In 1983, an Israeli F-15 lost almost its entire right wing. Fuselage lift, speed and CAS explain how it managed to land.
In summary
In May 1983, an Israeli Air Force F-15D collided with an A-4 Skyhawk over the Negev Desert. The impact tore off almost its entire right wing. Despite a fuel leak and going into a spin, the pilot, Zivi Nedivi, managed to recover the aircraft. He landed at approximately 463 km/h (250 kt), almost twice the usual approach speed.
This feat was not achieved by engine power alone. The F-15’s very wide fuselage generates lift in its own right. Its shape, integrated with the air intakes and wing roots, acts partly as a lifting body. The high speed also multiplied the effectiveness of the remaining control surfaces. Finally, the mechanical controls, hydraulic actuators and the Control Augmentation System made it possible to use the differential horizontal stabilisers, the control surfaces and the remaining wing to counteract the enormous aerodynamic asymmetry.
The Negev accident destroyed a wing in a matter of seconds
In May 1983, several Israeli Air Force aircraft were taking part in a simulated air combat exercise over the Negev. Two F-15D Bazes were engaging A-4N Skyhawks acting as enemy aircraft.
One of the A-4s performed a climbing manoeuvre. Its pilot did not see the F-15 positioned above him. The two aircraft collided at an altitude of approximately 3,660 metres (12,000 ft). The Skyhawk was destroyed. Its pilot managed to eject.
The F-15D, serial number 957, piloted by Zivi Nedivi with instructor Yehoar Gal in the rear seat, lost most of its right wing. The break occurred approximately 0.6 metres from the fuselage (2 ft), according to published accounts of the incident. A significant amount of fuel leaked from the ruptured structure. The cloud of sprayed fuel obscured the damage from the cockpit.
The crew realised that one wing was damaged. They were unaware that it had virtually disappeared.
The aircraft initially rolled to the right at approximately 648 km/h (350 kt), then entered a spiralling descent. The instructor recommends ejection. Nedivi chooses to remain at the controls. He applies left rudder, moves the flight controls and abruptly increases thrust.
The F-15 regains a controllable attitude. It heads towards Ramon Air Base.
The situation deteriorates again when the pilot reduces speed to prepare for landing.
At around 482 km/h (260 kt), the aircraft begins to autorotate once more. Nedivi re-engages the afterburner and regains control for a second time.
The message is clear. Below a certain speed, the aircraft can no longer fly.
The loss of a wing creates a far more serious imbalance than a loss of lift
A wing does not merely serve to support the weight of an aeroplane. It also produces aerodynamic moments around the centre of gravity.
On an intact aeroplane, the left and right wings generate comparable forces. Their roll moments cancel each other out. When the right wing is lost, the left wing continues to produce a vertical force situated several metres from the fuselage axis.
This force lifts the left side and causes the aircraft to tilt to the right. It generates a huge roll moment. Drag also becomes asymmetrical. The aircraft therefore tends simultaneously to roll and pitch about its vertical axis.
It is not enough simply to generate sufficient total lift. It is also necessary to prevent this lift from causing the aircraft to flip over.
Furthermore, the loss of the right wing deprives the F-15 of its right aileron and right flap. It abruptly alters the position of its centre of pressure. It disrupts the airflow around the air intake, the tail and the rear fuselage. The aerodynamic characteristics specified by the manufacturer are no longer valid.
The pilot is therefore no longer flying a degraded F-15. He is flying a new aircraft, the flight envelope of which no one has established.
The F-15’s fuselage generates significant aerodynamic lift
The F-15 is not a pure winged body, like certain experimental NASA aircraft. It has large wings measuring 56.61 square metres with a wingspan of 13.05 metres. However, its design differs significantly from that of an aircraft with a narrow cylindrical fuselage.
Its central fuselage is wide. The two engines are mounted side by side. The rectangular air intakes are situated on the sides of the fuselage. The wing roots extend naturally into this vast central surface.
Viewed from above, the nose, the air intakes, the fuselage shoulders and the section between the engines form a continuous surface. When the aircraft flies at a positive angle of attack, this geometry deflects the airflow downwards. Higher pressure develops beneath the fuselage. Lower pressure develops on the upper surfaces.
The result is a vertical force. The fuselage provides lift.
This central lift offers a key advantage in an asymmetrical situation. It acts close to the longitudinal axis. It therefore supports part of the weight without creating the same lever arm as the remaining left wing.
There are no reliable publicly available figures giving the exact percentage of lift produced by the fuselage of the crashed F-15D. Such a percentage would, moreover, vary with speed, angle of attack, the configuration of the air intakes and the state of the flow. It would be incorrect to claim that the fuselage entirely replaced the lost wing.
However, it did generate sufficient vertical force to reduce the load borne by the left wing and the tail surfaces. Subsequent tests carried out by McDonnell Douglas confirmed that this configuration did indeed possess a controllable centre of balance.
High speed transformed the fuselage into an effective lifting surface
Lift can be described by a simple equation:
Lift = ½ × air density × speed² × area × lift coefficient.
The key point is that lift varies with the square of the speed.
Close to the ground, an F-15 approaching at 241 km/h (130 kt) encounters a dynamic pressure of approximately 2.7 kilopascals under standard atmospheric conditions. At 463 km/h (250 kt), this pressure reaches approximately 10.1 kilopascals.
It is therefore 3.7 times higher.
This increase does not replicate the effect of a torn-off wing. However, it does increase the forces generated by each surface still exposed to the air. The fuselage, the left wing, the horizontal stabilisers, the vertical stabilisers and the control surfaces become much more effective.
This speed also provides greater control authority. A small deflection of the control surfaces produces a greater aerodynamic force. The pilot therefore has sufficient time to apply corrective inputs to counteract roll and yaw.
The trade-off is severe. Drag increases sharply. The landing distance skyrockets. The tyres, brakes and landing gear are subjected to stresses far beyond those of a normal approach. Above all, the slightest reduction in speed causes dynamic pressure to drop rapidly.
This is exactly what happened at around 482 km/h (260 kt). When Nedivi slowed down, the corrective forces became insufficient. The F-15 went into a spin again.
The engines did not carry the aircraft but maintained its speed
The oft-repeated saying that ‘a powerful enough engine can make a brick fly’ is dramatic. It is also incomplete.
The two Pratt & Whitney F100s did not directly replace the right wing. Their thrust was directed mainly forwards. It therefore produced only a small vertical component at the angles of attack encountered.
Their role was nonetheless vital. The afterburner made it possible to maintain speed despite the enormous drag from the damaged fuselage. It prevented the aircraft from falling below the threshold at which the fuselage and control surfaces would lose their effectiveness.
The thrust also increased the pilot’s ability to break out of the spin. By accelerating, the F-15 regained more dynamic pressure. The control surfaces were once again able to generate sufficient moments.
Thrust and lift must therefore not be confused. The engines provided the necessary energy. The remaining aerodynamic surfaces converted this speed into lift and control forces.
Differential use of the two engines might also have produced a yaw moment. However, the F-15’s engines are positioned very close together. Their nozzles are separated by only about 1.30 metres (4.25 ft). Their ability to directly correct a severe lateral imbalance is therefore limited. The available technical accounts focus primarily on the use of afterburner to accelerate.
The F-15D did not have a fully integrated fly-by-wire system
The F-15A, F-15B, F-15C and the original F-15D were not equipped with a fully integrated digital fly-by-wire system comparable to that of subsequent generations.
The pilot operated hydraulically assisted mechanical controls. Movements of the control stick and rudder pedals were transmitted to mechanisms that controlled hydraulic actuators. These actuators moved the ailerons, rudder and horizontal stabilisers.
Hydraulic power was essential. Without hydraulic pressure, the primary control surfaces could no longer provide the necessary forces at high speed.
This architecture retained a mechanical link between the pilot and the controls. It was, however, supplemented by an analogue electronic system: the Control Augmentation System, or CAS.
The CAS was not an autopilot capable of devising a new control law following the loss of a wing. It could not identify damage. It did not recalculate the aircraft’s aerodynamic model.
Its role was to improve stability and response to commands. Sensors measured pitch, roll and yaw movements, amongst other things. The system then applied corrections via servo controls integrated into the control system.
The CAS enhanced roll control via the tail surfaces
The F-15 has two large, fully movable horizontal stabilisers. Their total surface area is 11.15 square metres. They can move together to control pitch. They can also move in opposite directions to generate a roll moment.
This function is crucial following the loss of a wing. The stabilisers remain positioned on either side of the fuselage. They therefore provide lateral control capability even when one of the ailerons is missing.
On the F-15, increased roll control is achieved via the CAS servos associated with the stabilisers. NASA’s technical documentation specifies a ratio of 0.3 degrees of differential stabiliser deflection per degree of aileron deflection, subject to the limits imposed by the actuators and the flight envelope.
The differential stabilisers could thus complement the remaining left aileron. They produced a torque opposite to that created by the surviving wing.
The yaw CAS also provided damping via both rudders. Its authority extended up to 5 degrees of rudder deflection. An interconnection between the ailerons and rudders, known as the aileron-rudder interconnect, helped to coordinate lateral movements.
In the 1983 incident, the pilot applied a strong left rudder input. The hydraulic system and the increased stability made this manoeuvre feasible at very high speeds.
It would, however, be an exaggeration to claim that the CAS alone saved the aircraft. Detailed control data for F-15 number 957 have not been published. It is impossible to precisely distinguish the contributions of the pilot, the CAS, the stabilisers, the control surfaces and the remaining wing.
One thing remains certain. Without intact rear control surfaces, without hydraulics and without sufficient control authority, the load-bearing fuselage would not have been sufficient.
Landing at 463 km/h was the only realistic option
A standard F-15 approach was typically carried out at around 241 km/h (130 kt), depending on the aircraft’s weight and configuration. Nedivi now knew that his aircraft would become uncontrollable at around 482 km/h (260 kt).
He therefore decided to touch down on the runway at approximately 463 km/h (250 kt).
At this speed, the kinetic energy is considerable. As it too varies with the square of the speed, an aircraft approaching at 250 knots has approximately 3.7 times more kinetic energy than at 130 knots, for the same weight.
Ramon Air Base has an emergency arresting system. The pilot lowered the F-15’s tail hook to catch a transverse cable. The tail hook caught the cable, but the load was too great. It was torn from the aircraft.
The aircraft had nevertheless slowed to around 278 km/h (150 kt). Nedivi then applied the brakes. The F-15 came to a halt about ten metres from the safety net installed at the end of the runway.
Less than five minutes elapsed between the collision and touchdown.
It was only after the aircraft had come to a halt that the crew realised the true extent of the damage. The F-15D was subsequently repaired and returned to service.
Tests showed that the feat hinged on a minuscule margin
Following the accident, McDonnell Douglas engineers had no data corresponding to such significant structural damage.
They used an F-15 model in a wind tunnel. They gradually removed various sections of the right wing in order to study the characteristics of the damaged configuration.
The calculations showed that a controllable flight condition did indeed exist. However, it lay within an extremely narrow control window. According to the historical study published by NASA, the margin was only around 37 km/h around the equilibrium speed, or plus or minus 20 knots. The permissible variation in angle of attack was also limited to approximately plus or minus 20 degrees around the point of compensation.
These figures do not mean that the pilot could freely alter his angle of attack by 40 degrees. They indicate that the new aerodynamic equilibrium existed only within a narrow range. A change in speed, an excessive control input or a disturbance could cause the aircraft to leave this range.
Nedivi discovered this zone without a mathematical model, without simulation and without any specific indication on his display. He discovered it by reaction, observing that acceleration made the controls effective and that deceleration triggered the spin.
The feat reveals the limitations of the ‘indestructible F-15’ narrative
The Israeli F-15’s landing with a single wing is genuine. However, there is nothing supernatural about the explanation.
The wide fuselage generated central lift. The high speed multiplied the available aerodynamic forces. The horizontal stabilisers could generate roll. The control surfaces corrected yaw. The CAS improved damping and provided additional control to the surfaces. The engines kept the aircraft above its critical speed.
All these conditions had to be met.
The incident does not prove that an F-15 can normally fly without a wing. It proves that a specific F-15, with a specific wing separation, hydraulic systems still functioning and an intact tail, had a degraded flight condition that was extremely difficult to maintain.
A slightly different wing separation could have cut off the controls, caused a fire, damaged an engine or shifted the centre of pressure further. A reduction of just a few tens of kilometres per hour would have been enough to make recovery impossible.
The real lesson, therefore, does not lie in any supposed invulnerability. It lies in the rare combination of a forgiving aerodynamic design, strong control authority and a human decision taken in a matter of seconds. The aircraft did not defy the laws of physics. The pilot managed to exploit the last available loophole in them.
Live a unique fighter jet experience