what does the government watchdog do?

Basic design principle

The design principle of the Sizewell 'B' reactor pressure vessel is that the pressure induced stress present within the pressure vessel wall is less than the yield stress, so that all of the ills that pressure vessels are prone to can be surely and safely controlled. The questions arise 'what defects have been found?' and 'How do we know that they are safe?'. The answers to these questions are, wrongly, kept secret from public knowledge.

back to index

The 'Master Curve' upper shelf

During operation the Sizewell 'B' reactor pressure vessel temperature is sufficiently high to ensure ductile behaviour and the internal pressure is 15.5MPa. Failure, even when pressurized, stone cold and brittle, does not constitute a long range hazard. Expensive though. Considerable resources have been expended on the avoidance of brittle fracture.

back to index

The government view

In the case of Sizewell 'B', it is said by the HSE that the probability of explosion is one in one million per year from all causes, with the reactor pressure vessel contributing just one tenth of that,(Ref.3). That is to say, one in ten million per year. Can this figure be substantiated?, and if it can be, is it safe enough?

back to index

What we really know about it

The number of reactor pressure vessels in the world is about five hundred and they have been operating for about twenty years, making a total of experience of ten thousand vessel years; without a single catastrophic explosive failure so far. If we make the usual statistical assumption of ninety-five percent confidence, then this experience provides a reliable assurance that the catastrophic explosion rate is less than one in one thousand reactor years, or thereabouts. Not one in ten millions.

back to index

The government special case procedure

The Sizewell 'B' nuclear power station is licensed by the Health and Safety Commission, Health and Safety Executive, Nuclear Safety Department, Nuclear Installations Inspectorate, (Ref.4). This organization has provided itself with Safety Assessment Principles (Ref.3) which it uses to determine issues of large scale safety which it encounters during its licensing activities. For the most part these safety assessment principles are logical and well founded. However, when cases are encountered which can be neither justified nor denied, resort is made to a 'special case procedure'. The sense of the current version is as follows:-

"Special Case Procedure
A8.9 There are components in a nuclear installation whose safety is difficult to demonstrate in such a way as to readily satisfy the accident frequency requirements of the SAPs, the reactor pressure vessel of a pressurised water reactor being an example. This possibility is catered for in the SAPs by having a principle that allows for such items to be justified on a special case basis and this route has been used on a number of occasions.
A8.10 The two particularly important safety aspects to be addressed are that: the structure is as defect free as possible; and a demonstration that the structure is defect tolerant. In order to achieve this, several related but independent arguments must be used. For example, the arguments could include a demonstration that:

{ sound design concepts and proven design features have been incorporated;
{ potential failure modes have been analysed;
{ proven materials have been used;
Rev. 3 194
{ there has been a high standard of manufacturing;
{ a high standard of QA has been applied;
{ the component has been the subject of pre- service inspection, and will be the subject of in- service inspection, to detect defects at sizes below those which have the potential for causing or developing into a failure mode;
{ provision is made for in-service plant and material monitoring; and
{ a leak-before -break safety case has been made.

A8.11 Where the special case procedure is applied, or where any safety system is required to achieve a high reliability, the licensee has an independent assessment of the item carried out."

This means that in licensing Sizewell 'B', the government decided that the overall risk of explosion of the reactor pressure vessel in service was not worth the cost of any further improvement.

back to index