Experimental studies and Monte Carlo simulations for beam loss monitors

Abstract

Beam loss detection is essential for the machine protection and the fine-tuning of the accelerator to reduce the induced radioactivity. Monte Carlo simulations are also vital for the choice of beam loss monitor (BLM) type, for predicting and understanding the BLM response to beam losses. At the China Spallation Neutron Source (CSNS), the cylindrical ionization chamber (IC) filled with $\mathrm{Ar}/{\mathrm{N}}_{2}$ gas mixture is the main type of the BLM to detect the beam losses. This paper presents the detailed beam loss experiments and fluka simulations for the CSNS BLM system. It includes the response functions of the BLM detectors by taking into account the contributions of different secondary particles to the energy deposition in the sensitive volume of the BLM. Dedicated experiments were compared with fluka simulations. It was found that in the low energy section of the linac the usual ion chamber based BLMs are not sensitive enough. A BLM based on ${\mathrm{BF}}_{3}$ enclosed by a high-density polyethylene (PE) moderator is effective to detect the beam losses through detecting thermal neutrons. Its signal is about 3 orders of magnitude higher than that of $\mathrm{Ar}/{\mathrm{N}}_{2}$ BLM for a beam energy of 15 MeV. The simulations provide the results of the intrinsic delay time for the ${\mathrm{BF}}_{3}$ monitor and the thickness optimization for PE to reduce the delay. Finally, the simulated spatial resolution of loss location by detection of neutrons is also evaluated for a beam energy of 15 MeV, which presents a resolution of $\ensuremath{\sim}2\text{ }\text{ }\mathrm{m}$ in our experimental configuration.