Skip to content
From the High Desert book cover

From the High Desert

A Cultural History of Art Bell

Thumbnail for January 28, 1997: Three Mile Island Disaster - Scott Portzline

January 28, 1997: Three Mile Island Disaster - Scott Portzline

Jan 28, 1997
2h 37m
0:00 / 0:00
Scott Portzline, a researcher who spent 13 years investigating the Three Mile Island nuclear incident and reviewed over 30,000 pages of documents, joins Art Bell to present evidence suggesting the 1979 disaster may not have been an accident. Portzline details how emergency feedwater valves were found inexplicably closed, critical paperwork went missing, and a local cult had warned members to leave Harrisburg before April 1979, with a newspaper editorial specifically urging Three Mile Island to increase security.

The technical chronology of the disaster reveals how close Pennsylvania came to catastrophe. Fuel melted twice inside the reactor, temperatures exceeded 5,000 degrees, and the Rogovin Commission concluded the state was within 30 minutes of a release that could have killed 130,000 people and rendered an area the size of Pennsylvania uninhabitable. Portzline describes how the FBI opened and then denied a sabotage investigation within 24 hours.

The conversation shifts to present-day vulnerabilities at nuclear plants nationwide, including vehicle barriers that remain open half the day, spent fuel pools accessible to truck bombs, and 149 of 150 known sabotage incidents committed by insiders. Portzline's testimony before Congress and the NRC exposed security gaps that remain deeply unsettling nearly two decades after the original disaster.

Key Moments

  1. March 28, 1979 - 4 a.m. turbine trip and stuck PORV: Portzline gives the technical chronology: at 4 a.m. Wednesday, March 28, 1979, the turbine tripped after a blockage in the condensate polisher (filter) system the operators had been fighting for the prior 12 hours. Eight seconds later the reactor scrammed; the pilot-operated relief valve atop the reactor opened to vent steam and radioactive coolant - and failed to shut, beginning a loss-of-coolant accident with GPU and Metropolitan Edison publicly downplaying it as a 'minor incident.'

  2. Emergency feedwater valves were closed - manually: Portzline identifies the specific failure that escalated the accident: emergency feedwater valves required to be open were found closed, with no automatic or control-panel cause - investigators concluded a human had shut them. They went undiscovered for eight minutes until a worker arriving early happened to walk to the panel, opened them, and never reported it. A test two days prior had closed the valves; the paperwork that would prove they were reopened, Portzline says, was thrown away.

  3. Core melted twice - 50% collapsed to reactor bottom: Portzline says an Idaho laboratory only discovered four or five years before this 1997 broadcast that more than 50% of the TMI Unit 2 core collapsed molten to the bottom of the reactor vessel. Fuel hit 5,000-plus degrees, control rods and fuel rods burst, pellets melted, fell, and re-assembled into a geometry that started a second nuclear chain reaction. Vacuum and steam voids between the molten fuel and the stainless-steel vessel are credited with preventing burn-through to the basement.