P. Heitzenroeder, J. Citrolo, F. Dahlgren, H. Neilson, 1
Princeton Plasma Physics Laboratory
Princeton, NJ 08543
D.B. Montgomery2
University of California, San Diego
ITER Home Team Office
LaJolla, CA 92093
C. Sborchia3
ITER Naka Co-Center
Naka-gun, Ibaraki-ken, 311-01, Japan
The hybrid central solenoid design was developed as a possible alternative to the segmented central solenoid during a study of the segmentation of the ITER central solenoid. Both designs provide similar improvements in shaping flexibility and accuracy. However, the hybrid design avoids the most difficult challenge of the previous segmented solenoid design: the placement of superconducting joints in its high field (13 T) central bore. It also has several other features of significance. The hybrid maximizes the use of NbTi superconductor, and this may be beneficial in reducing costs and improving failure recovery.
The proposed design is called the hybrid because it adopts beneficial features from the reference and the segmented solenoid design, and therefore utilizes much of the investments already made in them. The hybrid consists of a layer wound central solenoid and two pairs of discrete pancake wound coils which can be independently powered.
The main solenoid section is layer wound like the reference design, but its overall length is reduced from 12 meters to 9.5 meters. Its shortened length provides space for a pair of discrete coils spaced from its two ends. The solenoid leads are located in the gaps between it and the discrete coils. The maximum field in this location is 5.7 T in contrast to 13 T in the bore of the coil where the leads for the solenoid of the segmented design are located.
A second pair of discrete coils is added above and below the TF coils, just outboard of the central solenoid. This second pair of coils is an important differentiating feature of the hybrid design. The hybrid geometry reduces the field levels at the PF 2 and PF 7 coils to 5T. Consequently, NbTi can be used in these coils rather than Nb3Sn. NbTi is less costly, and offers the possibility of winding a replacement in place in the event of a coil failure. All discrete coils are made of stacks of pancake-type windings similar to those proposed for the segmented solenoid.
This paper describes and compares engineering details of the two designs and discusses the tradeoffs between them.
1Work supported by U.S. DoE Contract No. DE-AC02-76-CHO373.
2Work supported by the Office of Fusion Energy Science, Energy Research, DoE.
3Work supported by