ABSTRACT
The safe operation of nuclear power stations is of extreme importance and it is the duty and responsibility of operators to demonstrate to themselves, to a regulator and in consequence to the public that their plant is safe. One of the most crucial components in this process is the main reactor pressure vessel which must be shown to have negligible probability of failure, conventionally classified as incredibility of failure (IOF). BNFL Magnox Generation operates eight nuclear power stations all but two of which have steel reactor pressure vessels. For these, sets of Ground Rules have been established to assist with the demonstration of safety. A key component in these is that during normal operation the material of the vessel must be on the upper shelf of the fracture toughness transition curve. Then in the unlikely event that fractures were initiated from a defect, the subsequent behaviour would be ductile and stable rather than brittle and potentially unstable.
Currently the temperature for the onset of the upper shelf (OUST) is determined from tests on standard high constraint fracture toughness specimens. By agreement with principles established by TAGSI, the OUST is defined as that temperature for which there is 5% probability of cleavage before a set amount of crack growth, Δa, generally taken to be Δa < 0.2 mm.
In order to determine this parameter, distributions of toughness are derived from tests in the transition range spanning the region from fully cleavage to fully ductile and statistical analyses of these data provide the OUST value.
The OUST derived in the above way is, however, likely to be pessimistic for plant applications in which the crack-tip constraint for a structurally significant defect in a pressure vessel is lower than that for a crack in a test piece. To allow for this the effects of constraint on fracture mode and the toughness need to be quantified. A methodology is currently under development to accomplish this for application to specific locations of pressure vessels. The methodology is based on Local Approaches to fracture and involves separately correcting the cleavage and ductile toughness distributions obtained from specimens, better to reflect the constraint on a defect in a pressure vessel. The Beremin model forms the basis of the cleavage correction calculations and the Gurson model the basis of the ductile ones. The results of calculations for various postulated structurally significant defects are presented to demonstrate the potential reduction in OUST and hence temperature margin benefit due to constraint.
The OUST is also affected by neutron irradiation in service. Exposure to neutron irradiation produces an elevation in the OUST. However neutron dose is attenuated through the vessel wall so the elevation in OUST is not uniform at a particular location. The variation of OUST due to 130attenuation may be combined with the change due to constraint to give an indication of the combined OUST.
The quantification of this is presented as an overall benefit, that is reduction in OUST, compared with the calculations for fully constrained, fully damaged conditions currently performed in safety cases. The procedure followed to combine the effects is described.
