At 97 percent power a standby service water pump tripped the instant its own diesel generator started at a Mississippi reactor, where that single trip left the emergency core spray system inoperable

A plant room under fluorescent light, concrete floor, a pump the size of a small car sitting on a pad.

It has not run in weeks. That is the point of it.

On a winter evening a crew starts a scheduled test, the kind meant to confirm the equipment works when nobody is asking.

It happened at a boiling water reactor in Claiborne County, Mississippi, operated by the same company that runs a similar unit across the state line in Louisiana.

The diesel behind it starts correctly. The pump trips at once.

Nothing here was broken mechanically.

What followed was electrical.

How one pump trip takes a whole safety train with it

A boiling water reactor of this generation carries three separate divisions of emergency power, each with its own diesel generator and its own switchgear.

The third division exists for one system, a high pressure core spray that floods the vessel from above if coolant is lost and the pressure stays high.

That spray system needs cooling water of its own, which is what the standby service water pump provides.

The pump draws from a single medium voltage bus, and that bus is fed through one supply breaker.

When the pump drew a sudden surge, the protection scheme read it as a short circuit and opened that breaker, which did exactly what a breaker is designed to do.

Opening it isolated everything downstream, including the spray train.

What the third division actually protects against

Core spray is not the first line. It is the one that matters when the others cannot act.

In an accident where coolant escapes but the vessel stays pressurized, low pressure systems cannot push water in against that pressure.

The high pressure spray can, and it does so from above the fuel rather than below, which is why it gets its own division and its own diesel.

Losing it while the reactor runs at power is a defined regulatory condition, not an engineering judgment call, and it starts a clock.

The plant entered a limiting condition for operation and notified federal regulators inside the required window.

A qualified system going down at power is a reportable event rather than an emergency.

The log entry, and which plant it belongs to

The event is recorded in the federal event notification system with a time of 2009 on the first of February this year.

It happened at a boiling water reactor in Claiborne County, Mississippi, operated by the same company that runs a similar unit across the state line in Louisiana.

Those two plants share a reactor generation, a three division layout and a high pressure core spray system, which is precisely how they get confused in coverage.

The reactor was at 97 percent power, not full rated power, and it stayed there throughout.

The notification cited the rule covering an occurrence that could have prevented a safety function, and reported no release, no fire and no injury.

Two Entergy reactors in two different states is a real trap.

What worked, and where the chain could have broken

The emergency diesel generator did its job. It detected the loss of supply and energized the bus within seconds, exactly as the start logic requires.

What the diesel could not restore was a surveillance requirement that the tripped breaker had already broken.

The backup reactor core isolation cooling system was separately verified operable, and all offsite power sources remained available.

The physical chain has three links. A pump fault, a breaker response, and a surveillance gap, and only the first is still unexplained.

A winding fault, a bearing seizure and a grounding problem each carry a different corrective action, and until the mechanism is known the lesson is incomplete.

Aging components are exactly what a decommissioning handover is built to catalogue, as at a Palisades transfer.

The event time, the power level and the systems declared inoperable appear in the event log.

The breaker performed correctly, and correct behavior still removed a safety function.

What an event like this adds to the record

Regulators track this kind of report because it degrades a safety boundary rather than because anything visible happened.

The value is in the record, not the outcome. It shows how the defense in depth layers responded when one of them was pulled.

The plant kept running, the diesel carried the bus, and the second cooling path was confirmed available, which is the design working under a real fault.

It also shows a design dependency worth naming. A bus breaker’s availability and the operable status of a separate safety train were bound together by one supply path.

New builds are reexamining exactly those dependencies, which is part of why small reactors are drawing new customers.

The same day’s notifications and the reporting rules are listed in the preceding log.

The fault surfaced during a test, and that distinction is the whole point of running tests.