A single bunch selector for the next low � continuous wave heavy ion beam

Abstract

The MEBT line of EURISOL heavy ion postaccelerator and the SPIRAL2 deuton/ion accelerator should transport a continuous wave (cw) beam from a 88.05 MHz RFQ (! respectively 0.035 and 0.04) to a drift-tube linac. A high frequency chopper is being studied to select only 1 bunch over N, respectively 10 < N and 100 < N as asked by the physicists. It requires pulse voltages up to 2.5k kV, rising in less than 6 ns at a repetition rate up to 8.8 MHz. These figures are at the border of what can be provided by the travelling wave fast choppers and the capacitive-type chopping technologies [1,2,3,4]. We have reviewed the current fast and slow chopping structures and their associated pulse generator [5]. Some preliminary RF simulations to adapt the present chopping devices to our requirements are presented. The main limitations of these technologies when applied to select one bunch in cw ion accelerators are also shown. Our first studies and results to solve the arising problems are discussed. $%-/.7*+-$.%"The MEBT line of the EURISOL post-accelerator and the SPIRAL2 accelerator should transport continuous wave (cw) beams from a RFQ operating at the 88.05 MHz frequency ("%!= 11.36 ns) to a drift tube linac (DTL). Both superconducting linacs are designed to have effective acceleration respectively for radioactive heavyions with a ratio A/q from 4 to 8, and either for deuton or heavy ions with a A/q of 3 and 6. The maximum beam current is respectively 1 !"!and 5 mA. The space charge effects are negligible in the EURISOL case. Many EURISOL experiments will require to chop the beam for periods from 10 ns to 1 ms, which will be made by means of a high frequency chopper located in the MEBT line. Due to the physicist requirements the EURISOL and SPIRAL2 choppers should operate like a single bunch selector: one bunch over N should pass through the chopper output with the repetition rate resumed in table 1.After having adapted existing 50-# travelling-wave (TW) electrodes to our cases, power constraints will be observed due to the cw mode of the beam and the physicist requirements. Two alternative ways might answer the requirements. We also decided to investigate another way to decrease power consumption: a 100-# meander microstrip line. The length of the strip line can lead to attenuation and dispersion of the pulse signal along its propagation. +2"5.7("+.%#-/!$%-#""Fast rising times can be achieved either by travelling wave structures [2,3,4] or by capacitive-type (C-type) chopper [6]. This scheme is used when rise time lower than 10 ns are searched. Chopper structure working at voltage level similar to those required for EURISOL are being studied for proton injectors foreseen to operate in the 300 to 400 MHz frequency range and within rise/fall time of the deflecting pulse shorter than 2 ns. In preliminary studies the 50-# SPL electrode was adapted to EURISOL and SPIRAL2 cases. For EURISOL and SPIRAL2, N-1 bunches over N should be suppressed where N exceeds respectively 10 and 100. These values imply a duty cycle of the high vo...

Fast beam chopper at SARAF accelerator via RF deflector before RFQ

“…More recently, chopping schemes have been developed for next generation high intensity proton driver systems, primarily to adopt linac beams to the beam accumulators and/or synchrotron accelerators [3][4][5]. These chopping systems are mostly implemented in the MEBT sections where the ion energy is of several MeV and the velocity of a few per-cent of the speed of light will match the phase velocity of deflecting elements comprising of slow wave structures such as meander type delay lines.…”

Proton beam commissioning of the SPIRAL2 Single Bunch Selector

“…The SBS system is based on the architecture studied for the post accelerator of the Eurisol Design Study [1]. It uses a static magnetic field to bend the beam out of axis, the initial deviation is increased by the following quadrupoles and the beam is stopped on a scraper able to handle full beam power.…”