Error-Field Correction

Detecting and cancelling small, unintended non-axisymmetric magnetic fields that can lock modes and trigger disruptions.

Where error fields come from

No magnet system is perfectly axisymmetric. Coil misalignments, joints, feeds, and nearby ferromagnetic material produce small non-axisymmetric field components - error fields. Though tiny compared with the main field, resonant components can penetrate the plasma and seed locked modes, especially at low density where the plasma cannot shield them by rotation.

Why small fields matter

A resonant error field can brake plasma rotation and lock a magnetic island in place. A locked mode grows, degrades confinement, and frequently ends in a disruption. Low-density, low-current phases - such as breakdown and early ramp-up - are especially vulnerable, so error-field correction is most active exactly when the plasma is most fragile.

Correction method

Dedicated correction coils apply a canceling non-axisymmetric field. The amplitude and phase of the applied correction are tuned - often empirically through compensation scans, and in operation through feedback on locked-mode detectors - to minimize the resonant field the plasma sees. The same coils frequently serve resistive-wall-mode control.

Real-time versus pre-set

Some error-field correction is fixed feedforward, calibrated once for the machine's static imperfections. But the optimal correction changes with plasma current, density, and rotation, so a feedback component tracks the plasma's response - typically driving a locked-mode signal toward zero - and adapts the correction through the discharge.

A quiet, essential loop

Error-field correction rarely makes headlines but prevents a large fraction of low-density locked-mode disruptions. It is a good example of a control function whose success is measured by the absence of events rather than any visible action.