Self-calibrating magnetic field diagnostics in beam emission spectroscopy

Magnetic field diagnostics in tokamaks using the motional Stark effect in fast neutral beams have been based on two kinds of polarimetry which we call ‘‘static’’ and ‘‘dynamic.’’ A detailed analysis shows that static polarimetry presents a number of advantages over dynamic polarimetry, provided it is made complete in the sense that a sufficient number of polarization analyzers are installed and different parts of the spectrum are explored to yield full information on the set of unknowns inherent in the problem. A detailed scheme of complete static polarimetry is proposed, including the case where an in-vessel mirror with changing characteristics (coating by impurities) is placed in front of the optical detection system. The main merit of this scheme relies on the fact that it is self-calibrating with respect to both the characteristics of the mirror and the transmission of the different polarization channels, the latter item implying that it is uniquely based on relative measurements of spectra. Further advantages are a greater flexibility with regard to different kinds of diagnostics and the circumstance that the technical equipment is less involved. The above scheme is based on a detection system of moderate etendue exploiting a large spectral domain, which is the regime where static polarimetry usually operates. It is also possible, however, to work with large etendue and a small spectral domain, such as commonly adopted in dynamic polarimetry. Using such a regime, static polarimetry loses the advantages mentioned above but gains, as a new advantage, the benefit of a comparatively lower level of photon noise.