Diamond Light Source - Annual Review 2022/23

74 75 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 2 2 / 2 3 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 2 2 / 2 3 Diagnostics Developments Single-crystal CVD diamond (sc-diamond) detectors are an established transmissive synchrotron beamline diagnostic instrument used for X-ray beam position and beam intensity monitoring. Silicon carbide (4H-SiC) devices are recently commercialised alternatives which have the potential to provide the same diagnostic information but with amuch larger transmissive aperture (see Fig. 6). . The position-sensitive regions for the sc-diamond and 4H-SiC devices are 3mm x 3mm and 9mm x 3mm respectively. An experimental comparison of the performance of the two devices has been carried out, comparing signal uniformity across the surface of the detectors, kHz intensitymeasurements, and kHz beampositionmeasurements. Results have shown equivalent operational performance in beam position measurement and flux linearity such that the silicon carbide XBPMs are a promising instrument to add to the suite of diagnostics available to beamlines. Machine Operation and Development RichardWalker, Technical Director 2 022/23 was our 16 th year of operation, and unlike the previous two years which were affected by the pandemic, was carried out entirely in normal operating mode: 6-day running per week, from 09:00 Wednesday to 09:00 Tuesday interspersed with Machine Development days. A total of 214 days (5145 hours) were scheduled for User Mode operation, including 5 beamline start-up days. All operation was in standard multibunch mode (900 bunch train) with total current of 300 mA, apart from 6 days of “hybrid”mode in May 2022, consisting of a 686 bunch train with a high charge (3 nC) bunch in the middle of the dark gap. The annual operating statistics are shown in Fig. 1. The overall MTBF continued to be good at almost 112 hrs, and each of the five operating Runs achieved a MTBF in excess of 72 hrs, which is the target minimum. However, the 97.3% uptime was disappointingly below the target of 98%. Around two- thirds of beam trips were recovered in under 2 hrs, but with several longer outages for a variety of different reasons, the longest of which was 9 hours caused by a power outage, the overall Mean Time To Repair (MTTR) was 2.5 hours. Insertion Device Developments In June 2022 a specially extended 4-week machine shutdown took place to allow a major modification of straight 4 of the storage ring. Two new insertion devices were installed (see Fig. 2), a 2m in-vacuum cryogenic permanent magnet undulator (CPMU4) for the I04 beamline, replacing the original in- vacuum permanent magnet device, and a 1.6m in-vacuum hybrid permanent magnet undulator (HPMU) for the I04-1 beamline, replacing a 0.7m ex- vacuumpermanent magnet undulator. To create the extra space needed for the HPMU, the downstream girder had to be replaced with a new, shorter, girder. Beams from the two undulators were previously separated by a 3-magnet chicane. In order to preserve the same canting angles, with different spacing of the undulators, it was necessary to convert this into a 4-magnet chicane. The modification was a great success with both beamlines reporting significantly enhanced flux compared to previously. Another major piece of workwas the successful upgrade of the cryo-coolers on the first superconducting wiggler which is used for the I15 beamline, which was carried out by Insertion Device Group staff. Radiofrequency Developments Further progress has been made with the RF upgrade programme which is aimed at increasing the resilience of machine operation. The third normal conducting HOM-damped cavity has now been commissioned with its 120 kW solid-state amplifier and digital low-level RF system. The amplifier was installed on a platform between Control & Instrumentation Areas as shown in Fig. 3. This is now a complete RF system identical to the ones that will be used for Diamond-II, using similar platforms. Another development was the installation of a new pulser for the linac gun which produces a much ‘cleaner’ pulse train than the old pulser as shown in Fig. 4 and which also allows arbitrary bunch trains to be produced. The next development will be to use the new pulser to carry out multi-bunch top-up in user mode operation of the storage ring, greatly reducing the length of time the beam is disturbed compared to the current single bunch top-up. Power Supply Developments The 1200 magnet power supplies used on the Diamond accelerators continue to operate reliably with only one planned intervention that interrupted user operation to correct a fault, and no beam trips during user time. Two new power supplies associated with a CPMU and a HPMU in cell 4 were installed and commissioned. Development of new digital controller power supply cards continued (see Fig. 5), which will eventually replace the obsolete PSI controller and ADC cards. Twelve were firstly deployed on the Linac to provide operational experience on the actual machine. Subsequently the first production batches of 100 and 160 cards have been delivered and will be progressively installed in different locations, to confirm they can operate with all of the different types of power supply. Figure 1. Mean Time Between Failures (MTBF) and Uptime by operating year. Figure 2. Straight 4 in the storage ring showing the two new insertion devices, CPMU (left) for the I04 beamline and HPMU (right) for the I04-1 beamline. The new girder is visible following the undulators. Figure 3. CAD model showing the arrangement of the new RF cavity, its solid-state amplifier and connecting waveguide. Figure 4. Bunch trains produced by the electron gun using the old and new pulsers. Figure 5. The two new power supply controller cards, the Digital Power Supply card (left) and the System Interface Card (right). Figure 6: The two X-ray beam position monitors used for this comparison. Top: an XBPM using a 20 μm thick sc-diamond plate; Bottom: an XBPMwith a 10 μm thick 4H-SiC frame with 2.3 μm thick window.