Vogtle & the AP1000: a real-world cost case study

Projects that last as long as building nuclear reactors in the U.S. are analogous to extreme endurance competitions in which overall progress is the sum of all stages of the race. There is no magic reset button that allows recovery of time lost early in the race, even though the specific events that resulted in…

_Vogtle 3 fabricated shield building panel lift

Courtesy Southern Company_

Projects that last as long as building nuclear reactors in the U.S. are analogous to extreme endurance competitions in which overall progress is the sum of all stages of the race. There is no magic reset button that allows recovery of time lost early in the race, even though the specific events that resulted in the lost time might have been forgotten by observers.

For the past dozen years, we have been repeatedly served stories bemoaning the growing costs and often delayed schedules associated with the Vogtle expansion project. At least the Vogtle project owners have had the fortitude to press through numerous challenges and seem likely to complete the obstacle-strewn endurance event. VC Summer owners chose to abandon their project after spending $9 B. Two responsible executives served time for lying to investors and the SEC about the project’s cost and schedule performance.

It’s worthwhile to learn and remember as many lessons as possible from the AP1000 project experiences. This article will focus on just one of those learning opportunities.

#### Knockdown Blow From Aircraft Impact Rule Imposition

The importance of regulatory stability was one of the lessons that was supposedly learned after the US nuclear plant building boom came to a halt amid spiraling costs and lengthy project delays. The industry told its cost regulators at public utility commissions and the federal safety regulators at the NRC that changing rules during construction led to very expensive changes and rework.

Unfortunately, the message that some apparently heard was that regulatory changes before the start of physical construction were okay. The four AP1000 reactors that started physical construction in the US in 2013 have taken more than two years longer – so far – to build than the four AP1000 reactors that were built in China. The Chinese reactors were first of a kind anywhere in the world while the US reactors were the first of a kind in the US with a reference plant that could be a base on which to improve techniques. (Note: Construction duration begins with first concrete pour and ends with start of commercial operations. None of the US AP1000’s have entered into commercial service.)

The US Nuclear Regulatory Commission published a fateful decision in the Federal Register on June 12, 2009. Titled Consideration of Aircraft Impacts for New Nuclear Reactors; Final Rule, the announcement required a major redesign of the AP1000. The decision was a seminal event in the sad saga of the U.S. AP1000 projects. It is the reason that US AP1000 plants are in a fundamentally different containment and shield structure.

Imposing a requirement to consider – and design for – a direct commercial aircraft impact on projects for which designs were already certified and for which Public Service Commission approvals had already been obtained was roughly equivalent to forcing a marathon runner to stumble and twist her ankle. Even though the injury wasn’t completely debilitating and the race could be run after taking time to treat and tape the injury, the damage to the final time for the race had no real chance of being overcome.

That major change in reactor design requirements was made despite an official NRC finding that “compliance with the rule is not needed for adequate protection to public health and safety or common defense and security.” (Emphasis added.)

The Aircraft Impact Assessment rule was not applied to any existing facility or to any facility for which a construction permit had already been issued. Here is a key excerpt from that final rule.

In making these additions, the NRC is making it clear that the requirements are not meant to apply to current or future operating license applications for which construction permits were issued before the effective date of this final rule. This is because existing construction permits are likely to involve designs which are essentially complete and may involve sites where construction has already taken place.

Applying the final rule to operating license applications for which there are existing construction permits could result in an unwarranted financial burden to change a design for a plant that is partially constructed. Such a financial burden is not justifiable in light of the fact that the NRC considers the events to which the aircraft impact rule is directed to be beyond-design-basis events and compliance with the rule is not needed for adequate protection to public health and safety or common defense and security.”

The unrecognized implication of issuing this rule and choosing to apply it to projects that had not yet been issued a construction permit is that AP1000 projects had already passed several important decision milestones that made them vulnerable to exactly the kind of unjustifiable financial burden that the NRC stated that it was trying to avoid.

Aside: I once had a conversation with a senior NRC regulator who was unapologetic about the effects of the new rule on the initial AP1000 projects. He told me that the staff had clearly communicated its intent to publish the rule and its recommendation that it should be applied to any new projects. He told me that Westinghouse should have begun revising its design well before the rule was published and that it should not have been marketing a design it knew would not be acceptable under the planned requirement. Needless to say, this conversation left me incredulous. But I was able to find several other regulators who confirmed the story on the condition that they not be named. End Aside.

Aside: It is important to understand that Westinghouse had first submitted the AP1000 design certification application (DCA) on March 28, 2002 and that the NRC had issued a design certification on March 6, 2006 based on the 15th revision to the initial DCA. The AP1000 is an evolution from the AP600, which received its final certification in December 1999. End Aside.

Though the detailed design was not yet complete, and though no concrete pours or metal forming had begun by June 2009, the plant design was complete enough to provide the basis for a cost estimate. That estimate was solid enough to pass muster with several boards of directors and two public service commissions.

That is no small task for multi-billion dollar projects, especially those that are as controversial and as visible as firm priced orders for the first new nuclear plants in the United States in more than thirty years.

Here is a quote from a document describing the Vogtle 3 and 4 projects to Southern Company customers.

• Georgia Power filed for an Application for Certification of Vogtle units 3 and 4 with the Georgia Public Service Commission (PSC) in August 2008.

• The Georgia PSC approved the need and cost-effectiveness, granting approval to implement the proposed Vogtle expansion in March 2009.

Remember, the notice applying the requirement for an aircraft impact assessment, and making appropriate design changes to mitigate the projected effects of such an impact was not issued until three months after the Georgia PSC had invested six months into the review and approval of the project as described in the NRC issued Design Certification Rule for the AP1000.

It is difficult to put a precise figure on the time that the project consortium invested into the cost and schedule analysis before submitting it to the PSC, but it was probably no less than a year’s worth of work.

As it turned out, meeting the new rule ended up requiring three more iterations on the design document and raised several issues of contention including concerns about the revised heat transfer capacity, concerns about the innovative shield building construction process devised, and concerns about the validity of the testing regime for the steel concrete composite modules that make up the shield building.

Those and other related issues were not completely resolved until the issuance of the final design certification document in late 2011, more than 2.5 years after the PSC had approved the project using the previously approved design revision.

During that delay, engineers who should have been working on the detailed design so that it could be as complete as possible before construction start were diligently working on t