42 Stalling & Spinning

Introduction

The wings generate the lift on an airplane to overcome the airplane’s weight. Lift production depends on the airplane’s airspeed, the density of the air in which it is flying, the size (i.e., area) of the wing, as well as the wing’s angle of attack. The angle of attack is of particular interest because a wing can only produce lift with good aerodynamic efficiency over a relatively small range of angle of attack before it stalls. At low angles of attack, the flow will pass smoothly over a wing, as shown in the image below; this aerodynamic condition is the normal mode of flight operations. The highest angle of attack that a wing can operate is called the stall angle of attack (or just the stall angle) and corresponds to the angle to the flow when the boundary layer first separates from the wing’s upper surface.

Flow visualization images showing the smooth streamlines typical of attached flow (left image) and the detached or separated flow and turbulence that is characteristic of the onset of stall (right image).

The onset of flow separation anywhere on a wing will limit its maximum lifting capability; this behavior is commensurate with an increase in drag. When a wing fully stalls, there is also a pronounced nose-down pitching moment associated with an aft movement of the center of pressure produced by the effects of the separated flow, as well as significant unsteady flow and buffeting from the creation of turbulence. Therefore, the onset of a stall on a wing usually represents a limiting factor in the operational flight envelope of an airplane and generally sets its lowest controllable airspeed. In some cases, the onset of stall may precipitate a spin where the airplane enters into a rapid downward spiral in the stalled state. Therefore, it is crucial to understand what stall is on a wing, how it occurs, and the factors that can affect the onset of the stall.

Objectives of this Lesson

  • Understand the basic flow physics associated with the onset of flow separation and stall on an airfoil and a finite wing, as well as its implications.
  • Know how to calculate the stall airspeed of an airplane under different conditions of weight and density altitude and for different maximum lift coefficients.
  • Understand the importance of various “high-lift” devices such as flaps on airplane wings to raise the maximum lift coefficients and reduce stall airspeeds.
  • Appreciate the significance of a spin, and why the onset of stall is a prerequisite to develop a spin.

Stalling Airspeeds & Angles of Attack

In terms of aerodynamics, for vertical equilibrium on an airplane in straight-and-level flight then

(1)   \begin{equation*} L = \frac{1}{2} \rho V_{\infty}^2 S C_L = W \end{equation*}

where \rho_{\infty} is the density of the air in which the airplane is flying, S is the wing area, and C_L is the total wing lift coefficient. The assumption here is that all lift to overcome the weight W is generated by the main wings, even though the horizontal tail may carry some small amount of lift (positive or negative).

Stall Airspeeds

For estimating the stall airspeeds of an airplane, i.e., before verification by flight testing, it is normal practice to assume that only the main wings carry the lift of the entire airplane. For the density of the air, recall that \rho = \rho_0 \sigma where the value of \sigma comes from the ISA model, so an alternative variant of Eq. \rewf{lift1} is that

(2)   \begin{equation*} L = W = \frac{1}{2} \rho_0 \sigma V_{\infty}^2 S C_L \end{equation*}

Rearranging this equation allows the lift coefficient needed for the wing at a given flight airspeed, airplane weight, and operational density altitude (i.e., air density), i.e.,

(3)   \begin{equation*} C_L = \frac{2 W}{\rho_0 \sigma S V_{\infty}^2} = \frac{2}{\rho_0 \sigma V_{\infty}^2} \frac{W}{S} \end{equation*}

The ratio of the airplane’s weight to its lifting wing area, W/S, is called the wing loading. Notice also that the lift coefficient is proportional to wing loading but decreases with the square of the airspeed. The lift coefficient also increases with density altitude for a given airspeed and flight weight.

The minimum airspeed that would allow for the airplane’s flight is called the stalling airspeed, which corresponds to the angle of attack and lift coefficient at which the wing will stall. To understand why this condition is important, consider a typical relationship between lift coefficient and angle of attack, as shown in the figure below, which also considers the effects of high lift devices such as flaps and leading edge slats. The maximum lift coefficient is C_{L_{\rm max}}, so it is apparent that C_{L_{\rm max}} increases from about 1.4 in the clean configuration to nearly 3.0 with full flaps and the leading edge slat extended.

The maximum attainable lift coefficient depends on whether the flaps or slats are retracted or deployed. If the flaps and slats are retracted, then the wing is said to be in its “clean” configuration.

Although the value of C_{L_{\rm max}} may not be exactly known for any wing without wind tunnel or flight testing, the corresponding stall airspeed in straight and level unaccelerated flight can be solved for by using

(4)   \begin{equation*} V_{\rm stall} = \sqrt{ \frac{2}{\rho_0 \sigma} \left( \frac{W}{S} \right) \frac{1}{C_{L_{\rm max}} }} \end{equation*}

Conversely, if the stall airspeed of the wing is known, which would need a careful measurement of the equivalent and true airspeeds (not the indicated airspeed) at which stall occurs, then it is possible to calculate the corresponding C_{L_{\rm max}} of the wing, i.e.,

(5)   \begin{equation*} C_{L_{\rm max}} = \frac{2 W}{\rho_0 \sigma S V_{\rm stall}^2} \end{equation*}

Stall Angle of Attack

The linear region of the lift curves shown in the previous figure is where the flow is fully attached to the wing so that in this region, the relationship between lift coefficient and angle of attack can be written as

(6)   \begin{equation*} C_L = C_{L_{\alpha}} \left( \alpha - \alpha_0 \right) \end{equation*}

where C_{L_{\alpha}} is the lift curve (in appropriate units of per degree or per radian). The parameter \alpha_0 in Eq. 3 is called the zero-lift angle of attack. It reflects that the zero lift coefficient will be obtained when the wing is at a small negative angle of attack and is typical for a wing with normal amounts of camber. It will be noticed. However, the value of \alpha_0 becomes increasingly negative by applying flaps and then with flaps and slats together.

The airspeed in level flight that corresponds to a given lift coefficient can be determined, i.e.,

(7)   \begin{equation*} V_{\infty} = \sqrt{ \frac{2}{\rho_0 \sigma} \left( \frac{W}{S} \right) \frac{1}{C_L} } = \sqrt{ \frac{2}{\rho_0 \sigma} \left( \frac{W}{S} \right) \frac{1}{C_{L_{\alpha}} \alpha} } \end{equation*}

which is a useful result in that airspeed can now be related to the angle of attack, albeit approximately based on the linear lift curve slope assumption, at least under straight-and-level, unaccelerated flight conditions. This latter result is beneficial for piloting purposes because the pilot generally monitors the airplane’s airspeed, not the wing angle of attack. However, some high-performance airplanes and airliners may have an angle of attack sensor and cockpit indicator to help in stall avoidance.

The value of C_{L_{\alpha}} for a high aspect ratio wing at low flight Mach numbers (below 0.3) is typically about 0.1 per degree or 5.7 per radian but somewhat larger at higher subsonic Mach numbers, i.e., according to the Glauert rule. Decreasing the aspect ratio of the wing will also decrease the value of C_{L_{\alpha}}; the actual value may not be known exactly a priori before the airplane is built and flown, even though it can be estimated based on mathematical aerodynamic models and perhaps with the aid of wind tunnel testing.

At higher angles of attack, the lift coefficient reaches a maximum value and then decreases markedly because of the onset of flow separation and stall. For example, for a plain wing of moderate to high aspect ratio at low Mach numbers (i.e., one without any high lift devices), the maximum lift coefficient is typically between 1.4 and 1.6. This result means that the corresponding maximum angle of attack of the wing before stall (based on typical values of lift curve slope) at the low subsonic Mach numbers typical of takeoff and landing will be between 14^{\circ} and 16^{\circ}.

Although airspeed has been related to the stall angle of attack, it is important to appreciate that a wing will stall at any airspeed if the angle of attack reaches the stall angle. For example, if the airplane is maneuvering and attains a steady load factor n >1 such that now L = nW, then the new stall airspeed would be

(8)   \begin{equation*} V_{\rm stall} = \sqrt{ \frac{2}{\rho_0 \sigma} \left( \frac{nW}{S} \right) \frac{1}{C_{L_{\rm max}} }} = \sqrt{n} \sqrt{ \frac{2}{\rho_0 \sigma} \left( \frac{W}{S} \right) \frac{1}{C_{L_{\rm max}} }} \end{equation*}

which shows that the stall airspeed increases with the square root of the load factor. The stall angle of attack, however, is the same regardless of the load factor, and it is not a function of airspeed (other than through the effects of Mach number and/or Reynolds number).

Summary of Effects on Stall Airspeed

In summary, from the foregoing, it can be concluded that:

  1. Stall airspeed will increase with the increasing weight of the airplane.
  2. Stall airspeed will increase with increasing density altitude, i.e., with a lower air density.
  3. Stall airspeed will decrease with increasing values of wing C_{L_{\rm max}}, which could be achieved by the application of wing flaps and/or leading-edge slats.
  4. Stall airspeed will decrease with increasing wing area.
  5. Stall airspeed will increase with an increasing load factor.

General Flow Physics of Stall

At higher angles of attack before stall, the adverse pressure gradients produced on the wing’s upper surface result in a progressive increase in the thickness of the boundary layer. The effect of this is to cause some deviation from the linear lift versus angle of attack behavior, which is manifest at first by “rounding” of the lift curve, as shown previously. With a further increase in the angle of attack, the flow will separate from the upper surface, causing a stall.

On many wings and airfoils, the onset of a stall occurs gradually with an increasing angle of attack, as shown in the figure below, which is called a trailing edge stall. But on other wings, particularly those with sharper leading edges, such as those found on high-speed aircraft, flow separation may occur quite suddenly on reaching a critical angle of attack, usually referred to as a leading-edge type of stall.

Diagram of how a thickening boundary layer and the onset of flow separation eventually stalls flow conditions.
The development of a thickening boundary layer and the onset of flow separation will eventually limit the lift production on an airfoil and will also increase its drag, which is called stalled flow or just “stall.”

After the stall, much less lift is generated on the wing, and drag is substantially greater. Also, under these separated flow conditions, steady flow no longer prevails with turbulence being shed into the wake, i.e., which manifests as a form of buffeting. For example, measurements made on stalled wings in the wind tunnel will generally show large fluctuating forces and pitching moments under stalled flow conditions.

Two-Dimensional Stall Characteristics

Much of the detailed understanding of the aerodynamics of wings and wing sections comes from testing in wind tunnels, the effects of the wing section (i.e., the airfoil shape), wing aspect ratio, and wing planform having the most important effects. In their compendium of measurements, Abbott & von Doenhoff have documented a summary of the characteristics of many airfoil sections made at Reynolds numbers of 3 to 9 million and Mach numbers up to 0.2, which contains a wealth of measurements for the engineer.

It has been found that airfoils generally fall into three static stall categories, namely: thin-airfoil stall, leading-edge stall, and trailing-edge stall. Abbott & von Doenhoff show that thin-airfoil and leading-edge stalls are somewhat abrupt types of stall and can be sensitive to changes in airfoil shape, whereas trailing-edge stall is a more gradual type of stall and less sensitive to shape. Most conventional subsonic wings fall into the trailing edge stall category at low to moderate Mach numbers, although the more abrupt form of a leading-edge stall is not uncommon. However, for supersonic wings, which have relatively sharp leading edges, their stall characteristics at low airspeeds is usually an abrupt leading-edge stall.

The figure below presents a summary of measurements showing the combined effect of thickness and camber on the C_{l_{\rm max}} of an airfoil. Clearly, “thin” airfoils (low values of thickness to chord ratio of between 4% and 8%) produce much lower values of C_{l_{\rm max}}. Still, it is interesting to note that increasing thickness above 12% begins to reduce the value of C_{l_{\rm max}}. The best airfoils in terms of attaining the highest C_{l_{\rm max}} have a thickness-to-chord ratio between 12% to 15%, which is common on wings used for airplanes.

Summary of results showing combined effect of thickness and camber on the C_{l_{\rm max}} of an airfoil.

Using some nose camber for a given thickness-to-chord ratio is also desirable in increasing C_{l_{\rm max}}, as shown in the figure below. However, there is a point of diminishing returns in that some small camber over the nose region of an airfoil is desirable, but too much gives only minor gains in C_{l_{\rm max}}. Indeed, the use of significant nose camber is usually undesirable for two reasons. First, large amounts of camber increase the pitching moments on the airfoil section and the wing as a whole, which is not always desirable. Second, too much camber is associated with an increase in profile drag, which is never a good outcome.

Results showing effect of nose camber on the C_{l_{\rm max}} of an airfoil.

The C_{l_{\rm max}} of an airfoil section also reduces with increasing free-stream Mach number, as shown in the figure below. This behavior is because increasing the Mach number makes the airfoil’s pressure gradients more adverse, increasing the boundary layer thickness on the upper surface at a lower angle of attack. The consequence of this effect is a reduction in the value of C_{l_{\rm max}} and a reduction in the corresponding stall angle with increasing Mach number. Notice also the effects of Reynolds number (chord-based), for which lower Reynolds numbers will result in lower values of C_{l_{\rm max}} for a given Mach number.

Results showing effect increasing free-stream Mach number on the C_{l_{\rm max}} of an airfoil at different chord Reynolds numbers.

Therefore, in addition to the foregoing factors that have been discussed that will affect the stall airspeed, then the stall angle of attack and corresponding airspeed will generally:

  1. Decrease with with increasing flight Mach number.
  2. Decrease with decreasing chord-based Reynolds number.

Three-Dimensional Stall

The effects of wing planform (e.g., taper) and twist (washout) also affect the local lift coefficient distribution along the wing (i.e., the distribution of C_l along the span of the wing) and also the total attainable maximum lift coefficient. Therefore, if the local C_l is higher at some point on the wing than another, the wing is more likely to stall at this location.

While no complete generalization of all the possible results is available for wings, it has been found that higher aspect ratio wings tend to reach higher overall maximum lift coefficients. However, using too much planform taper can reduce the wing’s maximum lift coefficient as a whole. Likewise, too much sweepback also tends to decrease the maximum attainable lift. Both effects are fairly predictable using various types of aerodynamic theories

The main problem in wing design is that at the point of stall, a finite wing often develops somewhat complicated three-dimensional flow patterns, an example being shown in the images below based on wind tunnel flow visualization. Sometimes the pattern is not symmetrical between the left and right side panels of a wing, which in practice tends to produce a rolling moment on the airplane and a departure from controlled flight. In a worst-case scenario, the airplane may develop a spin, taking a significant altitude to recover from (see later).

Interpretation of surface oil flow visualization of a wing spanning the walls in a wind tunnel showing the formation of three-dimensional “cells” the point of stall.

The three-dimensional stall developments on wings can be studied during flight testing by placing strips of yarn called “tufts” all over the wing’s upper surface. The tufts are arranged in orderly rows spanwise and chordwise along the wing, an example being shown in this video. It will be seen that when the tufts are blown straight back, the flow is fully attached. However, suppose the tufts change direction and lift from the surface. In that case, the flow at those locations is inevitably separated, which is likely an indicator of the incipient stall on the entire wing.

The stall patterns that develop on any given wing depend on many factors, but some general observations have emerged, as shown in the schematic below. For rectangular wings, the stall will generally start at the root of the wing and work forward and outward toward the mid-span. This behavior is a desirable stall pattern on a wing because the onset of flow separation gives some buffeting, and the pilot will recognize the stall onset before the stall progresses too far. The airplane also tends to pitch down naturally at the stall, reducing the wing’s angle of attack and suppressing further stall developments. Finally, the wing’s outer parts have a fully attached flow, so there will be no tendency for the airplane to roll, and the ailerons will be fully effective.

Examples showing the effects of wing planform on the three-dimensional stall developments.

A highly tapered wing has the opposite problem in that the wing tips will tend to stall first. With swept wings, the tips also tend to stall first. In these cases, not only will the aileron effectiveness be reduced because of the separated flow, but the airplane will often tend to show uncommanded rolls to the left or right. Indeed, in this cases, the application of controls to correct the roll may make the stall worse, and the airplane may experience a control reversal, i.e., if the airplane rolls to the right, for example, the use of normal corrective control inputs make the roll to the right much worse.

As previously mentioned, the use of large amounts of wing sweepback (which is used to delay the onset of compressibility effects to higher cruise airspeeds) is also known to produce stall toward the wing tips, which is a concern in the design of commercial airliners. There are two main reasons why swept wings tend to stall first at their tips. The first is that the use of sweepback alone tends to produce a spanwise distribution of the local lift coefficient that is biased toward the wing tips.

To establish where a wing will stall first, the product of the local section lift coefficient and the wing chord at that position can be divided by the product of the total wing lift coefficient and the mean aerodynamic chord, as shown in the graphs below. Where the resulting values are highest, then the wing can be expected to stall there first. Clearly, the use of too much planform taper only exacerbates this effect of increasing the lift coefficients at the wing tip, making the wing more likely to stall there. However, the effects can be mitigated to some extent by using wing twist (washout) and careful use of leading-edge camber.

The effects of sweepback on the spanwise distribution of local lift coefficient.

At higher angles of attack, sweepback tends to encourage the boundary layer to grow thicker in the spanwise direction, thus making it more susceptible to separating at the wing tips. Usually, the wing is carefully designed to minimize such tendencies, but other steps are usually necessary. For example, various fences and vortex generators are often used to help keep the flow attached. In addition, some commercial airplanes have “auto-slats” so that if the wing was to approach the stall, then the slats automatically extend to prevent the stall from developing.

Importance of High Lift Devices

Airplane wings are usually designed for the best efficiency in the cruising flight condition. In fact, on commercial transport airplanes, the supercritical wing is explicitly designed to be efficient at transonic flight conditions. However, relatively thin, swept wings tend to have much poorer aerodynamic performance at lower airspeeds and Mach numbers. In the “clean” condition, the airplane tends to exhibit fairly high stalling airspeeds and is susceptible to adverse stall qualities.

However, as previously explained, it is possible to increase the maximum lift coefficient of a wing and reduce the stall airspeed by using high-lift devices such as trailing-edge flaps and leading-edge slats. The photo below shows the triple-slotted trailing edge flap system in the fully deflected landing configuration. Notice also the condensation trails marking the locations of trailing vortices from the side edges of the flaps and the inboard aileron. The main advantage of reducing stall airspeeds is that the airplane can fly at lower airspeeds, which reduces takeoff and landing distances significantly. Of course, nearly every type of airplane can benefit from using high lift devices to lower its stall airspeed and reduce its takeoff and landing distances.

The triple slotted trailing edge flap system of a Boeing 747 in the fully deflected landing configuration.

The figure below shows a schematic of a wing section with high-lift devices. Usually, the flaps operate in conjunction with the leading-edge slats if they are available. In many cases, the flaps are designed to deflect rearward and downward (called Fowler flaps), which increases the effective area of the wing and increases its camber in the region of the flaps.

Cross-section of a wing with high lift devices. The flaps and leading edge slats work together to increase the maximum lift coefficient of the wing for takeoff and landing.

The flaps and slats are partly deflected for takeoff to reduce the takeoff airspeed and takeoff distance. After takeoff, the flaps on the airplane are progressively retracted as airspeed builds and the slats retracted. For landing, the flaps are fully deflected; they are gradually extended in stages as the airplane descends from altitude and slows down to its final landing approach airspeed. The extra drag from full flap deflection helps to slow the airplane down to landing airspeeds and steepens its final approach angle to the runway, both of which help the pilot successfully land the airplane.

Leading-edge slats usually work in synchronization with the flaps. Like the trailing edge flaps, their purpose is to increase the wing’s maximum lift and angle of attack capability before it stalls and allow the airplane to fly at lower airspeeds. The usual types of slats initially move forward and downward and increase the camber of the wing section, and when fully deployed, also open up a small gap between the slat and the leading edge of the main wing. This gap is critical because it helps to energize the boundary layer on the wing’s top surface, as shown in the figure below; the mixing of the boundary layers has a particular name called a confluent boundary layer.

Other leading-edge devices include Kruger flaps, which rotate forward to increase the leading-edge camber of the wing; such devices are often found on the inboard parts of wings, such as on the Boeing 747. Such devices are highly effective in increasing the attainable maximum lift coefficient and angle of attack before the wing stalls. While flaps by themselves are effective in increasing C_{L_{\rm max}}, the use of flaps and slats together may increase C_{L_{\rm max}} to values of over 3.0. Consequently, if the slats and flaps are both fully deployed, the airplane can typically fly at airspeeds that may be 50% less than without them, i.e., compared to flying in the clean condition with flaps and slats retracted. In addition, the use of flaps and slats substantially changes the wing’s camber, which modifies the zero-lift angle of attack and the nature of the lift curve. In some cases, the system’s lift curve slope may change, especially with the high effective wing camber introduced by the slats.

Worked example – Calculation of Stall Airspeed

A new jet airplane has been estimated from wind tunnel testing to have a maximum wing lift coefficient of 3.2 with the slats deployed and the flaps fully down. If the wing loading is 76.84 lb ft^{-2} then estimate the stalling airspeed in level unaccelerated flight at a pressure altitude of 5,000 ft on a standard ISA day.

The stall airspeed can be estimated using

(9)   \begin{equation*} V_{\rm stall} = \sqrt{ \frac{2}{\rho_0 \sigma} \left( \frac{W}{S} \right) \frac{1}{C_{L_{\rm max}} }} \end{equation*}

We are told that C_{L_{\rm max}} = 3.2 and that W/S = 76.84 lb ft^{-2}. At 5,000 ft on a standard day we can use the ISA model to determine that \rho = 0.002048 slugs ft^{-3}. Substituting gives

(10)   \begin{equation*} V_{\rm stall} = \sqrt{ \frac{2 \times 76.84}{0.002048} \left( \frac{1}{3.2} \right) } = 153.1 \mbox{ft s$^{-1}$} \end{equation*}

or about 91 kts, which is relatively low and shows the importance of getting a high value of C_{L_{\rm max}} on a wing.

Stall Control Devices

Because the effects of a stall on a new airplane are not entirely predictable in advance of the first flight, the stall is carefully examined during flight testing. This is done to determine the stall airspeeds and the stall qualities, which is how the aircraft feels or “handles” to the pilot at the point of stall. Not all aircraft are created equally in this regard.

Some aircraft have been found to experience undesirable flight characteristics near stalls, such as uncommanded rolling to the left or the right. What is desired for flight safety is a recognizable and repeatable stall with no adverse characteristics; this outcome would be considered as being good or acceptable handling qualities at the point of a stall for aircraft certification. In some cases, “stall fixes” are needed to give the airplane predictable and repeatable stall characteristics.

Vortex Generators

Even a casual observation of the wing on a modern airliner will show numerous regions where little “vanes” are attached to the wing surface, as shown in the photograph below. These devices are called vortex generators or VGs. They are designed to create small vortices to energize the boundary layer flow and keep the flow attached to the wing to higher angles of attack. While the VGs create some drag, the drag is more than offset by the maximum lift benefits.

The arrows point to vortex generators (VGs) on the leading edge of the wing of a commercial airplane. The VGs are typically angled flat plates that often look like miniature delta half-wings.

The VGs are usually just small angled plates set at a steep angle to the flow, often looking like little delta half-wings, which generate flow separation that rolls up into a concentrated vortex, as shown in the schematic below. This vortex is like a tip vortex, but it is of a much smaller scale and mixes in with the thickened boundary layer. The consequence is that the boundary layer profile, on average, becomes “energized” and is less susceptible to flow separation. The penalty, however, is a slight increase in drag.

Stall Strips

It may seem odd to force a wing to stall, but that is precisely the purpose of stall strips. A photograph of a stall strip is shown below, and they are relatively common on many types of aircraft, from small general aviation airplanes to commercial jets to fighter aircraft. The strips are usually a few inches long and perhaps 1/4 inch (6 mm) in height.

A stall strip mounted on the inboard leading edge of a wing.

Stall strips are used to promote flow separation at specific locations on the wing during high-angle of attack flight to improve the stall qualities of the airplane. They are typically employed in pairs symmetrically distributed on the inboard sections of both the left and right wings, but only sometimes. The usual approach in flight testing is to place the stall strips on the wing’s leading edge, where the desired outcomes are realized, i.e., to force stall to occur, as suggested in the figure below. On many propeller-driven airplanes, the stall strips are placed at different locations to account for the asymmetric flow produced on the wing by the wake swirl from the propeller(s).
Many aircraft have been built and flown to find that their stall qualities and high angle of attack flight characteristics are less than acceptable. The solution is usually a combination of stall strips and vortex generators, although more drastic action may be required in some cases. Stall strips are usually employed near the root of the wing to force flow separation to occur there first. Then, with the simultaneous application of vortex generators over the outboard wing panels, the airplane’s stall characteristics are usually improved.

Wing Fences

Wing fences, also known as boundary layer fences or even stall fences, are fixed aerodynamic devices that project perpendicular to aircraft wings and wrap around the leading edge, as shown in the schematic below. Many early jet aircraft, with their highly swept wings, used fences to control wing stall. Some wings may have several fences, which reflects the difficulties in correcting the stall qualities of some swept wing  designs.

WIng fences can be used to limit the growth in the spanwise thickness of the boundary layer, thereby improving aileron effectiveness and stall qualities.

If there is any significant spanwise flow along the wing, the boundary layer thickness can grow toward the wing tips without the fences. The consequence is reduced aileron effectiveness, increased tendency for the wing tips to stall first, and a susceptibility to control reversal and spins. The fences limit the spanwise airflow along the wing, directing the flow over the control surfaces and preventing premature boundary layer growth and stall. 

Spinning

A spin can develop as a consequence of an asymmetric stall, which causes the airplane to rotate about a vertical axis and develop a steep, spiraling, nose-down flight path, as shown in the figure below. While spins can be provoked on airplanes, such for the purposes of flight demonstrations and pilot training, inadvertent spins are usually dangerous.

A spin can be a consequence of an asymmetric stall and causes the airplane to develop into a steep, spiraling downward path in a stalled condition.

In a spin, the wings are stalled and operate at a high angle of attack, even though the airplane’s attitude is distinctly nosing downward. Recovery from a spin requires counterintuitive control inputs from the pilot, which includes pushing forward on the control stick to lower the angle of attack and so un-stall the wing, with simultaneous use of the opposite rudder (to the direction of spin) to stop the rotational motion. The rapid rotation about the vertical axis and the nose-down attitude can be very disorienting to a pilot, and many “stall/spin” accidents have occurred.

The development of a stall is a prerequisite to a spin, so stall avoidance will generally prevent an airplane from spinning. Modern commercial airplanes have sophisticated stall warning and stall avoidance systems, so the likelihood of such airplanes inadvertently developing a stall or a spin is extremely remote. Nevertheless, their stall characteristics are always carefully explored during flight testing.

Smaller airplanes, however, generally do not have such stall avoidance systems. Therefore, FAA certification standards for smaller civil airplanes with up to a 12,500 lb maximum takeoff weight require that they demonstrate both stalls and spins. Such airplanes need to show a full recovery from stalls and a one-turn spin with the normal use of flight controls. If intentional spins are to be allowed by certification, then the airplane must demonstrate a six-turn spin. Many GA airplanes are certified for spins if they are operated in the utility category, which usually means reduced gross weight and forward center of gravity position. Spin testing for certification is a potentially dangerous activity, so the test airplane will usually be equipped with some secondary spin-recovery device, such as a tail parachute.

Summary & Closure

Determining the stall characteristics of an airplane is critically important because the airplane will be flown near stalled conditions during takeoff and landing. While much can be done at the design stage to estimate stall airspeeds, both in the “clean” condition (with flaps and slats retracted) and in the “dirty” state (with flaps and slats deployed along with the undercarriage), the stall airspeeds must be confirmed. In addition to stall airspeed and onset, it is important to understand the recovery characteristics of an airplane during a stall.

A well-designed airplane should have a clear and uncomplicated stall recovery procedure, which can be executed by a pilot with reasonable training and skill. The recovery characteristics should be consistent and repeatable regardless of the airplane’s altitude, airspeed, and load factor. Flight test data must demonstrate that the stall recovery procedure is effective and reliable, and that the airplane does not have any dangerous stall-related behavior, such as an uncontrollable spin or an uncommanded roll. Overall, the stall characteristics and recovery of an airplane must be carefully studied and evaluated to ensure that the airplane is safe to fly in all conditions.

Vortex generators and/or stall strips can often improve unacceptable stall qualities. While a spin is not a dangerous maneuver in itself, an inherent tendency for an airplane to spin near the point of the stall is an unacceptable flight characteristic that needs to be mitigated before the airplane could be considered safe to fly by an average pilot, i.e., to meet FAA certification requirements.

5-Question Self-Assessment Quickquiz

For Further Thought or Discussion
  • According to published sources, the Helio Courier airplane has a stall airspeed of only 23 kts. Does this seem reasonable and why?
  • Another device that is sometimes used to control an airplane’s stall characteristics is called a “wing fence.” Do some research to find out what these devices are and how they work.
  • A Boeing 767 loses the ability to lower the leading-edge slats for landing. Wing flaps are not affected. If the reduction in the C_{L_{\rm max}} of the wing is 30% without using the slats, estimate the percentage increase in the landing airspeed of the airplane. Make any reasonable assumptions you feel are needed.
  • A new airplane is found to have an undesirable stall characteristic that causes it to roll consistently to the left with a tendency to spin as it approaches the stall. Suggest some options as to how this adverse flight characteristic may be mitigated.

Other Useful Online Resources

To understand more about stalling and spinning, then follow up with some of these online resources:

  • The use of tufts to examine stall and spin patterns on the wing.
  • A video about what a spin is.