1

Basics of Flight - Lift

1. Definition - LIFT
There are several "definitions" of lift that may be found by searching through the relevant literature, but only one that is really fundamentally correct. Sometimes the lift is inappropriately defined as the force that is required to support the weight of the aircraft in flight. On first inspection, this seems to make sense and indeed it is true in the case of steady level flight. However, it is otherwise patently not the case, e.g. when the aircraft is performing a manoeuvre. The only truly correct definition which works for all possible flight conditions is: the component of aerodynamic force generated on the aircraft which acts perpendicular to the instantaneous flight direction. This means that the lift is inclined backwards in a climb, forwards in a dive and inwards in the case of a banked turn (Fig 1.1).

Fig 1.1 Lift vector for different flight conditions

1.1 Lift Coefficient

This was defined previously in the Fundamental Concepts notes, section 7, whereby: Lift (L) = V2 S CL (1.1) Where = air density, V = freestream velocity and S = reference area (for fixed-wing aircraft, S is usually taken as the gross wing planform area while for most other bodies it is the maximum cross-sectional area).

2. Generation of Lift
The generation of an aerodynamic force was covered in the Fundamental Concepts notes, Section 1.1. It may be recalled that the overall force may be attributed to two natural phenomena (Fig 2.1): the variable static pressure distribution, acting perpendicular to the surface. the variable shear stress distribution, acting tangential to the surface.

Fig 2.1 Static pressure & shear stress distributions around subsonic aerofoil It is clear that, as far as lift is concerned, the shear stress is secondary and contributes mainly to the drag force. The lift force itself is primarily due to the pressure distribution imbalance over the upper and lower surfaces. In particular, the production of a lower net pressure across the surface relative to the lower surface will produce an upwards lift force. It is therefore often convenient to only deal with the pressure forces when discussing the physical origin of the lift force. This will form the basis of all the following reasoning on the subject.

2.1 Aerofoil Section

Most of this discussion on lift will be centred on typical subsonic aerofoil sections and it is therefore worthwhile reviewing their basic geometry at this stage. This is shown in Fig 2.2, with the camber and thickness exaggerated to aid clarity (more typically around 3% and 12% of the chord length respectively). It is worth noting that any shape may be made to produce lift, provided it is given either camber or inclined to the flow direction (i.e. given an angle of attack). It is the need to optimise the lift/drag ratio which leads to the use of the classical streamlined aerofoil sections.

Fig 2.2 Subsonic aerofoil section configuration & definitions

2.2 Popular Simplified Explanation - Flawed

Many embark upon aeronautical studies to satisfy a desire to understand how and why aircraft fly. It is unfortunate that the most popular explanations given to students are fundamentally incorrect and that many retain these misunderstandings throughout the rest of their studies! The basic explanations are often given along the following lines. The most obviously flawed theory is the one which espouses the fact that in order to produce lift it is necessary to use a cambered lifting surface. This is clearly nonsense as symmetrical sections and even flat plates can be used to produce lift provided they are inclined at an angle of attack to the oncoming flow. A slightly improved theory accepts the fact that only either a cambered or inclined section is needed. If this is now inserted into a viscous airflow, stagnation points form at the leading and trailing edges as the flow is unable to negotiate the sharp turn at the tail (Fig 2.3).

Fig 2.3 Leading & trailing edge stagnation point formation This means that the respective distances from SLE to STE are different with respect to whether the flow travels over the upper or lower surface. In particular, if cambered or inclined, the upper surface distance is greater than the lower surface distance. OK so far - but this is where the explanation goes wrong. Teachers now often state that the flow must take the same time to get from SLE to STE over both surfaces, meaning that the mean velocity over the top surface must be greater than over the lower surface. This is then linked with Bernoulli's theorem to explain where the pressure difference comes from and hence - LIFT! However, there is no reason why the time taken should be the same. Indeed, experiments and numerical simulations show that this is patently not the case and that the time taken to move over the top surface is actually a lot less than that taken to travel over the lower surface! This is shown in Fig 2.4 for a fluid element C which leaves the trailing edge a long time before element D moving over the lower surface. This is where the simplistic theory is flawed.

Fig 2.4 Time taken for flow to pass over upper & lower surfaces

2.3 Valid Physical Explanation

A more fundamentally correct explanation for the origin of aerodynamic lift may be found by making use of the conservation laws of physics, namely mass flow (continuity) and energy or momentum (Euler or Bernoulli). Consider the 2-D stream tubes A and B shown in Fig 2.5 - these are originally of the same width when well upstream of the aerofoil and out of its influence. The common streamline between them is aligned with the leading edge stagnation point and so acts to divide the airflow over the top and bottom surfaces. The stream tube A encounters the upper portion of the aerofoil and is squashed through what is effectively a smaller area. Stream tube B is squashed less, resulting in a larger stream tube width, relative to A. From continuity (m = A V) this must mean that the velocity in A must be greater than in B.

Fig 2.5 Streamtubes A & B passing over upper & lower surfaces Either Euler's (dp = - VdV) or Bernoulli's (p + V2 = constant) theorems may now be applied to show that pA must be less than pB. This results in the pressure distribution shown in Fig 2.6. It can be seen that the majority of lift is due to the low pressure (suction) acting over the upper surface, particularly over the front 20% to 30%, where the flow accelerations are at their most severe. Another popular misconception is thus disproved, that it is the high pressure on the lower surface produces most of the lift. In fact, in the case shown, the pressure on the lower surface is below that of the ambient value, its just that the pressure on the upper surface is even lower than this its the pressure difference which is important.

Fig 2.6 Pressure distribution giving lift

2.4 Alternative Physical Explanation (Newtons Laws of Motion)

An alternative explanation, sometimes given in textbooks and on websites on flight mechanics, uses Newton's laws of motion as a basis. The reasoning is that the wing deflects the air downwards so that the wing imparts a downward component to it. To do so, the wing must be providing a downwards force on the air (Newton's 1st Law). Then from Newton's 3rd Law, the air must be providing an equal and opposite reaction force on the wing.

Fig 2.7 Downward deflection of airflow (downwash) This explanation is certainly better than the standard but it could be argued that this is the effect of lift rather than the cause of it. There is no doubt that air is deflected downwards, and indeed that a considerable amount of air movement is required in order to provide the requisite amount of wing lift. It could, however, be the pressure distribution which is pushing the wing up and that this is really the action which is taking place. The reaction to this would then be for the wing to push the air down, with a sufficient momentum change to balance out the lift produced. This would mean that the original argument is probably the "wrong way around"!

3. Circulation Theory of Lift

Yet another lift theory exists, though this is somewhat more mathematical than those already covered. A brief summary of the principles involved will, however, be given here for the sake of completeness. It involves the theory of circulation, which is defined as below. Circulation (K or ) = qsds , (3.1) for a closed loop C within a fluid, where: qs = tangential velocity component

6 ds = incremental length

Fig 3.1 Definition of circulation If we now consider the case of a lifting aerofoil, the velocity vector distribution over the section may be represented by Fig 3.3a. If the freestream velocity vectors are subtracted we are left with the distribution shown in Fig 3.3b, where the circulatory components can clearly be seen.

Fig 3.3a & 3.3b 2-D velocity vectors & minus freestream velocity vectors

Fig 3.4 Equivalent flowfield This means that the case shown in Fig 3.4 is equivalent to that shown in Fig 3.3, i.e. that the flowfield may be represented by the superposition of the freestream vectors and a set of

7 circulatory vectors. This leads to the higher airspeeds experienced over the upper surface (where the circulatory components are moving in the same direction as the freestream) and the lower airspeeds over the lower surface (where they move in opposite directions). This, of course, is exactly what happens in reality, as discussed earlier. The difference in airspeeds then leads to the pressure difference (from Bernoulli) leading to lift.

3.1 Magnus Lift

From the discussion on the circulation theory of lift it should be apparent that any rotating object placed in a moving airstream will produce circulation and hence aerodynamic lift. This is then known as the magnus lift and produces many familiar effects such as the top/back/side spin on tennis balls and the hook/slice of a golf ball. It may easily be understood physically: air traveling over the top surface is accelerated by the bull traveling in the same direction whereas air traveling over the lower surface is decelerated by the ball moving in the opposite direction. The speed difference then leads to a pressure difference and, hence, lift.

Fig 3.5 Magnus lift on a rotating body

4. Camber & Angle of Attack Effects

In the case of an increase in either camber or angle of attack there will be a resultant downwards movement of the leading edge stagnation point (see Fig 4.1). This will mean that the stream tube squashing effect (described above in 2.2) will be more pronounced. This will lead to a larger speed imbalance across the upper and lower surfaces (continuity), a bigger pressure imbalance (Euler or Bernoulli) and hence more lift. This will always be the case provided stall doesn't occur (see later).

Fig 4.1 Movement of LE stagnation point due to camber or angle of attack change The net effects are that the lift varies with camber and angle of attack as shown in Fig 4.2.

Fig 4.2 Lift curves for subsonic cambered & symmetrical aerofoil sections

4.1 Stall
The relationship with angle of attack is virtually linear until the stall angle is approached. The stall angle is defined as the angle at which the CL curve peaks, i.e. when CL,max occurs. Beyond this, there is a dramatic loss of lift until it becomes reasonably constant at much higher angles of attack. For the symmetrical section, the curve passes through the origin, i.e. there is zero-lift at zero-, as would be expected. The effect of camber is to vertically raise the curve so that there is a certain amount of lift present even at zero-. There is also a slight reduction in the stall angle.

Fig 4.3 Subsonic aerofoil stall

4.2 Pressure Variations with Varying Angle of Attack

This is shown below, in Fig 4.4.

Fig 4.4 Pressure variations with angle of attack The following should be noted: Negative (nose-down) pitching moment at zero-lift (negative a). Positive lift at a = 0o. Highest pressure at LE stagnation point, lowest pressure at crest on upper surface. Peak suction pressure on upper surface strengthens and moves forwards with increasing a. - Most lift from near LE on upper surface due to suction.

4.3 Centre of Pressure Movement with Angle of Attack

The centre of pressure is a point on the chord line through which the overall lift force may be considered to act. In the case of a cambered aerofoil section, its position is highly variable (Fig 4.6). As the angle of attack is increased, the peak suction pressure moves further and further forwards on the upper surface until the stall angle is approached. This must then also mean that the centre of pressure moves forwards with increasing angle of attack. Post-stall the upper surface pressure distribution flattens out and the centre of pressure moves rearwards. Consequently the centre of pressure position is not best suited for moment, trim and stability/control analysis; what is required is a fixed position and for these purposes the aerodynamic centre is used instead. This is covered in more detail in the Stability section of the notes.

10

Fig 4.6 Centre of Pressure Movement with Angle of Attack f or Ca mbered Section

4.4 Normal Force & Lift

While it is mostly convenient to deal with aerodynamic lift, there are occasions when it is best to make use of a different resolution of the aerodynamic force in which it is broken down into normal force and axial force instead of lift and drag. It should be recalled that lift and drag are the resolved components relative to the airflow direction (lift is perpendicular to it, drag is parallel to it). If the total force is resolved relative to the chord line of the wing instead, then the components are known as the normal force and axial force. The vector sum of the two components will always be the same, of course, i.e. equal to the total aerodynamic force. A pair of general trigonometric relationships between the components may easily be derived: L = N cos - X sin (4.3a) D = N sin + X cos (4.3b) Clearly, in cases where the chord line and airflow are coincident (i.e. when = 0o), then L = N and D = X. The use of N instead of L is especially useful when it comes to dealing with the stability and control aspects of an aircraft.

Fig 4.7 Resolution into Lift & Drag, Normal Force & Axial Force Components It is a relatively straightforward process to calculate a wing's normal force by measuring the static pressure distribution around it (Fig 4.8).

11

Fig 4.8 Static Pressure Tappings used to Measure Wing Normal Force The measured static pressures are then converted into pressure coefficient values, using: Cp = (p - p) / ( V2 ) (4.4) Cp is thus the ratio of the gauge static pressure to the freestream dynamic pressure. In conditions where p > p, Cp is positive. Conversely, where p < p, Cp is negative. Negative Cp values should therefore be expected on the upper surface of a lift-producing wing. By plotting local values of Cp against distance/chord (x/c), as shown in Fig 2, it is possible to determine a value for the overall normal force for the given set of conditions. This is done by simply calculating the enclosed area of the plot, giving a value for the normal force coefficient (CN). This may then be translated into a value for the normal force itself by making use of the standard aerodynamic force definition: N = V2 S CN (4.5)

Fig 4.9 Pressure Coefficient Plot and Resultant Wing Normal Force