[en] The gas flow characteristics of a novel geometry (pumped neutralizer) for decreasing the flow of gas from neutral beam neutralizers were measured and compared with a conventional (passive) neutralizer. A passive neutralizer is typically a duct attached to the ion source. For the pumped neutralizer the top and bottom surfaces of the duct are replaced by a Venetian blind geometry which opens into ball as vacuum pumping volumes. With guidance from a Monte Carlo program which models gas flow at low pressure, a one-half scale model with pumped neutralizer geometry was built and compared to a passive neutralizer with comparable dimension. With the vanes on the pumped neutralizer opened to 55 degrees, the line density of the pumped neutralizer was 1.6 times less than the passive neutralizer. The amount of gas flowing from the exit of the pumped neutralizer was from 2 to 5 times less than the amount flowing from the pumped neutralizer. Hence,the pumped neutralizer geometry appears to be a promising method of limiting the flow of gas from neutral beam gas cell neutralizers

No abstract available

[en] This final report summarizes the design and performance of the VUW-8028 Pierce-Wiggler electron beam systems, which can be used to power high frequency gyro-BWO's. The operator's manual for this gyro-BWO beamstick is included as appendix A. Researchers at Lawrence Livermore National Laboratory (LLNL) are developing a gyro-BWO with a center frequency of 250 GHz, 6% bandwidth, and 10 kV peak output power. The gyro-BWO will be used to drive a free electron laser amplifier at LLNL. The electron beam requirements of the gyro-BWO application are: Small beam size, .100 inch at 2500 gauss axial magnetic field; a large fraction of the electron energy in rotational velocity; ability to vary the electrons' axial velocity easily, for electronic tuning; and low velocity spread i.e. little variation in the axial velocities of the electrons in the interaction region. 1 ref., 13 figs

[en] The conductance and line density of a novel geometry (pumped geometry) for neutral beam neutralizers were measured and compared with a conventional (passive) neutralizer geometry in the transition flow regime. A passive neutralizer geometry is merely a duct attached to the ion source. For the pumped geometry, the top and bottom surfaces of the duct are replaced by a venetian blind geometry which opens into ballast vacuum pumping volumes. A one-half scale model with pumped neutralizer geometry was built and compared with a passive neutralizer geometry of comparable dimensions. A gas source was maintained at fixed pressure and the conductance and line density of the two geometries were measured. These measurements were repeated for several values of gas source pressure, covering the range of pressures typically used in neutral beam ion sources. With the vanes on the pumped neutralizer geometry opened to 55 deg and the gas source pressure ''low,'' i.e., in the range of pressures typical for ''bucket-type'' magnetic field ion sources, the amount of gas flowing from the exit of the pumped geometry was about 3 times less than the amount flowing from the passive geometry. As the source gas pressure was increased, the ratio of gas flow from the exit of the passive geometry to that of the pumped geometry increased also. At gas source pressures typical of magnetic field free ion sources, up to 5 times less gas flowed from the pumped geometry. On the other hand, the neutral gas line density of the pumped geometry was only 1.6 times less than that of the passive geometry, independent of the source gas pressure

[en] A simple method for fabricating filament chucks for multifilament ion sources is described

[en] A simple method for fabricating spiral filaments for multifilament ion sources is described

[en] The deflection and increased divergence of a neutral beam due to a magnetic field perpendicular to the gas-cell neutralizer has been measured. The hydrogen neutral beam energy was 5.0 keV. The observed neutral beam deflection and divergence were in good agreement with results obtained from solutions of a coupled set of kinetic equations for the transverse velocity distribution functions for energetic ions and neutrals in the neutralizer volume

[en] Recent modifications of the neutral beam injection system on the Phaedrus Tandem Mirror Experiment have produced substantial improvements in the injection performance. The modifications consisted primarily of more tightly apertured baffles (smaller by a factor of two) in the beamline vacuum chamber to reduce the beamline cold gas neutral density incident on the plug. At the same time, the neutral beam energy was increased from 7 keV to 8.8 keV. The decreased neutral gas load leads to decreased charge exchange losses of trapped hot ions in the plug. The system parameters before and after the modifications are summarized in Table I. With the smaller aperature baffles, the plug neutral gas density from the neutral beam sources decreases by a factor of two, the target diamagnetism immediately before injection increases by almost a factor of three, and the increase in plug diamagnetism due to the beam injection doubles. The most significant improvement is the increase in plug line density; the beamline modifications have allowed the first unequivocal measurement of line density buildup ( about 20% increase) on Phaedrus. Experimental evidence to date indicates that further improvement in performance may be obtained at still higher beam energies. Code calculations suggest that substantial improvement will be obtained by switching from hydrogen injection to deuterium injection. Most recent data are presented

[en] Until recently, neutral beam injection experiments on Phaedrus have been dominated by cold neutral gas streaming from the neutral beam sources. Charge exchange loss of hot plasma ions due to this cold gas source has limited beam induced diamagnetism and density buildup of the RF sustained target plasma to approximately 50% and 10% respectively. Major modifications of the neutral beam line vacuum system (in the form of additional and tighter baffles, and extension tubes which increase the distance between the sources and the plasma) were made in September, 1983. These modifications reduced the flow of neutral gas into the plug due to beamline gas by an order of magnitude, in agreement with Monte Carlo calculations. After the modifications were accomplished, it was discovered that the RF target plasma itself was characterized by a very high neutral pressure (edge pressure ≅ 5 x 10/sup 11/ cm/sup -3/). The origin of the higher neutral pressure problem, a proposed solution and the impact of the proposed solution on neutral beam injection are discussed