Velocity-space analysis of fast-ion losses measured in MAST-U using a high-speed camera in the FILD detector
A Fast-Ion Loss Detector (FILD) was installed for the first time at
the MAST-U spherical tokamak during its upgrade in 2021. A CMOS camera
was installed in the MAST-U FILD acquisition system to provide high spatial
resolution (1.1 MPx) to infer the velocity space of the fast-ion losses. This camera
allows up to 3.5 kHz acquisition frequency, making it possible to capture the
temporal variation in velocity-space over a wide range of time scales. In this
manuscript, the diagnostic commissioning and the validation of the first velocityspace
measurements using this high-speed camera are presented. The former
starts with the analysis of the main parameters that determine the FILD signal:
the radial position of the probe and the effect of the edge safety factor (q95) on
the measurements. The orbit-following code ASCOT predicts an inverse relation
between the FILD signal and the probe’s relative distance to the separatrix. This
prediction has been validated experimentally by scanning both the outer radius
of the plasma and the FILD radial position, enabling the measurement of fast-ion
losses in the flat-top phase of the discharge. Furthermore, ASCOT simulations
show a big impact of the q95 on the toroidal deposition of the first-orbit losses,
indicating that the signal in FILD can be maximised by running scenarios with
|q95| < 5. This prediction was validated by a scan in the toroidal magnetic
field that was performed experimentally. Prior to the velocity-space analysis, the
experimental resolution of the MAST-U FILD detector has been evaluated, and
shown to be in the order of 0.5 to 1 degree in pitch angle, and of 1 to 3 cm in
gyroradius, both depending on the range of pitch and gyroradius measured. The
analysis of the velocity-space of the losses shows that the gyroradii of the promptlosses
measured match those expected for the NBI injection energies within the
resolution of the diagnostic. The experimentally measured pitch angles have been
compared with ASCOT simulations, and it has been found that the agreement
is better for scenarios heated with the SS beam. This analysis has been applied
to the first discharge where type-III ELM-induced fast-ion losses were measured,
showing that the ELMs result in an increase in the FILD signal, and that the
losses are coming from passing orbits.