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.

An investigation into the falsification of pellet diameter data in the MOX demonstration facility at the BNFL Sellafield Site and the effect of this on the safety of MOX fuel in Use

Foreword

This report sets out the findings of the investigation carried out by HSE's Nuclear Installations Inspectorate into the falsification of quality assurance data associated with the production of MOX nuclear fuel pellets manufactured in the MOX Demonstration Facility at Sellafield.

The investigation was carried out under the control of the Deputy Chief Inspector responsible for regulating the safety at BNFL's sites. The investigation began shortly after BNFL notified NII of suspected falsification on 10 September.

It is the Executive's view that the report gives a thorough analysis of the issues surrounding the falsification of quality assurance data at MDF. It is clear that various individuals were engaged in falsification of important records but a systematic failure allowed it to happen.

It has not been possible to establish the motive for this falsification, but the poor ergonomic design of this part of the plant and the tedium of the job seem to have been contributory factors. The lack of adequate supervision has provided the opportunity. Despite this, self-discipline ought to have ensured that those involved followed the proper procedures.

One point worth noting is that in the new Sellafield MOX Plant, currently being commissioned, the inspection processes for MOX pellets, rods and assemblies are designed to be almost fully automated: this should prevent the falsification of data of the kind described in this report.

There are many lessons to learn, but the MOX Demonstration Facility is shut down and will not be allowed to restart until NII is satisfied that the recommendations in the report have been implemented.

If you have any comments, or would like further information on the issues discussed in this report, write to the Chief Inspector at the following address below:


Laurence Williams
Director of Nuclear Safety and HM Chief Inspector of Nuclear Installations
Health and Safety Executive
St Peter's House
Stanley Precinct
Bootle L20 3LZ


Summary

The MOX Demonstration Facility (MDF) at BNFL's Sellafield site manufactures MOX (mixed oxides of plutonium and uranium) fuel pellets and assembles these using various customer supplied components to make complete fuel assemblies for use in nuclear power reactors. Each fuel pellet produced passes through a fully automated laser micrometer which checks and records the pellet's diameter at three points along its length, giving a 100% automatic check on all pellets used in a fuel rod. Any undersized or oversized pellets are automatically rejected. Those which fall within the specified diametral tolerance pass onto the next stage where each undergoes further visual checks. As a confirmatory check on diameter and in accordance with the 1% Acceptable Quality Level (AQL) criterion set out in BS 6001, a sample of 200 pellets (approximately 5%) which have passed through both these stages is measured for a second time. This quality check is done using a similar micrometer, but the sample pellets are presented to the micrometer by process workers who type each measured diameter, e.g. 8.195mm, into a computer spreadsheet.

On 20 August 1999 a member of MDF's Quality Control Team identified similarities between the secondary pellet diameter data for successive Lots. After further investigations, on 10 September 1999 BNFL reported to NII that some of these secondary pellet diameter checks on the fuel manufactured for a Japanese customer appeared to have been falsified by copying some data between spreadsheets.

The Health and Safety Executive's Nuclear Installations Inspectorate (NII) promptly launched an investigation to establish both the extent of the falsification and the causes of the event. NII concluded that data had indeed been falsified but that this would not affect the safety performance of the fuel, given the automated primary diameter check on 100% of the pellets used in each fuel rod.

NII believes the failure to properly carry out the agreed manual checks of the pellet diameter to be a contractual issue between BNFL and its customer. However, because it also represents a deliberate breach of operating procedures the Inspectorate launched an investigation which centred upon:

  1. understanding just what had occurred in MDF and why;
  2. whether the fuel involved will be safe in use; and
  3. what needs to be done to prevent any recurrence.

NII's investigation into possible reasons for the falsification identified that although various individuals were at fault, a systematic failure allowed it to happen. In a plant with the proper safety culture, the events described in this report could not have happened.

NII commissioned an independent analysis by HSE's statisticians of the extent of the falsification. The results of this and further manual checks of data by NII showed that the initial investigation by BNFL, carried out under severe time pressures was too narrow: there had been a tendency to rush to early conclusions which understated the extent of the problem by assuming that the falsification was largely confined to one shift. Nevertheless BNFL agreed to carry out further, more detailed investigations and, following discussions with NII, has taken steps to address the contributory factors to this incident which the Company and the Inspectorate have identified.

NII is satisfied that in spite of the falsification of the quality assurance related data, the totality of the fuel manufacturing quality checks are such that the MOX fuel produced for Japan will be safe in use. With regard to MDF, the plant is shut down and will not be allowed to restart until NII is satisfied that the recommendations outlined in this report have been implemented to ensure, inter alia, that the deficiencies found in the quality checking process have been rectified, the management of the plant has been improved and plant operators have been either replaced or retrained to bring the safety culture in the plant up to the standard NII requires for a nuclear installation.


Conclusions

103. The events at MDF which have been revealed in the course of this investigation could not have occurred had there been a proper safety culture within this plant. It is clear that some process workers falsified records of the diameter of fuel pellets taken for QA sampling. One example of falsification has been found dating back to 1996. There can be no excuse for process workers not following procedures and deliberately falsifying records to avoid doing a tedious task. These people need to be identified and disciplined. However, the management on the plant allowed this to happen, and since it had been going on for over three years, must share responsibility.

104. Before NII will allow the restart of MDF, BNFL will need to address all the recommendations in the report to the Inspectorate's satisfaction.


Added to HSE website 18th February 2000

Updated 2010-12-07