HiLumi LHC FP7 High Luminosity Large Hadron Collider Design Study - - PDF document

CERN-ACC-SLIDES-2014-0068

HiLumi LHC

FP7 High Luminosity Large Hadron Collider Design Study

Presentation Possible Strategy for Scrubbing LHC for 25 ns Operation

Rumolo, G (CERN) et al

27 November 2013

The HiLumi LHC Design Study is included in the High Luminosity LHC project and is partly funded by the European Commission within the Framework Programme 7 Capacities Specific Programme, Grant Agreement 284404. This work is part of HiLumi LHC Work Package 2: Accelerator Physics & Performance.

The electronic version of this HiLumi LHC Publication is available via the HiLumi LHC web site <http://hilumilhc.web.cern.ch> or on the CERN Document Server at the following URL: <http://cds.cern.ch/search?p=CERN-ACC-SLIDES-2014-0068>

CERN-ACC-SLIDES-2014-0068

Possible strategy for scrubbing LHC for 25 ns operation

• G. Rumolo, G. Iadarola, H. Bartosik, G. Arduini

Electron cloud simulation team: H. Bartosik, O. Dominguez, G. Iadarola, K. Li, S. Rioja Fuentelsaz, G. Rumolo, F. Zimmermann Acknowledgments: S. Claudet, L. Tavian (heat load data and general support for cryo),

• J. Esteban-Müller and E. Shaposhnikova (stable phase shift data), C. Zannini (power loss

calculations from impedance), J. Wenninger, W. Höfle, G. Kotzian, D. Valuch, P. Baudrenghien,

• T. Lefevre, B. Salvant, V. Baglin, R. Cimino, M. Taborelli, BE-ABP, BE-RF, BE-BI, BE-OP,

TE-ABT, TE-CRG, TE-VSC LMC, 27 November 2013

Outline

• Recapitulation of the main facts (2011-2012 experience with 25 ns beams)
• A possible path for LHC start up in 2015
• Use of the doublet beam
• Motivation
• Options (and issues) for a doublet beam in the LHC

− SPS experience and comparison with simulations − Production schemes (SPS or LHC) − Possible issues in the SPS and LHC

• Conclusions

LMC, 27 November 2013 1

1 1.2 1.4 1.6 1.8 2 2.2 10

• 4

10

• 2

10 10

2

10

4

SEY Heat load [W/hc/beam]

2

Scrubbing in the arcs in 2011-2012

LMC, 27 November 2013

Dipole 50 ns Dipole 25 ns Quadrupole 50 ns Quadrupole 25 ns

After 50 ns scrubbing (2011)

1 1.2 1.4 1.6 1.8 2 2.2 10

• 4

10

• 2

10 10

2

10

4

SEY Heat load [W/hc/beam] 25 ns Dipole Quadrupole

3

With 25 ns scrubbing @450 GeV (2011 + 2012) 450 GeV

• Arc quads: still significant heat

load (SEY ≈ 1.3)

• Arc dipoles: e-cloud 10x – 15x

lower (SEY ≈ 1.45, integrated effect about same as quads)

• Beam still affected by e-cloud

(emittance blow up, lifetime)

Scrubbing in the arcs in 2011-2012

LMC, 27 November 2013

Based on what we learnt from the triplets analysis

1 1.2 1.4 1.6 1.8 2 2.2 10

• 4

10

• 2

10 10

2

10

4

SEY Heat load [W/hc/beam] 25 ns Dipole Quadrupole

4

4 TeV

• SEY threshold in dipoles

decreases (lower transverse beam sizes, photoelectrons) ?

• SEY increases with magnetic

field ?

• Either way, the dipoles

become again dominant and no further scrubbing occurs

Scrubbing in the arcs in 2011-2012

LMC, 27 November 2013

1 1.2 1.4 1.6 1.8 2 2.2 10

• 4

10

• 2

10 10

2

10

4

SEY Heat load [W/hc/beam] 25 ns Dipole Quadrupole

5

 Electron cloud not detrimental to beams in collision at 4 TeV (from emittance data), emittance blow up @450  Arc heat load would exceed capacity of the cryogenic system (~ x2) extrapolating to full 25 ns beam @7 TeV  Do we have a means to achieve better scrubbing of the dipoles at 450 GeV ?

Scrubbing in the arcs in 2011-2012

LMC, 27 November 2013

FAQ: how do we know it’s all electron cloud?

• The enhanced heat load at 4 TeV (in the dipoles) seems NOT to

a.

Degrade the beam further

b.

Lead to further scrubbing (at least, visibly on the time scale of a fill)

LMC, 27 November 2013 6

Is it electron cloud ?

① These observations are compatible with electron cloud:

a.

The beam is more rigid and other effects are dominant over the e-cloud

b.

Scrubbing in a cold dipole has saturated or become inefficient, at least with the same nominal 25 ns beam used to get to that point and at 4 TeV

FAQ: how do we know it’s all electron cloud?

② Comparison between 50 ns and 25 ns

• 50 ns heating of the arc beam screen is fully consistent with impedance model

LMC, 27 November 2013 7

Heat load measurement from cryogenics Estimation (impedance + synchrotron rad.)

50 ns

FAQ: how do we know it’s all electron cloud?

LMC, 27 November 2013 8

Heat load measurement from cryogenics Estimation (impedance + synchrotron rad.)

25 ns

② Comparison between 50 ns and 25 ns

• 50 ns heating of the arc beam screen is fully consistent with impedance model
• 25 ns heating is 10x the value expected from impedance and synchrotron radiation

FAQ: how do we know it’s all electron cloud?

LMC, 27 November 2013 9

② Comparison between 50 ns and 25 ns

• 50 ns heating of the arc beam screen is fully consistent with impedance model
• 25 ns heating is 10x the value expected from impedance and synchrotron radiation
• Lines of the 25 ns spectrum twice less dense than those of the 50 ns spectrum

FAQ: how do we know it’s all electron cloud?

LMC, 27 November 2013 10

③ Bunch-by-bunch stable phase shift measurements

• Clearly show the build up shape along the bunch train at 450 GeV
• Reveal the same structure, but with steeper increase and larger saturation value, at

4 TeV

Beam 2

Fill 3429: 11 trains of 72 bunches

11 Commissioning (low intensity/luminos ity) Vacuum conditioning (50ns/short trains 25ns) (5-7 days) Scrubbing with 25ns (2 days)

50ns (intensity ramp up + physics)

δdip≥2.2 δdip≈2.2 δdip≤2. 450GeV 6.5 – 7 TeV

Proposed strategy for LHC start up in 2015

Start up after LS1  Phase 1 (conditioning + 50 ns)

⇒ Unconditioned machine with vented and new surfaces ⇒ Need for decrease of SEY (scrubbing), but also decrease of desorption yield, η  “intense” conditioning phase ⇒ Further conditioning will benefit from enhanced synchrotron radiation at 6.5 – 7 TeV, especially during the intensity ramp up

LMC, 27 November 2013 11

12

Proposed strategy for LHC start up in 2015

Phase 2 (25 ns)

⇒ Several improvements (thanks to TE/CRG, BE/RF, BE/OP)  Increased available cooling power of the SAMs  Faster injections  Heat load and stable phase shift measurements available online for scrubbing monitoring and steering  New monitoring equipment (e.g., 8 new thermometers in 5 half-cells of sector 45 to disentangle quad and dipole heating, new vacuum gauges) ⇒ Possibility to use doublet beams (see next slides) ⇒ Thermal cycle on the beam screen before scrubbing run to remove excess gas on wall ? (V. Baglin, in Chamonix Proc. 2003 and 2004)

LMC, 27 November 2013 12

25ns scrubbing (5 days) Scrubbing with doublet beams (≈ 5 days) 25ns test ramps + commissioning (≈ 5 days)

450GeV 6.5 – 7 TeV δdip≤2. δdip≈1.45 δdip≤1.4

1 1.2 1.4 1.6 1.8 2 2.2 10

• 4

10

• 2

10 10

2

10

4

SEY Heat load [W/hc/beam] 25 ns Dipole Quadrupole

LMC, 27 November 2013 13 13

Doublet beam, hopefully … (2015) dipoles with scrubbing beam

Motivation for the scrubbing beam

Scrubbing with 5 ns doublets

• The 5 ns doublet beam exhibits a much lower multipacting threshold

compared to the standard 25 ns beam

14

1.1 1.2 1.3 1.4 1.5 10

• 6

10

• 4

10

• 2

10 10

2

SEY Scrubbing dose (50eV) [mA/m] 0.50e11ppb 0.60e11ppb 0.70e11ppb 0.80e11ppb 0.90e11ppb 1.00e11ppb 1.10e11ppb

• Std. 25 ns

10 20 30 40 50 60 70 2 4 x 10

11

Beam prof. [p/m] 10 20 30 40 50 60 70 1 1.5 2 2.5 3 3.5 Time [ns] Ne / Ne(0)

LHC dipoles

LMC, 27 November 2013

• The 5 ns doublet beam exhibits a much lower multipacting threshold

compared to the standard 25 ns beam

• Efficient scrubbing with the doublet beam expected from e- energy spectrum

for a wide range of intensities

• Population ≥ 0.8x1011 p/b preferable to cover similar horizontal region as the

standard 25 ns beam with nominal intensity

15

• 15
• 10
• 5

5 10 15 10

• 4

10

• 3

10

• 2

10

• 1

10 10

1

10

2

sey = 1.50 Position [mm] Scrubbing current (50eV) [A/m2]

0.70e11ppb 0.80e11ppb 0.90e11ppb 1.00e11ppb

• nom. 25 ns

200 400 600 800 1000 10

• 4

10

• 3

10

• 2

10

• 1

10 sey = 1.50 Energy [eV] Normalized energy spectrum 0.70e11ppb 0.80e11ppb 0.90e11ppb 1.00e11ppb

• nom. 25 ns

LHC dipoles LHC dipoles

Scrubbing with 5 ns doublets

LMC, 27 November 2013

• The 2.5 ns doublet beam exhibits lower multipacting threshold compared to

the standard 25 ns beam, but higher threshold compared to 5 ns doublets

• Similar e- energy spectrum as with 5 ns doublets
• E-cloud build-up is concentrated in central part of the chamber  less

favorable compared to the 5 ns doublets !!

16

• 15
• 10
• 5

5 10 15 10

• 4

10

• 3

10

• 2

10

• 1

10 10

1

10

2

sey = 1.50 Position [mm] Scrubbing current (50eV) [A/m2]

0.70e11ppb 0.80e11ppb 0.90e11ppb 1.00e11ppb

• nom. 25 ns

LHC dipoles

Scrubbing with 2.5 ns doublets

1.1 1.2 1.3 1.4 1.5 10

• 5

10

• 4

10

• 3

10

• 2

10

• 1

10 10

1

SEY Scrubbing dose (50eV) [mA/m] 0.50e11ppb 0.60e11ppb 0.70e11ppb 0.80e11ppb 0.90e11ppb 1.00e11ppb 1.10e11ppb

• Std. 25 ns

LMC, 27 November 2013

2012-13 experience with 5 ns doublet beam (SPS)

• First machine tests in the SPS at the end of 2012-13 run in order to
• validate the doublet production scheme at SPS injection
• demonstrate the e-cloud enhancement
• The production scheme based on splitting at injection has been successfully

tested

• Injection of two trains of 72 bunches with 1.7e11 p/doublet
• Cycle included the start of acceleration to estimate capture losses (around 10%)

17 4 2 8 6 1 2 3

Time [ms] 200 MHz RF Voltage [MV]

4

• 2

1st inj.

0.92 0.94 0.96 0.98 1 1.02

• 0.02

0.02 0.04 0.06 Time [µs] Beam profile [a.u.] Turn 0.92 0.94 0.96 0.98 1 1.02 100 200 300 400 500

LMC, 27 November 2013

• Stronger pressure rise with doublet beam indicates enhanced e-cloud build-up in

the SPS arcs

• Direct comparison of standard and doublet beam within the same supercycle

18 25ns std. (1.65e11p/bunch) (1.65e11p/doublet) 25ns “doublet”

the curves represent pressure gauges in the center of the SPS arcs

LMC, 27 November 2013

2012-13 experience with 5 ns doublet beam (SPS)

• Stronger pressure rise with doublet beam indicates enhanced e-cloud build-up in

the SPS arcs

• Direct comparison of standard and doublet beam within the same supercycle
• Clear enhancement observed also in the dedicated e-cloud monitors
• MBA type chamber  e-cloud only with doublet beam

19 LMC, 27 November 2013

2012-13 experience with 5 ns doublet beam (SPS)

• Stronger pressure rise with doublet beam indicates enhanced e-cloud build-up in

the SPS arcs

• Direct comparison of standard and doublet beam within the same supercycle
• Clear enhancement observed also in the dedicated e-cloud monitors
• MBA type chamber  e-cloud only with doublet beam
• MBB-type chamber  Build-up with doublet beam stronger and concentrated in

central region, as expected from PyECLOUD simulations

20 PyECLOUD simulation Measurements LMC, 27 November 2013

2012-13 experience with 5 ns doublet beam (SPS)

Schemes for doublet production

• “Long” bunch splitting at SPS injection (5 ns doublets)
• Extracting long bunches (~10ns) from the PS and capturing them in two

neighboring LHC buckets  5 ns doublet spacing

• Demonstrated in the SPS @26 GeV
• Bunch splitting at high energy in the SPS (5 ns doublets)
• By sudden phase jump by 180° and recapturing each bunch in 2 neighboring

buckets

− A controlled phase jump will be possible with new module presently under

development for operation with ions (to be tested in 2014)

− Preferably done at intermediate energy to clean up uncaptured beam before

extraction and have shorter bunches at extraction

• Not tested yet
• “Long bunch splitting” at LHC injection (2.5 ns doublets)
• Extracting long bunches (~5ns) from the SPS and capturing them in two

neighboring LHC buckets  2.5 ns doublet spacing

• Not tested yet

21 LMC, 27 November 2013

Possible issues (under investigation)

• In the SPS
• Transverse beam stability due to enhanced e-cloud  losses and emittance

growth

• Transverse damper (with new 200 MHz electronics will be able to damp the

common oscillation mode, but dead-time period during splitting)

• Acceleration: RF power limitation, longitudinal stability, LLRF
• Beam quality at extraction
• In the LHC (being followed up at the LBOC)
• Possible injection/capture losses and satellite production due to degraded

quality or long bunches

• BI  False readings of interlocked BPMs (up to 2-4 mm), workaround being

studied

• RF compatibility  OK for doublets
• ADT  OK for 5 ns doublets, dead-time during splitting for 2.5 ns doublets

22 LMC, 27 November 2013

• No additional impedance heating of sensitive components is expected with

the doublet beam (2.5 or 5 ns)

• Beam power spectrum is modulated with cos2 function
• For total doublet intensity equal to 25 ns intensity, lines can only be

weakened by the modulation

23

Beam induced heating with doublets

LMC, 27 November 2013

LMC, 27 November 2013 24

Proposed strategy for LHC start up in 2015: Summary

Commissioning (low intensity/lumin

• sity)

Vacuum conditioning (50ns/short trains 25ns) (5-7 days) Scrubbing with 25ns (2 days)

50ns (intensity ramp up + physics)

δdip≥2.2 δdip≈2.2 δdip≤2. 450GeV 6.5 – 7 TeV Intensity ramp up & physics with 25ns

25ns scrubbing (5 days) Scrubbing with doublet beams (≈ 5 days) 25ns test ramps + commissioning (≈ 5 days)

450GeV 6.5 – 7 TeV δdip≤2. δdip≈1.45 δdip≤1.4

LMC, 27 November 2013 25 Commissioning (low intensity/lumin

• sity)

Vacuum conditioning (50ns/short trains 25ns) (5-7 days) Scrubbing with 25ns (2 days)

50ns (intensity ramp up + physics)

δdip≥2.2 δdip≈2.2 δdip≤2. 450GeV 6.5 – 7 TeV

Back up strategy for LHC start up in 2015

Intensity ramp up & physics with 25ns With electron cloud (degraded beam, scrubbing) With low e-cloud filling patterns

25ns scrubbing (5 days) Scrubbing with doublet beams (≈ 5 days) 25ns test ramps + commissioning (≈ 5 days)

450GeV 6.5 – 7 TeV δdip≤2. δdip≈1.45 δdip≤1.4

Summary and Conclusions

• Scrubbing was observed to saturate, or become inefficient, in 2012
• 25 ns beams are still degraded from e-cloud at 450 GeV
• The heat load in the arcs could exceed by a factor 2 the cooling capacity at 7 TeV
• Proposed LHC start up plan in 2015. Steps:

① First scrubbing run (≈ 7 days)  50 ns operation ② Second scrubbing run with nominal 25 ns and doublet beam + test ramps (≈ 10 +

5 days)  25 ns operation

• Possible scrubbing beams for the LHC (being followed up at the LBOC)
• 5 ns doublets (at SPS injection or at high energy)  The most promising !!
• 2.5 ns doublets at LHC injection  Less efficient than 5 ns doublets, maybe to be

kept as a backup solution?

• Main issues are the high intensity and bunch splitting (if @high energy) in the

SPS, the electron cloud itself, no major showstoppers found yet in the LHC

• Back up scenario if doublet scrubbing fails  go into physics with 25 ns beams

(either low e-cloud filling pattern or with degraded beam & further scrubbing)

26 LMC, 27 November 2013

THA THANK YOU FO U FOR Y YOUR UR ATTEN TTENTION

27 LMC, 27 November 2013

Scheme (I) for doublet production

• “Long” bunch splitting at SPS injection (5 ns doublets)
• Demonstrated in MDs
• Possible issues in the SPS

− Transverse beam stability due to enhanced e-cloud  losses and emittance

growth

− Transverse damper (with new 200 MHz electronics will be able to damp the

common oscillation mode, but dead-time period during splitting)

− Acceleration: RF power, longitudinal stability, LLRF − Beam quality at extraction

• In the LHC (being followed up at the LBOC)

− Possible injection/capture losses due to degraded quality (if longitudinally

unstable in the SPS)

− BI  False readings of interlocked BPMs (up to 2-4 mm) − RF compatibility  OK for doublets − ADT  OK for this type of doublets

28 LMC, 27 November 2013

• Bunch splitting at high energy in the SPS (5 ns doublets)
• By sudden phase jump by 180° and recapturing each bunch in 2 neighboring

buckets

− A controlled phase jump will be possible with new module presently under

development for operation with ions (to be tested in 2014)

− Preferably done at intermediate energy to clean up uncaptured beam before

extraction and have shorter bunches at extraction

• Not tested yet
• Possible issues in the SPS

− Similar to scheme (I), but e-cloud effects should be less critical because the

splitting enhances the e-cloud only at high energy

− Splitting at high energy: LLRF, capture losses at high energy, longitudinal

stability during/after the splitting

• In the LHC (being followed up at the LBOC)

− Similar to scheme (I) − Capture losses and production of satellites due to long bunches from SPS

29

Scheme (II) for doublet production

LMC, 27 November 2013

• “Long bunch splitting” at LHC injection (2.5 ns doublets)
• Extracting long bunches (~5ns) from the SPS and capturing them in two

neighboring LHC buckets  2.5 ns doublet spacing

• Not tested yet
• Possible issues in the SPS

− Acceleration of the needed high beam intensity in the SPS: RF power,

longitudinal stability

− Transverse beam stability due to e-cloud

• In the LHC (being followed up at the LBOC)

− BI  False readings of interlocked BPMs (up to 2-4 mm) − RF compatibility  OK for doublets − ADT  Dead-time period during splitting, when Σ signal on pick up goes below

threshold

− Capture losses due to splitting and RF voltage dips at successive injections

30

Scheme (III) for doublet production

LMC, 27 November 2013

Doublet beam parameters

• We need to accelerate

1.6 x 1011 p/doublet in the SPS

• εx,y ~3 μm - assuming all benefits from

larger longitudinal parameters at PS injection

• Limited by RF power, intensity perhaps

achievable with a 3x slower ramp rate (most probably at the expense of degraded beam quality)

• Longitudinal emittance
• Refined estimations from simulations

underway (and including possible controlled blow up for instability considerations …)

31 LMC, 27 November 2013

10 20 30 40 50 60 70 10 20 30 40 50 60 70 Time [ns]

Production of a doublet beam

• Place (long) bunches on the unstable phase (at injection or with phase

jump) and lower RF voltage

32

• Long. beam profile

∆E

LMC, 27 November 2013

Production of a doublet beam

• Place (long) bunches on the unstable phase (at injection or with phase

jump) and lower RF voltage

• Ramp the voltage back up in order to recapture each bunch in the two

neighboring 200 MHz (or 400 MHz) buckets

33 10 20 30 40 50 60 70 g 10 20 30 40 50 60 70 Time [ns]

• Long. beam profile

∆E

25 ns 25 ns 2.5 or 5 ns

LMC, 27 November 2013

Scrubbing with high intensity 25 ns beams

• What is the potential of the high intensity 25 ns beam (up to 1.7 x 1011 p/b) to

improve the scrubbing of LHC at 450 GeV (although possibly degraded) ?

• The SEY thresholds in dipoles do not change much (but e-cloud could be initially

enhanced by the shift of the stripe position)

34 LMC, 27 November 2013

1 1.2 1.4 1.6 1.8 2 10

• 6

10

• 4

10

• 2

10 10

2

SEY Heat load [W/m]

1.10e11 ppb 1.30e11 ppb 1.50e11 ppb 1.70e11 ppb

LHC dipoles

• 20
• 10

10 20 10

• 4

10

• 2

10 10

2

Position [mm] Heat load density [W/m

2]

Scrubbing with high intensity 25 ns beams

• What is the potential of the high intensity 25 ns beam (up to 1.7 x 1011 p/b) to

improve the scrubbing of LHC at 450 GeV (although possibly degraded) ?

• The SEY thresholds in dipoles do not change much (but e-cloud could be initially

enhanced by the shift of the stripe position)

• The energy spectrum of the electrons on wall moves to higher energies

(possible impact on scrubbing in cold environment?)

35 LMC, 27 November 2013 200 400 600 800 1000 10

• 4

10

• 3

10

• 2

10

• 1

10 Energy [eV] Normalized energy spectrum

1.10e11 ppb 1.30e11 ppb 1.50e11 ppb 1.70e11 ppb

LHC dipoles

1.10e11 ppb 1.30e11 ppb 1.50e11 ppb 1.70e11 ppb

Scrubbing with high intensity 25 ns beams

• What is the potential of the high intensity 25 ns beam (up to 1.7 x 1011 p/b) to

improve the scrubbing of LHC at 450 GeV ?

• The SEY thresholds in dipoles do not change much (but e-cloud could be initially

enhanced by the shift of the stripe position)

• The SEY thresholds in the quads become higher

36 LMC, 27 November 2013

1 1.2 1.4 1.6 1.8 2 10

• 4

10

• 2

10 10

2

SEY Heat load [W/m] 1 1.2 1.4 1.6 1.8 2 10

• 6

10

• 4

10

• 2

10 10

2

SEY Heat load [W/m] 1.10e11 ppb 1.30e11 ppb 1.50e11 ppb 1.70e11 ppb 1.10e11 ppb 1.30e11 ppb 1.50e11 ppb 1.70e11 ppb

LHC dipoles LHC quads

SPS RF power during acceleration (I)

• Possible ways to alleviate RF power limitations
• Reduce ramp rate (example below for 3 times longer acceleration time Tacc)
• Slightly less power needed in Q26, but other problems anticipated for high

intensity (e.g. TMCI)

• 1.6x1011 p/doublet seems within reach  0.8x1011 p/b
• however controlled long. emittance blow-up will be needed  to be checked in

measurements

37

Q26 Q20

LMC, 27 November 2013

SPS RF power during acceleration (II)

• Possible ways to alleviate RF power limitations
• Reduce ramp rate (example below for 3 times longer acceleration time Tacc)
• Slightly less power needed in Q26, but other problems anticipated for high

intensity (e.g. TMCI)

• 2x1011 p/doublet out of reach with present 200 MHz RF system

38

Q26 Q20

LMC, 27 November 2013

PyECLOUD simulations – 2.5 ns doublets

• The 2.5 ns doublet beam shows a lower multipacting threshold compared to

the standard 25 ns beam, but higher threshold compared to 5 ns doublets

39

10 20 30 40 50 60 70 2 4 x 10

11

Beam prof. [p/m] 10 20 30 40 50 60 70 1 1.5 2 2.5 3 Time [ns] Ne / Ne(0) 1.1 1.2 1.3 1.4 1.5 10

• 5

10

• 4

10

• 3

10

• 2

10

• 1

10 10

1

SEY Scrubbing dose (50eV) [mA/m] 0.50e11ppb 0.60e11ppb 0.70e11ppb 0.80e11ppb 0.90e11ppb 1.00e11ppb 1.10e11ppb

• Std. 25 ns

LHC dipoles

LMC, 27 November 2013

Tests of the 5 ns doublet beam in the SPS

• First machine tests in the SPS at the end of 2012-13 run in order to
• validate the doublet production scheme at SPS injection
• btain first indications about the e-cloud enhancement
• The production scheme has been successfully tested
• for a train of up to (2x)72 bunches with 1.7e11 p/doublet
• injecting a second batch without degrading the circulating beam has been shown
• Cycle included the start of acceleration to estimate capture losses (around 10%)

40 3604 3600 3602 3598 3596 3594 3592 3590 1 2 3

Time [ms] 200 MHz RF Voltage [MV]

4

1st inj. 2nd inj.

4 2 8 6

• 2

5 10 15 20 0.5 1 1.5 2 2.5 Time [ns] Longitudinal beam profile [a.u.] First bunch (of 2 single) after the second inj. After 1st inj. After 2nd inj.

LMC, 27 November 2013

Production of a doublet beam

• Place (long) bunches on the unstable phase (at injection or with phase

jump) and lower RF voltage

41 10 20 30 40 50 60 70 10 20 30 40 50 60 70 Time [ns]

• Long. beam profile

∆E

LMC, 27 November 2013

Production of a doublet beam

• Place (long) bunches on the unstable phase (at injection or with phase

jump) and lower RF voltage

• Ramp the voltage back up in order to recapture each bunch in the two

neighboring 200 MHz (or 400 MHz) buckets

42 10 20 30 40 50 60 70 g 10 20 30 40 50 60 70 Time [ns]

• Long. beam profile

∆E

25 ns 25 ns 2.5 or 5 ns

LMC, 27 November 2013