Running at 83 percent power on a July evening, a Michigan reactor shut itself down when one locked-out electrical bus silenced both safety circuits at once
Just after 6:30 on July 3, alarms flooded the control room at a nuclear plant on the western shore of Lake Erie.
Every control rod dropped within seconds.
Nothing had failed inside the reactor vessel itself.
The Fermi 2 switchyard connects the plant’s generators to the grid through a pair of station service transformers, each feeding one divisional safety bus.
The chain started in the switchyard, moved through a single transformer, and landed on the circuits whose only job is to shut the reactor down.
How does a fault that far outside the core silence two independent safety systems at once?
One locked bus, two dead circuits
A reactor protection system is the last electronic word on whether a reactor keeps running. At Fermi 2, like most boiling water reactors, two separate buses feed that system independently, so a fault on one side cannot silence the other. The design assumes both sources are always available.
On the evening of July 3, both went dark at the same moment. The NRC event notification describes the sequence plainly: at approximately 6:27 p.m. EDT, Fermi 2 experienced a lockout of bus 101, resulting in the loss of transformer 64, followed by a momentary loss of both reactor protection system bus power sources and a reactor scram.
The word “momentary” is precise but not reassuring. Even a brief simultaneous loss of both protection bus sources is the kind of event the NRC requires reporting within four hours, not four days.
The switchyard hardware at the center of it
The Fermi 2 switchyard connects the plant’s generators to the grid through a pair of station service transformers, each feeding one divisional safety bus. Transformer 64 serves Division I. When bus 101, the 120-kilovolt switching bus that feeds transformer 64, locked out, Division I lost its source instantly.
Under normal circumstances that loss alone would not reach the reactor protection system, because Division II stays energized on its own feed. But the update filed with the NRC four days later clarified the gap: reactor protection system A experienced a loss of power while RPS B remained energized throughout the transient. The scram had already been initiated before that distinction mattered.
Plant documentation describes exactly this vulnerability: the only circumstance that could result in a sustained loss of power to transformer 1 is a fault affecting the 120-kilovolt switchyard bus 101. The July 3 event was precisely that fault.
What the record shows about this transformer
The July 3 reactor scram was not transformer 64’s first appearance in Fermi 2’s event history. An earlier NRC preliminary notification documented a remarkably similar chain: isolation of the same transformer caused Division I to lose power, leading to an automatic scram, and the plant was off the air for several weeks while the transformer was repaired.
The July 3 recovery was faster. The scram was uncomplicated, operators stabilized the plant, and reactor water level was held at normal levels. Decay heat moved through the main steam system via turbine bypass valves, a routine path requiring no emergency equipment.
No members of the public or plant personnel were reported injured. Fermi 2 is a boiling water reactor rated at roughly 1,122 MW, operating since the late 1980s on Lake Erie south of Detroit, and tracking data show it had accumulated eleven scram events since 2011 before this one.
Where the chain is weakest
The event highlights a specific structural tension in aging switchyard equipment. The single 120-kilovolt bus feeding the Division I transformer is the only path to that side of the safety system, and a fault severe enough to lock it out has no automatic bypass. Two alternatives exist for restoring power: isolating the defective breaker and reenergizing the bus, or transferring Division I to an alternate transformer.
Both require time, and time is exactly what a momentary simultaneous loss of both RPS buses does not provide. That tension between electrical redundancy on paper and physical sequencing in practice is something regulators have tracked across the boiling water fleet for years. For a parallel case, the investigation into relay failures at a pair of Texas reactors found a grinding subcontractor’s undersized rotor hiding inside every one of eleven events over three years. Different origin, same structural lesson: the weakest link is often unglamorous switchyard hardware.
What comes next for the plant and its grid
The NRC requires a formal licensee event report within 60 days, detailing root cause and corrective actions. That report will determine whether bus 101’s lockout traced to a failed breaker, a protection relay that operated incorrectly, a line fault, or some combination. The full NRC event notification confirms the sequence and the update filed four days later. Until the licensee event report is complete, the precise initiating cause remains under investigation.
Fermi 2 returned to service after stabilizing in hot shutdown, with decay heat removal proceeding through normal post-scram systems. The Michigan grid absorbed the loss without a public reliability event, a reminder that large interconnected grids are themselves a form of redundancy. For those tracking how the industry handles electrical and safety system design from the ground up, a Texas research reactor reached first criticality in under a year, sidestepping the switchyard vulnerabilities that come with infrastructure built across multiple decades.
A non-complicated scram at a well-maintained boiling water reactor is a routine federal notification, not a safety crisis. The plant’s protective systems performed exactly as designed, and what the July 3 event adds to the public record is a clean, documented example of how a single failure in the electrical yard can reach both branches of a redundant safety system in the same instant.
Hugo is an engineer with strong technical expertise. Multilingual from an early age, his writing combines technical clarity with a strong interest in science and energy.
