S. Caspi

The use of pressurized bladders for stress control of superconducting magnets

LBNL is using pressurized bladders in its high field superconducting magnet program Magnet RD3; a 14 T race track dipole, has been assembled and pre-stressed using such a system. The bladder, placed between the coil pack and the iron yoke, can provide 70 MPa of pressure while compressing the coil pack and tensioning a 40 mm thick structural aluminum shell. Interference keys replace the bladders functionality as they are deflated and removed leaving the shell in 140 MPa of tension. During cool down, stress in the shell increases to 250 MPa as a result of the difference in thermal expansion between the aluminum shell and the inner iron yoke. A number of strain gauges mounted onto the shell were used to monitor its strain during assembly, cool-down and testing. This technique ensures that the final and maximum stress in the shell is reached before the magnet is ever energized. The use of a structural shell and pressurized bladders has simplified magnet assembly considerably. In this paper we describe the bladder system and its use in the assembly of a 14 T Nb/sub 3/Sn magnet.

Magnet RaD for the US LHC Accelerator Research Program (LARP)

TUA2OR6 Magnet RD fax: 510-486-5310; e-mail: sagourlay@lbl.gov). G. Ambrosio, N. Andreev, E. Barzi, R. Bossert, V. S. Kashikhin, V. V. Kashikhin, F. Nobrega, I. Novitsky, D. Turrioni, R. Yamada, and A.V. Zlobin are with Fermilab National Accelerator Laboratory, Batavia, IL 3 M. Anerella, A. Ghosh , , R. Gupta, M. Harrison, J. Schmazle, and P. Wanderer are with Brookhaven National Laboratory, Upton, NY.

Design of HQ—A High Field Large Bore

In support of the Large Hadron Collider luminosity upgrade, a large bore (120 mm) Nb3Sn quadrupole with 15 T peak coil field is being developed within the framework of the US LHC Accelerator Research Program (LARP). The 2-layer design with a 15 mm wide cable is aimed at pre-stress control, alignment and field quality while exploring the magnet performance limits in terms of gradient, forces and stresses. In addition, HQ will determine the magnetic, mechanical, and thermal margins of Nb3Sn technology with respect to the requirements of the luminosity upgrade at the LHC.

{\rm Nb}_{3}{\rm Sn}

Future high-energy accelerators will need very high magnetic fields in the range of 20 T. The Enhanced European Coordination for Accelerator Research and Development (EuCARD-2) Work Package 10 is a collaborative push to take high-temperature superconductor (HTS) materials into an accelerator-quality demonstrator magnet. The demonstrator will produce 5 T stand alone and between 17 and 20 T when inserted into the 100-mm aperture of a Fresca-2 high-field outsert magnet. The HTS magnet will demonstrate the field strength and the field quality that can be achieved. An effective quench detection and protection system will have to be developed to operate with the HTS superconducting materials. This paper presents a ReBCO magnet design using a multistrand Roebel cable that develops a stand-alone field of 5 T in a 40-mm clear aperture and discusses the challenges associated with a good field quality using this type of material. A selection of magnet designs is presented as the result of the first phase of development.

Quadrupole Magnet for LARP

J 2LC02 SC-MAG#773 LBNL-49918 An Approach for Faster High Field Magnet Technology Development R. R. Hafalia, S. Caspi, L. C hiesa, M . Coccoli , D.R. Die tde rich, S. A. Gourlay, A.F. Lietzke, l .W. ONeill, G. Sabbi, and R.M. Scanlan Abstract- The Superconducting Magnet Program at LBNL has developed a magnet design supporting our new Subscale Magnet Program, that facilitates rapid testing of small superconducting racetrack coils in the field range of 10·12 Tesla. Several coils have been made from a variety of NbJSnlCu cables, insulated, wound, reacted, potted and assembled into a small reusable yoke and shell loading structure. Bladder and key technology have provided a rapid, efficient means for adjusting coil pre-stress during both initial assembly, and between thermal cycles. This affords the opportunity to test moderately long cable samples under magnet conditions on a time scale considerably closer to that for traditional short-sample cable tests. We have built and tested four coils with the initial aim of determining the feasibility of reducing overall conductor costs with mixed-strand cables. Details of cost reduction improvements, coil construction, magnet structure, and assembly procedures are reported, along with the relative performance of the mixed-strand coil. II. THE CONDUcrOR The Subscale Magnet Program utilizes superconducting Nb3Sn strands produced by Oxford and lac. T he cables were wou nd with 20 strands ofO.7mm diameter. The nominal cable cross-section used by the Subscale Magnet Program is 7.9mm x 1.3mm. The cable is sheathed in a continuous, woven, fiberglass sleeve. Coil modules are wound in 2 layers in flat racetrack coil configuration around an iron pole-island - each layer having 22 turns each. (Fig. 1). Each double-layer coi l module requires 22m of cable (5 kg). The two baseline coil modules (SCOI & SC02) were used in the inaugural test of the new Subsea Ie Magnet Test Faci lity. They were both wound with Nb3Sn cable - with the same strand as was used in LBLs 14.7 Tesla RD-3B large-scale magnet. The winding and assembly, prior to reacti on, took about 2-3 days. Index Terms- Common Coil Magnet, mixed-strand cable, Nb)Sn, racetrack coil. I. I NTRODUcrlON awrence Berkeley National Laboratory has implemented a Subscale Magnet Program. The new Subscale Magnet Test Facility was fabricated to supporJ the program. The facility includes a 38 lmm-lD x - 1524mm-deep vertical cryostat, wi th supporJ structure and magnet supporJ header. T he program tests the performance of various types of superconducting cables in a medium-to-high field environment as well as investigates magnet structural design modifications, L in small scale, before implementing them into our large-scale program. The magnet structural components were mass- Fig. I . 2 layers wound around iron pole. prod uced and standardized for rapid coil mod ule assembly. Three of the four coil modules initially built (two baseline coils, one mi x.ed strand coil and a comparison coil), have been tested. Presently, two more coil modul es with mi xed Nb3SnlCu strand coil variations have been reacted and are presently being assembled. Also, a coil mod ule with a new woven ceramic insulator is being reacted. Manuscript received August 6, 2002. Thi s was supported by the Director, Office of Energy Research, O ffi ce of High Energy and N uclear Physics, Hi gh Energy Physics D ivision , U. S. Department of Energy. under Contract No. The next coil module (SC03) was wound with a mixed- strand cable consisting of 2 1 strands - 14 Nb3Sn strands cabled with 7 strands of pure copper. Due to the difference in elastic modulus between the SC and the Cu strands, decabling was evident throughout the whole length of the cable. Coil winding tension was cut in half, from 178N to 89N, to minimize decabling. Even with the lower tension, popped strands were observed. DE-AC03--76SFO0098. R.R. Hafalia is with the Lawrence Berkeley National Lab. Berkeley, CA 94720 USA (telephone: 5 10-486-5712, (e-rnail: rrhafalia @Ibl.gov). Coi l module SC04 was wound with 20 Nb3Sn strands and was intended as a comparison to the SC03 mixed strand modul e.

Accelerator-Quality HTS Dipole Magnet Demonstrator Designs for the EuCARD-2 5-T 40-mm Clear Aperture Magnet

Magnet programs at BNL, LBNL and FNAL have observed instabilities in high J/sub c/ Nb/sub 3/Sn strands and magnets made from these strands. This paper correlates the strand stability determined from a short sample-strand test to the observed magnet performance. It has been observed that strands that carry high currents at high fields (greater than 10 T) cannot sustain these same currents at low fields (1-3 T) when the sample current is fixed and the magnetic field is ramped. This suggests that the present generation of strand is susceptible to flux jumps (FJ). To prevent flux jumps from limiting stand performance, one must accommodate the energy released during a flux jump. To better understand FJ this work has focused on wire with a given sub-element diameter and shows that one can significantly improve stability by increasing the copper conductivity (higher residual resistivity ratio, RRR, of the Cu). This increased stability significantly improves the conductor performance and permits it to carry more current.

An approach for faster high field magnet technology development

A superconducting magnet assembly has been built for an ECR (Electron Cyclotron Resonance) ion source at the 88-inch cyclotron at LBL. Three 34-cm ID solenoids provide axial plasma confinement and a sextupole assembly in the solenoid bore provides radial stability. Two large solenoids are spaced 50 cm. Apart with a smaller opposing solenoid between. The sextupole assembly is 92 cm long with winding inner diameter of 20 cm. And outer diameter of 27.2 cm. The design goal is to achieve a field on axis of 4 T and 3 T at the mirrors with 0.4 T between and a sextupole field of 2.0 T at 15-cm diameter in the confinement volume. Each solenoid uses rectangular conductor wish copper/SC ratio of 4; the three coils are wet-wound on a one-piece aluminum bobbin with aluminum banding for radial support. The sextupole uses rectangular conductor with copper/SC ratio of 3. Each of the 6 coils is wet-wound with filled epoxy on a metal pole; the ends of the pole are aluminum and the central 34-cm is iron to augment the sextupole field. The six coils are assembled on a 20-cm-OD stainless steel tube with a 1.4-cm thick 30.0-cm OD aluminum tube over the assembly for structural support. Thin metal bladders are expanded azimuthally between each coil and axially at tire ends to pre-load the assembly. The sextupole assembly fits inside the solenoid bobbin, which provides support for the magnetic forces. The magnet exceeds design requirements with minimum training.

Correlation between strand stability and magnet performance

A Nb/sub 3/Sn dipole magnet (D20) has been designed, constructed, and tested at LBNL. Previously, we had reported test results from a hybrid design dipole which contained a similar inner Nb/sub 3/Sn and outer NbTi winding. This paper presents the final assembly characteristics and parameters which will be compared with those of the original magnet design. The actual winding size was determined and a secondary calibration of the assembly pre-load was done by pressure sensitive film. The actual azimuthal and radial D20 pre-loading was accomplished by a very controllable novel stretched wire technique. D20 reached 12.8 T (4.4 K) and 13.5 T (1.8 K) the highest dipole magnetic fields obtained to date in the world.

Magnet system for an ECR ion source

In support of the development of a large-aperture Nb3Sn superconducting quadrupole for the Large Hadron Collider (LHC) luminosity upgrade, several two-layer technological quadrupole models of TQC series with 90 mm aperture and collar-based mechanical structure have been developed at Fermilab in collaboration with LBNL. This paper summarizes the results of fabrication and test of TQC02a, the second TQC model based on RRP Nb3Sn strand, and TQC02b, built with both MJR and RRP strand. The test results presented include magnet strain and quench performance during training, as well as quench studies of current ramp rate and temperature dependence from 1.9 K to 4.5 K.

Test results for a high field (13 T) Nb/sub 3/Sn dipole

Amongst the magnet development program of a large-aperture Nb3Sn superconducting quadrupole for the Large Hadron Collider luminosity upgrade, six quadrupole magnets were built and tested using a shell based key and bladder technology (TQS). The 1 m long 90 mm aperture magnets are part of the US LHC Accelerator Research Program (LARP) aimed at demonstrating Nb3Sn technology by the year 2009, of a 3.6 m long magnet capable of achieving 200 T/m. In support of the LARP program the TQS magnets were tested at three different laboratories, LBNL, FNAL and CERN and while at CERN a technology-transfer and a four days magnet disassembly and reassembly were included. This paper summarizes the fabrication, assembly, cool-down and test results of the six magnets and compares measurements with design expectations.