Editing Lockheed F-104 Starfighter (section)

===Early problems===

The F-104 series all had a very high [[wing loading]] (made even higher when carrying external stores). During the early stall tests, the aircraft demonstrated the tendency to suddenly "pitch up" once it reached an angle of attack of approximately 15 degrees. This "pitch up" would result in a rapid increase in angle of attack to approximately 60 degrees, accompanied by lateral and directional oscillation, and followed by sudden uncontrolled [[Flight dynamics (fixed-wing aircraft)|yaw and roll]]. At this point the aircraft would be essentially tumbling, descending at a rate of {{convert|12000|–|15000|ft/min}}.<ref name="Upton2003p39">Upton 2003, p. 39.</ref> To combat this, an automatic pitch control (APC) was added, which initiated corrective action at the proper time to prevent reaching an angle of attack high enough to cause pitch-up under any operating condition. The high angle of attack area of flight was protected by a [[stick shaker]] system to warn the pilot of an approaching stall, and if this was ignored, a [[stick pusher]] system would pitch the aircraft's nose down to a safer angle of attack; this was often overridden by the pilot despite flight manual warnings against this practice.<ref name="Bashow1986p20">Bashow 1986, p. 20.</ref>

[[File:F-104A flight envelope.jpg|thumb|alt=F-104A flight envelope diagram|F-104A [[flight envelope]]]]

Another serious design issue that the aircraft encountered was T-tail flutter. Dick Heppe, who served as the initial project aerodynamics engineer for the F-104 program, recalled that "without question, the single most difficult technical challenge encountered in the XF-104 and F-104A development programs was the catastrophic flutter problem of the unique T-tailed empennage configuration". Because the horizontal tail's center of gravity was well aft of both the bending and torsional axes of the vertical tail to which it was attached, it was highly vulnerable to flutter at transonic speeds. During a test flight of the F-104A, the surviving XF-104 was flying chase to observe, and encountered violent T-tail flutter that caused the tail to separate from the aircraft. The test pilot of the XF-104 successfully ejected. This behavior was remedied by increasing the thickness of the stainless-steel box covers used in the tail assembly.<ref name="Upton2003p45">Upton 2003, p. 45.</ref>

The J79 was a brand-new engine, with development continuing throughout the YF-104A test phase and during service with the F-104A. The engine featured variable-incidence [[axial compressor|compressor stator]] blades, a design feature that altered the angle of the stator blades automatically with altitude and temperature.<ref name="Bowman2000p31"/> A condition known as "T-2 reset", a normal function that made large stator blade angle changes, caused several engine failures on takeoff. It was discovered that large and sudden temperature changes (e.g., from being parked in the sun prior to becoming airborne) were falsely causing the engine stator blades to close and choke the compressor. The dangers presented by these engine failures were compounded by the downward ejection seat, which gave the pilot little chance of a safe exit at low level. The engine systems were subsequently modified and the ejection seat changed to the more conventional upward type.<ref name="Pace1992pp66,68">Pace 1992, pp. 66, 68.</ref><ref name="Spitzmiller2011p153">Spitzmiller 2011, p. 153.</ref>

Uncontrolled [[oscillation]]s of its wingtip-mounted fuel tank sheared one wing off of an F-104B; this problem was apparent during testing of the XF-104 prototype and was eventually resolved by filling the tank compartments in a specific order.<ref name="Drendel1976p22">Drendel 1976, p. 22.</ref>

Early F-104s also lacked modulated afterburning, and as a result combat operations could only be performed at one of only two settings{{snd}} either maximum military power or full afterburner. Effectively this gave the pilot the choice of two level-flight speeds: Mach&nbsp;1 or Mach&nbsp;2.2.<ref name="Donald2003p142"/>