DESIGN FOR AIR COMBAT
By the 1960s the disadvantages of the delta wing planform had begun to make themselves felt as the demand grew for the lifting of larger weapon loads off shorter runways and for greater air combat manoeuvrability. The swept wing had by this time shown itself to be the more versatile, and most designers had dropped the delta. Indeed, even Dassault, the leading exponent of the delta, rejected it when developing an aircraft with reduced landing speeds, improved manoeuvrability and heavier weapon load. This was the Mirage F.I. which, with a swept wing two-thirds the area of that of the Mirage III and only 9kN more thrust, is faster, can loiter subsonically and pursue supersonically for three times as long, can carry much more offensive load over almost twice the distance, is 80% more manoeuvrable and lands 20% slower.
1 With its high leading-edge sweep and very low aspect ratio, the delta is less efficient as a lifting planform. The low lift-curve slope means that it must be flown at much higher angles of attack to generate the same lift. However, the demands of tail clearance and pilot view limit the usable AOA, as shown in Fig 37. Furthermore, the delta is unable to use trailing-edge flaps unless it has a separate tail to trim out the large nose-down pitching moment. These factors combine to give tailless deltas very high landing speeds and poor airfield performance.
2 The delta's high span loading (W/b) results in very high lift-induced drag in subsonic flight. This
The Sukhoi Su-21 Flagon-F "all-weather" interceptor has a tailled delta layout. Early Flagons had shorter-span wings locking the leading-edge kink the high wing loading and very slender, indented fuselage are typical of early long-range interceptors (Swedish Air Force)
Fig 44 High lift-induced drag of delta wing. Low aspect ratio and loss of lift due to trimming result in a glide ratio of only 3.7 at 1.4 times stalling speed (in landing configuration).
is a critical drawback in air combat, since very high thrust must be available to avoid a severe drop-off in specific excess power. This is further compounded in manoeuvring by the large trimmed lift loss and associated drag arising from the increasing downloads on the trailing-edge controls as AOA is increased (Fig 44).
3 The low wing loading of the early deltas made them very sensitive to gusts in low-level high-speed flight. This was overcome on the TSR.2 by using, together with a low tail, a very small delta wing of only 65 m2 area with blown flaps and a high thrust/weight ratio. Low-speed manoeuvrability is however invariably compromised whenever high wing loadings (i.e. greater than 500kg/m2) are used on deltas.
4 Supersonic manoeuvrability is greatly restricted by the relative ineffectiveness of the delta's trailing-edge controls, known as elevons (elevators/ailerons), compared with separate tail surfaces. In addition, the aft movement of the aerodynamic centre during transonic acceleration, though reduced by the use of high sweep, is still large in actual distance though appearing small in terms of chord length. Allied to the restricted allowable range of centre-of-gravity position, {56}
Dassault Mirage 2000 has a variable-camber wing — the high-lift devices are driven by a quadruplex fly-by-wire control system — and has flown at speeds of less than 100km/hr and more than Mach 2.2. The use of wing/body blending and fuselage area-ruling is apparent. (Dassault-Breguet)
{57} arising from the limited trimming power of the elevons, this means that trim drag is high. An up-elevon deflection is needed to produce the restoring download.
Restricted CG range was a problem with the Convair B-58, in which fuel had to be transferred internally in order to maintain the CG near the aft limit throughout the flight, partly to minimise supersonic trim drag. The elevon deflection required for rotation and lift-off at maximum weight was less than –5°, due to the aft CG and the low-slung engines. In supersonic flight at altitude, however, the elevons were never at less than –5°.
In Mach 3 cruise the Lockheed SR-71 minimises its trim drag by pumping fuel aft to shift its centre of gravity. In addition, the forebody chines (see Chapter 4) locate the aerodynamic centre well forward.
5 The combination of high wing sweep and the large AOAs necessary for high lift means that the effective dihedral (i.e. rolling moment due to sideslip) can be excessively high at low speed. Thus if a small dihedral effect is achieved at high speed, then the effective dihedral at low speed or high lift, or even high speed at high altitude, disturbs the desired relationship between lateral and directional stability. Dutch roll may then become exaggerated, requiring low-mounted wings and yaw dampers to move the rudder in opposition to the yawing motion. The YF-12, with its 60° leading-edge sweep and high dihedral effect, has approximately zero wing lift when its nosewheel is in contact with the runway during take-off, in order to minimise cross wind e fleets.
6 Longitudinal (i.e. pitch) damping cannot be as high as for a conventional tailled configuration. Typical resulting problems include the pilot-induced oscillations suffered by the Saab Draken, ultimately remedied by the use of automatic pitch dampers.
Although the addition of a tailplane to a delta would appear to improve longitudinal control and pitch damping, it is in fact the tailplane's vertical position which is all-important. A high-mounted tailplane may improve matters only within the low-AOA pitching envelope, which isn't that troublesome anyway. But at high AOA the separated flow from the wing may actually make a high tailplane destabilising. Aeroelastic problems can also arise from mounting a tailplane on top of a very thin {58} vertical stabiliser. For these reasons all the successful tailled deltas — outstanding examples are the MiG-21 and Su-21 — have featured low-mounted tailplanes.