Dynamical and Thermal Testing of a FW Panel Attachment System for the ITER Vessel

The objective of this paper is to describe and evaluate mechanical tests of the First Wall (FW) panel Attachment System (AS) for the blanket system of the fusion reactor ITER according to the 2001 design. The tests were performed in SˇKODA VY´ZKUM, Ltd. with FE simulation support from Czech Technical University in Prague, Faculty of Nuclear Sciences and Physical Engineering. The goal of the tests was to check the stiffness, strength limit, and fatigue behavior of the bolted joint under loads simulating conditions during off-normal plasma operations in the reactor. The FW panels are attached to a bottom thick shield block by means of ten special studs located on a shaped key-way on a shield block surface. A special device for a long-lasting test of stud tensile pre-load relaxation over of 30 000 temperature cycles between 100° and 200°C was developed. Two methods were used to determine the real stud pre-load force drop during such temperature cycling. An optimum procedure for pre-loading of the AS studs has been developed. Four panel mock-ups (1080×250×50 mm) and one massive shield block having all the features of the real AS were fabricated from 316L stainless steel at VI´TKOVICE Research and Development, Ltd. The panel screwed to the shield with stud preload from 45 to 100 kN and was then loaded alternatively by 2500 cycles of radial moment (±24.5 kN·m), poloidal (longitudinal to panel axis) force (±108 kN), or poloidal moment (±53 kN·m) at room temperature. The stud bending stresses, stud pre-load relaxation, cyclic deformation leading to undesirable radial gap opening at the key-way and a possible plastic deformations of AS were studied. Additional FE simulations were used for better interpretation of measurements. The experimental results have shown that thermal cycling leads to a stud pre-load drop from 100 to 60 kN, whereas the dynamic cycling itself does not cause an additional loss of the pre-load. The individual loads applied do not cause a loss of radial contact in the keyway or a damage of AS even under a low stud’s pre-load of only 54 kN. A small radial gap in the key way was observed only under maximum poloidal moment with an extremely low stud pre-load of 45 kN.