This dissertation utilized time-frequency analysis (TFA) to study high power microwave generation (20–40 MW) on large orbit, axis encircling, coaxial, gyrotrons. The Michigan Electron Long Beam Accelerator (MELBA) was the power source, with e-beam parameters: V = −0.7 to −1.0 MV, Idiode = 1–10 kA, Itube = 0.1–3 kA, and e-beam pulse length = 0.2–0.6 μs. The issue of greatest concern is the identification of some causes of microwave pulse shortening, that is, the duration of the microwave pulse being shorter than the duration of the cathode voltage. For years researchers have applied the Fourier transform to heterodyned microwave signals to identify frequency components of the experimental data. The problem with this technique is that a signal with multiple frequency components can have the same spectrum as that of a signal with frequency components emitted at different times. Time-frequency analysis (TFA) methods introduced to this field in this dissertation provide an entirely different outlook in the community when it was recently applied to heterodyned high power microwave signals. Results show, with unprecedented clarity, mode hopping, mode competition, and frequency modulation due to electron beam voltage fluctuations. The various processes that lead to pulse shortening for the coaxial gyrotron experiment and perhaps to other high power microwave devices may finally be identified. Time resolved maximum intensity of the TFA has produced results very similar to the microwave power signal, verifying the utility of TFA in the analysis of the temporal evolution of power in each mode.; Time-frequency analysis utilizes the Fourier transform of the local autocorrelation function, resulting in a spectrum with a large variance. Spectral variance has been reduced by applying discrete prolate spherical sequences (DPSS). The main advantages of DPSS lie in the wide main lobe and small secondary lobes, placing spectral energy around the actual frequency while minimizing spectral leakage.
展开▼
