Path to beam loss reduction in the SNS linac using measurements, simulation and collimation
Explore the source record for details and available documents.
SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
At the Interaction Point (IP) of the SLC Final Focus, beam-beam deflection scans routinely provide a measurement of the sum in quadrature of the electron and positron transverse beam sizes, but no information on the individual beam sizes. During the 1996 SLC run, an upgrade to the Final Focus beam position monitor system allowed a first measurement of the absolute beam energy loss of both beams on each step of the deflection scan. A fit to the energy loss distributions of the two beams provides a measurement not only of the individual transverse beam sizes at the IP but also of the individual bunch lengths.
Energy losses of a gas cluster ion beam (GCIB) after collisions with residual gases were studied for Ar and CO{sub 2} GCIBs. From the energy distribution of an Ar GCIB after collisions at various acceleration voltages and cluster sizes, it is found that GCIB lost its energy when its energy per atom was high. In addition, the energy loss of CO{sub 2} GCIB was moderate compared to that of Ar GCIB at the same cluster size because of strong binging energy of CO{sub 2} molecules. The energy loss mainly occurs as decrease of its cluster size. The number of the scattered atoms or molecules linearly increased with the energy per atom or molecule. The reduction of cluster size is determined by the energy/atom and the cohesive energy of the gas.
Here, neutral beam injection is a major auxiliary heating method in the EAST experimental campaign. This paper gives detailed calculations of beam loss with different plasma equilibria using the guiding center code ORBIT and NUBEAM/TRANSP. Increasing plasma current can dramatically lower the beam ion prompt loss and ripple loss. Countercurrent beam injection gives a much larger prompt loss fraction than co-injection, and ripple-induced collisionless stochastic diffusion is the dominant loss channel.
Purpose: Due to the close proximity of the linear accelerator (linac) to the magnetic resonance (MR) imager in linac-MR systems, it will be subjected to magnet fringe fields larger than the Earth's magnetic field of 5x10{sup -5} T. Even with passive or active shielding designed to reduce these fields, some magnitude of the magnetic field is still expected to intersect the linac, causing electron deflection and beam loss. This beam loss, resulting from magnetic fields that cannot be eliminated with shielding, can cause a detuning of the waveguide due to excessive heating. The detuning, if significant, could lead to an even further decrease in output above what would be expected strictly from electron deflections caused by an external magnetic field. Thus an investigation of detuning was performed through various simulations. Methods: According to the Lorentz force, the electrons will be deflected away from their straight course to the target, depositing energy as they impact the linac copper waveguide. The deposited energy would lead to a heating and deformation of the copper structure resulting in resonant frequency changes. PARMELA was used to determine the mean energy and fraction of total beam lost in each linac cavity. The energy deposited into the copper waveguide from the beam losses caused by transverse magnetic fields was calculated using the Monte Carlo program DOSRZnrc. From the total energy deposited, the rise in temperature and ultimately the deformation of the structure was estimated. The deformed structure was modeled using the finite element method program COMSOL MULTIPHYSICS to determine the change in cavity resonant frequency. Results: The largest changes in resonant frequency were found in the first two accelerating cavities for each field strength investigated. This was caused by a high electron fluence impacting the waveguide inner structures coupled with their low kinetic energies. At each field strength investigated, the total change in accelerator frequency was less than a manufacturing tolerance of 10 kHz and is thus not expected to have a noticeable effect on accelerator performance. Conclusions: The amount of beam loss caused by magnetic fringe fields for a linac in a linac-MR system depends on the effectiveness of its magnetic shielding. Despite the best efforts to shield the linac from the magnetic fringe fields, some persistent magnetic field is expected which would result in electron beam loss. This investigation showed that the detuning of the waveguide caused by additional electron beam loss in persistent magnetic fields is not a concern.
Dedicated experiments were performed on MAST Upgrade to study beam-ion losses caused by charge exchange (CX) with edge neutrals. The fuelling was switched from the high-field side to the low-field side mid-discharge. Direct measurements suggest a strong increase in the neutral density around the plasma and a decrease in the beam-ion density, which is qualitatively explained by CX losses. Measurements by a resistive bolometer have suggested particle bombardment during neutral beam injection, providing a unique opportunity to separate CX from other loss mechanisms. To verify and quantify CX losses, the orbit-following code ASCOT, which accounts both for CX neutralization and reionization, was used to simulate beam-particle power loads on the bolometer. Simulations reproduce measured bolometer power loads during high-field-side fuelling, verifying CX losses of approximately 10% of the off-axis beam power. Toroidally symmetric simulations overestimate power loads on the bolometer during low-field-side fuelling, which is explained by toroidal asymmetry in the neutral density distribution, as is demonstrated by toroidally asymmetric simulations. Results suggest significantly higher CX losses during low-field-side fuelling, up to about 50% of off-axis beam power.
Prompt loss of beam injected fast ions approaches 100% in lithium tokamak experiment-beta (LTX-β) discharges, though significantly improved confinement is expected for the higher current plasmas made available by a recent upgrade to the Ohmic heating power supply. Modeling of fast ions using TRANSP/NUBEAM finds a maximum coupled beam fraction of 76% at the near-term limits of the LTX-β operating space. The full ion orbit code POET is employed to validate NUBEAM results against possible non-adiabatic effects on fast ion orbits, but corrections to the prompt loss fraction due to collisionless transport are found to be small. The graphical method code CONBEAM is used to investigate the topology of fast ion phase space as it relates to neutral beam deposition, and counter-injected NBI is considered as a way to access a region of high field side beam deposition. A metric is developed within the CONBEAM using a beam filament model to estimate the prompt loss fraction and shown to agree well with both POET and NUBEAM, enabling near real-time analysis and potential feedback to operators between plasma discharges.
Since the start of neutron production in October of 2006, the average SNS proton beam power level has increased from ~ 5 kW to over 500 kW. This increased has been realized by increases in the beam current, pulse length and repetition rate. A major concern in the power ramp up has been minimization of uncontrolled beam loss. The beam loss levels, and operational experience with residual activation will be discussed. Also the operational run availability is discussed.
In Korea, a heavy ion accelerator facility (RAON) has been designed for production of rare isotopes. The 90° bending section of this accelerator includes a 1.3-μm-carbon stripper followed by two dipole magnets and other devices. An incident beam is 18.5 MeV/n 238 U 33+,34+ ions passing through the carbon stripper at the beginning of the section. The two dipoles are tuned to transport 238 U ions with specific charge states of 77+, 78+, 79+, 80+ and 81+. Then other ions will be deflected at the bends and cause beam losses. These beam losses are a concern to the devices of transport/beam line. The absorbed dose in devices and prompt dose in the tunnel were calculated using the FLUKA code in order to estimate radiation damage of such devices located at the 90° bending section and for the radiation protection. A novel method to transport multi-charged 238 U ions beam was applied in the FLUKA code by using charge distribution of 238 U ions after the stripper obtained from LISE++ code. The calculated results showed that the absorbed dose in the devices is influenced by the geometrical arrangement. The maximum dose was observed at the coils of first, second, fourth and fifth quadruples placed after first dipole magnet. The integrated doses for 30 years of operation with 9.5 pμA 238 U ions were about 2 MGy for those quadrupoles. In conclusion, the protection of devices particularly, quadruples would be necessary to reduce the damage to devices. Furthermore, results showed that the prompt radiation penetrated within the first 60 - 120 cm of concrete.
Beam ion loss from the Compact Helical System (CHS) has been measured with a scintillator-type probe. The total loss to the probe, and the pitch angle and gyroradius distributions of that loss, have been measured as various plasma parameters were scanned. Three classes of beam ion loss were observed at the probe position: passing ions with pitch angles within 10o of those of transition orbits, ions on transition orbits, and ions on trapped orbits, typically 15o or more from transition orbits. Some orbit calculations in this geometry have been performed in order to understand the characteristics of the loss. Simulation of the detector signal based upon the following of orbits from realistic beam deposition profiles is not able to reproduce the pitch angle distribution of the losses measured. Consequently it is inferred that internal plasma processes, whether magnetohydrodynamic modes, radial electric fields, or plasma turbulence, move previously confined beam ions to transition orbits, resulting in their loss.