Locked-Mode Detection and Control

A locked mode is a non-rotating magnetic island that often precedes a disruption; detecting and unlocking it is a priority for the off-normal layer.

What locking means

A magnetic island in a rotating plasma normally rotates with it, which limits its growth. If the island's rotation slows and stops, usually by coupling to a static error field or the conducting wall, it is said to lock. A locked mode grows unimpeded and is one of the most reliable precursors of a disruption.

Detection

Because a locked mode is stationary, it produces a steady non-axisymmetric magnetic signal that saddle coils around the vessel can detect directly. A rising locked-mode amplitude is a strong, low-latency warning. It is one of the most trusted disruption precursors precisely because it is simple and direct to measure.

Preventing locking

The best control is to prevent locking in the first place. Maintaining plasma rotation resists locking, and error-field correction removes the static field that a rotating island would otherwise lock to. Keeping rational surfaces free of large islands through tearing-mode control also reduces the chance a mode grows large enough to lock.

Responding to a locked mode

Once locked, options narrow. Targeted current drive can attempt to shrink the island, and briefly applied rotating fields can try to unlock it. If the mode continues to grow, the gatekeeper commands a controlled ramp-down before it disrupts. Because locking often gives limited warning, fast, decisive response is essential.

In the Kronos program

For the Hyperion breeder, locked-mode amplitude is a primary entry in the exception matrix, tied to a fast response on the avoidance-first ladder. Prevention leans on rotation control and error-field correction, especially at low density during ramp-up. The detection and response are exercised against injected locked modes in the flight simulator.