Several indicators have pointed to the presence of an Electron Cloud (EC) in some of the CERN accelerators, when operating with closely spaced bunched beams. In particular, spurious signals on the pick ups used for beam detection, pressure rise and b
eam instabilities were observed at the Proton Synchrotron (PS) during the last stage of preparation of the beams for the Large Hadron Collider (LHC), as well as at the Super Proton Synchrotron (SPS). Since the LHC has started operation in 2009, typical electron cloud phenomena have appeared also in this machine, when running with trains of closely packed bunches (i.e. with spacings below 150ns). Beside the above mentioned indicators, other typical signatures were seen in this machine (due to its operation mode and/or more refined detection possibilities), like heat load in the cold dipoles, bunch dependent emittance growth and degraded lifetime in store and bunch-by-bunch stable phase shift to compensate for the energy loss due to the electron cloud. An overview of the electron cloud status in the different CERN machines (PS, SPS, LHC) will be presented in this paper, with a special emphasis on the dangers for future operation with more intense beams and the necessary countermeasures to mitigate or suppress the effect.
PyECLOUD is a newly developed code for the simulation of the electron cloud (EC) build-up in particle accelerators. Almost entirely written in Python, it is mostly based on the physical models already used in the ECLOUD code but, thanks to the implem
entation of new optimized algorithms, it exhibits a significantly improved performance in accuracy, speed, reliability and flexibility. Such new features of PyECLOUD have been already broadly exploited to study EC observations in the Large Hadron Collider (LHC) and its injector chain as well as for the extrapolation to high luminosity upgrade scenarios.
After a successful scrubbing run in the beginning of 2011, the LHC can be presently operated with high intensity proton beams with 50 ns bunch spacing. However, strong electron cloud effects were observed during machine studies with the nominal beam
with 25 ns bunch spacing. In particular, fast transverse instabilities were observed when attempting to inject trains of 48 bunches into the LHC for the first time. An analysis of the turn-by-turn bunch-bybunch data from the transverse damper pick-ups during these injection studies is presented, showing a clear signature of the electron cloud effect. These experimental observations are reproduced using numerical simulations: the electron distribution before each bunch passage is generated with PyECLOUD and used as input for a set of HEADTAIL simulations. This paper describes the simulation method as well as the sensitivity of the results to the initial conditions for the electron build-up. The potential of this type of simulations and their clear limitations on the other hand are discussed.
