Search

The Super-FRS (Super FRagment Separator) is a part of FAIR (Facility for Antiproton and Ion Research), which will be constructed at GSI, Germany by 17 countries. The Super-FRS comprises 24 superferric dipole magnets. The 2D and 3D magnetic field simulations of the prototype magnet are described in this paper. A passive trim slot and four chamfered removable poles are used to satisfy the required field homogeneity which is better than at 1.6 T, 0.8 T and 0.16 T in a wide elliptical useable aperture of 380 mm 140 mm. Measurement results at various field levels are shown in this paper as well. It can be seen from the comparison of calculation and measurement results that the magnetic designs of the magnet fulfils the requirements.

FAIR (Facility for Antiproton and Ion Research), planned to be built at the site of GSI will include the 300 Tm fast-ramping heavy ion synchrotron SIS 300. In the frame of a collaboration between IHEP (Protvino, Russia) and GSI (Darmstadt, Germany) IHEP developed, produced and tested a superconducting model dipole. The main parameters of the dipole are a two-layer coil, 100 mm inner coil diameter and 1 m length. The magnet is designed for 6 T central magnetic field with a 1 T/s operating field ramp rate. The nominal operating current of 6720 A (6 T central magnetic field) was reached after the third quench and critical current of 7738 A (6.8 T central magnetic field) was reached after the eighteenth quench. The critical current of the dipole did not reduce up to ramp rate of 1300 A/s (1.15 T/s). AC losses for different magnetic field cycles were measured too.

Various models for the main dipole of the SIS 300 have been manufactured by GSI and Brookhaven (GSI001), IHEP (a 1 m long 6 T model magnet currently under construction) and INFN (a 4.5 T curved magnet for the actual FODO lattice, also under construction). These different magnets reflect the design changes of the SIS 300 over the recent years. Based on the current lattice design studies for the main quadrupoles were started at GSI. In the presentation the actual status of the work on these magnets will be reviewed. Magnetic field quality, AC losses and quench protection aspects of the current design will be addressed.

A design study of the multipole corrector magnet for the SIS 100 is in progress. This magnet, which has to integrate functions of a quadrupole, a sextupole and an octupole magnet, contains nested three windings to save longitudinal space in the accelerator ring. Although the windings are made from a Nuclotron type cable as for the SIS 100 main dipole magnet, insulated superconducting wires which are connected in series to allow low current operation, are used for the cable. In this paper, we describe the design principle and mechanical structure of the magnet and present computation results.

The international FAIR project planned next to G.S.I. (Germany) will lead to the construction of two superconducting synchrotrons (SIS100 and SIS300). This article reports on the quench calculations and measurements performed on one short SIS100 dipole model (1.4 m), the first to be tested at the FAIR prototype test facility. Quench propagation velocities and time to reach the 0.2 V quench threshold were measured in the bus bars and compared to computed values. The hotspot temperature was measured for different Miits (integral of I2 .dt) values at 6 and 8 kA and compared to temperatures computed with the Miits equation. This article also reports the use of one inductive spot heater.

At present, one of the main options for beam bending dipoles of the SIS 300 synchrotron, under design for the FAIR facility at GSI, is a single layer magnet 7.8 m long, 100 mm in bore diameter, generating 4.5 T. This coil has two main features: it is curved, with a curvature radius of 66.67 m (the corresponding sagitta is 114 mm), and shall be ramped at 1 T/s. Both these characteristics demand challenging RD 2) the technological developments (at the industrial level) for defining and optimizing the dipole constructing methods; 3) the construction of curved dipole coil winding models; 4) the construction of the complete curved dipole; 5) the test of the curved dipole in a vertical cryostat; 6) the integration of the curved dipole into a horizontal cryostat, for the final test at GSI.

For testing rapidly-cycled superconducting magnets in the framework of the FAIR project (Facility for Antiprotons and Ion Research) a test facility was set up at GSI to test model and prototype superconducting magnets for different machines (SIS100, SIS300, Super-FRS). We are able to perform the following magnet tests: quench training, magnetic measurements, measurements of AC losses (calorimetric and V-I methods), temperature distributions in the magnet, and long-term stability tests. The forced-flow and bath-cooled scheme can be applied for the superconducting magnets. The facility consists of the following components: a refrigerator (400 W at 4 K); a distribution box for different cooling schemes (supercritical and two-phase flow); 2 feed-boxes providing electrical and cryogenic supply for the magnets; an all-purpose horizontal magnet cryostat which can be used for all kinds of different magnets (max. 3 m length and 0.8 m diameter); a power converter (11 kA) including the quench protection system; a quench detection system adapted from the LHC-series test facilities. The results of the AC loss measurements and the quench training for the first tested superconducting magnet are presented.

The field quality of the GSI SIS-100 magnet is influenced by the eddy currents generated in the conductive beam tube during ramping. The eddy-current losses are computed on the basis of a transient 2D finite-element model. A sufficient accuracy is guaranteed by adaptively refining the mesh and by the use of error-controlled adaptive time integrators. The 2D model is accomplished by an analytical model dealing with the closing of the currents at the end parts of the beam tube and thus incorporating the most significant 3D effect. The time dependence of the field quality reveals significant differences at low aperture fields, especially for the sextupole component at the beginning of the ramping. Also without conductive beam tube and for low field values, the relative sextupole and decapole components depend on the magnitude of the field, which indicates the presence of a nonlinear effect. When the magnet operates at an injection field, the magnetic flux density in the bridge of the iron yoke is already saturated. Hence, the saturation pattern in and around this bridge changes immediately at the start of the ramping, resulting in a changing field quality.

INFN (Istituto Nazionale di Fisica Nucleare, Italy) has started in 2006 the DISCORAP project, which foresees the design, manufacture and test of a dipole prototype for the SIS-300 synchrotron of the FAIR facility at GSI. In order to minimize the losses produced by the fast ramp rate (1 T/s) at which the magnet is operated, we are developing with European industries, a Rutherford cable which incorporates several technologies to reduce the losses, and namely a NbTi filament diameter on the order of 2.5-3.5 mum, Cu 0.5 wt%Mn interfilamentary matrix and stainless steel core. In this paper we present the design principles and the first experimental results; we also analyze the impact of the CuMn paramagnetism on the field distortion at the different operating magnetic fields.

The FAIR project planned at G.S.I. (Germany) requires the construction of several superconducting accelerator rings. One of the two synchrotron rings, named SIS300, will have dipoles with a central field of 6 T and a ramp rate of 1 T/s. In order to protect these fast-ramping magnets in case of a quench, each of them must be equipped with a diode stack that bypasses the magnet in case it quenches. In order to save time and money on R&D, we decided to use the same diode type as that mounted in LHC magnets. These diodes were previously tested at 77 K up to a fast neutron fluence (F) of 2.8E13 n/cm2 and a dose (D) of 2 kGy which are the expected radiation loads after 20 years of LHC operation. Having no definitive estimation of the losses in the SIS300, we arbitrarily targeted a fluence of 1E14 n/cm2. We irradiated six LHC type diodes at 77 K in six steps. The highest values measured were 1.31 kGy and 2.25E14 n/cm2. Before and after each irradiation step, the forward and reverse voltage characteristics of each diode were measured. This paper reports the results of these measurements, carried out at ITEP (Moscow).

The synchrotron SIS 300, developed for the Facility for Antiproton and Ion Research (FAIR) project at GSI, will use fast-cycling dipoles with 100 mm inner diameter coils, magnetic field amplitude of 6 T and a ramp rate of 1 T/s. This work presents the calculations of the magnetic and thermal characteristics of a 1 m model dipole with optimized 2D and 3D geometries. AC losses in the coil and iron yoke, as well as temperature margins, were analyzed. Production tolerances for the main geometrical parameters were also studied.

GSI, Darmstadt is preparing to build FAIR (Facility for Antiproton and Ion Research) which include SIS 300, a 300T - m fast-ramping heavy ion synchrotron. Dipoles for this ring will be 2.9 m long, producing 6 T over a 100 mm coil aperture and ramped at 1 T/s. The cable for these dipoles must have low losses and produce acceptable field distortions during the fast ramp. We plan to achieve this objective by using fine (~ 3 mum) filaments of NbTi in a wire with an interfilamentary matrix of CuMn to reduce proximity coupling and increase the transverse resistivity. The Rutherford cable have a thin stainless steel core and the wires will be coated with SnAg solder which has been oxidized, using a recipe similar to that developed at CERN, to increase the adjacent strand resistance Ra. Measurements of crossover strand resistance Re and Ra in cored cable with oxidized SnAg coating will be presented, together with data on critical current, persistent current magnetization and eddy current coupling in a wire with ultra fine filaments and a CuMn matrix in the interfilamentary region of the wire. These data will be used to predict losses and field distortion in the SIS 300 dipole and optimize the final design of cable for FAIR.

The synchrotron SIS100 is one of the two basic accelerators of the future Facility for Antiproton and Ion Research (FAIR) at GSI in Darmstadt. The SIS100 should provide acceleration of high intensity U28+ and proton beams for 0.5 s, with a pulse repetition rate of 0.6 Hz. In the accelerator magnetic system superferric 2 T dipoles of about 3 m length and 32 T/m quadrupoles of about 1 m length will be used. The magnet coils are made from hollow NbTi composite cable cooled with two phase helium flow at 4.5 K. The lattice comprises 108 dipoles and 168 quadrupoles. The elliptic beam pipe inner dimensions have been fixed at 130 times 60 mm2 for the dipole and 135 times 65 mm2 for the quadrupole. Both the dipole and quadrupole design approaches are based on improved versions of the original Nuclotron fast-cycling magnets and provide significantly less AC loss at 4.5 K, better quality of the magnetic field, and a higher long-term mechanical stability of the magnet coils. The results are based on the investigation of 1.4 m dipole and 0.4 m quadrupole Nuclotron-type magnet models. The status of the new magnets design and its manufacturing are presented. Essential features and new results are discussed.

The SIS 100 magnets of the Nuclotron-type are fast- pulsed superferric cold iron magnets. The dipole coil is a double layer saddle-shaped coil with 16 turns. The cable is a hollow tube NbTi composite cable of about 7.2 mm outer diameter, cooled with two phase helium flow at 4.5 K. A 2D structural analysis of the dipole cross section has been performed to determine the stress levels in the components of the coil package and to suggest an appropriate support structure for the conductor. Different design proposals with the focus on limiting stress in the CuNi cooling tubes and on feasibility of industrial manufacturing procedures have been conceived and analysed. The chosen coil support structure (substrate or "comb"), together with new insulation and modification of the winding scheme, seems to be an appropriate way to stabilize the coil better and offers important advantages like precise positioning of the conductor, more uniform distribution of loads on the CuNi tubes, an easier winding process on a guiding support, and simplified production procedures. Samples of the new coil structure were produced and tested. The obtained results are presented and discussed.

The last stage of the GSI Fast-Pulsed Synchrotron Project for FAIR (Facility for Antiproton and Ion Research) is the SIS 300 ring, which will use superconducting dipoles with 100mm aperture, 6 T magnetic field amplitude and 1 T/s field ramp rate. Stringent requirements on physical parameters and on the dependable service of the dipole necessitated using a cable with increased current carrying capability and low dynamic losses. A suitable geometry of the superconducting cable was found in that used for the outer layer of the Large Hadron Collider superconducting main dipole, supplemented by a stainless steel core for the reduction of the transverse contact resistance between the strands. The shape of the partially keystoned cable demands numerical methods for the optimization of the geometry in adaptation to the wide aperture dipole. The main characteristics of the optimized 2D and 3D geometry of the dipole are presented. Two grades of iron yoke steel were analyzed for use, in view of their resulting field quality. Thermal analysis of the dipole was carried out

New results from the investigation of 2 T superferric model dipoles operating at 4 T/s, 1 Hz are presented. The works are performed within the R&D program on the design of the SIS100 synchrotron at GSI in Darmstadt. One of the main research goals is minimization of overall AC power losses in the magnet at 4 K level. Different modifications of the dipole were proposed and tested experimentally. By the present time, the losses are reduced to about 17 W/m of the magnet length, while the reference Nuclotron dipole produces about 38 W/m in the same operating mode. Essential design details of the model magnets are described and further R&D steps are outlined

FAIR, a new international accelerator complex is now being developed by GSI (Darmstadt, Germany). The synchrotron SIS100 is a part of this complex and will be equipped exclusively with fast-cycling superferric magnets. The magnet system of this accelerator includes 168 identical quadrupoles with a maximum gradient of 33.4 T/m in an elliptical useable aperture of 120 mm times 65 mm. An important requirement of the quadrupole design is a field gradient uniformity better than plusmn0.06% for field gradient variation range of 3 T/m-3.3 T/m. This goal is achieved by introducing slots in the central part of the magnet poles, creating the effect of artificial saturation. Simultaneously, these slots reduce the eddy currents in the yoke, induced by the current ramp in the magnet coil. This is important because the whole magnet will be operated near 4.2 K, so the power loss in the yoke should be minimized. Additionally, the shape of the quadrupole pole ends was optimized to provide the necessary integral field quality

An important part of FAIR (Facility for Antiproton and Ion Research), being developed at GSI, Darmstadt, Germany, is the fast-ramping heavy ion synchrotron SIS 300. The main elements of the SIS 300 magnetic system are 6 T, 2.6 m long superconducting dipoles, to guide charged particle beams on a path with a 50 m curvature radius. A model superconducting dipole, manufactured for the UNK project, was chosen to study the influence of bending on the magnet characteristics. The straight collared coil of this model magnet was tested, then bent with a 50 m curvature radius, and then retested. The main measured and calculated characteristics of the coil for both tests are presented. Test results showed similar characteristics, both in training and in ramp rate dependence, as well as in field quality of the straight and bent coils. There is a good agreement between measured and calculated results of the completed study

The results of the technology development and first tests of the samples of the novel CICC are presented and discussed. The miniature CICC consists of thirty 0.5 mm NbTi strands inside a conduit with an outer diameter of 6.4 mm and has a projected operating current not less than 6 kA. The conductor design uses the well known and proven CICC technology. This new design looks very attractive, with several possible advantages, such as increased conductor stability margin and better mechanical rigidity of the conductor. Before N-CICC development the preliminary predictive analysis was done to justify the new conductor concept

The planned new international accelerator facility FAIR (Facility for Antiproton and Ion Research, ) at GSI will provide primary beams of all projectiles up to uranium with energies up to 1.5 GeV/u for nuclear structure physics. The maximum intensities of these projectile beams will be (1-3) 1012/s, leading to a maximum beam power of 58 kW. 10-20% of the primary beam will react in a high-power production target at the entrance of the Super-FRS (Super-conducting Fragment Separator, ). The Super-FRS is a large-acceptance two-stage fragment separator which will provide spatially separated isotopic beams of exotic nuclei. The non-reacting primary beam will be dumped at dedicated beam catchers, located at well defined places in the first part of the Super-FRS. The target area as well as the beam dump area is in high fluxes of high-energy particles, mainly neutrons and protons. Since magnetic elements located in these areas have to be reliably operated for the lifetime of the facility (~20 years) their design has to be radiation resistant. Presently we consider building the magnet coils using MIC (Mineral Insulation Cable) technology. In this paper we present radiation issues, magnet specification, and possible magnetic designs of the first quadrupole magnet placed 1 m downstream of the production target

The eddy-current losses in the yoke of a superconductive dipole magnet are simulated using a transient, magnetoquasistatic formulation discretized in 3D by the finite integration technique. The laminated yoke material is modeled by an anisotropic, nonlinear material characteristic. The system is linearized by the successive approximation technique with backtracking and solved by a curl-curl algebraic multigrid solver. The results show the complex interaction between ferromagnetic saturation and eddy currents

Fast cycling superferric magnets are planned for use in the new International Accelerator Facility for Antiprotons and Ion Research (FAIR) at GSI, Darmstadt. The efficiency of this magnet is basically defined by the AC loss at liquid helium temperatures in the construction elements of the dipoles and quadrupoles (iron yoke, coil, beam pipe restraints and suspension). A detailed knowledge of the 3D magnetic field, the eddy current distributions and their transient behavior is necessary to minimize the hysteresis and the eddy current losses through the use of an appropriate design. Methodical problems are considered for finite element calculations (ANSYS) of eddy currents in a laminated iron yoke. The results for window-frame dipoles and quadrupoles of the Nuclotron type are given. We present the influence of nonlinear and anisotropic magnetic and electrical properties of laminated steel and of bulk restraint elements

The data from the investigation of hollow superconducting cable based on keystoned NbTi wires are presented. The cable sample, consisting of 15 wires wound around a cooper-nickel tube 5 mm in diameter and cooled with two-phase helium flow has operated inside the magnet aperture at a current of 13.4 kA and the current ramp rate of 26.8 kA/s. The external magnetic field value was about 2 T and the field ramp rate of 4 T/s. The pulse repetition rate of 1 Hz was provided. The first 650 m of the new wire was manufactured based on the improved technology. A piece of hollow cable 40 m in length was fabricated. The outer cable diameter with electric insulation is 7.34 mm. A test winding for a 1.4 m superferric nucleotron-type window frame model dipole for SIS100 as part of the FAIR-project of GSI was fabricated and tested. The obtained results are discussed. The new approach to a 'cable-in-conduit' conductor cable design, based on the use of keystoned wires, is described.

Part of the HICAT medical accelerator, under construction at Heidelberg, Germany, is a huge gantry dipole. The magnet design was already reported at MT-17. This paper summarizes the main construction features. The geometric measurements are made with a laser tracker and include the determination of the iron core dimensions, the fiducialization and the alignment of the magnetic measurement equipment. Goal of the magnetic measurements was to shim the effective field boundary, to measure the integral field quality and the absolute integral field values. Dynamic field data in the central and in the fringe field are also reported.

A project to design and construct a new international accelerator was proposed by GSI (Darmstadt, Germany). This project includes the design of several new facilities. One of them is a collector ring (CR). The magnets of the CR operate at constant field (DC), but the level of this field varies for different operational tasks. The magnet system of the CR includes 24 deflecting dipole magnets. The maximum operating field in the aperture is 1.6 T, with a deflection angle of 15 degree, and a bending radius of 8.125 m. The most important requirement for this dipole is a high field uniformity (better than /spl plusmn/0.01% over a volume of 380 mm /spl times/ 150 mm). This requirement must be met for all levels of magnetic flux density in the aperture. Several magnet designs have been investigated including window frame and H-type versions with normal and superconducting coils. The best economical and technical parameters were obtained for a superferric H-type magnet with a wide air filter in the pole. The field homogeneity requirements were fulfilled in such a magnet for field range of 0.2-1.63 T. The effects of the field distortion in the end parts of the magnet, as well as influence of the end plates on the field quality, were also investigated.

Short models of the UNK superconducting dipole reached a maximum central field of 6.5 T with 0.8 T/s ramp rate. It allows one to consider this magnet design as a basis for development of a superconducting dipole, which will have low heat loss and could be used for the production of fast-cycling magnetic fields, with 6T amplitude and 1 T/s ramp rate, for the new GSI synchrotron SIS-300. The dipole geometry makes it possible to independently suppress both central field and integral field low-order harmonics. The specific shape of the coil end parts enables a decrease of the maximum field in the end parts to the level in the straight part of the coil. Analyzes of magnet AC heating during the SIS-300 cycle, as well as heat transfer to the helium coolant, are presented. The magnet temperature margin and methods to increase it are discussed.

New experimental results from the investigation of a superferric window frame type dipole magnet with the operating parameters: B=2T, dB/dt=4T/s and pulse repetition rate f=1Hz are presented. The magnet SC coil is made from a hollow multifilamentary NbTi cable, cooled with two-phase helium flow. Special attention is devoted to minimization of AC power losses in the magnet to the specified value of 13W/m. The value of 9W/m has been obtained for a model magnet with the yoke at T=50K. Different possibilities were investigated to reduce the losses in the case of a cold iron yoke (T=4.5K). Other problems, connected with the magnetic field quality, mechanical and cryogenic stability of the magnet under SIS100 operating conditions are also discussed.

The Heavy Ion Cancer Therapy Accelerator facility (HICAT) for the university hospital in Heidelberg is now under construction and will be completed in 2006. The facility consists of 2 ion sources, low-energy beam transfer lines, a linear accelerator, a medium energy beam transfer line, a synchrotron, high-energy beam transfer lines, and 3 treatment locations. One of the treatment locations will be equipped with a gantry. All magnets were designed at GSI. The design of the magnets will be reviewed. Wherever possible, a conservative design approach was used. Special design efforts were necessary for the huge 90/spl deg/ gantry dipole and the tiny internal quadrupoles for the IH-type drift tube linac, so that prototypes of these magnets were built and tested. The design, construction, and measurements of the quadrupoles for the IH-type drift tube linac will also be discussed in detail.

The new heavy ion synchrotron facility proposed by GSI will have two superconducting magnet rings in the same tunnel, with rigidities of 300 T /spl middot/ m and 100 T /spl middot/ m. Fast ramp times are needed. These can cause problems of ac loss and field distortion in the magnets. For the high-energy ring, a 1-m model dipole magnet has been built, based on the RHIC dipole design. This magnet was tested under boiling liquid helium in a vertical dewar. The quench current showed very little dependence on ramp rate. The ac losses, measured by an electrical method, were fitted to straight-line plots of loss/cycle versus ramp rate, thereby separating the eddy current and hysteresis components. These results were compared with calculated values, using parameters which had previously been measured on short samples of cable. Reasonably good agreement between theory and experiment was found, although the measured hysteresis loss is higher than expected in ramps to the highest field levels.

Dynamic processes in a window-frame dipole with superconducting windings and a cold, laminated iron yoke have been investigated experimentally at JINR (Dubna, Russia), and theoretically at GSI (Darmstadt, Germany). The main aim of these investigations was a reduction of energy losses in the yoke during ramping. These losses are produced mainly by energy dissipation due to eddy currents and hysteresis effects in the laminations. Both effects were investigated theoretically. As a result, special end blocks were proposed to reduce eddy current losses. For eddy currents simulation we used a model developed at GSI and St. Petersburg State Polytechnical University (Russia). This model is based on an integro-differential approach to eddy currents in laminated structures and a volume integral method for the magnetization vector in ferromagnetic objects. To test this model, a series of experiments were performed at GSI on a normal conducting Heavy Ion Synchrotron dipole magnet. The magnetic field in the aperture of this magnet was controlled with a point coil during the current ramping in the winding. The same transient processes were simulated using software, developed on the basis of the integro-differential method. The theoretical calculations were in reasonable agreement with the measurements.

Cored Nb-Ti Rutherford cables with high crossover resistance R/sub c/ are being investigated to minimize ac losses in a set of superconducting dipole magnets for the proposed rapid cycling heavy ion synchrotron facility at the Gesellschaft fur Schwerionenforschung (GSI) in Darmstadt, Germany. This ring of magnets is based on the Relativistic Heavy Ion Collider (RHIC) design. R/sub c/ in cored cables is significantly larger than the contact resistance between adjacent strands R/sub a/. The latter, however, dominates the voltage profile when the interstrand contact resistance is measured via the usual VI technique. This makes the extraction of the magnitude of R/sub c/ out of the voltage profile very difficult. Knowledge of R/sub c/ is important to predict ac losses, because if R/sub c/ and R/sub a/ were equal, perpendicular field induced losses due to R/sub c/ would be around 50 times greater than those induced by R/sub a/. Because of the importance of the crossover resistance, we have developed a novel measurement technique for R/sub c/ in cored Nb-Ti Rutherford cables. The procedure involves removing the edges of the conductors along the region to be measured. We have measured the cables using the new technique and the usual VI technique. We describe the novel measurement technique and present results for cables made with different core materials. While measuring R/sub c/, we found evidence that R/sub a/ is nonuniform over the length of a twist pitch and is in fact significantly lower at the cable edge. We report on this as well.

The proposed International Accelerator Facility at Gesellschaft fur Schwerionenforschung (GSI) has several new experimental devices that may require superconducting magnets to provide the required magnetic fields. Superferric magnets have been designed that allow the production of the required high gradients in the large aperture quadrupoles and also allow the insertion of sextupole and octupole inserts. Dipoles have been designed that fulfill the field uniformity and radiation resistance requirements. Most of the magnets are static field, but the New Experimental Storage Ring (NESR) needs magnets that ramp at 1 T/s. Solutions are presented that use warm iron to reduce losses.

Two new options for a hollow NbTi superconducting cable were considered. The first one is based on keystoned wires wrapped around a copper-nickel tube 5 mm in diameter. The second one is a hollow cable of rectangular cross section. The data from cable short sample tests are presented. Some problems with the production technology are discussed. This work is part of the R&D for the Future Accelerator Facility, at GSI in Darmstadt.

The concept for GSI's planned future facility is based on two superconducting synchrotrons, SIS 100 and SIS 200. The two accelerators are in the same tunnel and have the same radius R, for operation at BR=100 Tm and 200 Tm respectively. Superconducting magnets are necessary to reach the appropriate magnetic field and may considerably reduce the investment and operating costs, in comparison with conventional magnets. An R&D program was initiated to develop dipole magnets with maximum fields of 2 and 4 Tesla and dipole ramp rates of 4 T/s and 1 T/s, respectively. These requirements were chosen to achieve high average beam intensities. The SIS 100 dipole is a window-frame Nuclotron-type dipole and is being developed in collaboration with JINR (Dubna, Russia). This magnet has been operated at 4 T/s up to a field of 2 Tesla. Reduced losses and improved magnetic field quality are required for the SIS 100 accelerator. In a separate collaboration with BNL (Upton, USA), the one coil layer cos/spl theta/-type RHIC arc dipole, originally designed for operation at 3.5 Tesla with a rather slow ramp rate of 0.042 T/s, will be upgraded for the SIS 200 accelerator to operate at a ramp rate of 1T/s, up to a field of 4 T. R&D for a 6 Tesla dipole was started in collaboration with IHEP (Protvino, Russia), to further increase the rigidity of the SIS 200 ring to 300 Tm. Alternative schemes have been investigated. Besides the synchrotrons, the planned facility will consist of several storage rings and the Super Fragment Separator (SFRS), which have mainly DC magnets with large apertures. NSCL (East Lansing, USA) prepared a feasibility study for these superconducting magnets. The main results of the R&D are presented.

Experimental data of a fast cycling (f=1 Hz) 2T dipole magnet based on a superconducting NbTi multi filament hollow cable cooled with forced two phase helium flow at T=4.5K and iron yoke at T=80 K are presented. A new magnet design is proposed. The magnet yoke made of laminated steel consists of two parts: the internal smaller part has close mechanical and thermal contact with the coil while the outer part is separated from the cold mass with a gap of 1 mm and cooled with liquid nitrogen.

A new option of a superconducting dipole magnet for a fast cycling ( f =1 Hz) synchrotron is presented. The magnet is based on cos θ -type geometry, but its cold mass ( T =4.5 K) consisting of SC-coil and a reinforcing shell is separated from the iron yoke ( T =50–80 K) by a small vacuum gap. The arc-type one layer SC-coil is made of a hollow NbTi composite cable.

The new heavy ion synchrotron proposed by GSI will comprise two superconducting magnet rings in the same tunnel, having rigidities of 200 T.m and 100 T.m. Fast ramp times are needed, which can cause significant problems for the magnets, particularly in the areas of ac loss and field distortion. This paper discusses the 200 T.m ring, which will use Cos /spl theta/ magnets based on the RHIC dipole design. We describe options for the low loss Rutherford cable that will be used, together with a novel insulation scheme designed to promote efficient cooling. Measurements of contact resistance in the cable are presented and the results of these measurements are used to predict the ac losses, temperature rise and field distortion in the magnets during fast ramp operation.

A hybrid method, based on an integro-differential approach to eddy currents in laminated structures and a volume integral formulation for the magnetization vector, is applied to analyze dynamic processes in a synchrotron bending magnet. The results of the numerical simulation are compared with the experimental data.

Two new prototype dipole magnets for the proposed new fast cycling (f=1 Hz) synchrotron at GSI in Darmstadt have been designed, fabricated and tested. The magnets are based on a window frame iron yoke cooled by liquid nitrogen and a superconducting winding made from a hollow NbTi composite superconductor cable cooled with forced two-phase helium flow at T=4.5 K. The cold mass of the magnet is separated from the yoke by a small vacuum gap of 0.75 mm to 2.5 mm. A decrease of ac power losses by a factor of 2.3 in comparison with a standard Nuclotron dipole is obtained. The design features of two prototype dipoles as well as the test results are presented.

GSI plans to upgrade its accelerator facilities. A fast-pulsed synchrotron (rigidity 100 Tm, dipole field/ramp rate 2T/4T/s) is one of the main parts of the GSI future project. Superconducting magnets of the Nuclotron type are foreseen for this synchrotron. R&D work has been done in order to improve the DC field quality and to reduce the cryogenic AC losses. Linear 2-D calculations were used to optimize the pole shape of the dipole magnet. Consequently, a negative shimming was introduced. Subsequent nonlinear extension led to the introduction of air slits in the iron yoke which improved the field quality at higher levels. Nonlinear 3-D simulations were then used to optimize the homogeneity of the integral field, by varying the ratio between the yoke and coil length. We built a model magnet with this optimized iron lamination cross section. In order to reduce the AC losses, we used stainless steel end plates, low coercitivity iron, better insulated iron lamination sheets and reduced the superconductor filament size to 6 /spl mu/m. Various contributions to the losses of the magnet were analyzed. Numerical calculations of the eddy current effects due to field components perpendicular to the iron laminations at the end of the magnet showed that this part must not be neglected. The magnet test results are presented and compared with the expected field quality and losses.

Gesellschaft fur Schwerionenforschung (GSI) is planning to build FAIR (Facility for Antiproton and Ion Research) in Darmstadt, Germany . This facility will include SIS 300, a fast-ramping heavy ion synchrotron with a rigidity of 300 T-m, with 6 T, 100 mm coil aperture 2.6 m long superconducting dipoles, ramped at 1 T/s . The fast ramp rate requires a magnet design that minimizes AC losses and field distortions during ramping. A two layer cos /spl theta/ magnet design, using a cored Rutherford cable, has been chosen. The paper will present the features of the magnet design that is to be built, the expected AC losses and field quality, and the experimental and analytical work that supports the choice of this design.

The dipole magnet of the Heavy Ion Synchrotron SIS is operated at fields of up to 1.8 T. Thus the non-linearity of the magnetization and hysteresis effects of the order of 0.1% cannot be neglected. The effective length is not constant, but varies with the magnetization of the magnet. Eddy currents due to the planned high ramp rate of 4 T/s in the laminations, the magnet coil, and in the vacuum chamber affect the field quality of the magnet. Long time constants of almost one second due to field components orthogonal to the laminations were observed.

The magnets and power supplies of a raster scanning system for a particle therapy facility have been designed and carefully measured after delivery. Well known and proven techniques have been used in construction and production of the laminated window frame magnets and the pulsed power supplies to avoid failures of the system and to assure absolute security of patients during treatment. The scanning interval in both transverse directions is 20 cm. The vertical deflection includes an offset of 20 cm between the undeflected beam and the scanning area to have space for last beam diagnostic purposes just before beginning treatment. The scanning depth is established by varying the energy of the carbon, oxygen, or neon ions to energies up to 550 A MeV.

A versatile Magnet Measuring Device is described, that produces large maps of the three field components in a volume of 2.7 m/spl times/1 m/spl times/1 m. Its main components are: the machine itself, which is based on a large granite table with slideways and air bearing shoes; the probe arm (up to 4 m long); the 3D probe head with three orthogonal hallplates. The straightness and flatness of the table and slideways is /spl les/5//spl mu/m, the position accuracy is better than 50 /spl mu/m while measuring "on the fly" with the 4 m long probe arm. Four applications are discussed: two dipoles, an electron cooling device and a large solenoid detector. >

The new heavy ion synchrotron facility proposed by GSI will have two superconducting magnet rings in the same tunnel, with rigidities of 200 T/spl middot/m and 100 T/spl middot/m. Fast ramp times are needed, which can cause significant problems for the magnets, particularly in the areas of ac loss and field distortion. This paper discusses the 200 T/spl middot/m ring, which will use Cos/spl theta/ magnets based on the RHIC dipole design. We discuss the reasons for choosing Rutherford cable with a resistive core and report loss measurements carried out on cable samples. These measurements are compared with theoretical calculations using measured values of inter-strand resistance. Reasonably good agreement is found, but there are indications of nonuniformity in the adjacent resistance R/sub a/. Using these measured parameters, losses and temperature rise are calculated for a RHIC dipole in the operating cycle of the accelerator. A novel insulation scheme designed to promote efficient cooling is described.

Error fields in a dipole due to superconductor magnetization and conductor misplacements add unwanted multipole, mainly sextupole and decapole, terms to the desired dipole field. Two persistent mode sextupole correction coils inside the bore of model SSC dipoles have been built and tested. A shorted superconducting sextupole coil has a current induced in it by the error sextupole field such that no sextupole field can penetrate into the proton beam region. The correction sextupole coils are one layer thick and are wound from a single length of insulated composite Nb-Ti and copper wire 0.60 mm in diameter. Each of the six poles has ten turns and is mounted on a 1.75 cm radius stainless steel bore tube. Details of testing and trimming of the correction coils are described. Test results of the measured magnetic field within the model SSC dipoles with the correction coils in and out of persistent mode operation are presented. An electrical heater is used to drive the coil out of the persistent mode. Measurements of joint resistance and coil decay time constants are also given.

Prototypes of all the SIS (Heavy Ion Synchrotron) magnets were measured to check for possible field errors, to verify that all design specifications have been fulfilled, and to check for higher-order multipoles. Measurement results are presented and discussed for the dipoles, quadrupoles, sextupoles, and ring magnets. Test benches for the measurement of the ring magnetics are also described. >