Magnetics Diagnostic Interface

Flux loops and Mirnov coils are the fastest, most trusted sensors on the breeder; their interface feeds real-time equilibrium and stability control.

The AI-Native S.M.A.R.T. Generator Master Blueprint — eight layers (L0→L7), one control stack, wired to both machines. Telemetry rises in microseconds; control descends the same path.

The backbone of control

Magnetic diagnostics — flux loops (integrated voltage) and Mirnov coils (dB/dt) — are the breeder's fastest and most reliable measurements. They underpin equilibrium reconstruction, vertical stability, and disruption-precursor detection. Nearly every fast control loop on Hyperion begins here.

Signal handling

python

# flux loop: integrate coil voltage to flux
psi_loop = integral(v_loop, dt) + psi0      # drift-corrected
# Mirnov: dB/dt -> mode analysis
b_theta  = integral(v_mirnov, dt)
modes    = fourier_poloidal(b_theta_array)  # n,m mode content
omega    = track_mode_rotation(modes)       # rotating MHD

Flux loops require careful integration-drift correction; Mirnov coils are analyzed as a poloidal array to extract rotating-mode content whose growth is a disruption precursor. The array structure is what a graph model exploits when a channel drops out.

Owner and timing

Owner: L2 for acquisition/timestamping; consumed by L1 (stability) and L3 (reconstruction).
Timing: fastest diagnostic class, sub-ms, on the deterministic fabric.
Feeds: equilibrium reconstruction, vertical stability, disruption precursors.

Calibration ties each coil's geometry and gain to a reference; because these signals drive protection-adjacent loops, their calibration and health are tracked continuously and a suspect channel is masked before it corrupts the reconstruction. Design-and-simulation specification, validated against the twin. Related existing pages cover Mirnov coils and flux loops in isolation; this interface page focuses on the acquisition contract feeding control.