Office for Nuclear Regulation

This website uses non-intrusive cookies to improve your user experience. You can visit our cookie privacy page for more information.

Level 4 meeting on Heysham 2/Torness post-stress reversal, pre keyway root cracking graphite safety case NP/SC 7663

Executive summary

Purpose of intervention

This meeting was held so that EDF Nuclear Generation Limited (NGL) could present their progress in the preparation of a new safety case for the graphite core for Heysham 2 and Torness Power Stations. NGL see the need for three safety cases for each pair of power stations throughout their lives. This safety case is to cover the intermediate period that is considered to occur after the graphite bricks have experienced stress reversal, but before the initiation of keyway root cracking.

Stress reversal in the bricks is a result of the expected dimensional change behaviour in graphite under the effect of irradiation. Keyway root cracking is the predicted occurrence when the stresses in graphite bricks are such that the outer surface is in tension and stresses have reached a level such that crack initiation occurs at the keyways. It is difficult to observe stress reversal directly, so the point of stress reversal is a nominal one based on calculation and certain inferences from the inspection of the graphite core. It does not have particular safety significance and is largely a marking point in the ageing of the graphite cores. However, NGL expects stress reversal to occur in about 2018, with the possible onset of keyway root cracking in about 2025.

Interventions Carried Out by ONR

ONR and NGL hold regular Level 4 interactions on graphite technical topics. This meeting was held purely as an information exchange to bring ONR up to date with progress.

Explanation of Judgement if Safety System Not Judged to be Adequate

NA

Key Findings, Inspector's Opinions and Reasons for Judgements Made

A number of significant points emerged from the meeting, notably that NGL will be relying on a different method for calculating weight loss than is used at the other AGRs. They will be relying heavily on the modelling of graphite weight loss by a statistical approach based on extrapolation from the region covered by trepanned sample data. They will not be using their approved code, which is called FEAT-DIFFUSE6. This code appears to predict a weight loss gradient with depth that is not as steep as that observed in the actual samples removed from the reactor. The code does not allow for differences in the gas flows within the brick between those with the Heysham 2/Torness design and the other AGRs. They will though continue to develop the FEAT-DIFFUSE6 code, with the intention that it can be applied to Heysham 2 and Torness at some point in the future.

Heysham 2 and Torness power stations are the most recent Advanced Gas-cooled Reactors (AGRs) to commence operation and inspection results indicate that there has only been limited cracking in the bricks compared to other AGRs so far. Broadly therefore, I would be surprised if any significant graphite related safety issues developed at these power stations that had not been detected earlier at the older AGRs. Both ONR and NGL therefore consider that the earlier AGRs are likely to be the ones where phenomena such as keyway root cracking are detected first and that they can be used as predictors of the likely future development of graphite ageing at Heysham 2 and Torness.

Conclusion of Intervention

NGL provided some useful insights into how the graphite safety cases will be developed in this largely positive meeting. Notwithstanding my comments above, I consider that NGL will need to maintain a detailed focus on the ageing of the graphite. The safety cases should be supported by analysis and data obtained by sampling and inspection of the graphite bricks during the periodic shutdowns.