Between 1973 and 1990 4 units of the Russian NPP type WWER-440/230 were operated in Greifswald (former GDR). The operation was stopped after the German reunification, because the units did not completely follow western nuclear safety standards. Material probes from the pressure vessels were gained in the frame of the ongoing decommissioning procedure. The investigations of this material started with material from the circumferential core weld of unit 1. This weld was annealed after 13 cycles and operated further for 2 cycles. Additionally, starting with cycle 11, dummy assemblies were inserted to reduce the neutron fluence in the RPV wall. Firstly this paper presents results of the RPV fluence calculations depending on different loading schemes and on the axial weld position based on the Monte Carlo code TRAMO. The results show, that the use of the dummy assemblies reduces the flux by a factor of 2 – 5 depending on the azimuthal position. The fluence increase is reduced to 1/6 at the position of the maximum fluence. The neutron fluence at the different circumferential welds is closely related to their distance to the core. The circumferential core weld (SN0.1.4) received a fluence of 2.4·1019 neutrons/cm2 at the inner surface, it decreases to 0.8·1019 neutrons/cm2 at the outer surface. The neutron fluences at the both other welds are 2 resp. 4 orders of magnitude smaller according to their distances to the core. It should be mentioned that in this cases the fluence gradient can be negative through the wall. The material investigations were done using a trepan from the circumferential core weld. Master Curve and Charpy V-notch testing were applied. Specimens from 7 locations through the thickness of the welding seam were tested. The reference temperature T0 was calculated with the measured fracture toughness values, KJc, at brittle failure of the specimen. Generally the KJc values measured on pre-cracked and side-grooved Charpy size SE(B) specimens of the investigated weld metal follows the course of the Master Curve. The KJc values show a remarkable scatter. In addition the MC SINTAP procedure was applied to determine T0SINTAP of the brittle fraction of the data set. There are remarkable differences between T0 and T0SINTAP indicating macroscopic inhomogeneous weld metal. The highest T0 was about 50°C at a distance of 22 mm from the inner surface of the weld. It is 40 K higher compared with T0 at the inner surface. This is important for the assessment of ductile-to-brittle temperatures measured with sub size Charpy specimens made of weld metal from the inner RPV wall. This material does not represent the most conservative condition. Nevertheless, the Charpy transition temperature TT41J estimated with results of sub size specimens after the recovery annealing was confirmed by the testing of standard Charpy V-notch specimens. The VERLIFE procedure prepared for the integrity assessment of WWER RPV was applied on the measured results. It enables the determination of a reference temperature, RTT0 to index a lower bound fracture toughness curve. This curve agrees with the MC 5% fractile as specified in ASTM E1921-05. The measured KJc values are not enveloped by this lower bound curve. However, the VERLIFE lower bound curve indexed with the SINTAP reference temperature RTT0SINTAP envelops the KJc values. Therefore for a conservative integrity assessment the fracture toughness curve indexed with a RT representing the brittle fraction of a dataset of measured KJc values has to be applied.

This content is only available via PDF.
You do not currently have access to this content.