12 Things That Went Wrong To Cause The Three Mile Island Nuclear Meltdown

Construction on nuclear reactor Unit One (also called TMI-1) on Three Mile Island began in 1968 and went off without a hitch. On the other hand, Unit Two (TMI-2, built between 1969 and 1978) was plagued with issues. Not only were there delays in the construction process (which should have only taken six years, not nine), but once it started operating, it suffered a series of unscheduled shutdowns.

The workers operating Unit Two found themselves falsifying data that went to the Nuclear Regulatory Commission, to prevent the commission from simply shutting the reactor down for good. 

On March 28, 1979, the pump that sent hot water to the reactor's steam generator failed for reasons still not entirely understood (though it may have been the fault of a valve stuck closed, or water that leaked into an air line). In response, the plant's turbine immediately shut down, expelling a huge plume of high-pressure steam.

Without the steam generator to feed it water, the feedwater pump in Unit Two also quit working. This particular pump was designed to provide a non-stop flow of water to the steam generators that kept the nuclear core of the plant cool.

As long as the nuclear core is kept cool, it works just fine. However, when it's allowed to heat up, a catastrophic chain of events begins, resulting in the nuclear core going "critical" and melting down. The result is a release of nuclear energy that spreads throughout the atmosphere. When the pump stopped working, the plant went into emergency shut-down mode, and the chances of a meltdown at Three Mile Island increased.

With the whole system shutting down, the pressure inside the reactor began to fall. Usually, this sudden drop in pressure would trigger the relief valve that was venting the steam to close again. This time, though, the valve was stuck open, so steam kept right on pouring out of the plant, leaving the water levels lower and lower.

There were backup measures in place at Three Mile Island in case the feedwater pump malfunctioned. When the main feedwater pump failed, an emergency backup pump should have automatically turned on. However, the emergency pump's relief valve was stuck in a closed position.

A few days prior to the meltdown, tests were done on the backup pump and it was found to be in good condition. However, in order to conduct those tests, the valve had to be manually opened and closed. When the valve is properly open, the backup pump kicks into gear. But since the valve was stuck in a closed position, the backup pump did not turn on, and any water near the core kept venting out, making the core even hotter. 

The instrumentation in the control room failed to alert the operators of the growing problem. They could tell that the main pump had stopped, but believed that the backup pump was working properly. It didn't help that in the past, the emergency pumps had turned on in that reactor for no reason at all, so workers were conditioned to assume it wasn't anything serious. Even though the valve venting steam was stuck open, the control panel's confusing readings seemed to indicate it was closed. The operators had no idea that cooling water was still pouring out in the form of steam. There also wasn't any way to see how much water surrounded the nuclear core from the control room.