Siting and Military Capabilities: DoD’s Energy Siting Clearinghouse
Update on HOM studies for PIP II SC Linac...at Fermilab. In either case, the siting shown in Figure...
Transcript of Update on HOM studies for PIP II SC Linac...at Fermilab. In either case, the siting shown in Figure...
A. Sukhanov
Fermilab, TD SRF T&D
HOMSC 2018, Cornell UniversityOctober 1-3, 2018
Update on HOM studies for PIP II SC Linac
PIP-II Project
•Mission: deliver intense beam of neutrinos to the international LBNF/DUNE project
•Goals:
‣ Deliver >1 MW proton beam power from the Fermilab Main Injector over the energy range 60 — 120 GeV, at the start of LBNF operation
‣ Support ongoing 8 GeV program at Fermilab, including upgrade path for Mu2e
‣ Possibility of extension of beam power to LBNF >2 MW
‣ Possibility of extension to high duty factor/higher beam power
2
PIP-II Project
• PIP-II Linac
3
8
Figure 1‐1: Site layout of PIP‐II (north is to the right). New construction includes the linac enclosure, transfer line enclosure (including the beam abort area and a stub to facilitate future connection to the Muon Campus), linac gallery, utility building, and cryo/compressor building. The blue areas denote identified wetland areas.
PIP-II provides a variety of straightforward and cost effective upgrade paths. Delivery of more than 2 MW to the LBNF target will require replacement of the existing Booster. The most straightforward strategy would be to extend the 0.8 GeV linac to 1.5-2 GeV and to inject at this energy into a newly constructed 8-GeV rapid cycling synchrotron (RCS). Such a synchrotron could either be of a conventional sort, as currently deployed at the J-PARC facility in Japan, or could incorporate novel design features based on highly non-linear optical elements currently under study at Fermilab. In either case, the siting shown in Figure 1.1 is compatible with siting of such a RCS to the south of the linac, providing capabilities for injecting beam into the Main Injector.
Parameter
Particle species H-
Input beam energy 2.1 MeV
Output beam energy 800 MeV
Bunch repetition rate 162.5 MHz
RF pulse length pulsed-to-CW
Sequence of bunches Programmable
Average beam current 2 mA
Final rms norm trans. emittance
<= 0.3 mm-mrad
Final rms norm long. emittance
<= 0.35 mm-mrad
RMS bunch length < 4 ps
PIP-II Project
• PIP-II Linac
4
HWR SSR1 SSR2Elliptical 5-cell
Cavity Aperture, mm Eff. length, cm Eacc, MV/m Epeak, MV/m Bpeak, mT (R/Q), Ohm G, Ohm
LB650 88 70.3 16.9 40.3 74.6 341 193
HB650 118 106.1 18.8 38.9 73.1 610 260
CM #cav/CM #CM CM config CM length, m Q0 x1e10, @2K Rs, nOhm QL, x1e6
LB650 3 11 ccc 4.32 2.15 9.0 10.36
HB650 6 4 cccccc 9.92 3 8.7 9.92
LB650 HB650
HOM in PIP II (a.k.a. Project X)
• Presented at HOMSC 2012, published in Proceedings
5
Higher Order Modes in the Project-X Linac
V. Yakovlev, Fermilab
HOM in PIP II• Complex beam current spectrum• No HOM couplers and dumpers (QL up to 1e7)
• Investigation of HOM various effects in PIP II reported at HOMSC 12:‣ found no big issues
• Here we consider HOM effects in PIP II‣ new modes of linac operation (long trains of 162.5 MHz bunches, 10 mA)
- important for accelerator-driven sub-critical system demonstration‣ modifications to LB650 cavity design
- optimization of EM and mechanical parameters, production process
6
Frequency, MHz0 20 40 60 80 100 120 140 160 180 200
, mA
nI~
-210
-110
1
Frequency, GHz2.5 3 3.5 4 4.5 5 5.5
Ωm
ax
(R/Q
),
1
10
210
R0 31.89± 137.7
f0 0.1515± 1.425
R0 31.89± 137.7
f0 0.1515± 1.425
Beam Dynamics in PIP2 Linac with 10 mA
Motivation
• Transverse misalignment of SRF cavities and beam offset lead to excitation of dipole HOMs
‣ depends on bunch charge, mode (R/Q), Qext
‣ excited dipole HOMs introduce additional kick on beam particles
‣ increase of transverse beam emittance
‣ small effect in a steady state (CW beam, no variations in beam current and/or bunch timing pattern)
• Study transverse emittance variations due to transitions (beam turn on) and bunch charge variations in HB650 section of PIP2 linac
‣ (R/Q) of dipole modes in LB650 are much smaller compared to HB650
8
Dipole modes in HB650
• (R/Q) depends on beam velocity
•Mode 1376 MHz: (R/Q)=80kOhm/m2, Qext=1e7
9
Linac layout
•HB650 beta=0.92 section: 4 cryo-modules, 24 cavities (6 cavities per CM)
•Matrix tracking through linac elements
10
HB650 (6-cavity) Period
NC Doublet
CavityFieldMap
1.316m 1.525m
CavityFieldMap
CavityFieldMap
1.44m
SC Cryomodule
0.6m
0.2m
10.97m
CavityFieldMap
CavityFieldMap
CavityFieldMap
NC Doublet
1.429m
• 150mmspaceaddedforgatevalveflangeandmagnetflange.
HBSec9on
Cavity Gradient and Quadrupole Fields
•HB650 Vacc(beta_G) = 20 MV
•HB650 Quad gradient 12 T/m
11
0
5
10
15
20
25
0 20 40 60 80 100 120
Am
plitu
de (M
V)
Cavity #
MEBTHWRSSR1SSR2
LB650HB650
0
5
10
15
20
0 20 40 60 80 100 120 140 160G
radi
ent (
T/m
)
Position of the Center (m)
MEBTHWR
LB650HB650
Model• 60 pC bunches with 10% variation, 162.5 MHz bunch frequency
‣ train of 5e6 bunches
• Transverse misalignment of cavities R.M.S. = 0.5 mm
• Matrix tracking through linac
• Consider one dipole mode with highest (R/Q)
• Each bunch introduces voltage V[i] = jcq(R/Q)x
‣ bunch sees half of this kick voltage
• Total kick seen by bunch is sum of voltages from all previous bunches with proper time dependent factors: ~exp(jwTb)*exp(-wTb/2Q)
• Simulate 100 linac configurations (time consuming)
‣ random variations of transverse displacement of cavities and HOM frequency with 1 MHz RMS
•Calculate effective RMS emittance at the end of linac from variations of (x,x’) of bunches
‣ compare to nominal emittance 0.3 mm*mrad (relative emittance)
12
Tracking in linac
• bunch trajectories in linac
13
z , m0 5 10 15 20 25 30 35 40 45 50
x , u
m
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
Effective emittance
• Effective emittance at the end of linac for single linac configuration as a function of bunch number
14
bunch
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000310×
rela
tive
emitt
ance
cha
nge
0
5
10
15
20
25
30
-610×
Maximum emittance growth
Initial emittance growth after turn on, scales ~ Qext
Emittance variations due to Qb variations
Maximum emittance growth
• 100 configurations of linac with random cavity misalignment of 0.5 mm
‣ Median value of relative emittance growth is 4e-5
15
Relative emittance growth0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1
-310×0
2
4
6
8
10
Longitudinal emittance
• 10 mA (presented in LINAC’12, TUPB054)
16
Relative emittance growth, !"z
Pro
babi
lity
Blue – old design Red – new design
10 mA
New Design of LB 650 MHz 5-cell Cavities
New Designs of LB650 5-cell Cavities
•New designs of LB650 5-cell cavities have been developed at FNAL, INFN and VECC
‣ Optimization of EM and mechanical properties, production process
• 3-D RF simulation of cavities of each design (A. Lunin)
‣ calculate frequency spectra, QL, (R/Q) of monopole, dipole and quadrupole modes
18
Monopole modes
•QL and (R/Q)
19
Frequency, GHz0 0.5 1 1.5 2 2.5
LQ
310
410
510
610
710
810
910
1010FNALINFNVECC
Frequency, GHz0 0.5 1 1.5 2 2.5
, O
hmz
(R/Q
)
-210
-110
1
10
210
310FNALINFNVECC
Dangerous trapped mode in VECCdesign: f=1.663 GHz, (R/Q)=18 OhmThis mode is relatively far from mainbeam current harmonics of 162.5 MHz
Dipole modes
•QL and (R/Q)
20
Frequency, GHz0 0.5 1 1.5 2 2.5
LQ
310
410
510
610
710
810
910
1010FNALINFNVECC
Frequency, GHz0 0.5 1 1.5 2 2.5
2 ,
Ohm
/mt
(R/Q
)
-210
-110
1
10
210
310
410
510
610
710
810
Most possibly dangerous modes are:VECC: 0.974 GHz, (R/Q) = 38 hOhm/m**2df = 1 MHz FNAL: 0.973 GHz, (R/Q) = 35 kOhm/m**2df=2 MHz
Linac Layout
• LB650 beta=0.61 section: 11 cryo-modules, 33 cavities (3 cavities per CM)
21
LB650 (3-cavity) Period
CavityFieldMap
0.6m
0.2m
NC Doublet
1.145m 1.184m
CavityFieldMap
CavityFieldMap
0.9m
SC Cryomodule
5.371m
NC Doublet
1.258m
• 150mmspaceaddedforgatevalveflangeandmagnetflange.
LBSec9on
Cavity Gradient and Quadrupole Fields
• LB650 Vacc(beta_G) = 12 MV
• LB650 Quad gradient 9 T/m
22
0
5
10
15
20
25
0 20 40 60 80 100 120
Am
plitu
de (M
V)
Cavity #
MEBTHWRSSR1SSR2
LB650HB650
0
5
10
15
20
0 20 40 60 80 100 120 140 160G
radi
ent (
T/m
)
Position of the Center (m)
MEBTHWR
LB650HB650
Model• 30 pC bunches with 10% variation, 162.5 MHz bunch frequency
‣ train of 1e5 bunches
‣ about 50% of bunches removed for injection at 44.705 MHz
• Transverse misalignment of cavities R.M.S. = 0.5 mm
• Matrix tracking through linac
• Consider one dipole mode with highest (R/Q)
• Each bunch introduces voltage V[i] = jcq(R/Q)x
‣ bunch sees half of this kick voltage
• Total kick seen by bunch is sum of voltages from all previous bunches with proper time dependent factors: ~exp(jwTb)*exp(-wTb/2Q)
• Simulate 100 linac configurations (time consuming)
‣ random variations of transverse displacement of cavities and HOM frequency with 1 MHz RMS
•Calculate effective RMS emittance at the end of linac from variations of (x,x’) of bunches
‣ compare to nominal emittance 0.3 mm*mrad (relative emittance)
23
bunch
0 10 20 30 40 50 60 70 80 90 100310×
rela
tive
emitt
ance
cha
nge
0
0.01
0.02
0.03
0.04
0.05
0.06-310×
Effective emittance
• Effective emittance at the end of linac for single linac configuration as a function of bunch number
24
Maximum emittance growth
Initial emittance growth after turn on, scales ~ Qext
Emittance variations due to Qb variations
Maximum emittance growth, VECC design
• 100 configurations of linac with random cavity misalignment of 0.5 mm
‣ Median value of relative emittance growth is 5e-5
‣ 95% of configurations have relative emittance growth less than 1e-3
25Relative emittance growth
0 0.0001 0.0002 0.0003 0.0004 0.0005 0.0006 0.0007 0.0008 0.0009 0.0010
5
10
15
20
25
30
35
40
45
Maximum emittance growth, FNAL design
• 100 configurations of linac with random cavity misalignment of 0.5 mm
‣ Median value of relative emittance growth is 2.5e-5
‣ 95% of configurations have relative emittance growth less than 1.5e-4
26
Relative emittance growth0 0.0001 0.0002 0.0003 0.0004 0.0005 0.0006 0.0007 0.0008 0.0009 0.0010
10
20
30
40
50
60
70
80
Conclusion
• Study effects of dipole HOM excitation on transverse beam dynamics in PIP2 linac with 10 mA peak current
‣ Considered mode with largest (R/Q)=80 kOhm/m2, f=1376 MHz, Qext=1e7
‣ 0.5 mm random cavity misalignment
‣ 10% bunch charge variations
• Relative emittance growth is 4e-5 - should not be a problem in PIP2 linac with 10 mA
• Study effects of dipole HOM excitation on transverse beam dynamics in LB650 section of PIP2 linac with different designs of LB650 cavities
‣ Considered modes with largest (R/Q)
- VECC design, mode 0.974 GHz, (R/Q)=38 Ohm/m**2, df=1MHz
- FNAL design, mode 0.973 GHz, (R/Q)=35 Ohm/m**2, df=2 Mhz
‣ 0.5 mm random cavity misalignment
‣ 10% bunch charge variations
• Relative emittance growth is 5e-5 for VECC design and 2.5e-5 for FNAL design
• No considerable effects are expected for INFN design
27