Error Field Correction

Tiny asymmetries in the magnetic field can lock modes and disrupt the plasma; correction coils cancel them under feedback.

What an error field is

An ideal tokamak field is perfectly axisymmetric, the same all the way around the torus. Real coils are slightly misaligned and real structures are slightly asymmetric, producing small non-axisymmetric error fields. Even a very small error field can penetrate the plasma, brake its rotation, and lock a mode, leading to a disruption, especially at low density.

Correction coils

Dedicated correction coils apply a controllable non-axisymmetric field designed to cancel the error field. The correction has both an amplitude and a phase, since it must oppose the specific error field present. Getting the phase wrong makes the situation worse, so correction is tuned carefully.

Finding the correction

The optimal correction is found experimentally by scanning the applied field and watching for the plasma response, or estimated from a model of the machine's known asymmetries. In operation, a feedback loop can trim the correction using the plasma's own response, for example the onset of braking, as the error signal.

Shared coils

The same coils used for error-field correction are often used for edge-localized-mode control and rotation braking. These uses can conflict, since one wants to cancel non-axisymmetry and another wants to add it. Actuator allocation resolves the conflict according to which need is more pressing in the current discharge phase.

In the Kronos program

The Hyperion breeder includes error-field correction to keep rotation up and prevent mode locking, particularly during low-density phases such as early ramp-up. Its correction coils are shared with edge control, so the supervisor arbitrates their use. The required correction is estimated from the machine model in simulation ahead of operation.