AVS2018 Session VT-TuM: Large Vacuum Systems and Accelerator Vacuum Technology

The LCLS-II (Linac Coherent Light Source II), a revolutionary new X-ray laser, is currently being built at SLAC National Accelerator Laboratory. The LCLS-II Vacuum System consists of a combination of particle free and non-particle free areas and includes the beamline vacuum, RF system vacuum, cryogenic system vacuum and supporting systems vacuum. The Vacuum Control System includes the controls and monitoring of different types of gauges, pumps and residual gas analyzers (RGA). The Vacuum Control System uses Allen Bradley PLC (Programmable Logic Controller) to perform interlocking to isolate bad vacuum areas. In particle free areas, a voting scheme is implemented for slow and fast shutter interlock logic to prevent spurious trips. Additional auxiliary control functions and high level monitoring of vacuum components is reported to global control system via an Experimental Physics and Industrial Control System (EPICS) input output controller .This paper will discuss the design and status of the LCLS-II Vacuum Control System.

A sextant of Cornell Electron Storage Ring (CESR) is currently being upgraded with Double Bend Achromat (DBA) lattice and CHESS Compact Undulators (CCUs) in order to significantly boost the performance of Cornell High Energy Synchrotron Source (CHESS). With this upgrade, dubbed CHESS-U, CESR is converted from a counter-propagating two-beam ring to a single-beam ring, and the beam energy will be increased from 5.3 GeV to 6 GeV. The beam current for normal operation will also be increased, from 120 mA to 200 mA. The beampipe aperture of this new section is much smaller than the rest of CESR, and therefore the beam pipes and vacuum pumping are quite different from the rest of CESR too.

The beam pipes are mainly made of three types of aluminum extrusions, fitting inside quadrupole magnets, dipole magnets, and undulators respectively. The dipole extrusion includes an ante-chamber for distributed pumping using NEG strips (SAES St 707). An exception is the dipole chamber from which the undulator X-ray beam exits. This chamber is instead made of two machined halves that are welded together.

In this presentation, we will give a brief overview of the design of the vacuum system, and then show some important details in chamber production and installation, such as extrusion bending by stretch forming, chamber welding and distortions, NEG strip pumping, etc. Base vacuum performance will be reported as well. Details regarding the crotch absorber will be reported in a separate presentation.

A major upgrade project (CHESS-U) will elevate performance of Cornell High Energy Synchrotron Source (CHESS) to a state-of-the-art 3^rd generation light source. In the CHESS-U project, ~80 meters of Cornell Electron Storage Ring (CESR) will be upgraded with a Double Bend Achromat (DBA) lattice and CHESS compact undulators (CCUs), together with new CHESS front-ends and X-ray beamlines. Tremendous progresses have been made in the design, construction, and test of all required new accelerator components (such as magnets, vacuum chambers, etc.), and we are on schedule for the CHESS-U installation in late 2018.

One of the most challenging vacuum components is the crotch absorbers at the synchrotron radiation (SR) exit ports at the ends of dipole bending chambers. CHESS-U is designed for 250-mA maximum beam current at 6-GeV beam energy. The crotches need to safely absorb 5.1 kW of SR power generated by the dipole magnets at small angles of incidence, while allowing the undulator beams produced by the canted CCUs to pass. We present a design for a compact crotch which can effectively dissipate this high SR power and will be compatible with the ultra-high vacuum system of CESR.

The crotch is a composite of a 5-mm thick beryllium outer ring, which acts as a SR diffuser, and an axially water cooled copper cylinder, similar to the existing CHESS crotches that have been in operation for decades. The outer beryllium ring is bonded to the copper cylinder via vacuum furnace braze using Cusiltin (60% Ag - 30% Cu - 10% Sn) braze alloy. Thermal analysis has shown no surface temperatures higher than 325°C. However, the calculated maximum thermal stress (on the surface of the beryllium) is near beryllium’s ultimate strength. As a practical validation of the CHESS-U crotch design, an existing CHESS crotch was removed from CESR for microscopic inspection. Though the CHESS crotch was exposed to slightly higher SR power (5.7 kW) during operations, no sign of fracture, surface melting, or other visible defect was detected on the beryllium surface. To ensure the good bonding of the beryllium diffuser ring to the copper cylinder, scanning X-ray fluorescence (XRF) inspections of the brazed crotches were performed at CHESS A1 beamline. The XRF inspection verified the presence of a continuous braze layer (corresponding to 10-30 μm thickness of Cusiltin) between the beryllium and the copper, indicating good braze joint. Five CHESS-U crotch assemblies have been successfully fabricated and XRF inspected, to fulfill the need for 4 crotches required for the CHESS-U project.

The objective of the KATRIN neutrino experiment is the determination the effective mass of electron anti-neutrinos with an unprecedented sensitivity of 0.2 eV/c² by measuring the energy spectrum of b-electrons from tritium decays close to the endpoint of the spectrum. The decays take place in a high-intensity windowless gaseous tritium source (WGTS). The electrons are guided by strong magnetic fields through a beamline with super-conducting solenoids to the huge main spectrometer for energy measurement. Since only a fraction of 2·10^-13 of all electrons have a kinetic energy within 1 eV below the spectral endpoint, the sensitivity of the measurement depends on a low background rate in the spectrometer. Therefore the tritium flow from the source (10^-3 mbar) has to be reduced by at least 14 orders of magnitude before it reaches the spectrometer section. This large reduction in the 90-mm-diameter beamline is achieved with a combination of a differential pumping section (DPS), using turbo-molecular pumps (TMP) and a cryogenic pumping section (CPS) with an argon frost layer at around 3 K.

This talk describes the Test Particle Monte Carlo (TPMC) simulation of the beamline for the differential and cryogenic pumping sections both for the non-radioactive gas deuterium and for tritium with MolFlow+. Since the gas flow through the CPS is reduced by far more than 7 orders of magnitude, the simulation is done in several consecutive steps. The results are combined in the post processing of the TPMC results. In the post processing algorithm we also investigate the time dependence of the reduction factor by considering a finite sojourn time of the molecules on the cryogenic argon layer. This leads to a slow migration of deuterium in downstream direction. In a final step the half-life and migration of tritium is taken into account by a reduced sojourn time.

The NSLS-II facility now has 23 beamlines in operation and 5 others nearing operational readiness. This talk will focus on general vacuum requirements common to all synchrotron beamlines as well as hardware and design considerations needed to meet them. Topics to be covered include: beamline vacuum system design, leak checking, bakeout, commissioning, and operation. Particular attention will be given to the successes and challenges experienced to date.

Next generation pulsed power systems require a better understanding of current loss and the ability to predict the scaling of current coupling. Plasma formation occurs in these devices at both the anode and cathode at high current densities. These plasmas are formed from desorbed and ionized surface and bulk electrode contaminants, expand into the anode-cathode gaps of the transmission lines (a phenomenon referred to as plasma closure), and potentially lead to shunting of current away from the fusion targets. Contaminants are pervasive on typically manufactured hardware and common on pulsed power vacuum systems. Understanding the physics of neutral desorption from electrode surfaces is a critical requirement for accurately predicting plasma formation (current loss) in pulsed power machines, We will present Temperature Programmed Desorption (TPD) results of common electrode materials, including machined wrought stainless steel and additively manufactured 304L stainless steel.

Sandia National Laboratories is a multi-mission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC., a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Vacuum chamber RF surface resistance may play a significant role in beam instabilities of charge particle accelerators. Several dedicated choked TE010-mode resonators with 2 and 3 chokes were designed and built in ASTeC for surface resistance measurements of planar samples.

In this paper we report our study on improving the accuracy of surface resistance measurements and possible sources of errors. Three 7.8 GHz cavities made of copper, aluminium and niobium and different well known samples were employed to prove this method and demonstrate the errors. Good agreement between measured surface resistance values and those predicted by the employed theoretical model demonstrated the suitability of the described technique for RF surface resistance measurements.