Not All Airfoils Stall The Same, Part 1

Patrick Veillette, Ph.D. April 15, 2022

Illustration of the airflow along an airfoil that exhibits leading-edge stall behavior. Notice the formation of a “short bubble” at a critical point along the leading edge. When the AOA is increased to a critical value, the bubble “bursts,” causing a separation of airflow over the rest of the airfoil with no aerodynamic warning of the stall. Credit: Clinton Tanner, “The Effect of Wing Leading-Edge Contamination on the Stall Characteristics of Aircraft. ”

Airfoil design has undergone immense changes as manufacturers seek to produce aircraft that can cruise more efficiently at high transonic speeds. In spite of these advances, high-performance airfoils exhibit stall behaviors completely different from low-performance airfoils used on earlier generations of training and transport aircraft. Specifically, this includes lack of aerodynamic warning and an abrupt loss of lift. Furthermore, these negative effects are exacerbated when the leading edge is contaminated by irregularities such as icing.

One of the direct results of these diverse stall behaviors is that high-performance airplanes have unique characteristics nearing the margins of their respective maneuvering envelopes that are likely to deter from the timely recognition and recovery of an airplane upset.

Some instructional materials provide adequate explanations of the stall behavior of older airfoils in terms suitable for general aviation pilots. Unfortunately, this information continues to be widely disseminated without any regard for the advances in engineering knowledge and is not accurate when describing the behavior of advanced generations of airfoils.

The misconceptions of this outdated information can have potentially dangerous consequences. According to the FAA’s Advisory Circular 120-111 “Upset Prevention and Recovery Training,” “Evidence indicates that training programs using operating procedures from one airplane type may have a detrimental effect if carried over to a different airplane type. This can lead to an upset.” The extensive collection of industry authorities who contributed to the Advisory Circular noted, “The safety implications and consequences of applying poor instructional technique, or providing misleading information, are more significant in upset prevention and recovery compared with some other areas of pilot training.”

The stall characteristics of a wing are affected by a multitude of factors, some of which are affected by deliberate design choices while others occur due to environmental factors. Examples of the former include an airfoil’s shape, the wing’s planform and wing twist. Modern wings are often built with aerodynamic twist, meaning that the cross section of the wing at various locations along its span is composed of different airfoils. This is done to enhance the aerodynamic efficiency of a wing, but it can also cause a distinct variation of stall development at different span sections of a wing. Swept-wing planforms have the tendency to pitch up as the angle of attack (AOA) reaches critical levels due to their characteristic of having stall development begin near the wingtips. Furthermore, separated boundary layers near the wingtips on swept wings do not produce airframe buffeting that is associated with early stall warning on rectangular wing planforms.

Environmental factors such as heavy rain, Mach waves, crosswinds, anti-icing fluids, decayed aerodynamic seals and icing induce significant changes on an airfoil’s and/or wing’s stall behavior. Without exception, these changes universally disrupt the boundary layer’s behavior and lead to unexpected abrupt stalls that occur at much lower angles of attack, and often lead to unrecoverable conditions due to the excessive loss of lift.

Extension of leading-edge or trailing-edge devices as well as flight controls and spoilers will change the stall development. An airfoil’s stall behavior in turbulent flow and at abrupt pitch rates (i.e., the “dynamic stall”) is significantly different from the “standard” explanation of stalls. Exceptionally thin airfoils have a stall behavior that is unlike the two types of stalls that will be discussed in this article. These are just a sample of many factors that have immense influence on the stall behavior of a wing.

The remainder of this article will concentrate on static stalls, or, in other words, a stall produced by a gradual increase in the AOA. Furthermore, we will focus on the stall behavior at airspeeds common for terminal area flight operations (for the more scientifically articulate reader, at lower Reynolds numbers) without the extension of high lift devices or the utilization of boundary-layer control apparatuses. We will compare the effects on maneuvering envelope margins for aircraft that exhibit traditional trailing-edge stalls to the leading-edge stall that predominates on high-performance aircraft. This article will be a translation into “pilot lingo” of the insightful and detailed work done by flight-test engineering teams to provide a much greater understanding of the stall behavior of high-performance airfoils.

Trailing-Edge Stall

During your earliest ground school training you were introduced to the concept of the trailing edge-stall. The trailing-edge stall is prevalent among airfoils designed with the traditional “Clark Y” shape, possessing the typical bulbous shape over the top and a flat bottom. These airfoils are useful at creating lift at relatively slow speeds due in part to plenty of camber. They typically have thickness ratios greater than 16%.

Aeronautical engineers often utilize fine smoke filaments inserted into the airflow of low-speed wind tunnels for flow visualization. This helps them to visually observe the separation and transition on airfoils and flow structures of airfoils. A commonly viewed online video shows the testing of a NACA 4421 airfoil. What do these numbers mean? This designation is from the early days of the National Advisory Committee of Aeronautics. This airfoil has a maximum thickness of 21%. Its maximum camber of 4% occurs at the 40% chord point.

The layer of air along the airfoil’s surface that experiences the effects of friction is called the boundary layer. The air molecules immediately adjacent to the wing’s surface are brought to a near halt by this friction. As the airflow slows along the top of the airfoil, an adverse pressure gradient grows in strength. Before 9 deg., the lift coefficient increases almost linearly, meaning that as the AOA is increased, the lift coefficient increases smoothly.

At moderate angles of attack, the adverse pressure gradient causes the boundary layer to “separate,” starting at the airfoil’s trailing edge. The chaotic airflow within a separated boundary layer creates a large increase in drag and is inefficient at creating lift.

As the AOA is increased in increments, the region of separated flow causes the boundary layer to thicken, and it also moves forward along the airfoil toward the leading edge. Each incremental increase in the AOA at this stage results in large increases in drag. After 9 deg., the lift curve slope changes with increasing AOA.

Eventually, the separated region covers the majority of the airfoil’s upper surface.

The “critical angle of attack” is reached when any further increase in the AOA results in a decrease in its lift production. Note that this doesn’t result in a total loss of lift. The airfoil is still producing lift, albeit inefficiently and with a substantial increase in drag.

There are several sources to find the exact lift and drag behavior of the airfoil used on your aircraft. The classic engineering textbook Theory of Wing Sections, by Ira Abbott and Albert Von Doenhoff, was standard in every aeronautical engineer’s library and contains the lift and drag curves for a wide spectrum of airfoils. Some of that data can now be found with a quick search of databases on the internet.

The majority of flight training materials written for pilots state that the critical AOA never changes and is independent of airspeed. That is basically true if one is addressing solely the low-performance airfoils at relatively slower speeds. For private pilot training purposes that is a sufficient generalization. However, the sharp-eyed reader who looks at these graphs will notice that there are different critical angles of attack shown in the lift curve data. The variance depends on the Reynolds number, which is the parameter utilized in fluid dynamics to compare the momentum of a fluid to its viscosity. This behavior becomes especially pertinent in the discussion of upset prevention when discussing wings exposed to propwash. That is a great topic for another day.

Airfoils with “forgiving” trailing-edge stall characteristics will have a wide, rounded top of the lift curve. Small changes in the AOA near the critical angle of attack will not cause significant changes in the lift. This is a desirable characteristic for training aircraft.

One inherent disadvantage of thicker airfoils is the propensity to develop a strong Mach wave due to the acceleration of air over the bulbous leading edge during transonic flight. This creates performance-robbing drag. Thus, with the advent of the jet era, aeronautical engineers began to utilize thin airfoils as a design method to lessen the development of Mach waves for aircraft that would cruise at transonic speeds.

The Learjet 23 was a product of this aeronautical concept. Its first flight on Oct. 7, 1963, signified the beginning of a revolution in transport aircraft to cruise at relatively fast transonic Mach numbers. It was designed with a 9% thick airfoil section, no wing twist, and the characteristic tip tanks.

While the potential cruise speeds of these jets were attractive, accidents and serious incidents revealed pilots’ lack of understanding of the different aerodynamics of airfoils designed for high-speed flight. As a result, they were ill-equipped to properly diagnose, prevent and recover from phenomena such as Mach tuck, deep stalls, aileron snatch and unforgiving abrupt stalls with uncontrollable wing roll-off.

In complete contrast to the traditional trailing-edge stall that is a characteristic of slower aircraft, flight tests by Lear observed that as the angle of attack increased, the flow around the tip tank caused a high AOA at the junction of the tip tank and the wing leading edge, which caused a triangular shaped area of separated airflow. This impacted the airflow over the aileron. The airflow over one wing would suddenly separate entirely, causing a rapid and large roll-off, often beyond 90 deg. of bank. Furthermore, there was little or no aerodynamic stall warning buffet. It was also noted that the stall characteristics deteriorated markedly with small accumulations of ice on the wing leading edges.

The rapid roll-off and abrupt stall with no aerodynamic warning caught pilots by surprise, adversely impacting recognition or recovery. An example of this occurred on Dec. 4, 1978, to a Learjet 25C that was on approach to Anchorage International Airport (PANC) with seven persons on board. Light to moderate icing was forecast in clouds below 12,000 ft. along with turbulence and gusting winds in the vicinity of the airport. The flight path was normal almost to touchdown, when the aircraft suddenly pitched up and began to bank steeply from side to side. The Learjet rolled to the right until its right wing struck the ground. Tragically, five of the seven occupants were fatally injured. One of the survivors of the violent impact was then Senate Minority Whip Ted Stevens, who had flown C-46s over the infamous “Hump” route in Asia during World War II and was a recipient of the Distinguished Flying Cross. Sadly, his wife of 26 years was one of the accident’s victims.

The NTSB determined the probable cause of the accident to be an encounter with strong, gusting crosswinds during the landing attempt, which caused the aircraft to roll abruptly and unexpectedly. The ensuing loss of control resulted from inappropriate pilot techniques during the attempt to regain control of the aircraft. Suspected light ice accumulations on the aerodynamic surfaces may have contributed to a stall and loss of control.

Flight Testing Uncovers the Leading-Edge Stall

In 1982, the Learjet flight-test team was tasked with trying to make improvements to the stall characteristics of its earlier aircraft. One of its first challenges was to better understand these abrupt stall characteristics. Earlier NASA studies had discovered that airfoils with this design exhibited a smooth laminar flow over the leading edge at low AOAs. As the AOA is increased, an adverse pressure gradient increases to the point where a small amount of separation occurs near the leading edge. This separation causes the boundary layer to transition from laminar to turbulent, and at this point the airflow would form a small rotating “bubble” that helped to reattach the boundary layer to the airfoil. This bubble remains attached to the airfoil until the AOA is increased to a critical value, whereupon it suddenly breaks down for reasons not presently understood. In smoke wind-tunnel tests the bubble appears to burst. At this point the boundary layer fails to reattach to the airfoil, causing very rapid full chord airflow separation. In other words, this causes a completely separated boundary layer, i.e., a stall, over the entire airfoil. Furthermore, traditional symptoms of stalls such as unresponsive flight controls or aerodynamic buffeting are not present.

Flight-test teams led by Learjet’s legendary test pilot Pete Reynolds utilized small tufts of yarn across a wing to observe its boundary layer behavior. These teams noticed that the burst of this leading-edge bubble along a small portion of the wingspan will instantly extend full span and the entire wing can separate almost instantly.

The leading-edge stall is common on airfoils with medium thickness, between 10-16%, utilized on high-performance aircraft. How can you tell if your aircraft uses an airfoil that exhibits the leading-edge stall? First look up the airfoil section for your wing with a quick internet search, and then search for the “lift curve” for that airfoil. An airfoil that exhibits classic leading-edge stall behavior, such as the NACA 4412, has little or no rounding of the lift curve near maximum lift. The lift coefficient exhibits an abrupt loss of lift. This differs from the other types of airfoils where the loss of lift is gradual after stall.

By the way, this information hasn’t been hidden from the aviation profession in secure vaults. Reynolds disseminated it throughout the professional journals so that the rest of the industry could gain useful insight. One example is his Ten Years of Stall Testing, published through the American Institute of Aeronautics and Astronautics as AIAA Paper #90-1268.