Minfeng Xu

Second Test Coil for the Development of a Compact 3 T

The authors have reported the results of low n -value from a MgB2 test coil developed a year ago. A second test coil has been developed with wire of different structure and manufacturing process. Although the n-value related voltage of the second test coil was lower than the first test coil at designed current, it still showed low n-value. A third test coil has been wound with reduced mechanical stress. It also showed very similar n-value related voltage and n-value. Investigation of voltage distribution over the coil indicated that magnetic field was the major factor causing degradation of the n-value and resulting in n -value related voltages. Since the n-value related coil voltages were on the order of 0.1 μV/cm, the usual short sample Ic test (1 μV/cm was the definition of Ic ) might not detect the n-value related voltage and might not be able to investigate the cause of low n -value. Therefore, the medium length ( ~ 10 m) samples were tested and they showed the wires lengthwise nonuniformity both on n-value and Ic, which might be another potential cause of the low n-value of the coil. Along with the electrical investigation, the manufacturing process of the wire was carefully inspected for longitudinal uniformity. Some wire segment samples from the same batch exhibited nonuniformity in the particle size distribution resulting in nonuniform filaments. This might have occurred in the wire for the second and third test coils.

\hbox{MgB}_{2}

Typical coils with BSCCO tape are wound in a pancake or double-pancake style to minimize the strain in the tape by reducing or eliminating the edge-wise bending. Stainless steel reinforced tapes are frequently used in the winding process to increase the strength and reduce the strain due to winding tension and handling. However, an MRI magnet requires high current density in the winding pack. This high current density in the winding pack gives a higher field in the imaging volume and also allows for a reduction in the overall magnet size. Layer winding was preferred for a better tolerance control and for a reduction in the number of joints, which are known sources of resistance and therefore locations of instability in the coil. A mock-up coil was wound using a high-current-density type of BSCCO tape without the typical stainless steel reinforcement. The coil was layer-wound which involved a few inline lap joints embedded in the winding pack. The test of the coil reveals a few issues that need to be addressed. Investigations and analysis lead to a better understanding of the issues. This paper discusses the lessons learned and solutions for using non-reinforced tape in a layer-wound coil, while controlling insulation dimensions within the build.

Magnet

Conventional superconducting MRI magnet electromagnetic (EM) design involves the critical parameters of the magnet field and dimensional requirements, plus the low temperature superconductor (LTS) wire properties. When using HTS material for an MRI magnet, in addition to the conventional characteristics of a superconductor, the wire and the joint resistance need also to be considered. This is mainly due to the relatively low n-values of the BSCCO tape when compared to that of a conventional LTS wire, and it is also due to the more resistive joint nature for HTS joints. This paper discusses an iterative EM design procedure that calculates the resistance of the HTS wire/tape for the entire magnet, which is a strong function of the magnetic fields at different locations in the coils. The resistance is fed back to the regular MRI EM design for optimization, in order to meet resistance target and the ordinary MRI design goals. This iterative design process results in an optimized HTS MRI design with overall low resistance for magnet drifting and power supply requirements.

Experimental Layer-Wound Mock-Up Coil for HTS MRI Magnet Using BSCCO Tape

The authors have reported results of MgB2 test coils that exhibited anomalously low n-value. It was discovered that the major cause of the n-value related voltage was nonuniformity of the wire along its length. Based on this finding, the development of a compact 3 T magnet has been started. The magnet consists of six coils of 30 cm bore each. The fields will be 3 T at 4 K and 1.5 T at 20 K, respectively. The coils will be cooled by thermal conduction. One of the center coils was manufactured with the refined wire of improved lengthwise uniformity. Results of tests on this coil showed no measurable n-value related voltage. Superconducting joint development has been ongoing. Current peak multifilament joints show superconductivity up to 120 A at 14 K. Further trials to achieve the full short sample operating current at each temperature are ongoing. The cryogenic design for the magnet is based on the use of dual coolants either with hydrogen or helium and consists of two distinct and separate primary and secondary cooling circuits.

Iterative EM Design of an MRI Magnet Using HTS Materials

An active quench protection system is developed to safely protect a 1.5 T superconducting MRI magnet. The protection system includes a digital quench detection controller, a set of quench heaters and voltage taps, a battery backed DC power supply and solid-state switches. The quench controller takes the voltage signals from the voltage taps over superconducting coils and monitor a quench onset. To distinguish a quench signal from operational signals, the quench controller uses a pre-determined moving average voltage over 50 ms as the threshold signal for quench detection. The threshold signal is more than 10 times larger than any of the normal operation signals to avoid false quenching. Once detected moving average voltage exceeds the threshold, the quench controller will close solid-state relays and power quench heaters to accelerate the quench propagation over the entire magnet. The system has been tested in a MRI magnet test module.

Development of a Compact 3 T

The team at GE Global Research presents cryo assembly and component test results of a high-temperature superconducting (HTS) limb size magnetic resonance imaging (MRI) scanner using Sumitomos DI-BSCCO tape conductor, under an NIH research grant. The goal is to investigate the thermosiphon behavior for different MRI operating modes, validating the cryogenic robustness of this cooling approach and its performance limits. The magnet is indirectly cooled using cooling tubes with liquid neon filling and a single-stage cryocooler for reliquefying.

\hbox{MgB}_{2}

Superconducting devices operating within a power system are expected to go through transient overload conditions during which the superconducting coil has to carry currents above the rated values. Designing the coil to remain superconducting through any possible fault scenario can be cost prohibitive, necessitating operation beyond the critical current for short periods. In order to set operating limits and design adequate protection systems for superconducting devices connected to a power system, the region of safe operation of these devices has to be described with general capability curves. Existing standards that define limits for these over-current situations are based on copper winding experience that do not apply to devices with superconducting components because of the highly nonlinear interaction between magnetic fields, operating temperature, and current density in the superconductor, and the rapidly varying material properties at cryogenic temperatures. In this paper, the behavior of superconducting coils during over-currents is discussed and a simplified capability curve is described to help standardize device capabilities. These curves are necessary to aid power system designers in appropriately designing the system and associated protection systems.

Magnet

We introduce an advanced and optimized cryogenic cooling concept featuring minimum coolant inventory requirements for small high temperature superconducting (HTS) magnets based on results obtained with an experimental model. Experience gained from these experiments led to a new design that will be experimentally verified by the end of this year. Winding of the HTS magnet has already begun and will be completed shortly. New components, current status and cryogenic scope of this new engineering model are described.

An Active Quench Protection System for MRI Magnets

Open MRI magnets are generally designed with ferromagnetic poles to contain and shape the magnetic flux and to reduce conductor cost. Permanent magnet MR magnets have blocks of PM and bulk ferromagnetic materials on or close to the pole face. These electrically conducting regions are sources of eddy currents that affect the image quality because of their relatively long time constants and close proximity to the imaging volume. The impact on image quality can be minimized by appropriate segmentation and/or lamination of these components. Detailed eddy current diffusion models are necessary to quantify the field distortion and time constants of the resulting field to perform design studies. The three dimensional frequency or time domain models required to accurately predict effects of eddy currents due to gradient fields are not computationally economical. This paper describes modeling of a PM imaging system using simplified 2D models with appropriate assumptions to evaluate the impact of these eddy currents. Experimental validation of some of the results with a prototype magnet is provided

The Cryogenics of a Thermosiphon-Cooled HTS MRI Magnet—Assembly and Component Testing

A limb size MRI magnet coldmass has been constructed using DI-BSCCO tapes from Sumitomo. The coils were wound with epoxy pre-impregnated fiberglass cloth (pre-preg) between layers to bond the wires. For radial dimensional control, temperature was elevated during the winding to thin out the epoxy and to adjust the pre-preg cloth layer thickness in order to control the coil build up. The wire tension was controlled within 1 kg with a set of moving pulleys. While the coils appeared solid after winding and curing, issues were found in the leads between coils. When the coldmass was cooled down to liquid nitrogen temperature, breaks in wire leads were found. Thermal expansion and contraction mismatch between the coil bobbin and the BSCCO tape was attributed to the leads break. The thermal stress was induced both in the oven curing and the cooling processes. Preliminary testing results at temperature are discussed. The magnet was designed to have a center field of 1.5 T operating at a liquid neon (LNe) temperature of 27 K.