The active control of plasma profiles, such as pressure and current density, is key to the realisation of steady state operation on ITER and is an important area of research on JET. Control requires accurate measurements and, recently, the measurement of the plasma current density on JET has been improved significantly. The Motional Stark Effect (MSE) diagnostic provides a local measurement of the pitch of the magnetic field and thus the safety factor (q). This is used as a constraint on the calculation of the magnetic geometry and the determination of the current profile. The diagnostic (fig. 1) is based on the Stark splitting of the Dα line excited by collisions between the fast neutral particles injected by the Neutral Beams used for heating the plasma, and thermal plasma ions. In the presence of a magnetic field B the Lorentz field E = v x B produces a separation of the Dα line into components polarised parallel (ρ) and perpendicular (ς) to the E field. By measuring the polarisation angle of one of these components, together with information from magnetic sensors, it is possible to determine the safety factor q.

Safety factor q

Number of turns the helical magnetic field lines in a tokamak make round the major circumference for each turn round the minor circumference, denoted by q. The use of the term safety factor refers to magnetic plasma stability and not to the safe operation of a tokamak.

During the Shutdown of 2001 the Neutral Beam source (PINI #1) used by MSE was upgraded in energy from 80 kV to 130 kV. As a result there is an increased Doppler shift of the Stark emission, which is thus spectrally resolved from that due to the other sources whose neutral beams intersect the detection line-of-sight at different angles. Another beneficial effect is the separation from interfering CII spectral lines in the outer edge of the plasma.

Experimental evidence of the enhanced performance especially at high power when all neutral beam sources are used is shown in fig. 2.a/b. Prior to the upgrade all detection channels were perturbed by about 7 degrees (fig. 2a) when one of the sources (PINI#7, not used for MSE) was also operated. Now, this effect is reduced to less than half a degree (fig. 2b) with smaller interference at higher toroidal magnetic field, compared with a statistical error of 0.1 degree.

A further use of this diagnostic is for the measurement of the radial electric field Er in the plasma. The influence of Er can become significant when plasma rotation is large and pressure gradients are steep, as in high performance discharges. A measurement of Er is also important for testing theories of improved plasma confinement mechanisms. On the other hand, the measurement of Er in the core of a hot (100 Million degrees), large (100 m3) plasma is extremely challenging. Recently, an estimate of Er has been obtained in JET by measuring the change in the polarisation angle of the MSE emission when two different sources are operated sequentially at different energies. The first profile measurements have been made and further experiments planned for 2003 aim at improving the technique.