ORNL, Oak Ridge, Tennessee
IUCF, Bloomington, Indiana
As SNS continues to ramp toward full intensity, we are acquiring a wealth of experimental data. Much effort is being applied to understand the details of the beam accumulation process under a variety of experimental conditions. An important part of this effort is the computational benchmarking of the experimental observations. In order to obtain quantitative agreement between the calculations and the observations, and hence a full understanding of the machine, a great deal of care must be taken to incorporate all the relevant experimental parameters into the calculation. These vary from case to case, depending upon what is being studied. In some of these cases, the benchmarks have been critical in unearthing flaws in the machine and in guiding their mitigation. In this paper we present the results of benchmarks with a variety of experiments, including coupling in beam distributions at low intensities, space charge effects at higher intensities, and a transverse instability driven by the impedance of the ring extraction kickers.
GSI, Darmstadt
ORNL, Oak Ridge, Tennessee
IAP, Frankfurt am Main
CEA, Gif-sur-Yvette
The GSI Univeral Linear Accelerator UNILAC can accelerate all ion species from protons to uranium. Hence its DTL section is equipped with e.m. quadupoles allowing for a wide range of field strength along the section. During the last years various campaigns on the quality of high current beams at the DTL exit as function of the applied transverse focusing have been performed. Measurements were compared with up to four different high intensity beam dynamics codes. Those comparisons triggered significant improvement of the final beam quality. The codes were used to prepare an ambitious and successful beam experiment on the first observation of a space charge driven octupolar resonance in a linear accelerator.
JLAB, Newport News, Virginia
In this paper, we present benchmarking results for high-class 3D electromagnetic (EM) codes in designing RF cavities today. These codes include Omega3P [1], VORPAL [2], CST Microwave Studio [3], Ansoft HFSS [4], and ANSYS [5]. Two spherical cavities are selected as the benchmark models. We have compared not only the accuracy of resonant frequencies, but also that of surface EM fields, which are critical for superconducting RF cavities. By removing degenerated modes, we calculate all the resonant modes up to 10 GHz with similar mesh densities, so that the geometry approximation and field interpolation error related to the wavelength can be observed.