A low-speed high-torque HTS machine is being developed at Siemens on the basis of previous steps (400kW demonstrator, 4MVA generator). The goal of the programme is to utilize the characteristic advantages offered by electrical machines with HTS-excited rotor, such as efficiency, compact size, and dynamic performance. To be able to address future markets, requirements from ship classification as well as potential customers have to be met. Electromagnetic design cannot be focused on nominal operation only, but has to deal with failure modes like short circuit too. Utilization of superconductor requires to consider margins taking into account that the windings have to operate reliably not only in "clean" laboratory conditions, but in rough environment with the stator connected to a power converter. Extensive quality control is needed to ensure homogenous performance (current capacity, electrical insulation, dimensions) for the large quantity of HTS (45 km). The next step was to set up and operate a small-scale "industrial" manufacturing process to produce HTS windings in a reproducible way, including tests at operating conditions. A HTS rotor includes many more components compared to a conventional one, so tough geometric tolerances must be met to ensure robust performance of the system. All this gives a challenging task, which will be concluded by cold testing of the rotor in a test facility. Then the rotor will be delivered for assembly to the stator. In following machine tests the performance of the innovative HTS drive system will be demonstrated.

The shielding of a DC magnetic field has been demonstrated to be complete at 4.2 K, at least up to 2 T, inside the 18 mm bore of a thick MgB2 bulk cylinder, 70 mm tall. No hysteresis has been detected along the cycle of the field application and removal, and no flux jumps have been detected either. The MgB2 cylinder was manufactured by the Reactive Mg Liquid Infiltration process, an 'in situ' technology able to give very dense MgB2 manufacts. According to the previously measured J(c)(B, T) characteristics of a MgB2 bulk superconductor of the same origin and quality, a critical current density of about 30 kA cm(-2) can be assumed at the fields of 2-3 T and at the temperature of 4.2 K of the reported shielding experiment. Accordingly, we estimate that the thickness of the MgB2 cylinder filled by the shielding currents is about 5.75 mm, lower than the present cylinder thickness of 8.77 mm.

Each superconducting coil of the ATLAS Barrel Toroid has to pass the commissioning tests on surface before the installation in the underground cavern for the ATLAS Experiment at CERN. Particular acceptance criteria have been developed to characterize the individual coils during the on-surface testing. Based on these criteria and the limited time of the test, a compressed test program was proposed and realized. In only a few cases some additional tests were required to justify the coil performance and acceptance. In this paper the analysis of the test results is presented and discussed with respect to the acceptance criteria. Some differences in the parameters found between the identical coils are analyzed in relation to coil production features

Each of the eight huge coils of the Barrel Toroid of the ATLAS detector consists of two double pancakes which are embedded in an aluminum alloy coil casing. The 57 mm times 12 mm sized conductor is a Rutherford cable with NbTi-Cu strands co-extruded with a high purity aluminum stabilizer. The race track coils have overall dimensions of 25 m times 5 m and the length of the conductor in the windings is 6.7 km. The coils are conduction cooled with forced flow helium. The nominal operating current is 20.5 kA and the nominal ramp rate is 4 A/s. During the test program of the individual coils the ramp losses are measured to confirm that they do not exceed the design cooling capacity of the ATLAS cryogenic system. The losses are determined from the amount of evaporated helium in the return flow. The ramp losses in the conductor consist of the hysteresis and coupling current losses in the Rutherford cable and eddy current loss in the pure aluminum stabilizer. Ohmic losses are generated in the coil casing which acts as a low resistive secondary of a transformer formed by the coil and the casing. In this paper the results of the loss measurements on the different coils, with different RRR (residual resistance ratio), are presented. Measurements are performed at various ramp rates. The results are in good agreement with the calculated losses, which are dominated by the loss in the coil casing

The Barrel Toroid (BT) provides the magnetic field for the muon detectors in the ATLAS experiment at CERN. The Toroid is built up from eight superconducting coils. Each coil consists of two 25 m times 5 m racetrack shape double pancakes impregnated and pre-stressed inside an aluminum coil casing. The 42-tons cold mass is cooled by forced-flow liquid helium circulating in aluminum pipes glued to its surface. The coils are tested on surface prior to their underground installation. The test program has started in September 2004 and finished in June 2005. This paper describes the test set up and various commissioning tests performed at the ATLAS Magnet Test Facility. It includes the aspects of test preparation, vacuum pumping, leak testing, cooling down, powering and warming up. The 8 coils have passed the tests successfully and have been assembled into the Toroid in the ATLAS cavern. The testing completes the production of the so far largest racetrack coils in the world

The Superconducting Barrel Toroid is providing (together with the two End-Cap Toroids not presented here) the magnetic field for the muon detectors in the ATLAS Experiment at the LHC at CERN. The toroid with outer dimensions of 25 m length and 20 m diameter, is built up from 8 identical racetrack coils. The coils with 120 turns each are wound with an aluminum stabilized NbTi conductor and operate at 20.5 kA at 3.9 T local field in the windings and is conduction cooled at 4.8 K by circulating forced flow helium in cooling tubes attached to the cold mass. The 8 coils of 25 m /spl times/ 5 m are presently under construction and the first coils have already been fully integrated and tested. Meanwhile the assembly of the toroid 100 m underground in the ATLAS cavern at CERN has started. The 8 coils are individually tested on surface before installation. In this paper the test of the first coil, unique in size and manufacturing technology, is described in detail and the results are compared to the previous experience with the 9 m long B0 model coil.

Hysteresis loss is an important factor when the performance of superconducting electric power devices is evaluated. The non-ideal voltage–current relation in Bi2Sr2Ca2Cu3Ox/Ag conductors is one of the reasons that critical state based loss relations do not accurately predict the AC loss in these conductors. In this paper the influence of the magnetic field dependent voltage–current relation on the hysteresis loss is discussed. An analytical high-field approach is used to demonstrate the effect of the finite steepness of the voltage–current relation on the induced current. Besides, a numerical technique is used to calculate also for applied magnetic fields below the penetration field the influence of both a field dependent critical current density and a field dependent steepness of the voltage–current relation. Both the magnitude of the hysteresis loss and the field dependence of the loss are influenced by the voltage–current relation. Especially for small applied fields the field dependence of the loss deviates from the cubic dependence that is predicted by the critical state model based relations. Results of calculations are compared with measured data. An intrinsic critical current density vs. magnetic field relation is determined in order to obtain agreement between measurements and calculations.

A method to calculate the AC loss of superconducting power devices from the measured AC loss of a short sample is developed. In coils and cables the magnetic field varies spatially. The position dependent field vector is calculated assuming a homogeneous current distribution. From this field profile and the transport current, the local AC loss is calculated. Integration over the conductor length yields the AC loss of the device. The total AC loss of the device is split up in different components. Magnetization loss, transport current loss and the loss due to the combined action of field and current all contribute to the AC loss of the device. Because ways to reduce the AC loss depend on the loss mechanism it is important to know the relative contribution of each component. The method is demonstrated on a prototype transformer coil wound from Bi/sub 2/Sr/sub 2/Ca/sub 2/Cu/sub 3/O/sub x//Ag superconducting tape. Differences between the model assumptions and devices are pointed out. Nevertheless, within the uncertainty margins the calculated AC loss is in agreement with the measured loss of the coil.

In some electrical apparatuses, superconducting tapes are exposed to the longitudinal magnetic field. In this work, AC losses were measured in twisted and untwisted Bi-2223 tapes carrying AC transport current in the AC longitudinal magnetic field. In twisted tapes, the transport, magnetization and total losses depend on the relative direction of the longitudinal magnetic field to the direction of the transport current, while the field direction does not influence the AC loss characteristics in untwisted tapes. In the Z-twisted tapes, the total AC loss is larger in the longitudinal magnetic field that is anti-parallel to the transport current than in the longitudinal magnetic field of another direction. Numerical analysis shows that this field direction dependence of the total AC loss results from the change in the current distribution. In the longitudinal magnetic field that is anti-parallel to the transport current, the total AC loss in the Z-twisted tape is more than that in the untwisted tape. This dependence on the field direction is reversed in S-twisted tapes. It is to be noted that the twist increases the total AC loss in a longitudinal magnetic field of a certain direction, while it reduces the AC loss in the transverse magnetic field.

The magnetisation loss in multifilament HTS tape superconductors can be reduced by twisting the filaments and increasing the matrix resistivity. In this paper the influence of filament twist on the coupling current loss of the tape is studied. Magnetisation measurements on multifilament Bi2223 tape conductors with different twist pitches are performed at 77 K in magnetic field applied parallel and perpendicular to the wide face of the conductor. Both frequency and magnetic field amplitude are varied. In perpendicular magnetic field the filaments are fully coupled. In parallel magnetic field the twisted filaments are decoupled, even for twist pitches around 10 mm and a pure silver core. However, for increasing magnetic field >0.04 T, the influence of the coupling current loss increases. The transition to full coupling and the coupling current loss are analysed with an existing analytical model for a tape superconductor.

An analytical approximation is developed for the magnetization of an infinitely long superconductor with an elliptical transverse cross-section. The superconductor is modeled in the critical state with a critical current density that is not dependent on the magnetic field. The aspect ratio of the ellipse is varied from one (=circle) to infinitely large. The magnetic field is applied perpendicular or parallel to the broadest face. The analytical expression is compared with a more detailed model that utilizes a numerically optimized contour for the boundary of the saturated zone. The two methods are compared and the maximum error is estimated at 2% for the optimized contour approach and 5% for the analytical approximation. The analytical model is compared with a magnetization loss measurement on a high-Tc superconducting tape with an aspect ratio of nearly 20. A good agreement is obtained for a magnetic field pointing perpendicular as well as parallel to the broadest face of the tape. An interesting result for the magnetic behavior determined for the ellipse is that it contradicts with the behavior that is predicted for an infinitely thin strip in perpendicular field. The difference is attributed to the two specific assumptions made in the thin strip model: the constant critical current density distribution across the tape and the magnetic-field profile that does not exclude unsaturated currents in the shielded zone.

Multifilamentary Bi2223 tapes are exposed to the longitudinal magnetic field as well as the transverse one in some electrical power apparatuses such as multilayer power transmission cables. Here, we define the longitudinal and transverse magnetic fields as the field components parallel and perpendicular to the tape axis, respectively. If the filament-bundle is twisted, it can couple to the AC longitudinal magnetic field to generate the longitudinal magnetization loss. Furthermore, the AC transport current flowing spirally in the twisted filament-bundle possibly influences the longitudinal magnetization. The longitudinal magnetization loss was measured in a twisted multifilamentary Bi2223 tape exposed to longitudinal magnetic field and carrying the transport current. The measured longitudinal magnetization loss in the twisted tape exposed to the longitudinal magnetic field is larger than that in another untwisted tape. Supplying the AC transport current changes the longitudinal magnetization loss in the twisted tape exposed to the AC longitudinal magnetic field. The influence of the transport current depends on the phase relation between the longitudinal magnetic field and the transport current. If their phase difference is 0°, the longitudinal magnetization loss decreases remarkably with increasing amplitude of the transport current. It means that the change in the current distribution due to the transport current results in the decrease in the power flow from the magnet power supply. But, a preliminary measurement of the transport loss shows that the total loss increases with increasing transport current.

In electric power applications, BSCCO superconducting tapes are exposed to an alternating magnetic field while simultaneously an alternating transport current is flowing through the conductor. The magnetic field has various orientations with respect to the tape surface. An engineering function to describe the AC loss in a BSCCO/Ag tape is developed. The expression describes the total AC loss in a wide range of magnetic fields (with different orientations) and transport currents. In this approach the specific properties of BSCCO tape e.g. the shape of the conductor cross-section and the magnetic field dependent voltage-current relation are taken into account. The equation is based on the analytical expressions available, and adjusted to measured data. The AC loss of a coil wound of BSCCO tape is calculated, the result is compared with measured data and possibilities to reduce the loss are considered. (C) 2002 Elsevier Science B.V. All rights reserved.

A high-quality superconducting resonator coil system has been developed in the Netherlands in cooperation with the companies SMIT Transformers and SMIT Draad. The coil system was manufactured in industry, using industrial tooling. It has a reactive power rating of 1 MVA at a frequency of 50 Hz and a temperature of 64 K. The system consists of four concentric solenoidal coils, including 2 km of Bi-2223 conductors from two manufacturers. The coil performance has been optimized by shaping the magnetic field around the coil edges with laminated ferro-magnetic C-cups to reduce the AC loss to about 1 kW at 1 MVA reactive power. The system is operated in a glass-epoxy cryostat at either 64 or 77 K. The resonator coil demonstrates all superconducting elements that are essential for manufacturing superconducting transformers. The construction of the system is finalized and the total system has been tested in various operational configurations. The results of the system test at full power are presented and analyzed.

Bi2Sr2Ca2Cu3Ox/Ag tape superconductors are used in applications like power cables and transformers. In these applications the superconductor is exposed to an alternating magnetic field that has different orientations with respect to the tape surface. In this paper the angle dependency of the magnetisation loss is considered from two points of view. First the measurement technique with pickup coils is analysed theoretically. Measured magnetisation loss in uni-directional magnetic field with various orientations and rotating magnetic field are compared. When the orientation is changed from perpendicular (0°) to parallel (90°) applied magnetic fields, the contribution of the perpendicular field component to the magnetisation loss is dominant up to 60°. A new model to describe the angle dependency of the magnetisation loss, based on the measured loss in perpendicular and parallel magnetic field is developed. Deviations between models and the measured loss are explained with the help of the theoretical analysis of the measurement technique. The new model is not only applicable for the magnetisation loss but also for other AC loss components.

The dynamic resistance loss, DC transport current combined with external AC magnetic field, in a Bi2Sr2Ca2Cu3O10/Ag superconducting tape is measured as a function of the orientation of the external AC magnetic field. We investigate the possibility of describing the AC transport current loss in an externally applied AC magnetic field from the much simpler dynamic resistance measurement. The orientation angles of the magnetic field with respect to the plane of the tape are changed from 0° (perpendicular configuration) to 90° (parallel configuration). The angular dependence of the dynamic resistance energy dissipation is described with an empirical formula. It is very similar to the angular dependence observed for the hysteresis loss. The dynamic resistance in parallel and perpendicular applied magnetic field is fairly well described with theoretical predictions based on the critical state model. The increase of the AC transport current loss due to a simultaneously applied AC magnetic field is described quite well with the dynamic resistance in the fully saturated state.

Magnetisation loss is an important factor in the design of superconducting transformers and motors. In these devices the tapes are usually placed face-to-face. Then the magnetisation loss is influenced by the mutual magnetic shielding between adjacent tapes. The shielding is investigated by measuring the magnetisation loss in stacks with various numbers of Bi-2223 tapes, exposed to a 48-Hz perpendicular magnetic field at 77 K. In a stack the penetration field is increased and the magnetisation loss below penetration is greatly decreased, compared to the behaviour of a single tape. The loss at high magnetic-field amplitudes is unaffected. The measured loss is compared to the loss calculated with two different models. The effect of shielding is qualitatively well described with an analytical model. However, predictions made with a numerical model display a better quantitative agreement with the measurement results.