A Primer on Nuclear Safety: 1.4.1 Complexity and Three Mile Island

1.4.1 Complexity and Three Mile Island
The Three Mile Island accident was the primary example of what concerned advocates of the probabilistic approach to nuclear safety. A series of improbable events, lead to a partial meltdown of core of the Three Mile Island Unite 2 reactor. The background of these events certainly reflected attitudes in both the NRC and among reactor operators that failed to take nuclear safety concerns with sufficient seriousness. This is evident in The Report of the Presidential Commission on the Accident at Three Mile Island which paints a damning picture of the safety problems at Three Mile Island and the relative indifference of the NRC, the reactor's manufacturer Babcox & Wilcox (B&W) and its operator. The report looks at the problems that the operators of TMI Unit 2 faced on March 28, 1979 and concluded that the operators have been overwealmed by problems which could and should have been foreseen by the reactor's manufacturer, the control room architect, the NRC, and the reactor's operator.   

The Commission Report noted:
5. TMI management and engineering personnel also had
difficulty in analyzing events. Even after supervisory personnel
took charge, significant delays occurred before core damage was
fully recognized, and stable cooling of the core was achieved.
(Translation: The TMI staff did not have the slightest idea what was happening.)

6. Some of the key TMI-2 operating and emergency procedures
in use on March 28 were inadequate, including the procedures for a
LOCA [Loss of Coolant Accident] and for pressurizer operation. Deficiencies in these
procedures could cause operator confusion or incorrect action.
(The were rooted in the plans for dealing with the accident. The staff was confussed by the bad plans they were expected to follow.)

7. Several earlier warnings that operators needed clear instructions for dealing with events like those during the TMI accident had been disregarded by Babcock & Wilcox (B&W) and the Nuclear Regulatory Commission (NRC).
(OK lets take the TMI Staff of the Hook, because the NRC and B&W had been warned about the problems, and ignored the warnings.)
a. In September 1977, an incident occured at the Davis-Besse plant, also equipped with a B&W reactor. During that incident, a PORV stuck open and pressurizer level increased, while
pressure fell. Although there were no serious consequences of that incident, operators had improperly interfered with the HPI, apparently relying on rising pressurizer level. The Davis-Besse plant had been operating at only 9 percent power and the PORV block valve was closed approximately 20 minutes after the PORV stuck open.  That incident was investigated by both B&W and the NRC, but no information calling attention to the correct operator actions was provided to utilities prior to the TMI accident. A B&W engineer had stated in an internal B&W memorandum written more than a year before the TMI accident that if the Davis-Besse event had occurred in a reactor operating at full power, "it is quite possible, perhaps
probable, that core uncovery and possible fuel damage would have occurred."
(There was a dress rehearsal for the Three Mile Island incident at the Davis-Besse nuclear facility a year and a half before the TMI accident and the NRC & B&W knew all about it, but did not warn reactor operators about the problems uncovered.)
b. An NRC official in January 1978 pointed out the likelihood for erroneous operator action in a TMI-type incident.  The NRC did not notify utilities prior to the accident.
(The NRC staff knew that accident management procedures were dangerously flawed but the NRC did not take actions to bring about changes.)
c. A Tennesse Valley Authority (TVA) engineer analyzed the problem of rising pressurizer level and falling pressure more than a year before the accident. His analysis was provided to B&W,
NRC, and the Advisory Committee on Reactor Safeguards. Again no notification was given to utilities prior to the accident.
(Yet another warning had been received by the NRC and B&W with no action taken.)
8. The control room was not adequately designed with the management of an accident in mind. (See also finding G.8.e.) For example:
a. Burns and Roe, the TMI-2 architect-engineer, had never systematically evaluated control room design in the context a serious accident to see how well it would serve in emergency.
(The design of the TMI control room made accident management more difficult. The architects did not consider accident management in control room design.)
b. The information was presented in a manner which could confuse operators:
(i) Over 100 alarms went off in the early stages of the accident with no way of suppressing
the unimportant ones and identifying the important ones. The danger of having too many alarms was recognized by Burns and Roe during the design stage, but the problem was never resolved.
(No wonder the operators were confused during the early stages of the accident with all of those alarms going off. The Architects knew this was going to be a problem, but did not fix it.)

(ii) The arrangement of controls and indicators was not well thought out.  Some key indicators relevant to the accident were on the back of the control panel.
(What a mess. No one could expect frightened, poorly trained operators who lacked adequate procedural guidance and who were confronted with this poorly designed and confusing control panel to manage this accident.)

(iii) Several instruments went off-scale during the course of the accident, depriving the
operators of highly significant diagnostic information. These instruments were not designed
to follow the course of an accident.
(The instrumentation of the reactor was not designed with the possibility of accidents in mind. There failure to provide useful information during the accident was but another evidence that the NRC and B&W were not seriously considering accident management in reactor design.)
(iv) The computer printer registering alarms was running more than 2-k hours behind the events and at one point jammed, thereby losing valuable
information.
(The information system was not designed to provide information during an accident and broke down.)
c. After an April 1978 incident, a TMI-2 control room operator complained to his superiors about problems with the control room. No corrective action was taken by the utility.
(The TMI operators had been warned about the problems with the control room and did not do anything to rectify the problem.)

The Presidential Commission report is serious reading for anyone who is interested in the history of reactor safety.  The emergence of the large light water civilian power reactor posed significant challenges for nuclear safety.  Debate over nuclear safety issues had rocked the nuclear industry during the late 1960's and early 1970's.  AEC reactor research czar Milton Shaw, with the undoubted backing of Admiral Hyman Rickover, and powerful Congressman Chet Hollifeld had insisted that light water reactors represented a mature technology,   Advocates of nuclear safety concerns including Alvin Weinberg were concerned that the complexity of power reactors was creating new and significant safety problems.  This concern was to be justified by the Three Mile Island incident. Although Dixie Lee Ray had out manipulated Milton Shaw at the AEC  6 years previously, Shaw's attitude toward nuclear safety, was behind the problems which the Presidential Commission had documented.