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.

Torness Reactor 2 2015 Periodic Shutdown: Graphite Core Inspection

Executive summary

Purpose of intervention

Torness (TOR) Reactor 2 (R2) began its periodic shutdown on 10 July 2015 as part of the licensee’s, EDF Nuclear Generation Limited’s (NGL), compliance with Licence Condition (LC) 30.  During the shutdown the licensee will perform examination, maintenance or testing of plant; in particular that required by the Maintenance Schedule which includes the graphite core.

Purpose of Intervention

I performed an intervention at Torness power station on 4 and 5 August 2015.  The purpose of the intervention was to inspect the adequacy of the licensee’s examinations and inspections of the graphite core and determine compliance with LC 28(1).

Interventions Carried Out by ONR

 I carried out the following interventions as part of my inspection:

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

Only 13 of the targeted 16 fuel channels were inspected using NICIE2 fuel channel inspection equipment.  This was because of equipment failures during the initial phase of inspections.  The station stated that the risks presented from not attempting to achieve the targeted 16 channels of inspection were As Low As Reasonably Practicable.  This was because no type III (>80% of brick height) axial cracking was observed in the 13 inspected channels which gives a greater than 99% confidence that the core has fewer than 5% cracked bricks.  Therefore, the considerable expense of extending the periodic shutdown to facilitate performing additional inspections was not proportionate to the small reduction in statistical uncertainty that a further three channels of inspection would provide.  I stated that although I was disappointed by the inspection shortfall I understood the claim that the risks were As Low As Reasonably Practicable.  Furthermore, I understood that having inspected a total of 32 channels over all previous outages, which accounts for approximately 300 bricks, only three type III cracks had been observed and only one of those was an axial crack.  Therefore, I thought it likely that the claim that the risk from not attempting 16 channels of inspection was As Low As Reasonably Practicable could be justified.

Given the equipment reliability issues found during my inspection I made arrangements to raise these issues with the group head of the graphite core inspection team at a forthcoming meeting.  

At the time of my inspection, seven of a minimum of 30, and a target of 35 trepanned samples had been retrieved.  I stated that because the relatively small trepanning data set suggests that there are different rates of weight loss in reactor 1 compared to reactor 2, I expected them to achieve a minimum of 30 samples from the core.  This was consistent with the graphite assessment panel agreed minimum and would reduce uncertainty over the behaviour of the two cores.

I inspected the calibration records and operating instructions for the graphite inspection equipment.  I was satisfied that appropriate processes were in place and had been followed to ensure that the equipment was set-up and calibrated properly before it was used for graphite inspections.

I spoke with and inspected the training records of staff that had performed the graphite inspections during the periodic shutdown.  I was satisfied that they had been adequately trained and understood the purpose of the work.

I was notified by the station that whilst conducting inspection of the steel core restraint system in quadrant C of the reactor core, a cracked graphite brick was observed.  The cracked brick was one of approximately 1600 that make up the outer graphite shield of the core.  Further inspections are being conducted to determine the extent of the problem.

Conclusion of Intervention

The target number of channel inspections set by the graphite assessment panel was not met with 13 out of 16 channels being completed.  This number was compliant with the maintenance schedule requirements but only by virtue of the fact that a new graphite safety case has been delayed and was not in place prior to the outage as expected.  Given that the shortfall was the result of a series of equipment reliability issues I judge that the licensee’s examinations and inspections of the graphite core were below standard, IIS rating 4.  These findings have been added to ONR issue 537

Discovery of cracks in the peripheral shielding graphite bricks at the outer circumference of the core will require some consideration by ONR specialist inspectors as part of any consent for return to service of Torness reactor 2.

A forthcoming assessment report will consider the results from the graphite inspection programme and consider whether the results are consistent with the graphite safety case.  Its conclusion will help to inform their decision as to whether to recommend that consent to return to service is granted.