Determination of the Magnetic Characteristics in the Injection Septum for the Metrology Light Source*

The pre-accelerator microtron supplies an electron beam at 105 MeV for the Metrology Light Source (MLS) of the Physikalisch-Technische Bundesanstalt (PTB) in Berlin. The beam is delivered via the transfer line to the injection septum and then into the storage ring. This septum magnet has its stainless steel vacuum beam pipe placed inside a laminated silicon iron magnet core. Hence, the pulsed magnetic field (half sine) used for the beam deflection must propagate through the thin metallic beam pipe. During the commissioning of the injection process, it became apparent that the calculated nominal pulse current for this energy and geometry had to be increased by 30 % to achieve proper beam transfer and accumulation. Two problems were apparent. Firstly, the injected beam trajectory had to be set at an angle away from the main beam axis. Secondly, the beam transfer from the septum entrance to exit was disturbed. As a first measure, the septum current pulse length was extended from 35 μs to 107 μs. Further on, the septum magnet was insulated from the transfer line beam pipe by a ceramic brake. This paper reports on measurements of pulsed magnetic fields inside the septum magnet and the improvements with the injection process. THE MLS STORAGE RING INJECTION The septum magnet is placed in a straight section of 2.5 m where the injection line leads into the MLS storage ring. Figure 1: Schematic of 1⁄2 MLS storage ring with injection Two kicker magnets are symmetrically attached to it where two other ones are placed in 6 m straight sections facing the ends of the achromatic magnet bows (Fig. 1). For efficient electron beam accumulation, decaying magnetic leakage field must not exceed theoretical ascertained limits. These are specified as maximal permissible integral induction along the septum rail and with local pick-up coil (Table 1). While the injection system can deliver beam at 10 Hz repetition rate, the most efficient and highest value beam accumulation was achieved with 2 Hz repetition rate for the septum system. Table 1: Specified Septum Design Parameters Parameter / Description Value Deflection angle α = 20° Bending radius ρ = 2 m Mechanical aperture 9.4 mm • 10.4 mm Magnetic length 700 mm ± 1.5 mm Mechanical width 800 mm ± 0.8 mm Injection energy (max. specified) E = 120 MeV Max. permissible integral induction in antechamber, along septum rail 5 • 10 Tm Max. permissible induction in antechamber, with local pick-up coil 1 • 10 T Hardware Integration of a Ready-Made System As the stray field requirements were stringent, it was decided to either purchase a system fitting the specifications or to take the effort for an in-house design based on previous experiences [1]. Because of the project progress and constraints in manpower, the preferred solution was to order an already developed septum system from the manufacturer Danfysik. Comparable septa showed good performance in other synchrotron light sources [2]. The septum passed factory acceptance tests and was installed in the MLS storage ring [3]. First Stage Commissioning Experience For the first accelerator commissioning term, the septum was powered with a short half-sine septum pulse current of τ = 35 μs length as ordered. The intention for the specified very short pulse was to reduce stray fields outside the septum rail effectively and permit eddy currents damping in the septum rail and attached beam pipe. At this stage, the required pulse current amplitude was as high as Î = 2380 A for successful beam transfer through the septum magnet, about 30 % higher than the nominal value. The injected beam appeared distorted on ___________________________________________ *Work supported by Physikalisch-Technische Bundesanstalt, Abbestr. 2 12, 10587 Berlin, Germany Proceedings of IPAC’10, Kyoto, Japan TUPEC026 02 Synchrotron Light Sources and FELs T12 Beam Injection/Extraction and Transport 1773 the fluorescent screen after the septum. To mitigate these effects, the pulse current length was extended to the technical limit of the pulser unit (τ = 107 μs). The beam spot on the fluorescent screen, after the septum, was more focused (Fig. 2 left: short 35 μs, right: long pulse 107 μs). Figure 2: Screen after septum; for short and long Î pulse. The pulse amplitude was Î = 1780 A for good injection conditions, near to the nominal value. The beam moved to the center, its distortion was reduced. Additional specifics were suspected as origins of the outward directed injection angle for optimal beam accumulation conditions. Eddy Currents in Beam Pipe and Lamination Since there was only one installed septum magnet available, but no test stand yet, the idea was that the eddy currents in the thin metallic beam pipe and in the lamination could be the primary reasons for the observed effects. The vacuum system could not be broken for field measurements at that time. However this assumption was in contradiction to the successfully passed factory tests. The penetration depth for the 0.3 mm thin stainless steel injection channel pipe was estimated with equation (1). The calculated eddy currents could not explain the described injection and accumulation problems [4].