(PART 1 OF 2 PARTS)

Preventing Tailspins By Robert T. Smith, EAA 1685 209 Highland Dr., Smyrna, Ga.

Iaircraft of aircraft without first examining the stall, because an must be stalled before it can be spun. If it will

T IS VIRTUALLY impossible to discuss th3 tailspinning

not stall, it cannot spin.

Most of us are familiar with the term "stalling angle of attack" and with the definition of the angle of attack, which is merely the angle between a wing's chord line and the direction of its motion through the air. Fig. A shows a wing moving through the air at a low angle of attack and Fig. B shows the angle of attack increased considerably. Fig. C shows an airfoil at its critical angle of attack where the stall is beginning. Air is a viscous fluid and wants to flow smoothly around an airfoil but, when the airfoil assumes a position such that the air flowing around it must turn a sharper corner than it is capable of, it breaks away from the airfoil's surface. In the case of an actual wing, the air usually starts breaking away at the trailing edge. It is possible, through wind tunnel tests, to precisely measure the angle of attack at which air begins to break away from any particular airfoil.

The ability of a wing to generate lift depends on its ability to let air flow smoothly over its surface. It is beyond the ability of a plain wing and beyond its critical angle of attack to keep the flow smooth. Separation and burbling then occur and this destroys lift, as well as greatly increasing drag. When air is moving very smoothly over an airfoil, the flow is said to be "laminar." The flow over the leading

edge of a smooth airfoil is generally laminar. Further back on the airfoil, the air flow becomes "turbulent", but lift is still being developed although drag in the turbulent layer of air is higher than in the laminar layer. It is important to build a smooth wing skin . . . to secure laminar flow which has less drag than turbulent flow.

T-IC.URE- B 32

FEBRUARY 1966

It is generally accepted that, on a smooth wing skin, the flow is laminar back to the wing's maximum thickness point, and turbulent thereafter. The so-called laminar flow airfoils have their maximum thickness from 40 parcent to 60 percent or more of the chord aft of the leading edge in an effort to maintain the laminar flow as far back over the wing as possible. This might prompt the choice of one of the NACA 66 series or 63 series airfoils for a high-speed airplane. However, unless the airplane is in the 400 mph class and has an extremely smooth surface, there will be little gain over what could be achieved with the old Clark Y. A really smooth surface on the Clark Y would, incidentally, be easier to obtain than on the laminar flow airfoils due to their difficult-to-produce curves. As the wing's angle of attack is increased, the laminar flow region becomes smaller and eventually the flow becomes turbulent. At the critical angle of attack, the airflow begins to separate and swirl around like water at a boat's stern. Lift is lost in the region of flow separation and drag is greatly increased. Because airfoils have wellrounded leading edges, the airflow separation does not begin at the leading edge but starts at the sharp trailing edge. Fig. D shows an airfoil that has its airflow greatly separated. Fig. C shows the same thing, but with the separation just beginning. Slots on the leading edge of an airfoil allow air to pass through and, because of the venturi cross section of the slot, it is considerably speeded up as it exits from the slot. It then flows back over the upper surface of airfoil and tends to retard separation. Thus, with slots, (Fig. E), the wing's critical angle of attack will be higher and the maximum lift with slots is greater.

Flaps (Fig. F), increase lift by increasing the curvature of the airfoil, and slotted flaps have the added advantage of allowing air under the airfoil to flow through the slot and over the flap to help retard separation just as leading edge slots do. It is possible to greatly incrsass maximum lift with leading edge slots and slotted flaps.

We have looked at stalls as though we were at the wing tip looking down the wing. Now, let's get above the wing and consider what the engineers call "spanwise" lift. We prefer the stall to start at the wing root and progress to the tip, so that we can retain aileron control well into the stall. If the stall starts at the tip, the aileron will become ineffecive immediately, and the airplane will probably roll abruptly at the stall. Some airfoils lose their lift gradually while others stall abruptly. The smoothstalling airfoils have a flow separation that progresses smoothly as the angle of attack is increased, while others maintain a fairly smooth airflow up to the critical angle of attack, then experience a widespread area of abrupt flow separation. The latter airfoils are said to have a "sharp break at the stall." If a wing design was such that the stall started at the tip and its airfoil had a sharp break at the stall, the aircraft would be almost uncontrollable laterally at the stall and inadvertent spins would become a strong possibility. Most designers twist the wing so that the tip sections have a lower angle of attack and, therefore, do not stall until well after the root has become stalled. This gives good aileron control well into the stall. Also, most designers pick an airfoil that has smooth stall characteristics with no sharp decrease in lift at the stall. A combination of twisting the wing and using a smoothly stalling airfoil means that the aircraft probably will be difficult to spin intentionally, and will not do so inadvertently.

Actually, "washout" combined with an abruptly stalling airfoil, or the use of a smoothly stalling airfoil with no washout, will also prevent inadvertent spins provided that the elevator power is satisfactorily restricted and if directional stability is good. Elevator power has less to do with the inadvertent spin than does good directional stability, which is translated to the amateur designer as meaning sufficient rudder area and adequate wing dihedral. The pilot controls stalls with the elevator. There must be enough elevator control to cause the wing to reach its critical angle of attack, or it will be impossible to stall the airplane. If we limit the up-elevator travel sufficiently, the airplane cannot be stalled. The "Ercoupe" is an example of this practice. By using a large amount of dihedral and limiting the up-elevator travel, the "Ercoupe" will not stall, and will have good directional stability at lower speeds near the stall.

r\.A»ii UP

If we have too much up-elevator travel it will be possible to achieve a much greater degree of wing stall than is common. We need enough elevator power (sufficient travel combined with enough area) to reach the stall if a conventional landing gear is used. If there isn't enough up-elevator power on a conventional geared airplane, there will be trouble in keeping the tail down on landings. The Beechcraft 17 suffers slightly in this respect when flying solo, and then pilots often made wheel landings. But, when loaded, the CG was far enough aft to help keep the tail down and it landed normally. Center of gravity location can have a lot to do with airplane performance at the stall. An airplane is usually designed so that when it stalls it will nose down of its own accord. However, if the CG is too far aft, the nosedown tendency will be reduced and, if we move the CG too far aft, we will reach a point where the nose cannot be put down with full down-elevator travel. Obviously then, we cannot recover from the stall. Never load an airplane so that its CG is behind the allowable limit. If you do, a stall could be fatal even when done with plenty of recovery altitude. It would then not be a problem of altitude so much as a problem of not having sufficient down-elevator power to nose the ship down and decrease the angle of attack below the stall. A CG located too far aft will also have a detrimental effect on spins, but it is important to know what an aft CG will do to the stall, because the spin is merely an aggravated stall; what cures a stall will also cure a spin. How much aft CG movement can we allow in a particular ship? Unless you care to study aircraft stability and control, I would suggest that you examine some production aircraft that are similar to the airplane you are building. Note their allowable CG movement, and elevator area and deflection. Design your aircraft to have similar CG movements, elevator deflections and areas, and you will be on reasonably safe ground. If the CG is too far forward, the aircraft will nose down abruptly at the stall, which is all right, and spinning characteristics will probably not be adversely affected either. But, if we carry the forward CG movement to an extreme, a point is reached where the nose cannot be raised after take-off, or where a conventional ship will not land on three points. An extreme forward CG is not as dangerous as an aft one, but stay in the limits just the same for good flying qualities. If you fly a conventional landing gear aircraft loaded so that its CG is forward of the limit, you may not notice anything on take-off because engine power may tend to temporarily offset the forward CG. But later, you may not be able to keep the tail down on landing, and a nose-over might result. For your own design, again study airplanes at the local airport and ask the local airport operator if you can see his list of Aircraft Specifications as published by the FAA. These give the CG limits. (END

®

OF PART 1) SPORT AVIATION

33

(PART 2 OF 2 PARTS)

Preventing Tailspins By Robert T. Smith, EAA 1685 209 Highland Dr., Smyrna, Ga.

ET'S ASSUME that we have a nice L ship with good stall characteristics, and its CG is located in the proper place. A spin is a stall that is held with full up elevator, and in most modern production aircraft just relaxing a small amount of this back elevator pressure will cause the aircraft to recover from the spin. For these ships to continue a spin, it is necessary for the pilot to continue to hold the elevator hard against the back stop. A spin also has autorotation which simply means it rotates by itself, but it will not start doing so by itself in a well-designed aircraft. A spin is entered with the engine at idle, wings level and nose just about at the landing attitude position or slightly higher, say, about 15 deg. nose-up attitude if one insists on technicalities! Or, use 30 deg. nose-up attitude . . . and it still won't make a lot of difference in the spinning characteristics. From this attitude of flight, bring the stick or wheel back until the aircraft begins to stall. As it begins to stall, ease in rudder in the desired direction of spin, meanwhile continuing to the back limit smoothly with the elevator. As the rudder approaches its limit, the aircraft will start yawing in that direction and the nose will start down through the horizon. Before you stall the aircraft, pick out a good reference point on the ground so that you can count the turns. As the aircraft reaches one-half of its first turn, the nose will approach

its maximum down attitude which will vary widely with aircraft, but will usually be about 45 deg. nose-down. As you hold full back elevator and full rudder, the aircraft will swing on around to the reference point for the completion of its first rotation in the spin, and between one-half to one full turn the nose probably will spring back up towards the horizon. In some aircraft, the nose will come back up to the horizon or above it at the end of the first turn. As the second turn of the spin starts, the nose will swing back downwards, and will stop at some nosedown attitude in the neighborhood of 30 to 50 deg. nose-down. Hereafter, the spin will continue with a steady rate of rotation and in this constant nose-down attitude. Some aircraft will reach this stabilized condition before the end of the first turn; others will require nearly two turns or more. All will reach a steady spin condition which will be characterized by a constant nose-down attitude and a constant rate of rotation. Fig. G shows an aircraft in a steady spin. Individual airplanes will have widely varying spinning characteristics as to how far the nose will be down, and how fast they may rotate. What keeps the rotation going? As you will notice from Fig. G, the wings are banked slightly in the direction of rotation. Since the outside wing is going faster, it is developing slightly more lift than the inside wing, and tends to keep trying to "fly up" over

SPIM

the inside wing. This keeps the aircraft rotating. The aircraft is stalled, as I have gone to considerable pains to point out, and I have just now said that one wing is developing more lift than the other one! They're both supposed to be stalled and, since lift isn't developed in the stall, how can this be? A stalled wing still develops some lift, small though it is. It is this small increment of lift that is causing the outside wing to lift up and over the inside wing. Fuselage shape has much to do with autorotation. A fuselage with large side area and generous rudder area tends to dampen and reduce autorotation. A slim fuselage and small rudder will not offer much resistance to

autorotation, and it has even been found in tests that such fuselages may even add to autorotative couples. This means that the spin recovery will not be as easy as with a large fuselage with plenty of rudder area. The matter of fuselage and tail design, as regards spinning, is a complex matter and there are many NACA reports on it which resulted from government studies in the 1930s and 1940s. It often has happened that flat topped fuselages were more prone to spinning than those with rounded tops, as the sharp corner of a flat top fuselage would cause air separation as it yawed in spins and so reduced the effectiveness of the fin. When all the rudder area is above the horizontal tail, there is often rudder blanketing and, for good spin control, a generous amount of rudder area should be below the horizontal tail. In general, dorsal fins on top of fuselages have had much less effect in stopping or minimizing spins than have fins or strips running along the ton longeron ahead of the stabilizer. These horizontal fillets or projections ahead of the stabilizer act as air dams, making the flat part of the fuselage below them more effective as spin-inhibiting areas. Heavy airplanes normally spin more viciously and lose more altitude per turn than

do lightly loaded ones, such as most amateurs build. When the pilot, passenger, engine and fuel tanks are all close together, there is less danger of centrifugal force acting on them to pull the plane into a flat spin than is the case where such weights are well spread out along the length of a fuseFIG.

G

lage.

(Continued on next page) SPORT AVIATION

41

to the resistance of the fuselage as

SECTION X-X &TRAX.R

VEVJTRAV- PMM

SECTION! X-X

SEPARATED FV.OW ON AHT> -

FIG. H

PREVENTING TAILSPINS . . .

(Continued from preceding page)

Recovery from a steady spin is effected by applying full rudder in the direction opposite to spin rotation. You enter the spin, for example, by applying right rudder, and you'll get right hand rotation. As soon as the rudder is full opposite to spin rotation, go forward on the stick or wheel. How far forward you must go will depend on the airplane. On most modern production ships, just relaxing the back pressure unstalls the ship and leads to recovery. But on some, such as the Cessna T-37 jet trainer, the stick is carried hard against the forward stop until the nose pitches down, indicating that the angle of attack is being reduced and that the spin is being recovered. As the spin is stopped, the rotation will cease in about one-fourth to one turn, and you will find yourself in a steep dive. Merely level the wings with the ailerons and start a steady pull to level flight, making this pull gradual and firm enough to avoid excessively high speeds or G loads, but not so hard that another stall is induced. Opposite rudder aids recovery by slowing the rotation, and in some ships it will practically stop the rotation. Forward stick is needed to get the wing below the critical angle of attack and is naturally the only control that will do this. I mentioned the adverse effects an extreme aft CG would have on the stall, and further mentioned that it

would adversely affect the spin. How?

First, the spin will be flatter (nose

higher than normal). Second, the elevator power may not be sufficient to reduce the wing's angle of attack below the stalling point, in which case 42

MARCH 1966

the aircraft will continue to spin until it contacts the ground. Needless to say, it will not be in a "normal landing attitude!" An extreme forward CG will cause the aircraft to spin with its nose further down than normal and at a faster rate of rotation. I doubt that recovery will be more difficult than in a spin with the CG in its proper place, but you may have to pull desperately on the elevator to get out of the dive you'll be in when recovery is made. Aircraft with high lateral mass loadings . . . heavy along the wings . . . such as a twin engine ship with the added weight of engines and perhaps tip tanks out on the wing will be critical on the recovery so far as the rudder is power to stop or slow the rotation, and the elevator is power to lower the nose to reduce the angle of attack below the stalling angle of attack is concerned. A single engine ship with a lot of fuel in the wings might also be in this category. If the tank in one wing was full while the one in the other wing was empty, the rudder's ability to stop spin rotation might be marginal. A light, twin engined jet produced by one manufacturer and used extensively by the military originally suffered in this respect because nearly one-third of its weight was in the wings, and it also had a flat, shallow fuselage. They solved the problem by adding a ventral fin, increasing the rudder arm (moving the rudder further aft), and by adding forward fuselage strakes. Fig. H shows a crosssection of a fuselage nose with strakes acting in a spin. These strakes build up positive air pressure on the side of the fuselage nose entering the air, and help to reduce rotational moments by adding

it rotates through the air. The strakes are merely flanges of extruded aluminum riveted to the fuselage sides. They probably are of no interest to the homebuilder unless he built a twin engine ship (or a light jet) with a lot of weight in the wings and a flat, shallow fuselage. In a ship with high lateral mass loadings (heavy along the wings), a severe requirement is placed on the elevator and, unless care is taken in the design, elevator power may be marginal in effecting recovery, particularly if the ship is spun with one wing heavier than its mate, and the heavy wing on the inside of the spin. The reason behind this is that the ship would spin with a lot of bank in the direction of spin, and the spin rotation rate would become pitch rate, which is what the elevator controls. The elevator would then have to do two things . . . slow the spin rate and reduce the angle of attack below the

stall. If you design a ship with a lot of weight in the wings, be sure that this weight is evenly distributed between the two wings before you spin it. If you are going to build a twin engine ship, I would discourage you from ever trying spins at all, as the higher rotational rates for such a ship in a spin might cause the stresses in the wing due to rotational inertia loads to go beyond the design limits of the wing structure. This might cause the engines, for example, to exert destructive loads on the wing structure. Most of the homebuilt designs that I have seen appear to be ships which would possess good spin characteristics. If you will remember that good stalling characteristics automatically lead to good spinning characteristics, you will be safe. If you also will remember that a ship must be stalled before it can be spun, you will never have spin trouble while test flying your creation. The military jet trainer I mentioned earlier had hairy spinning characteristics, although it had excellent stall characteristics and had to be forced into a spin. Its bad spin was due to its high wing loading which had nothing to do with its stall or spin entry and, in fact, it required a lot of rudder deflection (over 50 percent) just to start the spin autorotation. If your ship has a nice, gentle stall, with good aileron control well into the stall and adequate rudder control in the stall, you can always prevent a spin merely by recovering from the stall before a spin develops. A spin is an aggravated stall! Control the stall, and the spin will never be a problem. (Continued on bottom of next page)

WHAT ARE THEY? ANSWERS TO FEBRUARY The first picture is the popular American 101 "Eagle" built around 1929 and powered with a 100 hp Kinner engine. The other picture depicts the Bourdon "Kitty Hawk"

MYSTERY AIRCRAFT (later Viking) built in 1927 in Rhode Island. This model was powered with the 107 hp Siemens-Halske engine imported from Germany.

DUCTED FAN INFORMATION Of particular interest to those seeking information on application of the ducted fan principle to aircraft, the most comprehensive presentation on the subject will be found

THE LATEST IN GROUND-EFFECTS The Frigidaire Division of General Motors Corp. is currently advertising their latest refrigerators as having what they call the "Ride-Aire" feature. What "Ride-Aire" amounts to is a ground-effects machine! This is built right into the appliance by utilizing the already well-shrouded air space under the unit. A fitting is provided in the lower front whereby the blower attachment hose from a vacuum cleaner is attached. This generates a cushion of air on which the heavy refrigerator can be moved about with only gentle pushing or pulling. This could possibly portend the ultimate and most practical, if not the least expected, use of the ground-effects principle, still another direct descendant of the airplane. ®

in the book . . . THEORY OF PROPELLERS by Theodor-

sen (1948), Chapter IX, Page 98 ... published by McGrawHill Book Publishers at 330 W. 42nd St., New York, N.Y. 10036. ®

PROPELLERS FOR LOW HORSEPOWER ENGINES Arrow Propeller Company has printed a small brochure that lists all engines of 2 to 35 hp. This brochure shows the required diameter of the propeller to be used on each engine. If you are interested in determining the propeller for your project (if your project is within this horsepower range), write for the brochure. The brochures are free, and are available from: Arrow Propeller Co., P. O. Box 8, Paducah, Ky. 42001. ®

; "Is there a welder aboard?"

S

EITHER THE GEAR NEEDS HEAT-TREAT I NO,

OR YOU NEED A DIET!"

PREVENTING TAILSPINS . . . (Continued from page 42)

Remember, a "good" stall is one that starts at the wing root, and progresses toward the tip. It is one that does not happen abruptly, but comes on slowly. If you design a ship with a stall where aileron and rudder control are good well into the stalled re-

gion, you need never worry about accidentally spinning and, in fact, you can practically forget about what sort of a spin your ship might have. As long as you control the stall, you are controlling the spin, because the stall is the beginning of a spin. If you do

not stall, you cannot spin!

®

WHAT? A Monthly magazine for FUN Aviation? Rockets, missiles or a i r l i n e s poop? Nope! Just fresh news am! sparkling pictures. Homebuilts, airport of the month, antiques, coming e v e n t s , air shows, f l y - i n s , cartoons, photo

spreads, x-country directory, new products, book reviews, i n f o r m a t i v e ads and lots more for the general a v i a t i o n pilot.

Our 5th year of publication. send $2.00, cash, cbeck or money order for 12 big issues

THE FLYER 120 So. State St. Westerville, Ohio SPORT AVIATION

43