•*«1

*¥'v

•^Cr.i;^ vV^'t,.*^' . ;••'».-.,•?-

S.'-'-'i*

• '. -^-^

«^1^

-^^iEw,

Figure 1 — Curtiss Tanager. Note the floating tip aileron on the lower wing. The Tanager was the winner of the 1929 Guggenheim Safe Airplane Contest.

SAFETY OF SLOW FLYING AIRCRAFT By Robert T. Jones (EAA 114394) NASA, Ames Research Center Moffett Field, CA 94035

F,

rom the beginnings of aviation engineers have believed that slower speeds ought to make flying safer. Thus Alexander Graham Bell felt that the Wright Flyer was much too fast for safety. Speaking before the Washington

Academy of Sciences, he said. 'The Wright brothers' successful flying machine travels at the rate of about 37 miles per hour, and judging from its great flying weight (nearly two pounds per square foot of supporting surface), it is unlikely that it could be maintained in the air if it had very much less velocity. But should an accident happen to a body propelled through the air with the velocity of a railroad train, how about the

safety of the occupants?" Early airplanes with wing loadings of two pounds per

sq. ft. and often flimsy structures killed pilots with distressing regularity. It was not until the advent of the WWI trainers, such as the Standard J-l and the Jenny, that flying became even reasonably safe. These airplanes had 32 MARCH 1984

strong, sturdy structures, wing loadings around five pounds per sq. ft., minimum speeds of about 45 mph and very powerful controls. The Jenny and the Standard had ailerons powerful enough to support a man standing on the wing tip, and they often did so. Still, it seemed that if the airplane could be slowed down it would be safer. In 1929 there appeared the "Guggenheim Safe Airplane Contest". Entries were required to land and take off from a plot 500 ft. square surrounded by a 35 ft. obstacle and were required to demonstrate a minimum flying speed of no more than 35 mph. Two contestants, the Curtiss "Tanager" and the Handley Page "Gugnunc" met these requirements. Figure 1 shows the Tanager which won the contest. It is said that the Tanager could be flown hands off in gusty air, had full span flaps, Handley Page slots and full floating ailerons, giving control beyond the stall. In spite of very advanced technology and superior low speed performance, neither the Tanager nor the Gugnunc survived very long after the contest. The Tanager suffered a hard landing and was damaged but was not repaired. It is believed to have been ultimately destroyed in a fire. Some of the technology employed in these airplanes is now seen on modern STOL airplanes. It

is an unfortunate fact of life that the hull insurance rates on STOL aircraft are significantly higher than on more conventional, faster types. Shortly after the Guggenheim contest, it was discovered that many lightly loaded airplanes could be landed successfully by simply pulling the stick back and letting the airplane sink right on, with no attempt to control a flare. NACA engineers modified a Verville trainer by adding a strong landing gear with exceptionally long oleo shocks. Many landings were made in just this fashion, however, the test pilots considered the maneuver somewhat risky and I believe that on one occasion a down gust was encountered and the airplane was damaged. Details of these experiments can be found in NACA and TR No. 489 entitled "Air Conditions Close to the Ground the Effect on Airplane Landings." This report is well worth study since it contains information on atmospheric turbulence. Since these early days, great strides have been made in flight safety. Travel by commercial aircraft is considerably safer than travel by automobile. A good candidate for the world's safest airplane might be the Boeing 747. It is certainly remarkable that an airplane having a wing loading of 130 lbs. per sq. ft. (equal to a stack of four grand pianos) and a landing speed of 150 mph could show such outstanding safety in day by day operation throughout the World. What could be the reasons for this? I would list the following: 1. Comprehensive planning for each flight. 2. Strong, reliable structure 3. Perfect ease of control, uninfluenced by atmospheric turbulence General aviation aircraft cannot, of course, duplicate the wing loading and landing speed of the 747, nor could we afford the 10,000 ft. runway needed to land. (A six foot scale model to have the same wing loading and landing speed of the 747 would have to be made of solid lead.) However, by analyzing the problem carefully enough and getting rid of preconceived notions, we ought to be able to improve the safety of small airplanes, especially the really slow "ultralight". Of course the aeronautical engineer can influence only the last two items in my list. Consider first Item 3, ease of control and freedom from turbulence. We seem to believe that if an airplane flies slowly enough, it ought to be easy to control by an inexperienced pilot. Thus, ultralight aircraft are supposed capable of being flown by pilots with no previous experience. However, a dynamical analysis of a landing maneuver indicates that a heavier, faster airplane may require less skill than a slow, lighter one. Consider two airplanes: one with a wing loading of 5 lbs., about that of the Jenny; another

with a wing loading of twenty, about that of a Bonanza. The Jenny may land at about 40 mph while the Bonanza will land at 80. Consider a complete pattern sequence, downwind, base, final, touchdown and landing roll. Suppose each airplane executes the same sequence of bank angles (say 30 deg. bank in each turn), the same sequence of vertical and horizontal accelerations. The Bonanza will require four times the space for its maneuver and four times the landing run. In spite of its higher speed, however, every step of the maneuver takes twice as long! Plenty of time is available to check the flap setting, gear down, what obstacles may be on the runway, etc. Things happen a lot faster for the Jenny. Downwind, base, final and the landing run take half as long. Better get it right the first time. Moreover, if the air is gusty, the Jenny will suffer twice the vertical accelerations and about four times the vertical displacements that the Bonanza experiences. DISTANCE INTO GUST, m 5

10

2 E UJ LLI

II V)

04 ULTRALIGHT GLIDER

6L_

Flgure 2 — Vertical displacements caused by 4.6-m/sec

downward gust.

Figure 2 shows a detailed calculation of three airplanes

having different wing loadings, each flying through the

same down gust. The gust is supposed to have a downward velocity of 15 ft. per second and to extend for a distance of fifty feet. The commercial jet with a wing loading of 100 lbs. drops about an inch and experiences an acceleration of -.2 GS. The small airplane having a wing loading of 10

-—c^^TrC

Hang gliding in 1910. From "Histoire L'Aeronautique" L'lllustratlone, Paris. SPORT AVIATION 33

lbs. drops about two feet and receives -.6 GS, not quite

enough to lift the pilot out of the seat. The ultralight, however, drops a full 20 feet and with a negative acceleration of 1.2 GS, would put the pilot up against the canopy.

Photo Courtesy the Author

Figure — Max, the author's pet crow, on final and about to touch down on a fence post.

Figure 3 shows Max, our family pet crow, making a pinpoint landing on a fence post in front of the camera. In spite of his light wing loading of about one pound, Max negotiates wind gusts rather successfully. With his small size and quick response, Max has several things going for him. Considering dynamical motions with an appropriately shortened time scale, the significant parameter is the "specific mass ratio", which is the ratio of the mass of the aircraft to the mass of air entrained by the wing. Max's specific density ratio is about that of a 707. Max does not always win against the air currents, however. Once in a high wind, he tried landing on our chimney and fell in. I had to take the damper out to rescue him. As noted above, the history of slow flying aircraft has been rather negative. Nevertheless, an airplane that can land at low speed in a small space under perfect control remains a highly desirable goal. We should not, I think, give up. Designing a strong, sturdy structure certainly

Figure 4 — Control Linkage For Floating Tip Ailerons 34 MARCH 1984

requires no new knowledge, and, in fact, the new materials

available should make this easier. The problem of perfect control in gusty winds is more difficult. I believe we should adopt as a principle that the control power ought to increase as the flight speed is reduced, especially elevator and ailerons. I believe that an airplane intended to be flown by inexperienced pilots should be stall proof and spin proof. My own Ercoupe has this quality. A properly rigged Ercoupe will climb about 300 fpm with the wheel back against the stop — buffetting, but with marginal aileron control. The Ercoupe achieves this by an early center section stall combined with a limited up elevator. Unfortunately, the Ercoupe solution is probably not available to the ultralight. The radius of curvature in pulling out of a dive is about one-fourth that of the Ercoupe. This means, of course, that the ultralight will pull out of the dive with about 1A the loss of altitude, but it also means that a much greater up elevator angle is needed to cancel the curvature of the flight path. For further details of this calculation, the reader should consult NASA TM X 73,229 entitled "Dive Recovery of Hang Gliders", 1977. An ultralight with a stall limiting elevator will thus not have enough control to pull out of a dive in the minimum distance. There are, however, other ways of producing good behavior at high angles of attack. One of these is the use of floating tip ailerons, as on the Tanager. In the early 1930's NACA made many wind tunnel tests and in all cases, good control beyond the stall of the main wing was achieved. Figure 4 shows the type of linkage used to control these ailerons. It will be noted that the tips are free to float upward with the relative wind and hence never stall. Moving the control, however, establishes a difference angle between the two ailerons and hence provides lateral control. Figure 5, taken from NACA TR no. 424 (1932), compares the roll control power of floating tips with conventional ailerons at increasing angles of attack. The floating tips give good control (and favorable yawing moments) far beyond the point where normal ailerons cease to function. Quite a variety of floating tip controls were tested by NACA in the early thirties. TN's 316 (1929), 336 (1930) and 458 (1933) give the results of these tests. Figure 6, taken from TN 458, shows some of the arrangements tested. The multiple "tip feathers" were linked together as in a Venetian blind. All of these devices were quite effective aerodynamically, giving powerful control with favorable yawing moments. Floating tip controls are not without problems and one of these is structural. It is related that an early Bowlus sailplane equipped with such controls was flown by

.10

.08 t (

. ——— -c^.0

\

I

\y

nc

\

\

X

g

ROLLING MOMENT COEFFI

O

—T>

k

v_ ^-^

S

i

———————————————————————————————

10°

20°

30°

i

40°

ANGLE OF ATTACK, DEC. O— FLOATING TIP AILERONS A — CONVENTIONAL AILERONS

FIGURE 5 WIND TUNNEL TESTS COMPARING CONVENTIONAL AND FLOATING TIP AILERONS — FROM NACA TR 424.

Charles Lindbergh. After scraping the tips on the ground a few times they became pretty well bent out of shape and would no longer float. Lindbergh concluded that they were a bad idea. A still earlier application of moving tip controllers was on the Hill Pterodactyl, circa 1920. Figure 7 shows the Pterodactyl in its preliminary glider form. Here the tips were not free floating, but by virtue of their rearward position on the swept wing, were used as pitch control in addition to their function as ailerons. Pulling the stick back to increase the angle of attack reduces the relative angle of the tips and hence acts to prevent tip stalling. Flights in rough air, however, produced a kind of galloping motion, with the heavy tip controllers coupling with motions of the control stick. Evidently, the tip controllers were not counterweighted. An early criticism of the floating tips was that since they provided no lift, they reduced the effective span of the wing. This is certainly true in a simple geometric sense, but not in a structural sense since they do not stress the wing in bending. Structurally, they could be added to the tips of a wing already having the required span. While properly counterbalanced tips ought to be safe from high

speed flutter, stall buffetting was observed in some of the early wind tunnel tests. In all cases, application should be thoroughly tested in flight. Ultralight airplanes have wing loadings comparable to that of a bird, but they have much larger dimensions. The motions of large, light airplanes are dominated by what are known in the trade as "rotary derivaties". Consider a thirty ft. span ultralight and a one ft. span bird making the same 30° banked turn. The radius of turn will be the

Figure 6 — Floating Tip Ailerons by NASA

same in the two cases but the wing of the ultralight will extend much closer to the center of the turn where the airspeed is zero. The relative loss of lift on the inner wing is therefore much greater for the ultralight. This "rolling moment due to yawing rate" explains why making turns was so difficult in the Kremer prize contest. Powerful ailerons with favorable yawing moments should help in this. Another effect, not quite so obvious, is the extreme curvature of the flight path during pull out from a dive. As mentioned earlier, this means that the ultralight ought to have a powerful elevator control. One way of providing this is the use of an all-moving tail with a "leading" tab or flap as applied in certain Piper and Beech airplanes. The theory of the leading tab, all-moving surface was first applied to the vertical tail, and is described in NACA Wartime Rep. L 496 (originally issued in Jan. 1943) as an advanced restricted report. The leading tab can stabilize the surface even though the pivot is behind its normal aerodynamic center. Thus, the surface will float against the wind with the result that stick free stability is greater than stick fixed stability. In most applications the pivot is placed very near the aerodynamic center of the surface and the leading flap provides additional lift and adjusts the control force. Such an all-moving elevator could provide the extra control power needed by the ultralight. Finally, the pilot should be able to exercise more control over the flight path than the wind gusts do. This task becomes more difficult for airplanes in the ultralight category. Normally, the pilot exercises control by changing the angle of attack, reducing or increasing the lift. Of course, the wind gusts do the same. The measure of control here is the ratio of lift curve slope to the initial or steady flight lift (coefficients, if you prefer), and normally the

measure is the same for both the wind god and the pilot. One way of giving the pilot an advantage is to provide the SPORT AVIATION 35

MAW PLANE AREA

221.0 S}.FT

RUDDER AREA,TOTAL COWTROL TIP AREA

10

I

HiLL TAILLESS oLIDER

* I

M I

* I

I* I

I4.SSQ.FT S&.O&j.FT

20

J

FttT

Figure 7 Hill Pterodactyl, 1920

wing with a flap and link this to the elevator control. In this way the lift change produced by the pilot can be made greater than that produced by the gust. I cannot recall any applications of this idea and hence it should be approached with caution. I suspect that the linkage, elevator to flap, ought to be nonlinear; giving small flap angles with controls near neutral. Otherwise, use of the elevator at high speeds could result in high G forces. As mentioned earlier, if the ultralight is to be flown by inexperienced pilots, it should not snap or spin inadvertently. This requirement, however, poses a difficult requirement for the rudder control. When a wing is at or near the stall, a large mass of slow turbulent air accumulates above the wing. Pushing one wing tip ahead with the rudder causes this mass of air to rush toward the downstream tip, clearing up the forward tip and worsening the stall on the downstream tip. A typical accident occurs when the pilot, making a low, slow turn onto final, tries to tighten the turn by using the rudder. The ball slides off center and the airplane snaps onto its back. Thus, the rudder plays a prominent role in producing a spin. The Ercoupe, which has no independent rudder control cannot be made to spin, nor can it be made to sideslip; the ball is always at or very near the center. Of course, when rudder pedals are installed, the spinproof quality is lost. I find the Ercoupe without rudder pedals a delightful airplane to fly. Landings in a crosswind are not nearly as exciting as they can be in a tail dragger. Of course, those pilots who want every flight to be a test of personal skill may not be happy with this arrangement. When I first acquired my Ercoupe, it turned out to be spin proof but not stall proof. The fabric on the wings had been replaced by a metal covering, which evidently moved the center of gravity rearward, and the 36 MARCH 1984

elevator was not limited at the correct angle. In this condition the Ercoupe would stall sharply, dropping a wing and heading straight for the ground. It would not spin, however. I suspect that a great many Ercoupes are now flying in this condition. I later replaced the wings with fabric covered ones and set the elevator to the specified limit and this restored the original stall proof, spin proof characteristic. Landing an Ercoupe in a crosswind is easy. One merely keeps the wings level and accepts the drift angle dictated by the crosswind. The tricycle gear then straightens the airplane out as soon as it touches down. However, the angle of drift in an ultralight is apt to be much greater than it is with the Ercoupe, and for this reason it is difficult to say whether the Ercoupe solution is best for the ultralight. Of course, the ultralight should not be flown in strong winds anyway, but we cannot count on this. Laterally swivelling main wheels, as used on the early Bleriot monoplane might help. Clearly some practical flight experiments are required here and the question of whether, or how much, rudder control will have to be left to such experiments. Ultralight sport flying is rapidly gaining in popularity, but the rate of accidents has become a concern of both ultralight pilots and the general public. The availability of modern sophisticated radio equipment as is used to control model airplanes should permit us to explore thoroughly the outer limits of any new idea or new design. The suggestions made in this article are intended to encourage designers and manufacturers of such aircraft to undertake further experiments and more rigorous proof tests to improve their safety.

OX-5 Standard over the Missouri countryside

References NACA T.R. No. 489 — Air Conditions Close to the Ground and the Effect on Airplane Landings by F. L. Thompson, W. C. Peck and A. P. Beard; April 1934. TMX 73,229 — Dynamics of Ultralight Aircraft — Dive Recovery of Hang Gliders by Robert T. Jones; May 1977. NACA Tech Note 458 — Wind Tunnel Research Comparing Lateral Control Devices, Particularly at High Angles of Attack. XI Various Floatng Tip Ailerons on Both Rectangular and Tapered Wings by Fred E. Weick and Thomas A. Harris; 1933. NACA Wartime Rep. L-496 — Theory and Preliminary Flight Tests of an All Movable Vertical Tail Surface by Robert T. Jones and Harold F. Kleckner; 1943.

ABOUT THE AUTHOR — Robert T. Jones (EAA 114934) is one of the world's leading aerodynamicists. A 45 year veteran of NACA/NASA's research into the outer limits

of aeronautics and astronautics, he went to work for Fred Weick at Langley in 1932 and moved to Ames Research Center at Moffett Field, California in 1946. Over this long career Robert T. Jones has made discoveries that have changed the course of aeronautics . . . for which he has deservedly been honored many times by various organizations and universities. He is perhaps best know to the layman for his yawed or oblique wing concept for future supersonic airliners that has received so much coverage in the aviation press. In 1976 NASA published the Collected Works of Robert T. Jones, a 1025 page compendium of the NACA/ NASA reports written by Mr. Jones over his long career. Information on this publication can be obtained from the National Technical Information Service, Springfield, VA 22161. Reference should be made to Report Number TM X-3334. SPORT AVIATION 37