AESF 5310
Advanced Aerodynamics
Lecture 16: 16 Nov 2022
Lockheed Martin
F-35 Lightning II Instructor: Prof Larry Li 1
Today’s agenda [AS9.6−9.7] Developed theory for PM
expansion fans in 1907−08
• Compressible flow
• Prandtl-Meyer expansion fan [AS9.6]
• Application of shock-expansion theory [AS9.7] Ludwig Theodor
Prandtl Meyer
(Prandtl’s
student)
Lockheed SR-71 Blackbird at
L16 2
Imperial War Museum Duxford (UK)
Prandtl-Meyer expansion fan [AS9.6]
Earlier we saw that when a supersonic flow is turned towards itself (e.g. by
placing a wedge into the flow), we get compression via oblique shocks.
But what happens if the flow is turned in the opposite direction?
M2 = ?
L16 3
Prandtl-Meyer expansion fan [AS9.6]
Earlier we saw that when a supersonic flow is turned towards itself (e.g. by
placing a wedge into the flow), we get compression via oblique shocks.
But what happens if the flow is turned in the opposite direction?
We get expansion via a ‘fan’ → Prandtl-Meyer expansion fan
1 -Unlike a compressing flow, an expanding
𝜇1 = sin−1
𝑀1 flow does not lead to a shock wave.
-Instead, a continuous expansion occurs via
∞ # of expansion waves, called Mach waves.
-The flow turns gradually, with each
1 successive expansion wave turning the flow
𝜇2 = sin−1
𝑀2 by a very small (infinitesimal) amount.
-Each individual expansion wave is thus
isentropic, so the flow across the entire
expansion fan is also isentropic.
L16 4
Prandtl-Meyer expansion fan [AS9.6]
Earlier we saw that when a supersonic flow is turned towards itself (e.g. by
placing a wedge into the flow), we get compression via oblique shocks.
But what happens if the flow is turned in the opposite direction?
We get expansion via a ‘fan’ → Prandtl-Meyer expansion fan
-The Mach # downstream of the expansion fan increases 𝑀2 > 𝑀1 as the pressure,
density & temperature decrease → The flow accelerates across the fan isentropically
L16 5
Prandtl-Meyer expansion fan [AS9.6]
Consider a single Mach wave (e.g. the first of many in an expansion fan)
expanded through an infinitesimal angle 𝑑𝜃:
(𝜃 = turning angle) 1
with Mach angle: 𝜇 = sin−1
𝑀1
𝜇 = Mach angle
𝜇 + 𝑑𝜃
𝑉 𝜇
Incoming supersonic flow 𝑑𝜃
(𝑀1 > 1)
Neglect the influence of
boundary layers because the
Reynolds number is high
The velocity tangent to the Mach wave is the same on
both sides:
𝑉 cos 𝜇 = 𝑉 + 𝑑𝑉 cos 𝜇 + 𝑑𝜃 Trig. identity
= (𝑉 + 𝑑𝑉)(cos 𝜇 cos 𝑑𝜃 − sin 𝜇 sin 𝑑𝜃)
L16 6
Prandtl-Meyer expansion fan [AS9.6]
Because 𝑑𝜃 ≈ 0, cos 𝑑𝜃 ≈ 1 sin 𝑑𝜃 ≈ 𝑑𝜃
1 𝑑𝜃
𝑉 cos 𝜇 = (𝑉 + 𝑑𝑉)(cos 𝜇 cos 𝑑𝜃 − sin 𝜇 sin 𝑑𝜃)
𝑉 cos 𝜇 = 𝑉 cos 𝜇 + 𝑑𝑉 cos 𝜇 − 𝑉 sin 𝜇 𝑑𝜃 − 𝑑𝑉 sin 𝜇 𝑑𝜃
Product of two small
Cancel 𝑉 cos 𝜇 on both sides and neglect the higher order term
terms = very small term
𝑑𝑉 sin 𝜇 1 𝑑𝑉𝑑𝜃 ∴ may neglect altogether
= 𝑑𝜃 = 𝑑𝜃
𝑉 cos 𝜇 2
𝑀 −1
[AF9.28]
𝑑𝑉
𝑑𝜃 = 𝑀2 − 1 [AE9.32]
𝑉
Recalling the PM expansion fan is isentropic:
𝑑𝑉
Express in terms of 𝑀 and 𝑑𝑀 [AE9.39] and then integrate [AE9.32]:
𝑉
𝑀2
Turning angle across 𝑀2 − 1 𝑑𝑀
𝜃=න 𝜃 = 𝑣 𝑀2 − 𝑣(𝑀1 ) [AE9.43]
entire expansion fan: 1
𝑀 1 1 + 2 𝛾 − 1 𝑀2 𝑀
L16 7
[AE9.40] The PM function
Prandtl-Meyer expansion fan [AS9.6]
𝑀
𝑀2 − 1 𝑑𝑀
where: 𝑣(𝑀) ≡ න
1 PM function [AE9.41]
1 1+ 𝛾 − 1 𝑀 2 𝑀
2
This is also called the PM angle, which is the angle by which a sonic flow
(𝑀1 = 1) must turn (i.e. expand) to accelerate to a given 𝑀 > 1
These angles are in degrees, not radians.
PM function or Mach angle
PM angle
L16 8
Prandtl-Meyer expansion fan [AS9.6]
Procedure: Given 𝑀1 and 𝜃, how do we find 𝑀2 ?
We use Appendix C (PM function and Mach angle)
Recall: These are in degrees, not radians.
Appendix C
Given 𝑀1 > 1 𝑣(𝑀1 )
Appendix C
𝑣 𝑀1 + 𝜃 = 𝑣 𝑀2 𝑀2
This is usually given
by the geometry of
the problem
Expansion-fan theory can be
M2 = ? combined with oblique-shock
theory to calculate the lift and
L16 drag forces on a supersonic airfoil 9
Shock-expansion theory [AS9.7]
If an airfoil has a wedge-like shape (i.e. with straight line segments), we can combine oblique-
shock theory and expansion-fan theory, as long as any shocks formed are not detached.
For example: Expansion
fan
Flat-bottomed wedge
1 2 of half-angle 4o
B
3
𝑀∞ = 1.8 A 4°
𝑃∞ = 6 × 104 Pa 4°
8°
4 C
Shock-expansion theory gives: 𝑀2 = 𝑀4 = 1.652 𝑀3 = 2.225
𝑃2 = 𝑃4 = 7.4 × 104 Pa 𝑃3 = 3.06 × 104 Pa
Lift is the total vertical component of 𝑷𝟐 , 𝑷𝟑 , 𝑷𝟒 acting on the respective areas (per unit span):
𝐿′ = −𝑃2 ȁ𝐴𝐵ȁ cos 4° − 𝑃3 ȁ𝐵𝐶 ȁ cos 12° + 𝑃4 ȁ𝐴𝐶 ȁ cos 4° = 2.147 × 104 𝑐 N/m
(𝒄/𝟐)(𝟏/ 𝐜𝐨𝐬 𝟖°) 𝒄 = chord
𝐿′ 𝐿′
So the lift coefficient via shock-expansion theory: 𝐶𝑙 = 1 =1 = 0.158 To be compared
𝜌𝑉 2 𝑐
2 ∞
𝛾𝑃 𝑀2 𝑐
2 ∞ ∞
10
with LPE later
Shock-expansion theory [AS9.7]
If an airfoil has a wedge-like shape (i.e. with straight line segments), we can combine oblique-
shock theory and expansion-fan theory, as long as any shocks formed are not detached.
For example: Expansion
fan
Flat-bottomed wedge
1 2 of half-angle 4o
B
3
𝑀∞ = 1.8 A 4°
𝑃∞ = 6 × 104 Pa 4°
8°
4 C
Shock-expansion theory gives: 𝑀2 = 𝑀4 = 1.652 𝑀3 = 2.225
𝑃2 = 𝑃4 = 7.4 × 104 Pa 𝑃3 = 3.06 × 104 Pa
What if we projected the pressures in the horizontal direction?
Answer: We would get the drag force! In inviscid supersonic flow, the drag on a 2D airfoil is finite
because shocks cause an increase in entropy and hence a loss of total pressure,
resulting in a new source of drag: wave drag. Therefore, in supersonic flow, d’Alembert’s
L16 11
paradox does not occur. We will talk more about wave drag later in this course.
Why do we need supersonic airfoils?
Airfoils that work well in subsonic flow generally do not work well in supersonic flow,
and vice versa…
Subsonic airfoil in subsonic flow Subsonic airfoil in supersonic flow
M>1
M<1
Rounded LE prevents
flow separation across Rounded LE produces a
a wide range of 𝛼 bow shock, increasing drag
Supersonic airfoil in subsonic flow Supersonic airfoil in supersonic flow
Flow Separation
M >1
M >1
Sharp LE increases the Sharp LE prevents the
L16 risk of flow separation formation of a bow shock 12
Why do we need supersonic airfoils?
Airfoils that work well in subsonic flow generally do not work well in supersonic flow,
and vice versa…
• To prevent shock-induced separation at transonic
speeds, max thickness is usually at mid-chord.
• Thin and pointy → No normal shock or bow shock,
thus reducing drag.
X-15 (M=6.7)
L16 13
Protective
Supersonic airfoils sleeves on
LE and TE
Aerodynamics Concept Box: Lockheed F-104 Starfighter
The F-104 was a supersonic interceptor designed in
the late 1950s and flown beginning in the 1960s by
several countries. Its wing had an aspect ratio of
2.45 and an airfoil which was only 3.34% thick. The
airfoil section was a modified bi-convex (bi-convex
upper surface and essentially flat lower surface) with
sharp leading and trailing edges.
Most supersonic airplanes have airfoil sections
which are a compromise between subsonic and
supersonic characteristics, because the airplanes
must fly in both speed regimes. In the case of the F-
104, the designers used an airfoil that had good
supersonic capabilities, but rather poor subsonic
characteristics (as the design objectives required
Mach 2.0 flight). This meant that the F-104 required
large blown flaps for landing, in addition to a drag
chute. Despite this, the approach speed of the
airplane was nearly 200 knots, and the landing
distance was quite large. Fighters designed in the
1970s (like the F-16 or F-15) used subsonic airfoil
sections (although they were quite thin) to give
improved subsonic characteristics, at the cost of
L16 supersonic performance. 14
F-104 Starfighter walk-around
Supersonic airfoils (12:30 mark for airfoil):
[Link]
ch?v=eB8l_fCKDSw
Aerodynamics Concept Box: Lockheed F-104 Starfighter
The resulting sheet of hot air blown down
the upper surface of the flap delays flow Internally blown flap system
separation, allowing for a larger flap angle
and increasing the lift coefficient. Incoming air from the
wing’s upper surface.
Hot engine bleed air is fed into a
titanium tube and blown through a
slot nozzle located at its top.
However, if the engine
Flap with
failed on final approach,
re-energized
the lift coefficient would
boundary layer.
drop, causing the approach
L16
speed to increase to 15
around 240 knots!
Not all supersonic airplanes use supersonic airfoils
Aerodynamics Concept Box: Lockheed Martin F-22 Raptor
The F-22 uses modified NACA 6-series airfoils: NACA 64A005.92 at the root and NACA 64A004.29 at the tip. This is surprising because most
aircraft today, particularly advanced military aircraft, almost never use traditional airfoils like the NACA series. Today’s advanced computers make
it possible to design and optimize new airfoils suited to the specific needs and operating conditions of an aircraft, and designers usually prefer to
develop a new airfoil shape to improve a vehicle's overall performance. Boeing has been custom-designing airfoils for its aircraft for decades, so
it seems odd that Lockheed chose an airfoil dating back to the 1940s rather than designing a new airfoil better tailored to the specific
performance requirements of its premiere product.
But then again, the NACA 6-series was designed for low-drag at transonic speeds, where the F-22 spends most of its time, and so it makes sense
that Lockheed chose this airfoil. In addition, the digits 05.92 and 04.29 indicate that the root and tip airfoils are only 5.92% and 4.29% thick. We
would expect rather thin airfoils like these on a high-speed aircraft to reduce drag. [Source: [Link]
Homework 6
L16 17