Toward Integrated Design and Modeling of High Field Accelerator Magnets

Abstract

The next generation of superconducting accelerator magnets will most likely use a brittle conductor (such as Nb 3 Sn), generate fields around 18 T, handle forces that are 3-4 times higher than in the present LHC dipoles, and store energy that starts to make accelerator magnets look like fusion magnets. To meet the challenge and reduce the complexity, magnet design will have to be more innovative and better integrated. The recent design of several high field superconducting magnets have now benefited from the integration between CAD (e.g. ProE), magnetic analysis tools (e.g. TOSCA) and structural analysis tools (e.g. ANSYS). Not only it is now possible to address complex issues such as stress in magnet ends, but the analysis can be better detailed an extended into new areas previously too difficult to address. Integrated thermal, electrical and structural analysis can be followed from assembly and cool-down through excitation and quench propagation. In this paper we report on the integrated design approach, discuss analysis results and point out areas of future interest.

Mechanical Design of the SMC (Short Model Coil) Dipole Magnet

“…For this optimization, the rod diameter has been set to 28 mm. The optimization process has been based on the radial stress component at the boundary between the first cable turns and the central pole, being the point of the highest stress level eventually leading to coil motion [12]. The longitudinal pre-load is the input parameter; it is expressed in terms of rod displacement (mm) at RT, induced by the hydraulic piston.…”

LARP Nb[sub 3]Sn QUADRUPOLE MAGNETS FOR THE LHC LUMINOSITY UPGRADE

“…The presence of iron pads in close contact with the coil determines, for the same current, an 8% higher gradient than TQC. As shown in FIGURE 4 (right) and TABLE 2, a total of 7 TQS tests were performed, with 6 assembly and loading operations and 1 re-loading operation [22][23][24][25]. The target gradient of 200 T/m was consistency reached in the TQS02 tests, with a maximum gradient of 231 T/m (and a peak field of 11.8 T) obtained in TQS02c.…”

Towards Computing Ratcheting and Training in Superconducting Magnets

“…The goal of the study was to understand and reproduce the main features described in the previous section: the occurrence of quenches in the pole turn, the progressive axial elongation of the coil (ratcheting), and the increase in quench current during the test (training). With respect to previous analyses [6], this study added the frictional energy dissipated due to conductor motion. Moreover, we performed a series of computations that included consecutive current cycles of loading and unloading, and combined the dissipated energy with a thermal model, in an attempt to find out the temperature increase and to reproduce qualitatively a training curve.…”

“…On one side, a detailed analysis of the voltage signal recorded right before and after a quench has been performed, with the goal of investigating the causes (flux-jumps or stick-slip mechanical motions) and the locations of the quench [5]. On the other side, 3D finite element models have been implemented to study coil movements during excitation, with the goal of contributing to the interpretation of magnet performances, and, ultimately, to improve training [6]. We present in this paper an attempt to simulate and reproduce a magnet training behavior through a finite element model.…”