DEC Fault Ride-Through

L1 isolates a faulted collector stage within the deterministic window so the direct-conversion train keeps running through single-stage upsets.

Ride through, don't trip out

A collector-grid arc, a supply glitch, or a transient overvoltage on one DEC stage should not stop the whole conversion train. L1 implements fault ride-through: detect the stage fault, isolate it, and keep the remaining stages converting, then attempt controlled re-insertion once the fault clears. This preserves operation through disturbances that would otherwise force a full shutdown.

Detection and isolation

Each stage is monitored for fault signatures — arc current, voltage collapse, rate-of-change limits. On detection, L1 removes drive from the affected stage and reconfigures the grid schedule so the remaining stages absorb the redistributed load without exceeding their limits. The whole sequence completes inside the fast window so the fault cannot cascade to neighboring stages or back into the plasma loss channel.

Why determinism matters here

• An arc evolves fast; late isolation can damage the stage.
• Load redistribution must respect the surviving stages' envelopes.
• Re-insertion is gated on verified fault clearance.
• The plasma-side coupling must stay bounded during the transient.

Ride-through is protection that preserves function, the DEC analogue of breeder beam notching: a fast, reversible action that contains a fault while keeping the machine productive. It sits at protection priority in arbitration for the grid actuators.

Coordination with conversion

After isolation, the modulation loop re-optimizes the reduced grid set for the current spectrum, and the twin's power module tracks the degraded-but-operating state. If faults accumulate beyond a safe threshold, L1 escalates to a controlled burner power reduction rather than an abrupt trip, keeping the machine within safe bounds throughout.