Thrust Reversers and Thrust Vectoring

Cascade and clamshell reversers, reverse-thrust accounting, pitch vectoring and the three-bearing swivel duct

Lauda Air Flight 004 — 26 May 1991

A Lauda Air Boeing 767 takes off from Bangkok for Vienna. During the climb, the thrust reverser on its left engine deploys without command. The disturbed airflow over the wing costs it lift; the aircraft stalls, enters an uncontrolled dive and breaks up in the air over Phu Toei National Park. All 223 people on board — 213 passengers and 10 crew — are killed (Lauda Air Flight 004). Investigators concluded that what followed the deployment "was uncontrollable for an unexpecting flight crew". Boeing modified the reverser system, adding sync locks that stop the reversers deploying unless the main landing gear's truck-tilt angle shows the aircraft is on the ground. A thrust reverser exists to make an engine push the wrong way — on the runway, for a few seconds, and never anywhere else.

Reversers. On landing, wheel brakes do most of the stopping, but reversers add a large force at high speed — when brakes are least effective and the runway is wet or icy. Two families dominate:

Cascade reversers on high-bypass turbofans: a sleeve slides back, blocker doors close the bypass duct, and the fan air leaves through rows of turning vanes that point it forward and outward. The core jet usually keeps pushing rearward.
Target (clamshell) reversers, mostly on low-bypass engines and business jets: two doors close behind the nozzle and deflect the whole jet forward.

F_{retard} = \dot m_F V_F \cos\theta + (\dot m_F + \dot m_C)V_0 - \dot m_C V_C

The retarding force of a cascade reverser: the forward component of the fan stream leaving at angle θ from straight ahead, plus the ram drag of all the air swallowed, minus the core jet still pushing rearward.

Derivation

Accounting for reverse thrust

Draw the usual control volume around the engine. Air enters at the landing speed

V_0

; its momentum flux

\dot m V_0

must be destroyed whatever happens next — the ram drag, now on our side. The fan stream leaves through the cascades at speed

V_F

, angled

\theta

from straight ahead, so its momentum flux has a forward component

\dot m_F V_F\cos\theta

that pushes the engine backward. The core jet still leaves rearward at

V_C

and pushes the engine forward. The net retarding force is the sum above. As the aircraft slows, ram drag shrinks, and at low speed the reversed air can be swallowed again by the inlet — so reverse power is reduced as the aircraft decelerates.

📐A cascade reverser on the landing roll

Illustrative numbers for a large twin at reverse power, touching down at 70 m/s: 350 kg/s of fan air leaving the cascades at 180 m/s, 35 kg/s of core gas at 350 m/s (Lab 1):

Cascades at 45° from straight ahead: fan 44.5 kN + ram drag 26.9 kN − core 12.2 kN = 59.2 kN retarding, about half the engine's take-off thrust.
At 30° the fan gives more (69.3 kN in total); at 60° less (46.2 kN).

Ram drag is a large share at touchdown speed and fades as the aircraft slows.

💡Vectoring

Turning the jet sideways gives a force across the flight path and, through the distance to the centre of gravity, a moment — even at low speed and high angle of attack, where control surfaces lose authority:

2-D pitch vectoring: the F-22's F119 nozzles turn the jet ±20° in pitch (the F119).
Multi-axis vectoring: the Su-35S has thrust-vectoring nozzles for supermanoeuvrability (thrust vectoring).
Lift vectoring: the F-35B's three-bearing swivel module turns its main jet through 95° in 2.5 seconds, beside the shaft-driven lift fan of Lesson 3-4 (the LiftSystem).

M = n\,T\,\ell\,\sin\delta, \qquad \frac{\Delta T_{axial}}{T} = 1 - \cos\delta

Pitching moment of n engines vectored by δ at an arm ℓ from the centre of gravity, and the fraction of axial thrust given up.

Derivation

Moment and cost of vectoring

Turning a jet of thrust

T

through

\delta

splits it into an axial part

T\cos\delta

and a side force

T\sin\delta

. The side force acts at the nozzle, a distance

\ell

behind the centre of gravity, so

n

engines give a moment

nT\ell\sin\delta

. The axial loss is

T(1 - \cos\delta) \approx T\delta^2/2

for small angles: the moment grows with

\delta

, the loss only with

\delta^2

— vectoring is cheap at small angles.

📐Trimming with the nozzles

Two engines of 150 kN (illustrative), 6 m behind the centre of gravity (Lab 2): 10° of vectoring gives 313 kN·m of pitching moment for 1.5 % of the axial thrust; trimming 400 kN·m needs 12.8°, costing 2.5 %. The three-bearing swivel duct (Lab 3) is a lesson in rotation matrices. In a simplified model, two joints cut at the same angle turn in opposite directions and bend the duct by up to four times the cut angle — 95° for a cut of 23.75° — while the front bearing rolls the whole bend so the jet points straight down. Hovering needs 90°: 143.7° of counter-rotation and a 160° roll.

For a successful technology, reality must take precedence over public relations, for nature cannot be fooled.

– Richard Feynman, Report of the Presidential Commission on the Space Shuttle Challenger Accident, Appendix F (1986) (Wikiquote)

⚠Interlocks

Reverser systems now use several independent locks and require evidence that the aircraft is on the ground — weight on the wheels, gear truck tilt, wheel speed — before deployment is possible. The same thinking applies to any device whose function is dangerous in the wrong flight phase.

💡Where you will use this

Lesson 9-3 follows the reverser's loads into the nacelle and pylon. Lesson 10-1 returns to vectoring in adaptive-cycle fighter engines. The rotation-matrix tools of Lab 3 are the same ones used for spacecraft attitude and robot arms.

📚Further reading

Thrust reversal and thrust vectoring on Wikipedia.
Lauda Air Flight 004, the accident and its investigation.
Jack D. Mattingly and Keith M. Boyer, Elements of Propulsion: Gas Turbines and Rockets (2nd ed., AIAA, 2016), on exhaust systems.

↓But there is a problem

The jet that stops and steers the aircraft is also the loudest, hottest and most visible thing about it — heard over whole cities near airports, and seen from miles away by an infrared seeker. In 1952 a 28-year-old mathematician named James Lighthill showed why jets are so loud, with a law that every engine designer since has had to obey.