Davis-Besse Leaked Reactor Cooling Water, Again

June 16, 2016 | 6:00 am
Dave Lochbaum
Former Contributor

The Davis-Besse nuclear plant in Oak Harbor, Ohio experienced more than its share of borated water leaking from the reactor coolant system and leaving boric acid behind.

Fig. 1 (Source: NRC)

Fig. 1 (Source: NRC)

The Bad Leak

On August 6, 1999, the NRC sanctioned the plant’s owner for leakage that corroded parts of the pressurizer spray valve. Davis-Besse is a pressurized water reactor. Water flowing through the reactor core during operation is heated to over 550°F, but prevented from boiling by pressure of over 2,000 pounds per square inch. The pressurizer is a large metal tank partially filled with water connected to the piping between the reactor vessel and steam generator. The pressurizer accommodates the expansion and contraction of water within the reactor coolant system as it heats up and cools down. The pressurizer also allows the operators to control the pressure of the reactor coolant system water by turning on electric coils in the pressurizer’s bottom portion or by opening the spray valve to inject cool water into the tank.

Borated water leaking from the spray valve evaporated after it leaked, leaving behind boric acid. The boric acid corroded away some of the parts holding the valve together. Despite several opportunities to correct this deteriorating condition, the leakage was tolerated for awhile. The NRC reported that the corroded valve could have broken apart, creating a pathway that could not be stopped for reactor cooling water to drain out. The NRC determined that the violation normally warranted a $55,000 fine, but waived the fine in this case due to the extensive corrective actions taken by the owner.

The Worse Leak

On April 21, 2005, the NRC fined the plant’s owner a current record of $5.45 million for leakage that corroded a football-sized hole in the reactor vessel’s head. The water leaked through a small crack in one of the 4-inch diameter metal tubes passing through the head to allow the control rods within the reactor vessel to be connected to their electric motors mounted on a platform above the head.

Despite several opportunities to correct this deteriorating condition, the leakage and corrosion was tolerated for an estimated six years until it was finally noticed in March 2002. The NRC reported that the corrosion could have caused the head to break open, creating a pathway that could not be stopped for reactor cooling water to leave the system. In addition, the corrosion could have enabled a control rod to be ejected from the reactor core through the opening.

The Latest Leak

On March 30, 2016, workers found boric acid residue on a vent line from a reactor coolant pump during an inspection during a refueling outage. Davis-Besse has two steam generators and each steam generator has two reactor coolant pumps. Each pump sends about 90,000 gallons per minute from the steam generator back to the reactor vessel.

An estimated half teaspoon of dry boric acid was found on a vent line. This specific vent line had been installed during the prior refueling outage in April 2014 to replace vent lines that had worn out due to vibrations caused by the pump’s operation. Workers shipped the section of the failed vent line to a laboratory for examination, since no hole or crack was visible. Laboratory analysis could not identify a pathway for the borated water to have leaked out, but found crack indications in a weld that was the likely suspect.

The replacement vent line had been examined using liquid penetrant (i.e., applying a colored dye to the surface, waiting a short while, wiping away the excess dye, and examining the surface for signs that dye had been sucked into tiny cracks by capillary action) and by pressuring the line to a pressure 50% higher than it would experience during reactor operation and monitoring that pressure did not drop for ten minutes.

The plant’s owner determined that these testing methods were not adequate to detect small cracks like the one that caused this leakage. So, they applied two more intrusive test methods to the three spare vent lines purchased at the same time as the line that failed. Two spare vent lines passed both tests; one vent line failed the helium tracer probe test. (Like anyone who bought helium balloons for a party or event learned, helium gas permeates through miniscule paths. Workers filled the vent line with helium gas and monitored the outside for signs that the gas escaped via tiny cracks.)

Workers inspected the vent lines on the reactor coolant pumps to determine if they had similar flaws. No additional problems were found.

Workers also revised the procurement procedures for vent lines to require the stringent helium tracer probe test to verify that vent lines obtained in the future do not have small imperfections.

The Past Has Passed

William Shakespeare wrote in The Tempest that “what’s past is prologue.” As with much of what Willie wrote, I have no real clue what this phrase means. But I’ve read other writers use the phrase to imply that what happened yesterday explains what happens now.

If so, that might explain why workers failed to extract the lessons from the pressurizer spray valve degradation in 1999 needed to detect the reactor vessel head degradation before 2002. By only fixing the pressurizer spray valve and not fixing the broader, programmatic problems that enabled it, workers were unable to avoid degradation from recurring.

If so, that might also explain why workers reacted so thoroughly to the discovery of a very small leak from the reactor coolant pump vent line in 2016. They replaced the faulty vent line. They verified that the other vent lines installed at the plant lacked similar faults. And they revised the purchase specification to lessen the chances that future vent lines will be installed with very minor imperfections.

When I came across the report of the recent leak in NRC’s ADAMS library, my initial reaction was “not again.” After reading the report, my reaction remains “not again” but in a good sense. Past reports revealed that leakage of borated water had not evoked the broader response necessary to fix the real cause. Too often, the response had been narrowly confined to the latest symptom of that problem, sustaining the cause and enabling it to manifest problems elsewhere.

It took more rehearsals than it should have to get to this solid response to borated water leakage. The past is passed. The recent commendable response to leakage of borated water at Davis-Besse suggests that there’s not a nasty surprise lurking in the future in the form of boric acid corrosion being tolerated too long before it causes an intolerable result.