1. PyECLOUD and Build Up Simulations at CERN
G. Iadarola, G. Rumolo
Thanks to:
F. Zimmermann, G. Arduini, H. Bartosik, C. Bhat, V. Baglin, R. De Maria, O. Dominguez, M. Driss Mensi, J. Esteban-Muller, K. Li, H. Maury Cuna, G. Miano, H. Neupert,
G. Papotti, E. Shaposhnikova, M. Taborelli, C. Y. Vallgren

2. Outline Why a new build-up code? Inside PyECLOUD: Overview
MP size management
Convergence and performances
PyECLOUD at work:
Build-up simulations for LHC 800mm common chamber

3. Outline Why a new build-up code? Inside PyECLOUD: Overview
MP size management
Convergence and performances
PyECLOUD at work:
Build-up simulations for LHC 800mm common chamber

4. From ECLOUD to PyECLOUD ECLOUD
Developed at CERN since 1997
(mainly by F. Zimmermann, G. Bellodi, O. Bruning, G. Rumolo, D. Schulte)
Pioneering work which defined a physical model for the EC build-up
FORTRAN 77 code
Scarcely modular
(difficult to maintain, develop and debug)

5. From ECLOUD to PyECLOUD ECLOUD PyECLOUD
Developed at CERN since 1997
(mainly by F. Zimmermann, G. Bellodi, O. Bruning, G. Rumolo, D. Schulte)
Development started in 2011
Pioneering work which defined a physical model for the EC build-up
Inherits the physical model of ECLOUD
FORTRAN 77 code
Python code
Scarcely modular
(difficult to maintain, develop and debug)
Strongly modular (much easier to develop and maintain)

6. From ECLOUD to PyECLOUD ECLOUD PyECLOUD
Developed at CERN since 1997
(mainly by F. Zimmermann, G. Bellodi, O. Bruning, G. Rumolo, D. Schulte)
Development started in 2011
Pioneering work which defined a physical model for the EC build-up
Inherits the physical model of ECLOUD
FORTRAN 77 code
Python code
Scarcely modular
(difficult to maintain, develop and debug)
Strongly modular (much easier to develop and maintain)
Several improvements introduced with better performances in terms of reliability, accuracy, efficiency, and flexibility

7. Outline Why a new build-up code? Inside PyECLOUD: Overview
MP size management
Convergence and performances
PyECLOUD at work:
Bunch energy loss estimation from build-up simulation
Build-up simulations for LHC 800mm common chamber

8. PyECLOUD flowchart
t=t+Δt
Generate seed e-
PyECLOUD is a 2D macroparticle (MP) code for the simulation of the electron cloud build-up with:
Arbitrary shaped chamber
Ultra-relativistic beam
Externally applied (uniform) magnetic field
Evaluate the electric field of beam at each MP location
Evaluate the e- space charge
electric field
Compute MP motion (t->t+Δt)
Detect impacts and generate secondaries

9. PyECLOUD flowchart
t=t+Δt
Generate seed e-
Evaluate the electric field of beam at each MP location
Evaluate the number of seed e- generated during the current time step and generate the corresponding MP:
Residual gas ionization and photoemission are implemented
Evaluate the e- space charge
electric field
Compute MP motion (t->t+Δt)
Detect impacts and generate secondaries

10. PyECLOUD flowchart
t=t+Δt
Generate seed e-
Evaluate the electric field of beam at each MP location
The field map for the relevant chamber geometry and beam shape is pre-computed on a suitable rectangular grid and loaded from file in the initialization stage
When the field at a certain location is needed a linear (4 points) interpolation algorithm is employed
Evaluate the e- space charge
electric field
Compute MP motion (t->t+Δt)
Detect impacts and generate secondaries

11. PyECLOUD flowchart Classical Particle In Cell (PIC) algorithm:
t=t+Δt
Generate seed e-
Evaluate the electric field of beam at each MP location
Classical Particle In Cell (PIC) algorithm:
Electron charge density distribution ρ(x,y) computed on a rectangular grid
Poisson equation solved using finite difference method
Field at MP location evaluated through linear (4 points) interpolation
Evaluate the e- space charge
electric field
Compute MP motion (t->t+Δt)
Detect impacts and generate secondaries

12. PyECLOUD flowchart
t=t+Δt
Generate seed e-
Evaluate the electric field of beam at each MP location
The dynamics equation is integrated in order to update MP position and momentum:
Evaluate the e- space charge
electric field
Compute MP motion (t->t+Δt)
When possible, “strong B condition” is exploited in order to speed-up the computation
Detect impacts and generate secondaries

13. PyECLOUD flowchart
t=t+Δt
Generate seed e-
Evaluate the electric field of beam at each MP location
When a MP hits the wall theoretical/empirical models are employed to generate charge, energy and angle of the emitted charge
According to the number of emitted electrons, MPs can be simply rescaled or new MP can be generated
Evaluate the e- space charge
electric field
Compute MP motion (t->t+Δt)
Detect impacts and generate secondaries

14. Outline Why a new build-up code? Inside PyECLOUD: Overview
MP size management
Convergence and performances
PyECLOUD at work:
Build-up simulations for LHC 800mm common chamber

15. Macroparticle size management
72 bunches – 25ns spac.
72 bunches – 25ns spac.
In an electron-cloud buildup, due to the multipacting process, the electron number extends over several orders of magnitude
It is practically impossible to choose a MP size that is suitable for the entire simulation (allowing a satisfactory description of the phenomenon and a computationally affordable number of MPs)

16. Macroparticle size management
72 bunches – 25ns spac.
72 bunches – 25ns spac.
A reference MP size Nref is used to “take decisions”:
Seed MP generation: the generated MPs have size Nref
Secondary MP emission: additional true secondary MPs are emitted if the total emitted charge is >1.5Nref
MP cleaning: at each bunch passage a clean function is called to eliminate all the MPs with charge <10-4Nref x

17. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.

18. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.

19. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.

20. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
MP reg.

21. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
MP set regeneration
Each macroparticle is assigned to a cell of a uniform grid in the 5-D space (x,y,vx,vy,vz) obtaining an approximation of the phase space distribution
The new target MP size is chosen such that:
ref. size [m-1] MP
10.
MP reg.
A new MPs set, having the new reference size, is generated according to the computed distribution
The error on total charge and total energy does not go beyond 1-2%

22. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
MP reg.

23. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
MP reg.

24. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
MP reg.

25. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
MP reg.

26. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
MP reg.

27. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
MP reg.

28. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
MP reg.

29. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
5.5e3
MP reg.

30. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
5.5e3
MP reg.

31. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
5.5e3
2.9e4
MP reg.

32. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
5.5e3
2.9e4
MP reg.

33. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
5.5e3
2.9e4
1.3e5
MP reg.

34. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
5.5e3
2.9e4
1.3e5
MP reg.

35. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
5.5e3
2.9e4
1.3e5
3.1e5
MP reg.

36. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
5.5e3
2.9e4
1.3e5
3.1e5
MP reg.

37. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
5.5e3
2.9e4
1.3e5
3.1e5
MP reg.

38. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
5.5e3
2.9e4
1.3e5
3.1e5
MP reg.

39. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
5.5e3
2.9e4
1.3e5
3.1e5
MP reg.

40. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
5.5e3
2.9e4
1.3e5
3.1e5
3.6e5
MP reg.

41. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
5.5e3
2.9e4
1.3e5
3.1e5
3.6e5
MP reg.

42. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
5.5e3
2.9e4
1.3e5
3.1e5
3.6e5
MP reg.

43. Macroparticle size management
The reference MP size Nref is adaptively changed during the simulation:
ref. size [m-1] MP
10.
45.
2.1e2
1.1e3
5.5e3
2.9e4
1.3e5
3.1e5
3.6e5
MP reg.

44. Outline Why a new build-up code? Inside PyECLOUD: Overview
MP size management
Convergence and performances
PyECLOUD at work:
Build-up simulations for LHC 800mm common chamber

45. SPS MBB bending magnet, SEYmax = 1.5, nominal 25ns beam, E=26GeV
Convergence study - Number of electrons
ECLOUD
PyECLOUD
SPS MBB bending magnet, SEYmax = 1.5, nominal 25ns beam, E=26GeV

46. SPS MBB bending magnet, SEYmax = 1.5, nominal 25ns beam, E=26GeV
Convergence study – Electrons ditribution
ECLOUD
PyECLOUD
SPS MBB bending magnet, SEYmax = 1.5, nominal 25ns beam, E=26GeV

47. SPS MBB bending magnet, SEYmax = 1.5, nominal 25ns beam, E=26GeV
Processing time
Time step
ECLOUD
PyECLOUD
0.2 ns
29 min
12 min
0.1 ns
1h 27 min
13 min
0.05 ns
1h 45 min
24 min
0.025ns
3h 7 min
40 min
0.012ns
4h 15 min
1h 6 min
SPS MBB bending magnet, SEYmax = 1.5, nominal 25ns beam, E=26GeV

48. Outline Why a new build-up code? Inside PyECLOUD: Overview
MP size management
Convergence and performances
PyECLOUD at work:
Build-up simulations for LHC 800mm common chamber

49. PyECLOUD at work
Several studies at CERN are/have been employing the new code:
Proton Synchrotron (PS):
Study on EC dependence on the Bunch Profile (C. Bhat)
Benchmarking of shielded pickup measurements (S. Gilardoni, G. Iadarola, M Pivi, G. Rumolo, C. Y. Vallgren)
Super Proton Synchrotron (SPS):
Scrubbing optimization studies (G.Iadarola, G. Rumolo)
Intensity upgrade studied (G.Iadarola, G. Rumolo)
Benchmarking of Strip Detector measurements (H. Bartosik, G.Iadarola, H. Neupert, M. Driss Mensi, G. Rumolo, M. Taborelli)
Large Hadron Collider (LHC):
Benchmarking of bunch-by-bunch energy loss data from stable-phase shift (J. F. Esteban Muller, G.Iadarola, G. Rumolo, E. Shaposhnikova)
Map formalism study for scrubbing optimization (O. Dominguez, F. Zimmermann)
Pressure observations vs. simulations benchmarking (O. Dominguez, F. Zimmermann)
Background study for 800mm chamber close to ALICE (V. Baglin, O. Dominguez, G. Iadarola, G. Rumolo)
Heat load benchmarking for the cryogenic arcs (G. Iadarola, H. Maury Cuna, G. Rumolo. F. Zimmermann)
Benchmarking of Instability Simulations at LHC (H. Bartosik, G. Iadarola, G. Rumolo)

50. >104 simulations run so far
PyECLOUD at work
Several studies at CERN are/have been employing the new code:
Proton Synchrotron (PS):
Study on EC dependence on the Bunch Profile (C. Bhat)
Benchmarking of shielded pickup measurements (S. Gilardoni, G. Iadarola, M Pivi, G. Rumolo, C. Y. Vallgren)
Super Proton Synchrotron (SPS):
Scrubbing optimization studies (G.Iadarola, G. Rumolo)
Intensity upgrade studied (G.Iadarola, G. Rumolo)
Benchmarking of Strip Detector measurements (H. Bartosik, G.Iadarola, H. Neupert, M. Driss Mensi, G. Rumolo, M. Taborelli)
Large Hadron Collider (LHC):
Benchmarking of bunch-by-bunch energy loss data from stable-phase shift (J. F. Esteban Muller, G.Iadarola, G. Rumolo, E. Shaposhnikova)
Map formalism study for scrubbing optimization (O. Dominguez, F. Zimmermann)
Pressure observations vs. simulations benchmarking (O. Dominguez, F. Zimmermann)
Background study for 800mm chamber close to ALICE (V. Baglin, O. Dominguez, G. Iadarola, G. Rumolo)
Heat load benchmarking for the cryogenic arcs (G. Iadarola, H. Maury Cuna, G. Rumolo. F. Zimmermann)
Benchmarking of Instability Simulations at LHC (H. Bartosik, G. Iadarola, G. Rumolo)
>104 simulations run so far

51. PyECLOUD at work
Several studies at CERN are/have been employing the new code:
Proton Synchrotron (PS):
Study on EC dependence on the Bunch Profile (C. Bhat)
Benchmarking of shielded pickup measurements (S. Gilardoni, G. Iadarola, M Pivi, G. Rumolo, C. Y. Vallgren)
Super Proton Synchrotron (SPS):
Scrubbing optimization studies (G.Iadarola, G. Rumolo)
Intensity upgrade studied (G.Iadarola, G. Rumolo)
Benchmarking of Strip Detector measurements (H. Bartosik, G.Iadarola, H. Neupert, M. Driss Mensi, G. Rumolo, M. Taborelli)
Large Hadron Collider (LHC):
Benchmarking of bunch-by-bunch energy loss data from stable-phase shift (J. F. Esteban Muller, G.Iadarola, G. Rumolo, E. Shaposhnikova)
Map formalism study for scrubbing optimization (O. Dominguez, F. Zimmermann)
Pressure observations vs. simulations benchmarking (O. Dominguez, F. Zimmermann)
Background study for 800mm chamber close to ALICE (V. Baglin, O. Dominguez, G. Iadarola, G. Rumolo)
Heat load benchmarking for the cryogenic arcs (G. Iadarola, H. Maury Cuna, G. Rumolo. F. Zimmermann)
Benchmarking of Instability Simulations at LHC (H. Bartosik, G. Iadarola, G. Rumolo)

52. PyECLOUD at work
Several studies at CERN are/have been employing the new code:
Proton Synchrotron (PS):
Study on EC dependence on the Bunch Profile (C. Bhat)
Benchmarking of shielded pickup measurements (S. Gilardoni, G. Iadarola, M Pivi, G. Rumolo, C. Y. Vallgren)
Super Proton Synchrotron (SPS):
Scrubbing optimization studies (G.Iadarola, G. Rumolo)
Intensity upgrade studied (G.Iadarola, G. Rumolo)
Benchmarking of Strip Detector measurements (H. Bartosik, G.Iadarola, H. Neupert, M. Driss Mensi, G. Rumolo, M. Taborelli)
Large Hadron Collider (LHC):
Benchmarking of bunch-by-bunch energy loss data from stable-phase shift (J. F. Esteban Muller, G.Iadarola, G. Rumolo, E. Shaposhnikova)
Map formalism study for scrubbing optimization (O. Dominguez, F. Zimmermann)
Pressure observations vs. simulations benchmarking (O. Dominguez, F. Zimmermann)
Background study for 800mm chamber close to ALICE (V. Baglin, O. Dominguez, G. Iadarola, G. Rumolo)
Heat load benchmarking for the cryogenic arcs (G. Iadarola, H. Maury Cuna, G. Rumolo. F. Zimmermann)
Benchmarking of Instability Simulations at LHC (H. Bartosik, G. Iadarola, G. Rumolo)

53. Outline Why a new build-up code? Inside PyECLOUD: Overview
MP size management
Convergence and performances
PyECLOUD at work:
Build-up simulations for LHC 800mm common chamber

54. 800mm vacuum IP2
Vacuum team has reported pressure rise in 800mm common vacuum chambers on both sides of ALICE, with significant impact on background
Ø 80cm
Ø 20cm
27m

55. Vacuum observations
Ramp
Fill /07/2011

56. Vacuum observations Ramp
Fill /07/2011
Pressure rise is strongly correlated to the injection of the last two batches from the SPS and already appears at 450GeV

57. 800mm chamber – two beams simulations
In the considered chamber both counter-rotating beams circulate at the same time
This changes the picture since at different sections, different “effective bunch spacings” (delay between following bunch passages) are observed

58. 800mm chamber – two beams simulations

59. 800mm chamber – two beams simulations

60. 800mm chamber – two beams simulations

61. 800mm chamber – two beams simulations

62. 800mm chamber – two beams simulations

63. 800mm chamber – two beams simulations

64. 800mm chamber – two beams simulations

65. 800mm chamber – two beams simulations

66. 800mm chamber – two beams simulations

67. 800mm chamber – two beams simulations

68. 800mm chamber – two beams simulations

69. 800mm chamber – two beams simulations

70. 800mm chamber – two beams simulations

71. 800mm chamber – two beams simulations

72. 800mm chamber – two beams simulations

73. 800mm chamber – two beams simulations

74. 800mm chamber – two beams simulations

75. 800mm chamber – two beams simulations

76. 800mm chamber – two beams simulations

77. 800mm chamber – two beams simulations

78. 800mm chamber – two beams simulations

79. 800mm chamber – two beams simulations

80. 800mm chamber – two beams simulations

81. 800mm chamber – two beams simulations

82. 800mm chamber – two beams simulations

83. 800mm chamber – two beams simulations

84. 800mm chamber – two beams simulations

85. 800mm chamber – two beams simulations

86. 800mm chamber – two beams simulations

87. 800mm chamber – two beams simulations

88. 800mm chamber – two beams simulations

89. 800mm chamber – two beams simulations

90. 800mm chamber – two beams simulations

91. 800mm chamber – two beams simulations

92. 800mm chamber – two beams simulations

93. 800mm chamber – two beams simulations

94. 800mm chamber – two beams simulations

95. 800mm chamber – two beams simulations

96. 800mm chamber – two beams simulations

97. 800mm chamber – two beams simulations

98. 800mm chamber – two beams simulations

99. 800mm chamber – two beams simulations

100. 800mm chamber – two beams simulations

101. 800mm chamber – two beams simulations

102. 800mm chamber – two beams simulations

103. 800mm chamber – two beams simulations

104. 800mm chamber – two beams simulations

105. 800mm chamber – two beams simulations

106. 800mm chamber – two beams simulations
To check if our model of e-cloud can explain the observed behavior of pressure rise, we have simulated the electron cloud build up in the 800mm chamber before and after the last two injections
Fill /07/2011
Simulated conditions

107. 800mm chamber – two beams simulations
Before the last two injections (144 bunches per injection) about 1/4 of each ring is still empty
2nd turn
1st turn
LSS2

108. 800mm chamber – two beams simulations
Before the last two injections (144 bunches per injection) about 1/4 of each ring is still empty
2nd turn
1st turn
LSS2

109. 800mm chamber – two beams simulations
Before the last two injections (144 bunches per injection) about 1/4 of each ring is still empty
2nd turn
1st turn
LSS2

110. 800mm chamber – two beams simulations
Before the last two injections (144 bunches per injection) about 1/4 of each ring is still empty
2nd turn
1st turn
LSS2

111. 800mm chamber – two beams simulations
Before the last two injections (144 bunches per injection) about 1/4 of each ring is still empty
2nd turn
1st turn
LSS2

112. 800mm chamber – two beams simulations
2nd turn
1st turn
LSS2

113. 800mm chamber – two beams simulations
Before the last two injections (144 bunches per injection) about 1/4 of each ring is still empty
2nd turn
1st turn
LSS2

114. 800mm chamber – two beams simulations
Before the last two injections (144 bunches per injection) about 1/4 of each ring is still empty
2nd turn
1st turn
LSS2

115. 800mm chamber – two beams simulations
e-cloud buildup is observed when both beams are passing in the 800mm chamber
2nd turn
1st turn
LSS2

116. 800mm chamber – two beams simulations
e-cloud buildup is observed when both beams are passing in the 800mm chamber
2nd turn
1st turn
LSS2

117. 800mm chamber – two beams simulations
e-cloud buildup is observed when both beams are passing in the 800mm chamber
2nd turn
1st turn
LSS2

118. 800mm chamber – two beams simulations
e-cloud buildup is observed when both beams are passing in the 800mm chamber
2nd turn
1st turn
LSS2

119. 800mm chamber – two beams simulations
e-cloud buildup is observed when both beams are passing in the 800mm chamber
2nd turn
1st turn
LSS2

120. 800mm chamber – two beams simulations
e-cloud buildup is observed when both beams are passing in the 800mm chamber
2nd turn
1st turn
LSS2

121. 800mm chamber – two beams simulations
e-cloud buildup is observed when both beams are passing in the 800mm chamber
2nd turn
1st turn
LSS2

122. 800mm chamber – two beams simulations
A slow decay of the e-cloud is observed when only one 50ns beam is passing in the 800mm chamber
2nd turn
1st turn
LSS2

123. 800mm chamber – two beams simulations
A slow decay of the e-cloud is observed when only one 50ns beam is passing in the 800mm chamber
2nd turn
1st turn
LSS2

124. 800mm chamber – two beams simulations
A slow decay of the e-cloud is observed when only one 50ns beam is passing in the 800mm chamber
2nd turn
1st turn
LSS2

125. 800mm chamber – two beams simulations
A slow decay of the e-cloud is observed when only one 50ns beam is passing in the 800mm chamber
2nd turn
1st turn
LSS2

126. 800mm chamber – two beams simulations
A slow decay of the e-cloud is observed when only one 50ns beam is passing in the 800mm chamber
2nd turn
1st turn
LSS2

127. 800mm chamber – two beams simulations
A slow decay of the e-cloud is observed when only one 50ns beam is passing in the 800mm chamber
2nd turn
1st turn
LSS2

128. 800mm chamber – two beams simulations
A slow decay of the e-cloud is observed when only one 50ns beam is passing in the 800mm chamber
2nd turn
1st turn
LSS2

129. 800mm chamber – two beams simulations
A slow decay of the e-cloud is observed when only one 50ns beam is passing in the 800mm chamber
2nd turn
1st turn
LSS2

130. 800mm chamber – two beams simulations
A slow decay of the e-cloud is observed when only one 50ns beam is passing in the 800mm chamber
2nd turn
1st turn
LSS2

131. 800mm chamber – two beams simulations
A slow decay of the e-cloud is observed when only one 50ns beam is passing in the 800mm chamber
2nd turn
1st turn
LSS2

132. 800mm chamber – two beams simulations
A slow decay of the e-cloud is observed when only one 50ns beam is passing in the 800mm chamber
2nd turn
1st turn
LSS2

133. 800mm chamber – two beams simulations
A slow decay of the e-cloud is observed when only one 50ns beam is passing in the 800mm chamber
2nd turn
1st turn
LSS2

134. 800mm chamber – two beams simulations
A slow decay of the e-cloud is observed when only one 50ns beam is passing in the 800mm chamber
2nd turn
1st turn
LSS2

135. 800mm chamber – two beams simulations
A slow decay of the e-cloud is observed when only one 50ns beam is passing in the 800mm chamber
2nd turn
1st turn
LSS2

136. 800mm chamber – two beams simulations
A slow decay of the e-cloud is observed when only one 50ns beam is passing in the 800mm chamber
2nd turn
1st turn
LSS2

137. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

138. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

139. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

140. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

141. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

142. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

143. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

144. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

145. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

146. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

147. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

148. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

149. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

150. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

151. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

152. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

153. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

154. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

155. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

156. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

157. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

158. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

159. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

160. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

161. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

162. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

163. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

164. 800mm chamber – two beams simulations
No memory effect between following turns is observed
2nd turn
1st turn
LSS2

165. 800mm chamber – two beams simulations
To check if our model of e-cloud can explain the observed behavior of pressure rise, we have simulated the electron cloud build up in the 800mm chamber before and after the last two injections
Fill /07/2011
Simulated conditions

166. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

167. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

168. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

169. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

170. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

171. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

172. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

173. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

174. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

175. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

176. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

177. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

178. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

179. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

180. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

181. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

182. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

183. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

184. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

185. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

186. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

187. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

188. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

189. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

190. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

191. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

192. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

193. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

194. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

195. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

196. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

197. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

198. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

199. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

200. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

201. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

202. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

203. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

204. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

205. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

206. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

207. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

208. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

209. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

210. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

211. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

212. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

213. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

214. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

215. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

216. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

217. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

218. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

219. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

220. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

221. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

222. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

223. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

224. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

225. 800mm chamber – two beams simulations
At the end of injection the two rings look quite completely filled, the largest holes being the two abort gaps
2nd turn
1st turn
LSS2

226. 800mm chamber – two beams simulations
Memory effect is observed between turns, which can strongly enhance the electron cloud
1st turn
2nd turn
LSS2

227. 800mm chamber – two beams simulations
Memory effect is observed between turns, which can strongly enhance the electron cloud
1st turn
2nd turn
LSS2
This is consistent with the observation of pressure rise only after the last two injections

228. Summary
A new Python code for the simulation of the e-cloud build-up has been developed
Based on the physical models ECLOUD, PyECLOUD shows significantly improved accuracy, flexibility and efficiency
Several studies have already been conducted at CERN with the new code
Future plans
Arbitrary shaped chamber with non-uniform SEY (already implemented, test ongoing)
Non uniform magnetic field map (e. g. quadrupoles, combined function magnets)
Integration with HEADTAIL for self-consistent simulations

229. Thanks for your attention!

230. Before the last two injections (1380 - 288 bunches per beam)
800mm chamber – two beams simulations
Hybrid spacings
50ns beam
Before the last two injections ( bunches per beam)
Hybrid spacings
SEY 1.6
No multipacting
Simulations indicate different e-cloud densities at different sections of the considered chamber (due to different “crossing conditions”) but build-up along al the tube

231. PyECLOUD
We have decided to write a new fully reorganized build-up code, in a newer and more powerful language, considering that the initial effort would be compensated by a significantly increased efficiency in future development and debugging.
The employed programming language is Python:
Interpreted language (open source), allowing incremental and interactive development of the code, encouraging an highly modular structure
Libraries for scientific computation (e.g Numpy, Scipy, Pylab)
Extensible with C/C++ or FORTRAN compiled modules for computationally intensive parts