The ITER vacuum vessel leak tightness requirements, proving and achieving them in construction
Pearce, R J H1; Wykes, M P E2; Bansley, R2; Frederici, G1; Jones, L P D F1; Worth, L3
1Germany;
2France;
3United Kingdom

ITER, the largest magnetic confinement fusion device to date, will be constructed in Europe at Cadarache in France. It is a global international project which aims to demonstrate the scientific and technical feasibility of fusion power.
The ITER vacuum vessel is an all welded, 60mm thick, double walled, torus-shaped structure of 19.4m outer diameter and 11.3m height. The internal volume will be 1600m3. It will be constructed in nine toroidal sectors which will be shipped to the ITER site for in situ joining. The interspace between vacuum vessel walls is used for high pressure water cooling and heating. The vessel will be housed within a large cryostat providing thermal insulation for the super conducting coils.
The overall vacuum leak tightness required of the vacuum vessel is reviewed. This is assessed considering operational experience of other tokamaks. In particular a qualitative assessment is made of the detrimental effects of leakage on operations including considering the effects of leakage on plasma breakdown and on radiated power losses during the plasma burn phase. The requirement to minimise leakage from the vessel interspace into the cryostat is also concidered.
A risk based approach is used as the basis for recommending both the quantity and type of leak testing to be used in the construction of the ITER vacuum vessel. Data and experience on construction leaks on vacuum components for JET and other large vacuum devices is considered as an input to this analysis. Other considerations are the costs, practicality and time impact of testing; these are weighed against the impact of leaks only being only discovered at a later stage of ITER construction.
It is recommended that each of the nine vessel sectors are factory leak tested on completion, with an acceptance leak criteria of <1x10-9 Pam3s-1 and for the testing to include a baking cycle and a hot leak test. The technique to be employed involves detecting leaks into the interspace between the vessel walls. Due to the very high surface area within this interspace (1700m2) the leak testing technique has be sensitive and responsive in the presence of a high gas load. A method by which this can be achieved is presented and qualified theoretically and experimentally.
back