1. CERN, LIU-SPS ZS Review, 20/02/2013 1 Brief review on electron cloud simulations for the SPS electrostatic septum (ZS) G. Rumolo and G. Iadarola in LIU-SPS ZS Review, 20/02/2013 ZS with LHC-type beams Study of electron cloud build up thresholds in a ZS-like geometry with LHC25 beams: –Without external fields –With the voltage from the ion traps Some conclusions New simulations run with PyECLOUD with the correct model/geometry

2. Some practical information on the ZS 20mm V=-220kV -6kV -3kV Anode Cathode Ion-traps 2 From J. Wenninger, Introduction to Slow Extraction to the North Targets 2000 W.75 Re.25 septum wires. Wire diameter 50  m (first ZS) to 100  m. Wire spacing 1.5 mm.

3. Instructions when high intensity LHC beams are in the SPS:  Retracted girder (not anymore after 2010 to allow || MDs & conditioning)  HV 0 kV (not anymore after 2010, 20-100 kV applied, reduces outgassing)  Ion traps on Observations at the ZS with 25 ns high intensity beam (see also talks from Bruno and Karel):  Above certain intensities (nominal beam, more than 1 batch injected) vacuum spike in the ZS observed  The increased vacuum levels can provoke a vacuum interlock, which stops the ion traps and hence the beam in the machine. In this sense, “ZS limits the LHC beam”  By increasing the bunch length the vacuum does not degrade.  Sparking occasionally occurs (ramp, ejection) Ion trap voltage drop and current measured off the plates Does not appear to be the principal problem, rate decreasing ⇒ Some hints that e-cloud could build up in the ZS, even if the presence of a voltage should clear electrons…. ZS during SPS operation with high intensity LHC beams 3

4. In absence of voltage from the ion traps  significant electron cloud builds up for  max > 1.5  the electron cloud between bunches is uniformly distributed over the chamber cross section Electron cloud simulations the ZS geometry without external fields 4 ec = 10 10 e - /m 46 mm 140 mm

5. Assuming a voltage of 3 kV between the bottom and top plates  the electron cloud is suppressed at least up to  max =2.0 (different build up curves are all below the one for  max =2.0) Electron cloud simulations including the voltage from the ion traps 5 E ec = 10 3 e - /m 46 mm 140 mm

6. Assuming a voltage of 3 kV between the bottom and top plates  the electron cloud is suppressed at least up to  max =2.0 (different build up curves are all below the one for  max =2.0)  the electrons are fully cleared between subsequent bunches and there is no visible dependence on the SEY of the plate. Electron cloud simulations including the voltage from the ion traps 6 E 46 mm 140 mm Zoom of previous plot in the first 0.3  s

7. For V=100 V the SEY threshold lies also around 1.5 Assuming  max =1.7 and scanning the voltage between the bottom and top plate  the electron cloud is found to be fully suppressed for 500 V ≤ V < 4 kV  V=100 V is not sufficient and a strong electron cloud is formed (with a faster rise time than V=0 but a faster decay, too, due to the clearing voltage) Electron cloud simulations scanning the voltage values 7 V = 100 V

8. Changing the voltage, we actually change the clearing efficiency  electrons are cleared more efficiently (i.e. more quickly) with higher voltages  However, if the voltage becomes too low (<500 V), the intra-bunch clearing is insufficient and multipacting cannot be avoided  the quoted clearing voltage values also depend on having assumed in the model a maximum SEY at E max =230 eV and R 0 =0.7 Electron cloud simulations scanning the voltage values 8  max =1.7, zoom on the first 50 ns

9. Electron cloud simulations electron dynamics and distribution 9 No voltage applied

10. Electron cloud simulations electron dynamics and distribution 10 V = 3 kV

11. All the simulations have been re-run with the PyECLOUD, which allows for the use of the correct chamber shape (both correct geometric and electromagnetic boundary conditions) The geometry of the ZS is confirmed to be prone to electron cloud build up with LHC25 beams at nominal intensity → in absence of the voltage from the ion traps, there is electron cloud build up for maximum SEY above 1.5 A difference of potential between top and bottom plate is certainly effective to clear the electrons from residual gas ionization between bunches  voltage should be at least a few hundreds V (100V certainly not enough) Summary and conclusions 11