EAP Aircraft: A Blueprint for Advanced Flight Controls

Introduction

The Experimental Aircraft Program (EAP) was a technology demonstrator built in the mid 80’s to examine design and flight controls for advanced fighter aircraft. This demonstrator aircraft included composite structures and an advanced fly-by-wire system. The EAP program was a successor to the successful Jaguar fly-by-wire program which proved the first active control technology principles.. With minor modifications, the same basic airframe and basic control structure were later folded into one of the world’s top fourth generation fighter program, the Eurofighter. To show how radical this deign was we can see the many different control surfaces that were used on the EAP demonstrator aircraft below, all of which were computer controlled.

Background

One of the other major design choices to improve aerodynamic performance for a fighter aircraft was to use an airframe with a reduced static stability. This technique had been used in various 70’s fighter aircraft such as the F-16 and the J-37 Viggen. Static margin is the relationship between the center of gravity and the aerodynamic forces. An arrow has a positive static margin as most of the aerodynamic effects act on the feathers, which are behind the center of gravity (balance point) of the arrow. No static margin would look like an arrow with feathers in the middle, and negative static margin would put the feathers near the front of the arrow.

Relaxed static stability aircraft have a negative static margin at subsonic speeds. This allows them have lower positive static margins at supersonic speeds which translates into better maneuverability. The images below show the force balance for a conventional vs a relaxed-stability aircraft at subsonic speeds. For RSS aircraft, the total lift acts ahead of the center of gravity.

The rearward shift of the aerodynamic center from subsonic to supersonic flight causes an increase in the static margin. As the aircraft was already unstable, this shift produces an aircraft with a static margin that is still less than the conventional aircraft. This reduced static margin also decreases the force that is needed from the horizontal stabilizer, which decreases trim drag. Below is a force balance diagram of the supersonic flight condition for both conventional and reduced-static stability aircraft.

With relaxed static stability, it is more challenging to build a robust control system, and most aircraft with relaxed static stability need some form of augmented control. Increasing static stability or short-period damping are important for relaxed-static stability flight control designs.

To meet performance requirements, the EAP was developed as an unstable Canard-Delta configuration with the all-moving canards located ahead and above of the main delta wing. The minimum time to double amplitude of 180ms, which meant that without computer control, the aircraft would double its pitch angle in just under 1/5 of a second.

One of the main innovations was the improvement of control laws for the fly-by-wire system. In a fly-by-wire system when a pilot adjusts the control column, a transducer generates a signal that is directed to the flight control computer. This computer processes the control signal in tandem with other aircraft state data. Consequently, it sends a refined control signal to the actuators that move the control surfaces. This creates a dual-loop control system where the outer loop represents the pilot’s perception of the aircraft’s motion, and the inner loop encompasses the interaction between the air data sensors and the flight control computer.

Fly-by-wire systems are the preferred system in many modern high-performance aircraft, spanning both military and civilian sectors. Due to its electrical nature, DFBW must be backed by a robust and redundant electrical power system. It also permits reconfigurable control laws in-flight and introduces task-tailored control modes, enhancing aircraft adaptability.

One of the advantages of digital fly-by-wire aircraft are that it can implement control elements that are too complex for analog system. It can also be used to modify the dynamic performance of an aircraft, as in the case with relaxed-stability aircraft like the EAP. Some other advantages include redundant information paths for data as well as variable schedules for optimal control. There is also a significant weight and maintenance advantage over mechanical systems. Fewer moving parts mean reduced wear and tear, translating to decreased maintenance needs. Safety can be increased by providing flight envelope protection to keep the pilot from placing the aircraft into an unrecoverable condition.

For the EAP, The canards, intake cowl, and wing leading-edge devices were controlled directly from the flight control computers. The trailing-edge elevons and rudder were driven from aft-mounted actuator drive units. The canards provided minimum trim drag in subsonic and supersonic configurations due to the relaxed stability as explained above.

Avionics

The avionics system on the EAP used a multiplexed MIL-STD-1553 bus. This is a dual-redundant serial communication protocol that is standard for aerospace vehicles. Many of the aircraft’s electrical and control components were connected to the main avionics bus. The cockpit displays were the biggest avionics bus subsystem. 2 Cockpit interface units process the data from pilot switches and other commands before distributing it to the bus.

In the diagram below we can see how the flight computer stack connected to the rest of the system through the 1553 bus. Above the horizontal bus line are all of the radios, navigation systems, and pilot cockpit displays. Below the horizontal bus line are the flight control computers, the air data computers and the management computers.

The PASCAL programming language was used for the utility management system, the waveform generators, and the cockpit interface units. The CORAL 66 programming language was used in software development. This language was related to the ALGOL60 language and also used on the first implementation of the VAAC Harrier flight control system.

Flight Control Design and Architecture

All flight control laws were digitally computed by the quadruplex full-authority DFBW system. The basic pitch mode was a pitch-rate demand with the dynamics that are tuned to a conventional Angle of Attack(AOA) response for the short-period. Increasing stick demand was blended progressively into a normal acceleration demand, and then an AOA demand above a specified corner point.

A complementary filter was combined with the mixed-airstream sensor data to derive the incidence for pitch stiffness augmentation. This prevented unexpected transients when passing through the wake of another aircraft, where using only air data would give erroneous results.

A Differential PI Controller (DPIA) was used in a Single-Input-Single-Output(SISO) structure to stabilize the pitch axis. A DPIA controller is a scheduled Proportional-Integral controller where all proportional feedback loops and direct command paths are differentiated.

The differentiation provides signals that are zero in steady-state conditions. This means that the downstream gains of the differentiation can be scheduled with respect to AOA/Angle of Sideslip(AOS) without creating unwanted feedback loops. The integration also provides smoothing which allows gains scheduled upstream of the integration to include noisy signals. For A DPIA controller the proportional loops have no steady-state influence.

DPIA controllers can easily avoid integral windup as the surface and rate limits place a limit on the possible integrator values as shown in the block diagram below. Integral windup happens when a saturated control causes an accumulation of error over a large period of time. This structure caps the upper limit to how much weight the integral term can accumulate thereby avoiding a windup condition.

The wing leading and trailing edges were scheduled with incidence angle and Mach number. This allowed the camber of the wing to be adjusted for maximum performance. Below is a diagram for the F-5 wing flaps, not the EAP, but it gives you an idea of how the wing camber can change for different conditions.

The flight control system of the EAP compensated for the effects of dynamic pressure by normalizing the control surface effectiveness within the control laws. This was accomplished with nonlinear control demand functions upstream of the actuation system commands. By providing a linear pitching moment command point within the control law independent of the aircraft’s operating points, this method simplified the gain scheduling.

Performance and Capabilities

The PI controller had the effect of giving a very linear low-frequency response. The trim distribution, control power, and stability variations for the aircraft were all nonlinear. The control system allowed for precise control of the unstable aircraft which drastically increased maneuverability.

The quad-redundant computer increased safety as the aircraft would still be flyable even if some of the computers failed. The canards, flaperons, and rudders had a two-fail operational redundancy. Another useful features was an automatic wings-leveling function that educed pilot workload when the wings were close to level. The flight control modes had seamless transitions with no difference between wheels-up and wheels-Down handling. A similar system, if implemented in the YF-22 prototype would have prevented a crash that happened just a few years later.

The control system also added a static-stability function where the pilot stick offset was scheduled by airspeed. This meant that the pilot would have to move the stick more aft as the speed got slower, This mimicked the speed stability characteristic cue of a conventional aircraft which would be more familiar to the pilots. A linear high-frequency mode was incorporated to avoid structural coupling from feeding back into the control signals and causing an unwanted oscillation.

Influence on Future Aircraft

The EAP program helped to validate the design that would eventually become the Eurofighter. A few of the minor differences between the EAP and the Eurofighter was that the wing was changed to a normal delta wing and the speedbrakes were moved up to a single dorsal brake on the top of the aircraft.

The EAP also used carbon-fiber composites as well as an aluminum-lithium alloy for the elevons. The carbon fiber composite made up 25% of the structure by weight. The wing skins, canards, and spars were made from prepreg tape layups. Below we can see a diagram of the surfaces that used carbon composites as well as the lithium alloy control surfaces.

These advances in composite airframes were among one of the first examples of advanced fiber composite aircraft along with NASA’s X-29 aircraft just a few years earlier.

Conclusion

Over the 40 years since the EAP aircraft, we have seen its advances put directly into the Eurofighter program. We have also seen the rise of digital fly-by-wire aircraft in both the commercial and military markets. One big takeaway from this program was that it is good practice to separate the regulator and command path designs when building flight control systems. The EAP was a great example of a successful demonstrator aircraft program. It showed the world what was possible with the cutting-edge of available technology and influenced many of the design choices for what became a future modern front-line fighter aircraft.

Sources

Keep In Touch

Twitter Linkedin-in