverability. Is that true or not?

The first thing that I did is to express that the programmers of the game have adjustments that they can adjust to make one plane do one thing and another plane do almost the same thing, but better, or worse, by some measure. What are those adjustable things?

I don’t know. That is certainly true, not subjective.

The programmers, or those who have their hands on the variables in the program, are the ones who control the reality of the simulation. That is not subjective, that is what they can do, or not do. True or false? If the reality of the simulation is similar to reality, then it is, if it isn't, then it isn't. How does one measure such things?

If they can tweak the settings to allow a plane to hold 4 to 5 g for 10 seconds, and they can tweak the settings in the game to allow a plane to hold 6 g for 4-5 seconds, or if they can’t then they can’t. I don’t know. That is not subjective. That is true; I don’t know.

Whatever information is chosen to be good enough information taken from history, and information that can be adjusted in the game, then they do that, and that is fine with me.

That is also not subjective. They decide what is good enough to use for adjusting, and if it can be adjusted in the game, then they decide to adjust to that chosen value: True or False? If it is fine with me, then I am expressing my subjective opinion; that it is fine with me that they pick whatever they want and plug whatever they want into the game, and then the game is what the game is, which is fine with me.

Then a user of the simulator can test, on a machine, to see if the adjustment to the code works as intended by the programmers on the system. Then two people using the simulator, on two machines, can test the same results of the change of that one variable. Maybe only one user, on one machine, can reach the intended adjusted variable, and maybe no one else can, just one user using the simulator on one machine can do that specific thing, that good, or maybe that one variable changed does not work on all the systems tested. Maybe the adjustment works the same exact desired way each time, at the same time, such as top speed, on each machine, with each user of the simulator, each reaching the same top speed at the same time, starting from the same starting point, at the same time, without any difference at all, all exactly the same, each time.

What are the documented facts concerning how well one user will reach one variable in how much time compared to all the other users of the simulator? I don’t know, it sounds like a move into subjectivity, or averages, or guesses, or means, modes, etc. I’m not being subjective about that; someone else may be subjective about that, not me. That is what that is: variables.

If the only reliable thing that the programmers can plug into the game is top speed numbers, and climb numbers: here this one does this top speed, good, this one does that best climb speed, that is good too, that is all, then that is what is done, to the limits of that which can be done, and the program works all the other things out based upon those two variables, and the programmers pick which is the best historical top speed numbers, and which is the best climb numbers. Who has a subjective problem with that, not me.

How long did the person in WWI spend at full throttle before documenting the top speed that is chosen? Did the pilot in WWI measure best climb speed or best climb angle or something in between? I don’t know how subjective the test was in history, and I’m not being subjective about that fact – I don’t know.

What happens at 6 g for 4 or 5 seconds in the game?

Back to the question:

Can you provide any guidance of how to derive E-M information based on scant historic data?

I will pick out something I already wrote, in case it was missed:

Obviously there is a problem in a game as a game may or may not model pilot g tolerance, or it may fudge that measure in ways that change the planes performance due to a change in the way the game changes pilot performance one way in one plane, and another way in another plane.

That was already noted.

A. Is the data good enough to use in the simulator? (6 g for 4 or 5 seconds)

B. Is there a method of adjusting the simulator to model pilot g tolerance?

C. Is there a method of adjusting aircraft g tolerance?

D. Is there a record of how much altitude and/or how much speed was lost while 6 g was held for 4 or 5 seconds - assuming that the historical data is good enough historical data?

E. Is there a record of the speed held while altitude is lost while 6 g is held for 4 or 5 seconds in that plane?

F. Can that plane, or any plane, hold 6 g for 4 or 5 seconds under any condition in the simulator?

If speed is known and g load is known then turn rate is knowable and turn radius is knowable. If any two variables are known, among speed, G force, turn rate, or turn radius, any of the other two variables can be known.

If the planes are modeled to break up under g stress then that is a variable.

If the pilots are modeled to black out under g stress then that is a variable.

If one plane just happens to be modeled to turn 6 g, and another just happens to be modeled to be limited to only 5 g, then at which speed is either turning their maximum g?

Can any of that important information be known if the programmers and users of the simulator are only testing top seed, climb rate, or even going out on a subjective limb and testing level flight sustained airspeed, sustained altitude, turn tests?

The most common parameters are top speed and time to height (best rate of climb - often lacking best climb speed). I understand that best rate of climb is analogous to best sustained sustained turn (at constant height).

I think that you may be entering subjectivity with that analogy. I am trying to do the opposite.

Best climb rate is a 1 g flight condition where the angle of attack on the wing will produce the best L/D ratio, or the most force of lift relative to the force of drag, and with the controls of the airplane carefully maintaining that angle of attack the plane will reach the highest possible altitude in the shortest amount of time.

Compare that to a climb at best climb angle. Please.

Then the pilot is tasked with the job of turning the smallest turn radius, highest g, and fastest turn rate, or something along the lines of the best turn possible, while maintaining the same altitude. Does the pilot maintain the same wing angle of attack as was done for best climb rate or does the pilot slow the plane down and go past best climb rate angle of attack, entering into the best climb angle region before stall?

Another illustration of that grey area between best L/D angle of attack and past that point is the difference between best glide angle and best sink rate angle of attack. At best glide angle the plane will go the farthest for a given altitude, and pitching the wing angle of attack further will not reach as far horizontally, but the plane does not lose altitude as fast if the wing is pitched past best L/D angle of attack, that is why it is called best sink rate.

Which is the best angle of attack to turn the highest turn rate? Probably the best climb rate, best L/D, angle of attack.

Which is the best angle of attack to turn the tightest turn radius? My subjective guess is that a little past best climb rate will offer a slower speed at almost the same g, and it may turn into a tighter turn, but, then again, the plane may lose some altitude. I think that you may be right, but I don’t know if the anecdotal assumption or analogy is precisely correct in each case of wing design, or each pilot.

How can you know for sure if all you do is test, record, and adjust top speed and best climb rate?

I did not quote, earlier, the looping data already provided in this thread, and going back to your earlier question:

Can you provide any guidance of how to derive E-M information based on scant historic data?


What is involved in a loop? A loop is very much a part of E-M performance data, and particularly important because the planes in WWI were high drag, low power, and compared to modern planes, or even many WWII planes, they are marginally able to perform loops; having relatively high drag, and relatively low power.

If in order to do a loop a plane has to almost reach maximum airspeed (to the limit of the air speed causing the engine to unload due to prop pitch limitation and therefore causing the engine to over-rev) and then the pilot has to maximize turn rate, by pitching the wing to best L/D angle of attack, and the wings are close to overstress at the maximum g gained at best L/D at maximum speed, and the pilot has to fight off black out, and the loop had better be over soon because the rate of energy loss (speed slowing down much faster than altitude being gained) is quick, and the plane is nearly stalled at the top of the loop, then you have corner velocity almost nailed down, at least as nailed down as the analogy between best climb rate and best “sustained” turn rate (speed and altitude being sustained while maintaining highest g in a turn), and included in the loop information is a measure of energy loss from maximum speed, through a maximum performance turn, down to stall, then loops offer some help, yes or no?

The limitations on corner speed include the pilots g tolerance limit at which it doesn’t matter if the plane limits are not reached, and so far all the fighter planes that were worth anything could exceed the pilot limits, and then there are plane structural limits, and as far as I’ve read in the information in this thread even the WWI fighter planes were stronger than the pilots who began to acclimatize themselves to g forces, 6 g for 4 to 5 seconds as one example.

Best climb rate is analogous to corner speed, both being at best L/D angle of attack.

Actual planes are subjective, more or less, as wing designs include some variations that test the limits of the math, and then a subjective person has to fly the thing, or simulate it.

I am trying to work the opposite direction of subjectivity, and being thrown under that bus is not appreciated, in the least, if that is what you intend to do with me, and this information.

I have found normal accelerations during mock combat for a few British aeroplanes but the data lacks other parameters to enable further conclusion. As an aside - I believe that flight simmers routinely pull many times more the normal acceleration experienced in reality (post#60 shows peaks of 3g for an SE5a in mock combat with an F2B). Non-flight model related limitations of sims compared to reality effect this.

The programmers will pick that which they pick for the reasons they alone have the power to decide, and that is, in fact, subjective.

1. Is pilot g tolerance modeled, if so how, if not why not?
2. Is plane g limit (structural limit) modeled, if so how, if not why?

I also wonder about the relationship between optimism/conservatism of flight models relative to fidelity of available data. Could it be "unfair" to model one aeroplane to higher fidelity than another? This may introduce advantages or disadvantages relative to the competition.

That is way too ambiguous to be objectively understood. without clarification. What is the meaning intended with the word choice: “fidelity”? Are you intending to use that word as a euphemism for “favoritism” or as already quoted: “purely personal prejudices”?

What is meant by fidelity?

What do you mean?
  • 0

#75 JG14_Josf

JG14_Josf
  • Posts: 7

Posted 26 June 2011 - 13:55

Chill31,

Reading your posts, we are all on the same page. My only intent in asking questions about Josf original post was to clarify that for the purpose of FM development and refinement, E-M is a secondary and subjective method for comparing ROF FMs.

I am not on the subjectivity page, leave me out of it, please. E-M is chosen, subjectively, to be secondary, by you, and whomever else chooses to do so, and if you can prove, objectively, otherwise, then you can, if you can’t, then what do you do instead?

Not that it matters at this point, I will clarify sustained turn rate. Sustained turn rate implies level flight. If you are diving to maintain corner speed or you pull Gs and bleed speed, you are not sustaining your turn as you will either hit the ground or stall.

A diving turn so as to sustain corner speed is a diving turn that sustained corner speed, and if the pilot chooses, subjectively, to pull out before hitting the ground, then that is what the subjective pilot chooses to do, just like a pilot will choose to stop sustaining an altitude sustaining level turn before running out of gas, at which point the pilot has to subjectively choose something else to do.

Good discussion and information by all. Thanks

That is a subjective opinion. I’m holding out for more objective measure.

Thanks.
  • 0

#76 piecost

piecost
  • Posts: 1318

Posted 26 June 2011 - 17:35

Josp, Thanks for the post it is certainly giving me something to think about. If I understand correctly then you are asserting that the E-M method provides a non-subjective comparison between aircraft as modeled in the sim (subject to testing errors). Thus, we can be sure how the RoF modeled aeroplanes compare to each other.

I may not have been clear in explaining the issue of inconsistent fidelity of available data. To take a recent example; there exists much wind tunnel data for the Bristol F2B such as the drag polar, lift curve slope CLmax, and even the impact of slipstream. This wealth of data is unusual.

When RoF model a more obscure aeroplane parameters such as CLmax have to be estimated from geometry and comparison with other aircraft. The estimation of the missing parameters is subjective and may be conservative/optimistic relative to real-life. The relative performance between aeroplanes is thus critically sensitive to the accuracy of estimation of missing performance parameters.

I have no feel for how flight model testing is performed and modifications are made to match expectations. I suppose that this is the time where favoritism/personal prejudices may creep in.

Finally, I would love to see E-M plots of all the planes. I have found the corner speed and estimated the maximum load factor of each plane and so can fill in the outside of the doghouse plots. Unfortunately, my initial testing of sustained turn produced such large scatter to make the plotting of a curve pointless.

I wonder if there is a way to use the autopilot to navigate around a circular flightpath?
  • 0

#77 JG14_Josf

JG14_Josf
  • Posts: 7

Posted 27 June 2011 - 17:32

piecost,

Already in the data offered in this thread is one measure of the difference between that which is calculated and that which is tested, if I understand the quote as follows:

Many measurements of the smallest radius of turn which could be obtained have been made either by taking the time of a turn at a known speed, or by camera obscura observation; The maximum acceleration has also been measured directly. It was found that for a number of different aeroplanes the minimum radius obtainable was above the theoretical minimum but by an approximately constant percentage (20% to 30%).

A. Theory
B. 20% to 30% off

Here are your words:

Unfortunately, my initial testing of sustained turn produced such large scatter to make the plotting of a curve pointless.

An objective measure is possible.

A. Theoretical or calculated performance envelope is the code - fixed.

B. Best possible flight test results; done by more than one person on more than one machine so as to find the maximum or best tested performance envelope, and to also find some average measure, and keeping the window open for repeated tests to confirm or deny the current measure of both the best maximum test result to date and the average measure of the performance envelope, based upon all the tests recorded to date.

You test; your test is off by a wide margin, but if you are the only one testing, then your test is the best test. Do you adjust the code so as to make your test results match up with the historical data?

Someone else tests and their test is even worse than yours.

Now you have two obvious measures of the difference between the code and the actual ability of the users to use the code, to reach the goal, or to simulate.

Suppose 10 people offer test results, and now there is a few who get closer to the intended coded results, even better than your best.

Now there is a new best, and a new average, and over time the best will offer better information, and the average will be a better representation of the average.

If the calculation varies from the test results by 20% to 30%, and the calculated measure is consistently optimistic, then that is what you have to work with objectively.

If one person on one machine manages to test in simulation, or in reality, performance much higher than anyone else can, then that is another place to begin looking for objective data; what does one pilot do that all the other pilots are not able to do, ever?

In reality it may be found that the one person doing the test had been using a different grade of gasoline, or a different timing on the engine, or tighter wires, or tighter fabric, or a better coordinated use of the controls, etc.

In simulation the outstanding performing test pilot may have a more accurate joystick, or faster machine, or slower machine, or corrupted code, or better coordinated use of the controls, etc.

Not knowing is: not knowing, and knowing is: knowing. Guessing helps by offering competitive paths to take toward the goal of knowing, but guessing isn’t knowing, so who claims that it is?

“I know that the simulation is exactly the way it was in history as far as top speed is concerned.”

Really?

Which test done in history? Why was that test picked? Which calculation based upon which blueprint?

In the end, my guess here, the idea, or the goal, is to simulate WWI air combat, and do so by producing and then using a WWI era computer flight simulator, so as to make the use of it fun to the people doing the simulating, and profitable, if possible, to the people employed in the venture.

If it is the best guess, then it is the best guess, why pretend otherwise?

If the user can prove that the best guess isn’t, then they can, why pretend otherwise?

If the goal is the same for producer and user, the best guess may be continuously improved upon, with a feedback loop, where users offer test results, and producers match up the test results with the chosen, and published, best guess historical data, and the calculated program code adjustments.

A. Top speed for plane A is based on this historical test.
B. Program code intends to reach A above.
C. User test results to date arrive at A, or exceed A, by one best test, and the average test results, by some average measure, are within 10% of A, so far.

Pretense of authority over knowing, absolutely, the best guess is preposterous, but actual authority over what will, or will not, be chosen and coded into the simulator is merely fact.

Authority over the choice to use the game, buy the planes, or not, is also a fact, and somewhere in between the decisions made by the producers and the decisions made by the users, in that connection, there has to be a feedback loop of some kind, if the idea is to reach the goal.

Which test is the best test?

Some people claim that top speed, best climb rate (which would be flown at best climb speed and not best climb angle), and level flight sustained turn tests, are the best, no other is even worth doing.

Which is the test that is the best test if you had only one to do, or which is a good one to start doing, and if there is time, do other tests afterwards?

Why pick that test?

Why did the actual pilots do what they did here:

Many measurements of the smallest radius of turn which could be obtained have been made either by taking the time of a turn at a known speed, or by camera obscura observation; The maximum acceleration has also been measured directly. It was found that for a number of different aeroplanes the minimum radius obtainable was above the theoretical minimum but by an approximately constant percentage (20% to 30%).

Why did they test looping performance?

What were the methods of testing?

In WWII the opposing sides captured opposing planes and flew them side by side with tests that measured specific combat performance variables; because that was the necessary information.

When, finally, John Boyd managed to get on the project he developed Energy Maneuverability and the results prove the point, that actual specific test results actually measure the necessary combat performance variables.

Here are your words:

I have no feel for how flight model testing is performed and modifications are made to match expectations. I suppose that this is the time where favoritism/personal prejudices may creep in.

I may have lost, on an old hard drive, the NavAir testing procedures used by the U.S. Navy, but it isn’t difficult to begin to get a feel for the EM tests, and then know their value, in context with this discussion.

One simple test is the level flight acceleration test. You can, or any simulator user can, get on-line, and preferably get on-line with someone else, and fly side by side, in formation, and pick an altitude, say 1000 meters, or just above, or just below the clouds, and as best as possible all the users in the test form up line abreast in level flight and everyone slows down pitching the nose up into a stall at full throttle.

Try to get the plane pitched far enough back to be climbing at best climb angle which may not be climbing at all, the plane may be stalled out and hovering with a very slow horizontal velocity, and now you have the beginning plot on the level flight acceleration test, and since there are more than just you doing the testing at that point in time and place, you have many competitive plots, and they will be all over the place, unless you have one exceptional pilot who can perform the test flawlessly.

Over time, with care, the knowledge of what the simulation can do, by the users, will be known. Those who can't reach that level of performance can't, other users can, and that is known.

Back to the test results, and assuming that you have at least two of the best, of the best users, and therefore you have right there, at that time, two planes, side by side, ready to record the best level flight test so far, you will have both planes at the slowest speed possible in level flight, at full throttle, and they will begin to push the nose forward, not climbing, but using the reduction in induced drag, as the wing moves from the highest possible angle of attack to the angle of attack that only maintains level flight, and each of the two best of the best simulation pilots in the game are now using the engine thrust to accelerate the plane, as fast as it can go, from that slowest speed in level flight point.

One plane may be well behind the other one at the start, since it has a lot of wing area, low wing loading, but perhaps not much in the engine power department, perhaps not as much in the power loading department, perhaps more, on paper, and since it has a lot of wing area it may be very high in the drag loading department.

Now you may begin to see Boyd’s formula working:

[(T-D)/W]V = Ps or in a more instructive way of writing the same thing: (T/W – D/W)V = Ps

One plane may catch up to the other plane as a huge amount of induced drag is unloaded from a slower starting speed, while the other plane, a heavier engine, less as good power loading (or not), but then both planes are side by side as the drag loading on the high wing area plane begins to accumulate more severely as a higher form drag increases square with velocity, and the better drag loaded plane (D/W) begins to pull away.

What can be known from such things is that the speed range where the lighter wing loaded, and better power loading plane accelerates better, it will turn tighter, and turn tighter longer, or turn tighter and lose less altitude, and turn tighter and even gain altitude, with a better climbing spiral at that speed range, while the other plane with a higher speed range where it has better acceleration may have an absolute higher rate of acceleration, just not at the lower speed range; therefore the climbing spiral, although performed at a higher speed, nets more altitude sooner.

Which will have the better climb rate?

Who does the climb tests?

Which climb speed is chosen to test best climb rate?

If the level flight tests results prove, over and over again, that the lighter loaded higher wing area, less powerful engine plane has better acceleration at a slower speed range, but, the higher loaded more powerful engine plane has an absolute higher rate of acceleration at a higher speed, due to better drag loading, then there are two speeds known, and usable to test that speed relative to best climb speeds.

The maximum rate of acceleration for the lighter loaded plane may be at X horizontal velocity, lower than the heavier wing loaded plane, which reaches a higher rate of acceleration at X+ horizontal velocity.

Now, having that understood, what does a best climb rate test look like between those two planes flying side by side?

What does it look like on the drawing board, or in the program, and what does it look like in the reality of the simulator, or in the reality that the simulator intends to simulate?

I wonder if there is a way to use the autopilot to navigate around a circular flightpath?

I think that it may be a good idea to find anyone who can fly consistently accurate flight tests, and even better to find two, not just one.

If the producers have users who demand higher fidelity, not something unheard of, the producers could challenge those users with specific tasks, or challenge the beta testers.

I am new to the simulation, it is fun. I have a laundry list developing:

1. Does the simulation simulate pilot g tolerance limits, if so how, and if not why not?

2. Does the simulation simulate aircraft g tolerance limits, if so how, and if not why not?

3. Does the simulation have the ability to record replay or track files, and if so then producers can ask for track files whenever a user demands higher fidelity, and serious track files, that record serious efforts to measure the lower fidelity in question, and those files can be used in the work that moves the product toward higher fidelity according to the best guesses offered by the users, as well as the best guesses of the producers of the simulation.
  • 0

#78 piecost

piecost
  • Posts: 1318

Posted 27 June 2011 - 20:03

Josp,

Another interesting post. The comparison technique is useful and is also employed on other aircraft types employing a carefully calibrated reference aircraft.

The sim has the advantage in that the mission builder allows the accurate placement of objects in x,y,z so it is possible to define a trajectory to fly around. I have employed fire/smoke objects in a circular pattern of known size to measure sustained turn.

I agree about harnessing the enthusiasm of the community to provide test data, especially when changes to existing FM are requested. I know that Chill31 has done extensive testing on the N17 and obtained results from other people, though I don't know how well that went (I must admit to not taking part).

In answer to your questions:

1. Pilot g limits are modeled, I don't know how it is done.

2. Aircraft g limits are modelled and I have estimated the ultimate load factor of each aeroplane (see: RoF test flight data, post#59). I include an updated plot below. I do not know if a limit load factor is modeled with exceedance weakening the plane (as employed on IL2 latest patch) - I believe not but haven't tested it.

3. The sim can record and playback (similar to IL2) and includes a time-bar so tracks can be rewound/fast-forwarded interactively.

Attached Files


  • 0

#79 piecost

piecost
  • Posts: 1318

Posted 28 June 2011 - 17:37

Correction to Ultimate load factor plot

Dr1 corrected should be 9, the Pflaz DIIIa needs checking
  • 0

#80 piecost

piecost
  • Posts: 1318

Posted 28 June 2011 - 18:01

E-M Plot for RoF Dr1 & Camel - lacking sustained turn curves

see RoF test flight data post#60 for details

Attached Files


  • 0


0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users