LBNL, Berkeley, California, USA
Three-dimensional Poisson solver plays an important role in the self-consistent space-charge simulation. In this paper, we present several efficient 3D Poisson solvers inside an open rectangular conducting pipe for space-charge simulation. We describe numerical algorithm of each solver, show comparative results for these solvers and discuss the pros and cons associated with each solver.
LANL, Los Alamos, New Mexico, USA
During high-power ion-linac operations, the low-beam-loss operational settings are typically extrapolated from the low-power mode through highly subjective empirical adjustments. Existing simulation tools in accelerator control rooms are too simplified to handle the complex beam dynamics and the strong space charge effects, therefore providing no useful guidance for the high power beam tuning. We have been developing a GPU accelerated multi-particle beam dynamics simulator to try to bridge the gap. By combining the GPU technology and the multi-particle beam dynamics simulation algorithms, we have created a realistic beam simulator that is both accurate and fast enough to be useful in accelerator control rooms. Once connected to the EPICS control system, the simulator can rapidly respond to any control set point changes and predict beam properties along an ion linac in pseudo real time. Its applications include virtual diagnostics during operations, test-bed for new operation/control schemes, operation optimization and operator training.
Northern Illinois University, DeKalb, Illinois, USA
ANL, Argonne, USA
Space charge is a leading concern in high-intensity beams, causing effects such as emittance growth, beam halos, etc. As the need for high-intensity beams spreads, the demand for efficient space charge analysis grows. We developed a self consistent space charge simulation method for this purpose [*]. In order to facilitate space charge analysis, we implemented a method that allows space charge map extraction and analysis from any tracking method [*,**]. We demonstrate the method by calculating the transverse space charge. We compare the method of moments and the fast multipole method as the tracking methods employed in the transfer map extraction process. We show results from analysis of the raw map elements as well as quantities obtained from normal forms.
[*] Erdelyi, Nissen, and Manikonda. A Differential Algebraic Method for the Solution of the Poisson Equation for Charged Particle Beams.
[**] Berz. Modern Map Methods in Particle Beam Physics.
LBNL, Berkeley, California, USA
Fermilab, Batavia, Illinois, USA
SLAC, Menlo Park, California, USA
Thanks to sustained advances in hardware and software technologies, computer modeling is playing an increasingly important role in the design of particle accelerators. This rise in importance is further fueled by the economic pressure for reducing uncertainties and costs of development, construction and commissioning, thus pushing the field toward an increased use of “virtual prototyping”. Until now, the development of accelerator codes has been left to projects without mandate and programmatic funding for coordination, distribution and user support. While this is adequate for the development of relatively small-scale codes on targeted applications, a more coordinated approach is needed to enable general codes with user bases that extend beyond individual projects, as well as cross-cutting activities. In light of this, it is desirable to strengthen and coordinate programmatic activities of particle accelerator modeling within the accelerator community. This increased focus on computational activities is all the more timely as computer architectures are transitioning to new technologies that require the adaptation of existing - and emergence of new - algorithms and codes.
TUO2LR02
LBNL, Berkeley, California, USA
In this paper, we will report on the progress of development for end-to-end modeling of accelerators. We will discuss about some challenges in the study and suggest some potential solutions.
HU/AdSM, Higashi-Hiroshima, Japan
TUO4LR01
Fermilab, Batavia, Illinois, USA
RadiaSoft LLC, Boulder, Colorado, USA
ORNL, Oak Ridge, Tennessee, USA
UMD, College Park, Maryland, USA
Fermilab, Batavia, Illinois, USA
TUO4LR03
UMD, College Park, Maryland, USA
Scaled experiments at the University of Maryland Electron Ring (UMER) facility have a long history of successes in understanding space-charge dynamics. Recent experiments included beam halo and losses, resonances with space-charge, longitudinal space-charge waves, solitons, and longitudinal confinement with induction cells. This talk presents progress in the design of next-generation UMER-based experimental research, with expanded capabilities to study nonlinear lattices. The work is complementary and in collaboration with the proposed IOTA ring at FNAL. UMER offers 3 advantages for this type of research: (1) the ability to assess the effects of intermediate space charge strength relevant to conventional Hadron rings on the performance on nonlinear lattices; (2) its low-cost printed-circuit magnets (enabled by the 10 keV electron beam energy) can be easily replaced to test different lattice concepts; and (3) the detailed diagnostics already in place facilitate benchmarking of simulation codes in this new regime.
CERN, Geneva, Switzerland
ORNL, Oak Ridge, Tennessee, USA
PY-ORBIT is a broad collection of accelerator beam dynamics simulation models, written primarily in C++, but accessed by the user through Python scripts. PY-ORBIT was conceived as a modernization, standardization, and architectural improvement of ORBIT, a beam dynamics code designed primarily for rings. Although this goal has been substantially achieved, PY-ORBIT has now incorporated additional capabilities. A major consideration in high intensity beam dynamics codes, such as PY-ORBIT and ORBIT, is the simulation of space charge effects. Computational space charge simulation is, of necessity, accompanied by noise due to discretization errors, which can compromise results over long time scales. Discretization errors occur due to finite step sizes between space charge kicks, due to graininess of the numerical space charge distribution, and due to the effects of spatial grids embedded in certain solvers. In order to simulate space charge, most tracking codes use solvers containing some or all of these effects. We compare the manifestation of discretization effects in different types of space charge solvers with the object of long time scale space charge simulation.
KEK, Ibaraki, Japan
THO3LR03
BNL, Upton, Long Island, New York, USA
To compensate for the beam-beam effects from the proton-proton interactions at the two interaction points in the Relativistic Heavy Ion Collider (RHIC), two electron lenses (e-lenses) have been installed and commissioned in 2014. In this report, the physics of electron lens is briefly introduced, followed by the electron lens hardware and electron beam commissioning results in 2014 RHIC run. Although in 2014, RHIC is operating with gold and 3He beams, and the luminosity is not limited by head on beam-beam interactions, we still aligned the electron beam with the hadron beam to get the first experience with the electron-hadron beam interaction. The demonstration of electron and gold beam overlap has been achieved via electron backscattered detector, as well as the demonstration of electron beam parameters that are sufficiently stable to have no negative impact on the gold beam life time. With the experience of using electron lens on hadron beam, head on beam-beam compensation can be commissioned in the following year with proton beams, with a lattice which phase advance has a multiple of 180 degrees between the beam-beam interaction and electron lens locations.
BNL, Upton, Long Island, New York, USA
To search for the critical point in the QCD phase diagram, RHIC has been colliding gold ions at a variety of beam energies ranging from 2.5 GeV/n to 9.8 GeV/n. During these low energy operations below the regular injection energy, significant lifetime reductions due to the beam-beam interaction in conjunction with large space charge tune shifts have been observed. Extensive simulation studies as well as beam experiments have been performed to understand this phenomenon, leading to improved performance during the 7.3 GeV run in FY2014.