Timothy A. Antaya

TPX superconducting tokamak magnet system 1995 design and status overview

The TPX magnet preliminary design effort is summarized. Key results and accomplishments during preliminary design and supporting R&D are discussed, including conductor development, quench detection, TF and PF magnet design, conductor bending and forming, reaction heat treating, helium stubs, and winding pack insulation.

The ITER Central Solenoid

The central solenoid for the International Thermonuclear Experimental Reactor (ITER), a fusion tokamak experiment with the goal of generating 500 MW of fusion power with high gain (Q>10), must provide most of the volt-seconds needed to induce and sustain a 15 MA plasma for burn times of >400 s. The 6.4 GJ central solenoid design requires a 45 kA conductor and has a peak field of 13 T. The central solenoid consists of six pancake-wound modules, stacked vertically, and held in axial compression by an external structure. The five-stage cable has 1/3 copper and 2/3 advanced Nb3Sn strands in a thick superalloy conduit and is cooled by the forced-flow of supercritical helium through the cable space. Key design issues include the qualification of a conduit with adequate fatigue strength, avoiding filament damage from transverse Lorentz loads, eliminating axial tension in the winding insulation, and qualification of space-saving intramodule butt joints

Internal Quench of Superconducting Magnets by the Use of AC Fields

The sensitivity of superconducting magnets to AC losses is well known. If superconducting magnets are this sensitive to AC fields, why not use AC fields for magnet protection, and in particular, for internal energy dump when a quench has been detected? The answer is the large reactive power needed to provide the rate of change of the fields required to quench of a large fraction of the magnet. In this paper we describe a novel approach where quench protection secondary windings external to the magnet are used to minimize the power to initiate the energy dump. The main requirement of these secondary windings is that the mutual inductance between the primary winding and the protection secondary windings has to be small, ideally zero. One means to provide for zero mutual inductance between the protection secondary winding and the primary winding is by designing a protection secondary winding that produces AC fields that everywhere in the volume are normal to the fields produced by the primary winding. Alternatively, appropriate windings can be made so that the coupling in one region is the opposite to that of another region, with zero total mutual inductance. The latter approach results in low voltages at the leads, but could result in high voltages within the coil. We describe several circuit topologies applicable to solenoidal and toroidal windings that satisfy these requirements. Calculations of the heating due to AC fields are presented, including eddy current heating in cable-in-copper-channel conductors. If the quench inducing coil is optimally designed and powered, the hysteresis of the superconductor dominates the heating. Thus, as a portion of the superconducting magnet quenches, the heating power shifts to those zones that are not yet in the current-sharing regime. This approach is an alternative to the use of resistive heating elements, which need to be placed on, or embedded in, the winding pack.

Test of a Conduction-Cooled, Prototype, Superconducting Magnet for a Compact Cyclotron

A 208-mm inner diameter, 62-mm-tall, wind-and-react Nb3 Sn prototype magnet was tested to demonstrate its suitability for use in a compact superconducting cyclotron. The magnet and its 270 kg iron return yoke were cooled together by conduction using a 2-stage Gifford-McMahon cycle refrigerator. This paper presents thermal, electrical and electromagnetic data collected during the tests. Although the radiation shield for the assembly cooled to 47 K within one day, a total of nine days were needed to cool the magnet assembly to below 4 K, at an ultimate heat load of 0.2 W. The dominant heat load at each stage of the cold head was due to thermal conduction along the current leads. The coil was charged without quench to 211 A, generating a pole tip magnetic induction of 3 T. The test demonstrates the technical feasibility to design, manufacture and operate compact superconducting cyclotron magnets at currents in excess of 200 A. The results are being used to improve the design for future high-field, conduction-cooled superconducting cyclotrons.

The G0 Spectrometer Superconducting Magnet System: From a Challenging Construction to Reliable Operations

We report on the design, fabrication, commissioning and operation of a large superconducting magnet system that is an important element of the 8 sector super conducting toroidal G0 Spectrometer located at Jefferson Lab (JLAB) in Newport News, VA. The purpose of the G0 experiment is the high precision measurement of polarized electron scattering by protons to isolate the strange quark content of normal baryonic matter by observing parity violation caused by the weak interaction. The G0 spectrometer has been operating for three years and first results are submitted for publication . The G0 SC torus is 4 meters long and 4 meters outside diameter and produces 3 Tesla in the 8 gaps that are accessible to particles. The realization of this 8 sector superconducting toroidal magnet required the development of a number of challenging large scale features including: large total open solid angle, high sector-sector field symmetry, the symmetry axis aligned perpendicular to gravity, the location of the liquid hydrogen (proton) target on axis in the magnet cryostat, and large surface area but thin titanium exit windows on one end of the cryostat. The cryostat consists of a super-alloy welded low permeability stainless steel shell (to minimize magnetization effects) and aluminum end caps. The 8 superconducting coils have unique characteristics including dry pancake wound copper stabilized NbTi conductors, encased in aluminum structure, mechanically preloaded and indirectly cooled by a set of parallel thermo siphon circuits. This magnet was built by BWXT under a fixed price performance contract that included fabrication to a defined ideal cold current spatial distribution. The commissioning and operations will be discussed

MCNPX characterization of compact superconducting cyclotron with 11 B target

A compact superconducting cyclotron accelerator has been built and is undergoing testing by Ionetix Corporation in Hampton, NH. This demonstration cyclotron will accelerate protons up to 12.5 MeV or deuterons up to 6.25 MeV. The design supports the use of internal or external targets. Characterization of the cyclotron using MCNPX-PoliMi was conducted in advance of planned neutron and gamma-ray measurements at the University of Michigan. The cyclotron simulations utilize neutron and gamma-ray spectra resulting from the 11B(d,n)12C reaction with 4-MeV deuterons. The 11B(d,n)12C reaction is of interest for active interrogation research due to its production of high-energy neutrons and gamma rays. Simulated quantities of interest include particle spectra, dose rates, and detector response to facilitate detailed analysis of shielding requirements and experiment design.

Active interrogation source based on deuteron reactions

Simulations of secondary neutron and gamma-ray production from the ¹¹B(d,n)¹²C and ²⁷Al(d,n)²⁸Si reactions were conducted using an MCNPX extension called MCUNED. MCUNED enables simulation of deuteron-based nuclear reactions. Deuteron libraries from the TENDL-2010 and TENDL-2011 packages were used. The simulated neutron and gamma-ray spectra were compared to experimental results provided by Taddeucci et al. and Massey et al. The method was found to be capable of simulating neutron production from ¹¹B(d,n)¹²C and ²⁷Al(d,n)²⁸Si, and gamma-ray production from ²⁷Al(d,n)²⁸Si, but not gamma-ray production from ¹¹B(d,n)¹²C. The TENDL libraries appear to be more reliable for higher mass targets because of the characteristics of the Talys code.

The Advanced Hydrotest Facility (AHF) large bore quadrupole focusing magnet system

The Advanced Hydrotest Facility (AHF) at Los Alamos will provide proton radiography of large-scale, dynamic events. The large bore (Case II) quadrupole focusing magnets are a subsystem in this facility, consisting of four complete imaging lines with a total of eight imaging plates and 52 quadrupole magnets. Each large bore quadrupole has an inner winding diameter of 660 mm and provides a gradient of 10.4 T/m with a 300 mm field of view. Each magnet is a two-layer saddle, contained by a three cm steel shell. The conductor is a Rutherford cable, soldered into a C-shaped copper channel. The magnets are cooled by the forced-flow of two-phase helium through coolant pipes. Since the winding was calculated to absorb bursts of 0.35 J/kg irradiation, both NbTi and Nb/sub 3/Sn designs are being considered.

TPX TF magnet structure design

The design of the Tokamak Physics Experiment (TPX) utilized 16 superconducting toroidal field (TF) magnet coils supported by a wedged structure. The TF magnet structure supports and locates the poloidal field (PF) coils and must restrain the twisting and expansion forces generated by the interaction of currents and magnetic fields. This structure must also leave significant open space to permit access to the plasma chamber. The geometry and alignment of the magnet structural components must be controlled to meet very tight tolerances with respect to the operating toroidal magnetic field errors. During the preliminary design phase, the TPX TF magnet structure configuration was revised from a welded and bolted field assembly to one utilizing bolted joints only. These electrically insulated joints were specifically designed for ease of assembly and to provide the flexibility to adjust coil alignments at assembly to accommodate for manufacturing tolerance stackups. We report on the final preliminary design which meets all project design requirements.