A report on the root causes of problems at the San Onofre nuclear power plant shows that officials considered making design changes to its new steam generators before they were installed but rejected some fixes in part because they would require further regulatory approvals.
Some of the generators began to malfunction a year after they were installed, and the plant has been shuttered for 13 months. The closure has already cost San Onofre's owners, Southern California Edison and San Diego Gas and Electric, more than $470 million.
Ratepayers across the region are shouldering some of the plant's costs and could be on the hook for hefty future repair bills.
DOCUMENT: Read the full report
The report, released Friday by the Nuclear Regulatory Commission, provides the most detailed picture to date of how the flawed system was designed. It was written by Mitsubishi Heavy Industries, which built the generators.
Both companies insisted Friday that they were not aware of the problems that crippled San Onofre when the generators were installed. Mitsubishi said that the design changes officials contemplated would not have made a major difference.
Mitsubishi, however, acknowledged in the report that it made an incorrect input into a computer code that resulted in underestimating the velocity of steam flow in the plant's replacement steam generators. Again, the company asserted that error did not cause the failure.
The report comes amid debate over who is to blame for the defects that led to the shutdown.
Sen. Barbara Boxer (D-Calif.) and U.S. Rep. Ed Markey (D-Mass.) obtained a leaked version of the proprietary Mitsubishi report. They wrote to the head of the U.S. Nuclear Regulatory Commission last month alleging that the report showed Edison and Mitsubishi were aware of design defects in the generators before they were installed and chose not to make fixes.
On Friday, after weeks of back and forth with Mitsubishi, the NRC released a redacted version of the report.
San Onofre was shuttered after a tube in the plant's replacement steam generator system leaked a small amount of radioactive steam Jan. 31, 2012. Eight other tubes in the same reactor unit later failed pressure tests, an unprecedented number in the industry, and thousands more tubes in both the plant's units showed signs of wear.
The wear was blamed on tube vibration caused by excessively dry and high-velocity steam and inadequate support structures, particularly in one of the plant's two units. Tube vibration and wear has been a problem at other plants, but the specific type of vibration seen at San Onofre had never been experienced before in the industry.
According to the Mitsubishi report, a design team formed by Mitsubishi and Edison recognized that the steam could be drier than in previous designs. The team made some changes and considered others to address the condition, but eventually decided the final design was "optimal, based on the overall ? design requirements and constraints."
One factor they considered, Mitsubishi said, was that major changes would require a license amendment. That could be a lengthy process that might require public hearings.
However, the company also said of the proposed design changes in a supplemental report, ?None of these alternatives had a large enough effect [on the dryness of the steam] to justify such a significant change.? In a letter to the NRC, an attorney for Mitsubishi said the changes were also rejected because they might have had ?negative safety impacts? on other portions of the design.
Boxer released a statement Friday expressing "serious concerns" about Edison and Mitsubishi's actions, and Markey said they "cut safety corners to avoid a new NRC licensing process."
Friends of the Earth, an environmental group that has been pushing for the NRC to require a license amendment before the unit can fire up again, said the report showed that Edison "clearly knew about design problems."
The two companies strongly disagreed.
Mitsubishi said it was impossible to predict the problems at San Onofre because that type of wear had never been seen at any other nuclear power plant.
As for code error, the report said Mitsubishi had used an "inappropriate definition of the gap between tubes" in its calculations. But the company said even using the correct code would not have predicted the type of wear that occurred.