S. Le Naour

Status of the LHC superconducting cable mass production

Six contracts have been placed with industrial companies for the production of 1200 tons of the superconducting (SC) cables needed for the main dipoles and quadrupoles of the Large Hadron Collider (LHC). In addition, two contracts have been placed for the supply of 470 tons of NbTi and 26 tons of Nb sheets. The main characteristic of the specification is that it is functional. This means that the physical, mechanical and electrical properties of strands and cables are specified without defining the manufacturing processes. Facilities for the high precision measurements of the wire and cable properties have been implemented at CERN, such as strand and cable critical current, copper to superconductor ratio, interstrand resistance, magnetization, RRR at 4.2 K and 1.9 K. The production has started showing that the highly demanding specifications can be fulfilled. This paper reviews the organization of the contracts, the test facilities installed at CERN, the various types of measurements and the results of the main physical properties obtained on the first batches. The status of the deliveries is presented.

Magnetization measurements on LHC superconducting strands

When using superconducting magnets in particle accelerators like the LHC, persistent currents in the superconductor often determine the field quality at injection, where the magnetic field is low. This paper describes magnetization measurements made on LHC cable strands at the Technical University of Vienna and the Institute of Physics of the Polish Academy of Sciences in collaboration with CERN. Measurements were performed at T=2 K and T=4.2 K on more than 50 strands of 7 different manufacturers with NbTi filament diameter between 5 and 7 micrometer. Two different measurement set-ups were used: vibrating sample magnetometer, with a sample length of about 8 mm, and an integrating coil magnetometer, with sample length of about 1 m. The two methods were compared by measuring the same sample. Low field evidence of proximity effect is discussed. Statistics like ratio of the width of the magnetization loop at 4.2 K and 2 K, and the initial slope dM/dB after cooldown are presented. Decrease of the magnetization with time, of the order of 2% per hour, was observed in some samples.

Critical Current Density in Superconducting

The knowledge of the critical current density in a wide temperature and applied magnetic field range is a crucial issue for the design of a superconducting magnet, especially for determining both current and temperature margins. The critical current density of LHC-type Nb-Ti strands of 0.82 and 0.48 mm diameter was measured by means of critical current and magnetization measurements at both 4.2 K and 1.9 K and for a broad magnetic field range (up to 11 T). For the magnetic field range common to both measurement methods, critical current density values as extracted from transport current and from magnetization data are compared and found fairly consistent. Our experimental data are compared to other sets from literature and to scaling laws as well

rm Nb-Ti

One of the main issues for the operation of the LHC accelerator at CERN is the field errors generated by persistent and coupling currents in the main dipoles at injection conditions, i.e., 0.54 T dipole field. For this reason we are conducting systematic magnetic field measurements to quantify the above effects and compare them to the expected values from measurement on strands and cables. We discuss the results in terms of DC effects from persistent current magnetization, AC effects with short time constant from strand and cable coupling currents, and long-term decay during constant current excitation. Average and spread of the measured field errors over the population of magnets tested are as expected or smaller. Field decay at injection, and subsequent snap-back, show for the moment the largest variation from magnet to magnet, with weak correlation to parameters that can be controlled during production. For this reason these effects are likely to result in the largest spread of field errors over the whole dipole production.

Strands in the 100 mT to 11 T Applied Field Range

In the superconducting main magnets of the Large Hadron Collider (LHC), persistent currents in the superconductor determine the field quality at injection field. For this reason it is necessary to check the magnetization of the cable strands during their production. During four gears, this requires measurements of the width of the strand magnetization hysteresis loop at 0.5 T, 1.9 K, at a rate of up to eight samples per day. This paper describes the design, construction and the first results of a magnetization test station built for this purpose. The samples are cooled in a cryostat, with a 2-m long elliptic tail. This tail is inserted in a normal conducting dipole magnet with a field between /spl plusmn/1.5 T. Racetrack pick-up coils, integrated in the cryostat, detect the voltage due to flux change, which is then integrated numerically. The sample holder can contain eight strand samples, each 20 cm long. The test station operates in two modes: either the sample is fixed while the external field is changed, or the sample is moved while the field remains constant, First results of calibration measurements with nickel and niobium are reported.

Persistent and coupling current effects in the LHC superconducting dipoles

Applications of 50 Hz superconductors like the transformer and the fault current limiter correspond to relatively low magnetic fields, so that AC losses and stability are mainly governed by the conductor self field. AC loss calculations as they are performed in most cases for superconductors, are based on the Bean critical state model which states that everywhere in a superconductor, the current density has a modulus equal to the critical current density J/sub c/. This model is applicable when the superconducting transition E(J) is very sharp, but sizeable discrepancies appear for 50 Hz superconductors, as they present a relatively smooth superconducting transition. AC loss calculations have been developed using the Maxwell equations combined with the actual E(J) relationship. The heat generation in the conductor is then used as an input for a numerical calculation of the temperature distribution through the superconductor. The stability limits are directly derived from the thermal model.<<ETX>>

Test station for magnetization measurements on large quantities of superconducting strands

The production of more than 60% of superconducting cables for the main dipoles of the Large Hadron Collider has been completed. The results of the measurements of cable magnetization and the dependence on the manufacturers are presented. The strand magnetization produces field errors that have been measured in a large number of dipoles (approximately 100 to date) tested in cold conditions. We examine here the correlation between the available magnetic measurements and the large database of cable magnetization. The analysis is based on models documented elsewhere in the literature. Finally, a forecast of the persistent current effects to be expected in the LHC main dipoles is presented, and the more critical parameters for beam dynamics are singled out.

Application of 50 Hz superconductors close to self field conditions

The Large Hadron Collider (LHC) contains eight main dipole circuits, each of them with 154 dipole magnets powered in series. These 15-m-long magnets are wound from Nb-Ti superconducting Rutherford cables, and have active quench detection triggering heaters to quickly force the transition of the coil to the normal conducting state in case of a quench, and hence reduce the hot spot temperature. During the reception tests in 2002-2007, all these magnets have been trained up to at least 12 kA, corresponding to a beam energy of 7.1 TeV. After installation in the accelerator, the circuits have been operated at reduced currents of up to 6.8 kA, from 2010 to 2013, corresponding to a beam energy of 4 TeV. After the first long shutdown of 2013-2014, the LHC runs at 6.5 TeV, requiring a dipole magnet current of 11.0 kA. A significant number of training quenches were needed to bring the 1232 magnets up to this current. In this paper, the circuit behavior in case of a quench is presented, as well as the quench training as compared to the initial training during the reception tests of the individual magnets.

Trends in cable magnetization and persistent currents during the production of the main dipoles of the Large Hadron Collider

Fast cycled superconducting magnets (FCMs) are an option of interest for the long-term consolidation and upgrade plan of the LHC accelerator complex. In the past two years we have conducted an R&D targeted at investigating the feasibility, operational issues and economical advantage of FCMs in the range of 2 T bore field, continuously cycled at 1 Hz. In this paper we report the main results on the development of strands and cables suitable for this application, providing details on the strands tested and the cable manufacturing and performance.

Retraining of the 1232 Main Dipole Magnets in the LHC

Throughout 2015 and 2016, the LHC is operated with a current in the main dipoles of 10980 A, equivalent to a proton–proton collision energy of 13 TeV in the center of mass. A total of 175 training quenches were needed in 2014 in the 1232 main dipole magnets installed in the LHC at CERN to reach operational conditions. Since 2008, a number of dipole magnets have been removed from the LHC and were, sometimes after repairs of nonconformities, retested in the CERN based SM18 magnet test facility up to ultimate current. Other magnets have been retested after long storage. The results confirm earlier findings that some magnets series are more prone to quenching than others after thermal cycle. The correlation between a short and long thermal cycle is under investigation. Special cases with many thermal cycles will be highlighted and a new magnet series, fully produced at CERN is introduced. Results of a quench heater fatigue test, assessing the long-term reliability of the quench heaters, will be given. The results of repairs following high internal splice resistances are discussed.