Transcript Necessary LIU studies in the injectors during 2012 Giovanni Rumolo H. Bartosik, Y.

Necessary LIU studies in
the injectors during 2012
Giovanni Rumolo
H. Bartosik, Y. Papaphilippou
in LHC Performance Workshop (Chamonix 2012), 9 February 2012
H. Damerau, A. Findlay, S. Gilardoni, B. Goddard, S. Hancock, K. Hanke,
G. Iadarola, B. Mikulec, E. Shaposhnikova
et al.
Outline
Focus of the talk  Review the progress made in 2011 with the
LIU MDs and extract which questions still remain to be answered
with beam in 2012
– 2011 LIU MDs by the machine – organization, distribution and
highlights
→ Linac2 + PSB
→ PS
→ SPS
– Questions to be followed up in 2012
– Priorities and organizational proposals
– Concluding remarks
2
Overview 2011
– 2011 distribution of the MD
days (priority to MDs in all
machines)
→ Floating MDs biweekly
→ Long dedicated blocks
during LHC TS (4x)
– Total number of available MD
hours was 408 (434 on
original schedule, same
planned for 2012)
– Parallel MDs
→ PSB + PS: at least one cycle
per supercycle always
available for MDs (or MTE)
→ SPS: usually one parallel MD
cycle available every week
day for studies (some times
used for set up)
3
Linac2 + PSB in 2011
Dedicated MDs
Parallel MDs
– Linac2 MDs
1. Remove hot spots found with
the 2010 radiation survey
2. Run with increased beam
current (180 mA)
– Types of PSB MDs
1.
2.
3.
Prepare new beams, tune and check
existing beams, for the downstream
machines
Define the limitations of the machine,
improve the performance
Specific machine studies in view of the
future upgrades
4
Linac2 + PSB in 2011
Dedicated MDs
Parallel MDs
– Linac2 MDs
1. Remove hot spots found with
the 2010 radiation survey
2. Run with increased beam
current (180 mA)
– Types of PSB MDs
1.
Time, resources!
2.
3.
Prepare new beams, fine tune and check
existing beams, for the downstream
machines
Define the limitations of the machine,
improve the performance
Specific machine studies in view of the
future upgrades
5
Linac2 + PSB
– LIU-PSB activities in 2012 (RF, hardware)
① Continue deployment of the digital RF control
① Test the newly installed Finemet prototype
cavity hardware
– Beam dynamics questions relevant for LIU-PSB
?
①
①
②
②
②
Continue tests with higher beam current from Linac2
Identify the impedance source responsible for the known instabilities
and specify transverse damper requirements
Determine resonance diagram with tune scans at 160 MeV to optimize
placement of working point at injection with Linac4
Optics model based on turn-by-turn data from the available BPMs
Study the efficiency of the resonance compensation schemes
Space charge induced emittance blow up
6
PS in 2011
Dedicated MDs
– PS MDs
1.
Parallel MDs
2.
3.
4.
5.
6.
7.
8.
Space charge studies at 1.4 GeV and 2 GeV
and tune diagrams
Adjustment of working point at low energy
with quads and PFW
Fast instability at transition with TOF-like
beams
Tune shift with intensity at injection and
extraction energy
Electron cloud at flat top with 25 and 50ns
beams
Limitations from longitudinal coupled bunch
instabilities during the ramp and flat top
Batch compression scheme
h=9 10  20 21
One-turn feedback with C11
7
PS: space [email protected]
Nb (x 1010 p)
ex,y (mm)
4st (ns)
DQy
LHCINDIVhigh
40
1.7
90
-0.2
LHC50nom (DB)
80
1.1
180
-0.26
LHC50ult (DB)
120
1.8
180
-0.26
LHC25 (DB)
160
2.5
180
-0.26
AD
400
9.0/5.0
185
-0.28
TOF
800
12.0/8.0
230
-0.31
LHC50 SB
rebucketed
150.0-190.0
2.5-3.0
130
-0.34
LHChighbright
70
2.0
95
-0.25
Double batch
LHC beams,
1.2sec @FB
Physics beams,
short FB
MD beams
For space
charge studies
@FB after
bunch rotation
8
PS: space charge@2GeV
E. Benedetto
Shortened bunch @2GeV
→ DQx = -0.19, DQy=-0.27
→ Three working points analyzed
o
Qx=0.15, Qy=0.196
o
Qx=0.17, Qy=0.23
o
Qx=0.17, Qy=0.30
Is DQy=- 0.26 really the limit for the
space charge tune spread at injection
for double batch injection into the PS?
Can we identify dangerous resonance
lines and compensate them?
- No losses
- Little emittance growth
TUNE SCAN @ 2 GeV
9
PS: longitudinal coupled
bunch instabilities
• Longitudinal coupled bunch instabilities with both 25ns and 50ns
beams observed (previously also with 75ns and 150ns beams)
 During the ramp
 At flat top when ramping down h=21 during bunch splitting
• What we have learnt about it
⇒ Instabilities during the ramp probably caused by the wide band
impedance of the 10MHz cavities
⇒ Coupled bunch mode spectrum changes between ramp and flat top
⇒ Little dependence on the number of bunches in the batch, but growth
rates scale like Nb/ez
⇒ Small improvement with 2 gap relays, parking of unused cavities
beneficial
• 2012 LIU-PS studies towards raising the thresholds
⇒ Additional feedback (decide if more hardware is needed after LS1)
⇒ PS/SPS transfer  perhaps allow for larger ez to be injected into the
SPS, optimize final bunch rotation
10
PS: Electron cloud
Systematic scans taken with
⇒ 50ns and 25ns beams
⇒ Simulation
ongoing
Av. effort
intensity = 1.64*10
ppb
G. Iadarola, C. Yin-Vallgren
50ns, 36b
0.04
1
1
Scans in bunch intensity and length
done!
Av. intensity
= 1.33*10
0.8
0
0.5
1
time [us]
0.4
5
0.2
0.6
11
ppb
0.15
25ns, 72b
0.1
0.4
1.5
2
0.2
0
1.5
Simulation
0.5
0
2
2012 MDs  Measurements in presence
0.05
of B field, study of double0 step bunch
0
0.5
1
rotation
time [us]
0.6
10
0
1.5
e-cloud signal [a.u. ]
0.8
0
Simulation
1
time [us]
0.5
1
Time [us]
2
1.5
2
0
0
Pick-up signal [a.u.]
0
e-cloud signal [a.u.]
Electron cloud indicator [a.u]
0.02
Pick-up signal [a.u.]
e-cloud signal [a.u. ]
11
Measurement
6
0.5
4
1
Time [us]
1.5
2
2
0
dmax
R0
Beam in the gap
1.6
0.5
5%
0
0.5
1
time [us]
1.5
2
11
PS: miscellaneous
– More LIU-PS machine studies in 2012
➀
➀
➀
➁
➁
➁
➁
Batch compression scheme h=9 10  20 21, acceleration,
transfer to SPS
Batch compression + bunch merging
One-turn feedback against transient beam-loading
Commissioning of transverse feedback system
Head-tail instabilities on the flat bottom
Transverse instabilities of short intense bunches at flat top
(electron cloud?)
Impedance identification for modeling
③ Injection studies
12
SPS in 2011
Dedicated MDs
39%
(very high
efficiency ~95%)
Distribution of SPS MDs in 2011
21%
Floating MDs
61%
(efficiency slightly
above 50%)
38%
Upgrade studies
Other MDs
41%
Beam set up (ion
and coasts)
– Upgrade studies included
→ Push the SPS performance with the present 25 and 50ns beams, as well
as with single intense bunches (up to 4 x 1011 ppb)
→ Development of low gamma transition optics (Q20)
→ Electron cloud mitigation techniques (a-C coating, scrubbing, clearing
electrodes)
→ Beam tests for a high bandwidth feedback system
13
SPS: single bunch limits
Limitations on single bunch
⇒ TMCI threshold
⇒ Emittance blow up along an LHC-type cycle
xV=0.25
TMCI threshold
@26 GeV/c (xV
=0)
Q26 – nominal optics
– 2–2.5 x 1011 ppb
injected
→ Losses around 10%
→ No emittance
growth wrt PS
extraction
– Above 2.5 x 1011 ppb
injected, large losses
(20%) and emittance
blow up
14
SPS: single bunch limits
Limitations on single bunch
⇒ TMCI threshold
⇒ Emittance blow up along an LHC-type cycle
xV=0.
1
Q20 – low gt optics
– Q20 allows for
injection of higher
intensity bunches (up
to 3 x 1011 ppb)
→ Low chroma
→ Losses below 10%
even above
2.5 x 1011 ppb
TMCI threshold
@26 GeV/c (xV =0)
– Trend from injectors
or space charge in
the SPS?
15
SPS: single bunch limits
Comparison across machines
⇒ Effect of a 3 sec flat bottom in the SPS
⇒ Small blow up + losses above 2.5 x 1011 ppb
Q20 – low gt optics
From measurements at 450 GeV/c
(previous slide)
Blow up above 1.7 x 1011 ppb, during flat bottom (3 to 10 sec) or during ramp?
Working point optimization needed?
16
SPS: multi-bunch beams
High intensity 50ns beams tested in MDs
⇒ Record intensity at flat top with Q20
⇒ Quoted transmission is from PS extraction to SPS flat top
1.9 x 1011 ppb injected
1.7 x 1011 ppb at flat top
Q20 – low gt optics
1.6 x 1011 ppb injected
1.5 x 1011 ppb at flat top
17
SPS: multi-bunch beams
High intensity 50ns beams tested in MDs
⇒ Longitudinally more stable on Q20
⇒ Margin for longitudinal emittance blow up, when needed
50ns beam at flat top
Continue work on optimization of multi- (1 batch, 1.5x1011ppb)
bunch beams on Q20 (also 25ns), especially
wrt longitudinal quality and CBI
– Without controlled
Q26 Injection tests into LHC Q20
12h sessions every week instead of 24h
every 2nd week could help (at least later in
the run)
no quadrupole
emittance blow-up
– 800MHz cavity
(V800=0.15V200)
oscillations
Courtesy T. Argyropoulos
18
SPS: electron cloud,
scrubbing or coating?
– Electron cloud in 2011
1.
2.
3.
Effects on the nominal 25ns beam have become
less evident from the start
More studies on mitigation techniques done
Progress for the high bandwidth feedback system
– 2012 studies
1.
2.
Scrubbing week in W13
☹ Perhaps negligible effect on the beam if we start like in 2011
☹ Most interesting scrubbing techniques not possible for now
(5ns or 10+15ns spacings from PS)
☺ Testing efficiency of scrubbing with uncaptured beam
☺ Monitor and qualify beam induced scrubbing under different
beam/chamber conditions (beam observables, direct electron
cloud observables)
☺ Validate simulation models on scrubbing times (like for LHC)
Some new setups for validation of a-C coating
19
SPS: miscellaneous
– More LIU-SPS machine studies in 2012
➀
➀
Tests with increased peak RF power
More on Q20
 Nonlinear optics model
 Split tunes (20, 26), coupling correction
 Instabilities (TMCI, ECI)
 Extension of Q20 to fixed target physics cycles
➁
➁
PS-SPS transfer studies
Terminate phase (1) for high bandwidth feedback studies (i.e.
close feedback loop and prove damping of head-tail modes)
Impedance identification
➁
20
All the machines
– Emittance preservation across the injector chain
G. Arduini, LMC 12/10/2011
Systematic measurements with
different beam types
Define where the emittance blow up
occurs
21
Closing remarks
Prioritized lists of needed/advisable LIU studies have been detailed out for
each machine
– PSB  More resources desirable for the key studies
→ Resonances at 160 MeV
→ Origin of instabilities, efficiency of transverse feedback in the enlarged
parameter range
– PS  Important questions
→ Space charge limit at injection
→ Feedback against CBI
→ Alternative production schemes – like batch compression
– SPS  Redistribution of the MD time + MD follow up meetings in the frame
of SPSU-BD WG in 2012 recommended
→ More frequent – and shorter – MD blocks to allow for more continuous
effort on Q20 optimization (with experts available)
→ 3 to 5-day dedicated block for scrubbing studies
22
THANK YOU FOR YOUR ATTENTION!