Measuring the beamlet divergence in NBI systems for fusion: A diagnostic comparison

Abstract

The ITER Neutral Beam Injection (NBI) system is based on negative hydrogen ions, which are extracted and subsequently accelerated in 5 steps to the full beam energy of 1 MeV by a 7 grid system. A low beamlet divergence is crucial for the efficiency of the ITER-NBI systems, since it affects the transmission of the beam through the duct. The divergence describes the width of the angular distribution, but in practice is often diagnosed from the spatial profile, which requires assumptions on the propagation behaviour of the beam. The RF-driven BATMAN Upgrade testbed, hosted at the Max Planck Institute for Plasma Physics in Garching Germany (IPP), is 1/8 of the ITER source size, and has a 3 grid system that extracts and accelerates the beamlet particles up to 45 keV. In a joint campaign between IPP and Consorzio RFX, an Allison type emittance scanner was operated at BATMAN Upgrade. The emittance scanner, which was constructed and operated by Consorzio RFX, measures the spatial and angular distribution in one direction as well as correlations between them. The direct measurements of the divergence from the emittance scanner are compared to spatially resolved thermographic measurements of the beam footprint on a carbon fibre composite tile. Comparison between the two methods shows the error margin in the conversion from the spatial profile to a divergence. The combined spatial and angular distribution measured by the emittance scanner yields the Twiss parameters, which not only describe the spatial size and divergence, but also the beamlet propagation behaviour. The propagation behaviour provides insight into the limitations of a divergence estimation from a single spatial measurements close to the grid system, and provides input for the design of future diagnostics.