Electron cloud effects on the LHC beam dynamics
Electron cloud effects on the LHC beam dynamics A. Romano, G. Iadarola, G. Rumolo Many thanks to: D. Cesini, P. Dijkstal, K. Li, E. Mètral, M. Schenk and INFN-CNAF institute in Bologna A. Romano et al. ABP Information Meeting 1
Outline • Basics on the electron cloud in particle accelerators Ø Electron cloud buildup and main effects on the beam dynamics • Simulation studies Ø Observations in the LHC vs simulation results • Conclusions A. Romano et al. ABP Information Meeting 2
Electron cloud build up • Primary (seed) electrons are generated inside beam chamber (gas ionization, photoemission) • Seed electrons are accelerated by beam field and produce secondary electrons when hitting the wall • If the Secondary Electron Yield (SEY) of the surface is large enough, it can drive an avalanche electron production exponential growth of electron density (multipacting regime) A. Romano et al. ABP Information Meeting 3
Electron cloud build up • Primary (seed) electrons are generated inside beam chamber (gas ionization, photoemission) • Seed electrons are accelerated by beam field and produce secondary electrons when hitting the wall • If the Secondary Electron Yield (SEY) of the surface is large enough, it can drive an avalanche electron production exponential growth of electron density (multipacting regime) • Electron distribution within the chamber is strongly influenced by the magnetic field LHC Arc Dipole A. Romano et al. LHC Arc Quadrupole ABP Information Meeting 4
Electron cloud effects The presence of an EC inside an accelerator ring can be revealed by several typical signatures Ø Machine observables • Heat load on the chamber’s walls • Vacuum degradation • Fast pressure rise and outgassing Ø Beam observables • Transverse instabilities & emittance growth • Tune shift & spread • Incoherent beam losses Understanding of beam observables relies on Py. ECLOUD-Py. HEADTAIL simulations • Simulations are very demanding in terms of time and computational resources typical study requires hundreds of CPUs (8 CPUs per job) and 3 -4 week to simulate 10 4 turns (INFN-CNAF clusters used for this purpose) • Recent work has been focused on increasing the performance of our simulation tools (1) G. Iadarola et al, “Evolution of python tools for the simulation of electron cloud effects”, in Proceedings of the 8 th International Particle Accelerator Conference, Copenhagen, Denmark, THPAB 043 (2017) A. Romano et al. ABP Information Meeting 5
Simulation studies An extensive simulation campaign has been carried out in order to improve the understanding of machine settings on the instability induced by the EC very long simulation run needed to approach the time scale of the observed instabilities Main effects studied (1) • EC in the main LHC arcs (dipoles and quadrupoles) • Beam energy • Different machine settings (chromaticity, octupoles and transverse feedback) Vert Chromaticity Beam energy effect Chromaticity effect (1) A. Romano et al. , “Electron cloud induced instabilities in the LHC”, presentation at the Joint Ecloud-Py. HEADTAIL Meeting, 12 May 2017, https: //indico. cern. ch/event/638087/ (2017) A. Romano et al. ABP Information Meeting 6
Observations vs simulation results 2015 (after scrubbing run): high chromaticity and octupoles settings, together with the full performance of the damper, were needed to ensure the beam stability at 450 Ge. V Beam lifetime measured for different settings of vertical tune and chromaticity (1) • Drop observed after increasing Q’V • Lifetime recovered by lowering the tune Tune footprints as obtained from Py. ECLOUDPy. HEADTAIL simulations for different tunes (2) • Large tune spreads due to high settings • Lower tunes needed to avoid Qy =. 33 lifetime improved Design fractional tunes (0. 28, 0. 31) 2016 operational tunes (0. 27, 0. 295) Simulation settings: • EC in the arcs • Q’H, V = 15/15 • Oct current = 26 A (1) A. Romano et al. , “Effect of the electron cloud on the tune footprint at 450 Ge. V”, presentation at the LBOC Meeting No 51, 27 October 2015, https: //indico. cern. ch/event/455596/contributions/1966373/ (2015) (2) A. Romano et al. , “Macroparticle simulation studies of the LHC beam dynamics in the presence of the electron cloud”, in Proceedings of the 8 th International Particle Accelerator Conference, Copenhagen, Denmark, TUPVA 018 (2017) A. Romano et al. ABP Information Meeting 7
Observations vs simulation results 2016 (beginning of the run): In spite of high machine settings, instabilities were observed at 6. 5 Te. V during collisions (stable beams) (1) Vertical emittance blown up during collision • Affecting only bunches at tail of trains • Instabilities occurred for bunch intensities between 0. 7 e 11 and 1. 0 e 11 EC central density (from Py. ECLOUD) vs bunch intensity • for lower beam intensity, the EC density can become sufficiently high to drive an instability EC in dipoles? (1) A. Romano et al. , “Instabilities in stable beams”, presentation at the ½ -day internal review of LHC performance limitations during run 2, 29 November 2016, https: //indico. cern. ch/event/589625/ (2016) A. Romano et al. ABP Information Meeting 8
Observations vs simulation results 2016: Tests were performed at injection energy in order to assess the EC impact on the beam stability and potential mitigation strategies Comparison of emittance measurements at 450 Ge. V for two test fills : EC free pattern and standard 25 ns Simulation of transverse instabilities due to the EC Ø Any sizable emittance blowup observed Ø High machine settings needed to reach a good beam quality Can we explain this horizontal emittance blowup? A. Romano et al. EC in quadrupoles alone can explain why instabilities can be seen in both planes ABP Information Meeting 9
Observations vs simulation results 2016: Tests were performed at injection energy in order to assess the EC impact on the beam stability and potential mitigation strategies Comparison of emittance measurements at 450 Ge. V for two test fills : EC free pattern and standard 25 ns Simulation of transverse instabilities due to the EC Ø Any sizable emittance blowupingredient observed to explain observed LHC instabilities detailed Quads are found to be a critical studies have been carried out in order to investigate the effect of different machine settings (chromaticity, octupoles and bunch-by-bunch transverse feedback) on the beam dynamics [1] Ø High machine settings needed to reach a good beam quality Can we explain this horizontal emittance blowup? EC in quadrupoles alone can explain why instabilities can be seen in both planes [1] A. Romano et al. , “Electron cloud induced instabilities in the LHC”, presentation at the Joint Ecloud-Py. HEADTAIL Meeting, 12 May 2017, https: //indico. cern. ch/event/638087/ (2017) A. Romano et al. ABP Information Meeting 10
Conclusions • Electron cloud could pose important challenges to the machine operation difficulty to ensure the beam stability and a good beam quality from injection to collision • Improvements of simulation tools were needed to exploit new scenarios • Several configurations (EC in dipoles and quadrupoles, chromaticity, octupoles, injection energy, flattop, transverse feedback) have been simulated in order to explain the underlying mechanism of the observed instabilities in the LHC and find potential mitigation strategies • Simulation results could explain several machine observations Ø Poor lifetime with the nominal tunes at injection Ø Instability in stable beams for lower than nominal bunch intensity Ø Horizontal instabilities at injection (found to be driven by quadrupoles) A. Romano et al. ABP Information Meeting 11
Thanks for your attention! A. Romano et al. ABP Information Meeting 12
Electron cloud induced transverse instability Electron density during a bunch passage Py. ECLOUD-Py. HEADTAIL simulation • e-cloud driven instability Py. ECLOUD-Py. HEADTAIL simulation Electrons are attracted by the circulating proton bunch and ”fly” through it resulting in regions of high electron density within the bunch itself transverse instability A. Romano et al. ABP Information Meeting 13
Py. ECLOUD-Py. HEADTAIL simulations The beam dynamics is modeled using Py. HEADTAIL(2) 1. The ring is split into a set of segments (IPs) 2. Linear periodic maps is used for transverse tracking from one IP to the next 3. Synchrotron motion is applied once per turn 4. At each IP the bunch/e-cloud interaction takes place IPs more than 104 simulated turns are needed …. Segment with linear transport (2) K. Li et al. , “Code Development for Collective Effects”, HB 16, Malmo, Sweden, paper WEAM 3 X 01, 2016 A. Romano et al. ABP Information Meeting 14
Py. ECLOUD-Py. HEADTAIL simulations The beam dynamics is modeled using Py. HEADTAIL(2) 1. The ring is split into a set of segments (IPs) 2. Linear periodic maps is used for transverse tracking from one IP to the next 3. Synchrotron motion is applied once per turn 4. At each IP the bunch/e-cloud interaction takes place New approach: Py. ECLOUD(3) used in combination with Py. HEADTAIL by means of an ad-hoc developed interface (2) (3) K. Li et al. , “Code Development for Collective Effects”, HB 16), Malmo, Sweden, paper WEAM 3 X 01, 2016 G. Iadarola, "Electron cloud studies for CERN particle accelerators and simulation code development", CERN-THESIS 2014 -047 A. Romano et al. ABP Information Meeting 15
- Slides: 15