Abstract

A moving wavelength particle in cell computer code has been used to study the interaction in emitting sole crossed-field amplifiers (CFAs). Results of the study of magnetrons using a related code have been published previously. The CFA results echo many of the features of the magnetron results. Application of the code to a cold cathode, pulsed CFA showed the existence of chaotic fluctuations in the space charge which we believe are related to noise in experimental CFAs. A major feature of these fluctuations is voids in the charge above the cathode which propagate slower than the RF wave (and, hence, travel backward in the moving reference frame). As the voids pass beneath the base of the spokes they modulate the amount of charge entering the spokes and enhance the chaotic behavior. The spokes recirculate from output to input becoming partially debunched in the process. Differing amounts and distributions of the space charge recirculated result in the computed power and phase pushing varying from pass-to-pass of the moving wavelength around the CFA anode. The fluctuations in field at the cathode and the pass-to-pass fluctuations are used as a measure of the fluctuation of the simulation. Two alternative designs have been investigated in an attempt to reduce CFA noise. A nonreentrant design eliminates the pass-to-pass fluctuations. However, the simulations show internal fluctuations commencing shortly after the space charge has built up. Experimentally, noise of this CFA was improved over the baseline design by from 4 to 8 dB depending on the operating frequency. Simulation of an alternate reentrant design with a thermionic cathode capable of a current density which will support a Slater solution of the sheath trajectories, showed elimination of the internal space-charge fluctuations. Experimentally, noise of this CFA was improved by 10-30 dB over that of the baseline CFA. Fluctuations observed in the simulations are believed to be a qualitative model of noise generation phenomenon and may help point the way to future reduction of such noise.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call