The looming gap in new-build standards9 March 2010
The 2008 Mattson report from the US DOE found that the process of completely identifying standards to be used in US nuclear new-build had to await the start of construction. To avoid construction delays, the timely resolution of standards issues will be critical. Today’s tools can help. By Jim August and Joe Hunter.
Design, materials and construction techniques have changed since the United States started its last new nuclear power plant more than 25 years ago. The National Institute of Standards Technology (NIST) and American National Standards Institute (ANSI) established the Nuclear Energy Standards Coordination Collaborative (NESCC) to identify and address the current standards needs of the nuclear industry . Working cooperatively with standards developers, the NESCC coordinates activities that relate to nuclear standards development. NESCC formulates, coordinates, and recommends priorities for revising existing standards to help develop new standards that support plant operation, licence renewal, new nuclear plant construction and advanced nuclear technologies development. NESCC identifies needs, gaps, and challenges to recommend work, including identifying funding required.
As a part of Congress NP 2010 initiative, the nuclear industry, supported by the Department of Energy (DOE), Nuclear Regulatory Commission (NRC), NIST (DOC) and the standards development organisations (SDOs) identified standards that could pose risks to nuclear construction.
The DOE report  reviewed the state of current standards used for nuclear construction that are referenced by NRC rules and codes. It evaluates what standards might be needed for new nuclear plants, whether they are current, available and acceptable for use as consensus standards, or whether they could pose delays. The report also assesses the capability of developing any new consensus standards required, quickly.
The surprise in the DOE’s ‘Mattson report’ is that organisations cannot identify many standards needs early. To a large degree, even with high-level nuclear safety design rule certifications, their design control documents (DCDs) and supporting requirements will not be uniquely determined until construction. However, Part 52 of the US federal code of legislation, (Code of Federal Regulations, Title 10 Part 52, Licenses, Certifications & Approvals (Combined Licenses)) failed to anticipate this requirement to the degree needed. Thus, NRC and industry need to either approve many construction standards early that will never be used, or develop methods that can review and approve newly-identified needed standards quickly. Methods proposed contrast with past construction practice, where NRC placed greater reliance (and trust) upon vendors to develop methods to identify requirements, perform voluntary inspections and determine their compliance. Conformity assessments then were often not clear.
The report identifies five codes and standards areas of weakness:
1. Different standards are referenced by NRC and industry.
2. SDO consensus standards are updated (by reaffirmation or revision) according to ANSI requirements.
3. Lessons learned from Generation I-II plant construction not reflected.
4. Technology advances such as digital controls require new standards.
5. New methods like modular construction deployment require new or modified standards.
Standards organisations and their users should consider how to alleviate concerns to improve nuclear standards use. Furthermore, they should look at how users apply standards they use to develop better methods to use them.
Three licence differences
The report cited three differences between the NRC and industry that warrant further consideration: different interpretations in selection, details or versions of the standards that apply.
1. Vendor DCD and applicant construction and operating licence applications (COLAs) do not reference NRC-referenced standards.
2. Industry-referenced codes and standards that NRC has not referenced.
3. NRC and industry reference different editions of the same standard or one superseded, withdrawn or historical.
For example, NRC Inspection Procedures (IP) reference standards that vendors’ and applicants’ DCDs and COLAs do not. NRC needs to review and approve other standards, while certifying a standardized design or issuing a COL. They must know which standards they will use to do this.
During combined licence application review, the NRC examines construction processes for details such as proposed standards. Ideally, they would resolve any differences they discover before construction begins. Project schedules challenge the timing of detailed design and selection of all construction standards. Custom manual methods that review standard content case-by-case present an ineffective review process that needs improvement. Thus, current methods that apply standards will not provide clear requirements before construction begins – although this was the intent of Part 52. Nor can industry significantly revise withdrawn historical standards that may still provide the best available guidance.
Are necessary rules in place?
Nuclear industry participation in the four primary nuclear SDOs – the American Nuclear Society (ANS), the American Society of Mechanical Engineers (ASME), ASTM, and the Institute of Electrical and Electronics Engineers (IEEE) – on construction-related standards committees fell during the years when no nuclear plants were built in the USA. Hence, to assure that necessary applicant-agreed standards are in place on the front end of this new US construction cycle, licensees and the NRC need increased standards support. Support includes support for improving standards processes. Improvements should consider how standards are developed, as well as maintained, updated and used.
Current codes and standards may not incorporate previous nuclear power plant construction experience lessons adequately. ASME stated that although their standards are up-to-date, in the past forward-looking plant operators, designers and NRC provided the feedback that raised issues that led to updating the boiler and pressure vessel code. ASME said that it has not taken a rigorous look at code interpretations and other issues that occurred during the current fleet construction to resolve all issues applicable to future plants. Nor have other SDOs cited fully reviewing lessons learned to incorporate, either. The lack of continuity between those who constructed the last generation of nuclear power plants and the people who will construct the next provides another reason to revisit how we manage and use standards. So are the necessary standards in place? No, definitely not!
“Additional nuclear standards required to construct new
When the IEEE Nuclear Power Engineering Committee (NPEC) queried nuclear vendors about the adequacy of current standards, they said that available standards were adequate, and no new standards would be needed. Upgrading standards would not delay new plants’ design and construction, they said. The Nuclear Energy Institute (industry) and NRC continue to discuss whether additional standards would be useful or even be needed. Yet NRC/IEEE representatives described additional nuclear standards required to construct new plants. Of those cited, several covered digital instrumentation and controls (I&C), cyber security, and safety system computers that manage design basis (DB) safety information, probabilistic risk assessments (PRA) and setting surveillance intervals from PRA. Other SDOs have similar lists, including standby emergency combustion turbines, modular construction techniques, and new welding alloys and techniques.
NRC, SDOs and industry should consider expanding their discussions to other areas of technology. It will take hundreds of consensus codes and standards to design and build new nuclear power plants. These each incorporate hundreds of discrete critical requirements. DOE has recommended that NRC, industry, and the SDOs undertake joint efforts to ensure that they agree on what standards apply to design/construction, version(s) and that the contents of those standards are valid. Given their importance to safe and timely construction, NRC (and SDOs) must then ensure that users consistently implement the applicable standards.
Revising national codes and standards to vet regulations, regulatory guides, and other documents requires time-consuming consensus development processes. These mechanisms cannot resolve construction issues quickly. Therefore, to the fullest extent possible, authorities should identify the gaps to close in codes and standards prior to commencing construction. The technology used to develop, update and manage the information content found within national standards is the largest gap.
Addressing standards processes before significant nuclear construction begins could allow time to develop better standard coordination methods. Those knowledgeable in new plant design, construction and operations (including regulation) should suggest how to address new nuclear plant standards adequacy and coordinated use, while identifying the new standards needed.
SDOs should provide better methods to speed the development, maintenance and use of their standards. More important than ever, they need better tools to meet industry needs. New Part 52 rules should improve coordination, control, update and use of standards compared to Part 50. To do this and reduce costs requires addressing structural process requirements today. Failure to address new reactor design standards application will increase long-term nuclear costs, risk extending the time to needed to develop/construct new nuclear plants and increase the investment required.
Standards provide highly structured guidance that establish requirements governing activities. In the US, OMB Circular A-119 and the National Technology Transfer and Advancement Act of 1995 require consensus standards to be used in rulemaking. Effective standards also specify performance requirements. Because there are so many nuclear design, construction and operations standards requirements, relationships between independently evolving standards from multiple SDOs are complex. NIST has estimated that building a traditional LWR takes hundreds of standards with thousands of requirements. Considering the rules that also apply and organisations like INPO or NEI that provide other requirements, the profusion of guidance is tedious.
Complicating the problem of standards use is the need to relate not only content, but also related version, identifying application and other aspects. Rules also need to identify applicable scope and accepted industry guidance. These interrelationships require identifying critical standard content. Thus, standards must specify requirements, which well-written standards do. Furthermore, the users of even poorly-written standards can identify the characteristics of their critical content. Thus, despite the intimidation of performing such an effort, the key to its success lies not in standards so much as in the format that identifies, presents and relates the critical standard content within the nuclear design framework.
To integrate the use of standards requires a relational application. The database model used to develop critical content relationships is the most important requirement. Upon completion, such an application could automate links between content, much like a closed-logic application tracking system, computer-aided design (CAD), or a similarly organised financial accounting transaction system. Such an application would link content directly so that every user could develop, trace or examine threads easier.
At the ASME conference ICONE 2009, the authors presented papers  that gave steps to address developing a new framework. Broadly, they are:
• Develop an approach – an information framework.
• Map critical standards/information content relationships.
• Design a simple application in test prototype overview.
• Implement a test programme to confirm the approach and validate design.
• Migrate the validated application to the internet.
• Resolve SDOs rights with intended users (for funding).
• Find a method to fund information maintenance long-term.
• Limit access to those willing to pay reasonable fees for the information content they use.
Nuclear organisation lobbyists attempt to influence the US Congress for the industry. Yet, the nuclear organisations are reluctant to ask forthrightly for funding. The means and methods to improve how designs are developed and maintained are with us, yet no one has sought Congressional help to fund this recommended improvement. To increase US nuclear energy use, we need a better foundation – better processes to support new plant design.
Well over 50% of nuclear operating costs arise from support. Thus, improving nuclear support productivity is an important strategy to lower nuclear costs and increase its use. Streamlining nuclear information processes will benefit nuclear design and lower costs.
In their current state, standards, rules and processes take substantial due-diligence effort to execute well in practice. Developing rules into useful performance-based guidance takes a significant time investment. Organising requirements for general use, like the periodic table or human genome project, would help provide clear guidance and performance-based new design requirements. That will lower entrance barriers for new reactor designers who seek to develop and licence new designers. Otherwise, only the most fully-funded, well-heeled pedigreed nuclear designs will ever have the needed expertise to meet expectations for developing new designs.
Jim August, chief engineer and vice president, CORE, Inc, 5915 Braun Way, Arvada, CO 80004. He participates in several US SDO committees, including ANS 28, gas reactors and the ANS Nuclear Facilities Safety Committee. Joe Hunter, CORE, Inc, 7 Dos Rios, Greeley, CO 80634, is a former US nuclear SRO. Core, Inc develops risk-based maintenance processes.Related ArticlesFollow the jumbo jet
Additional nuclear standards required to construct new