status & performance of the lhc (proton) injector complex

7th December 2006 LHC Machine Advisory Committee

Status & performance of the Status & performance of the LHC (proton) Injector ComplexLHC (proton) Injector Complex

G. Arduini – AB/ABP

Contributions from: S. Baird, H. Burkhardt, L. Ducimetiere, E. Gaxiola, B. Goddard, K. Hanke, S.

Hancock, E. Métral, B. Mikulec, F. Roncarolo, G. Rumolo, E. Shaposhnikova, R. Steerenberg, J. Tan, J. Wenninger,

F. Zimmermann et al.


OutlineOutline

The LHC Injector Complex: HW statusThe LHC beamsStatus of the Injector CommissioningPerformancePresent limitationsSummary and Conclusions


The LHC Injector ComplexThe LHC Injector Complex The LHC Proton Injectors:

– Linac 2 (1979)– Proton Synchrotron Booster - PSB

(1972)– Proton Synchrotron – PS (1959)– Super Proton Synchrotron – SPS

(1976)

Upgrade of the PS-SPS Complex to as LHC injector is almost complete.

Installation of the elements for the SPS fast extraction of Beam 1 to TT60 during last SD.

Main remaining item is the installation of TI2 (April-Aug 2007).


ConsolidationConsolidation

Important refurbishment program ongoing for the coils of the PS Main Magnets.– 26 during the 2004-2006 SD– 8 during the 2006-2007 SD– money available for

refurbishing ~50 magnets out of 100


ConsolidationConsolidation

Renovation of the water manifolds of the SPS MBB-type dipoles produced by one of the manufacturers (255):– 75 dipoles during the SD

2006-2007– The rest in ~3 slots in future

machine shut-downs


The LHC proton beam (25 and 75 The LHC proton beam (25 and 75 ns)ns) PSB@inj PSB@extr PS@inj PS@extr SPS@inj SPS@extr

p [GeV/c] 0.31 2.14 2.14 26 26 450K [GeV] 0.050 1.4 1.4 25.08 25.08 449.06

Trev [s] 1.67 0.572 2.29 2.1 23.07 23.05

Q (H/V) 4.3/5.45 4.2/5.2 6.22/6.25 26.13/26.18

tr 4.15 6.11 22.83

bunches/ring 0-1 0-1 1-6 1-6x121-6x4

2-4×12-722-4×4-24

2-4×12-722-4×4-24

Nb [1011 p] 13.820.4

13.820.4

13.820.4

1.151.7

1.151.7

1.151.7

Tbunch [ns] - - 326.88 24.9774.91

24.9774.91

24.9574.85

b [ns] 571 190 190 4 4 <2

*H,V [m] - <2.5 - <3 - <3.5

L [eV.s] ~0.7 1.40.9

1.40.9 0.35 0.35 <0.8

75 ns LHC beam / Nominal 25 ns LHC beam / Ultimate 25 ns LHC beam


……but not onlybut not only

During commissioning and machine physics,– Probe Beam, single bunch of 5x109, with low emittance– Single Bunch, from 5x109 up to 2x1011, at nominal

emittance During routine operation for proton physics

– Pilot Beam of 5 109, preferably at the physics emittance– Intermediate Beam (1 PSB bunch out of 6) corresponding

to:• 4 bunches in LHC at 75ns spacing• 12 bunches in LHC at 25ns spacing


……but not onlybut not only

43 and 156 bunch scheme consisting of small number of LHC bunches large bunch spacingno parasitic encountersno need of crossing angle

Initially “designed” for TOTEM expt.

Will be used during commissioning and Early Physics

R. Bailey & P. Collier, LHC_Project_Note 323_Rev.


Status of the InjectorsStatus of the Injectors

Machine Commissioned

LINAC2 Up to ultimate intensity

PSB Up to ultimate intensity

PS Up to nominal intensity

SPS Up to nominal intensity

TT40 Up to nominal intensity – including Interlocks

TT60 Pilot and LHC intermediate beam (12 bunches – 0.4x1011 p/bunch) – successful commissioning in 2006

TI8 Pilot and single LHC bunch

TI2 in 2007


Probe beam ”the smallest”Probe beam ”the smallest”

Injection oscillations and mismatch under control

Nbunch=5×109 p

SPS @ 450 GeV/c


Pilot and Commissioning beamsPilot and Commissioning beams

Production scheme reviewed in 2004 (h=1 instead of h=2 in PSB) it avoids “throwing away” a bunch/ring in the transfer from PSB to PS and it simplifies tuning of the PSB-PS transfer.

BeamNbunch

[1011p]

*H/V

[m]

L

[eV s]#bunches tb [ns]

Probe beam0.05-0.2

0.05< 1

<0.7\<0.7< 0.80.26

11

-

Pilot beam0.050.05

< 1<0.7-6\<0.7-14

<0.80.26

11

-

TOTEM/Commissioning beam

0.2-1.150.31.1

<3.50.95/0.85<1.8\<1.3

<0.80.40.4

1-4-161-4-16

525525

SPS @ 450 GeV/c

Controlled transverse emittance blow-up “pink noise” in the SPS TFB


LHC beam - multibunch LHC beam - multibunch

SPS @ 450 GeV/c

Nominal longitudinal emittance (<0.8 eV.s) achieved in 2002

Improvement resulting from:– Fine synchronization of the injection kicker magnet kick waveforms

(not nominal) still marginal.

– Independent correction of the injection of each (of the 4) batch from PS

– Non-linear chromaticity correction at flat-bottom

Transverse emittance still marginal in the vertical plane

2003 (LHCMAC 12/7/04) 2004

H [m] 2.6 ± 0.5 3.0 ± 0.3

V [m] 4.0 ± 0.2 3.6 ± 0.3

LHC beam - 25 ns


Beam Nbunch

[1011p]

*H/V

[m]

L

[eV s]

#b tb [ns]

Early -75ns 0.1-1.150.6

1.15

2.5\2.50.9±0.05\0.8±0.051.8±0.1\1.5±0.05

<0.80.60.6

1-6x41-4x24

75

Nominal 0.1-1.150.7

1.15

3.5\3.51.7\2

3.0±0.3\3.6±0.3

<0.80.60.6

1-6x121-4x72

25

LHC beam - multibunchLHC beam - multibunch


PSB limitationsPSB limitations

Space Charge limits:– LHC beam brightness.

Feasible for the NOMINAL beam in spite of the margin required to account for losses in PS and SPS (dashed line). Difficult to meet for the ULTIMATE beam, in particular for ring 3

Minimizing the losses in the downstream machines is mandatory as well as understanding the behaviour of PSB ring 3.

TSTLHC

1

1.5

2

2.5

3

3.5

4

4.5

5

100 150 200 250 300

Nbunch [1010]

NO

RM

. HO

R. E

MIT

TA

NC

E [ m

] Ring 1Ring 2Ring 3Ring 4All rings - average

TSTLHC

1.5

2

2.5

3

3.5

4

100 150 200 250 300

Nbunch [1010]

NO

RM

. VE

RT

. EM

ITT

AN

CE

[m

]

Ring 1Ring 2Ring 3Ring 4All rings - average


PS limitationsPS limitations

Losses mainly affecting more intense and/or shorter bunches probably related to space charge driven resonance trapping phenomena. Further optimization of the working point might reduce losses.

LHC beam in the PS

0

2

4

6

8

10

12

150 650 1150 1650 2150

TIME IN THE CYCLE [ms]

INT

EN

SIT

Y [

101

2 p

pp

]

0

5

10

15

20

25

30

p [

GeV

/c]

~1.31011 p/b - ~3% losses

0 500 1000 1500 20000

50

100

150

200

ns

kturns


PS limitationsPS limitations In 2006 observed intermittently

single bunch and coupled bunch (mainly horizontal) instability

– Nth ~ 0.6-0.7x1011 p/bunch– Growth times: few tens of turns– Creating “holes” of at least 12

bunches along the batch increases the threshold

– Dependence on bunch length (stronger instability for shorter bunches)

Found a bad voltage calibration for one of the 40 MHz RF cavities (spare) used for high energy splitting and bunch rotation resulting in shorter bunch length

Bunch length differenceC40-77 or C40-78 at 100 KV and C80 at 0 KV

10

10.5

11

11.5

12

12.5

13

13.5

14

14.5

15

0 10 20 30 40 50 60 70

Bunch number [#]

Bun

ch le

ngth

[ns]

Average (C40-77)

Average (C40-78)Stable beam

Unstable beam

Hardware problem was understood and solved at the end of the run but it showed how little is the margin for operation with nominal beam.


PS limitationsPS limitations Observations are consistent with e-cloud instability observed in

the past but only in a special mode of operation when the bunches where kept short (less than 10 ns) for several tens of ms

We need to build some margin (taking into account that we will have to open the machine vacuum in the future to continue the magnet renovation campaign – at least its first phase):– Review the RF gymnastics at high energy– Complete the commissioning of the transverse feedback– Add flexibility to the control of chromaticity by an independent

powering of the PS magnet pole-face windings (planned for 2007)


SPS limitationsSPS limitations

Low energy losses: reduction from >10 to 7-8

%– New RF voltage

programme (end of 2003)– New working point

compatible with larger momentum spread (end of 2004) and large vertical chromaticity required to fight the ECI. Results confirmed in 2006.

Need to understand better the blow-up and loss mechanisms at the beginning of the ramp (mainly on the tail of the 4th batch). Data collected in 2006 being analyzed.

-14

-12

-10

-8

-6

-4

-2

0

0 5000 10000 15000 20000

TIME IN THE CYCLE [ms]

RE

LA

TIV

E L

OS

S [

%]

16/10/2003

02/07/2004

15/11/2004

NEW WP

INJ. PLATEAU RAMP



Short lifetime even after optimization of the working point

“longitudinal” lifetime dominate (capture losses due to Re ZL and bunch shape) Impedance reduction and optimization of the bunch shape from PS

Difference in lifetime between the head and the tail of the batch

– recovers as the intensity decreases– Bunches are getting shorter particularly at

the tail of the batch …..while electron cloud signal disappears

E-cloud density variation during the bunch passage and synchrotron motion lead to periodic tune modulation and trapping-de-trapping on resonance islands halo and losses only cure so far: “scrubbing”



Fast vertical instability observed since 2002 and studied in 2003-2004 for single LHC bunches with low longitudinal emittance (L~0.2 eV.s < L LHC

=0.35 eV.s) for Nb>0.6×1011 p Driven by the machine Ztr. Expect instability threshold close to the ultimate intensity for L LHC

Since 2002 a constant increase on the 800 MHz RF voltage required to stabilize the beam has been also observed due to increased Im ZL

0.05y 0.8y



2006 results are preliminary

Significant contribution from the kickers (in particular extraction kickers).


Interdigital comb structure 20mm spacing


Impedance reduction campaign for the extraction kickers started with a prototype module installed in the machine

Impedance measurements performed in the laboratory effect on beam stability being estimated before continuing with the programme

Plan to build an impedance DB for the SPS (....and ultimately for the injectors)


Summary and ConclusionsSummary and Conclusions

From the point of view of the HW the LHC injectors are ready for the LHC and a consolidation programme is ongoing to allow them to serve reliably the LHC

The Commissioning and Early Physics Beams have been tested during machine studies and can be provided.

Need to turn their operation mode from occasional (MD/expert type) into routine during the 2007 run:– CCC is certainly an advantage– Need to improve beam diagnostics signals/data availability and

reliability (particularly lacking in 2006) in CCC and control tools


Summary and ConclusionsSummary and Conclusions

Nominal beam is feasible but we have to create more margin for a reliable operation in PS and particularly in the SPS where e-cloud together with impedance are the main limitations.

Operation above nominal is for the time being out of range and likely requires a more drastic upgrade programme of the injectors.

status & performance of the lhc (proton) injector complex

Documents

lhc beam brightness

downsthe lhc proton

ns lhc beam ultimate

nominal beam

ns75 ns lhc beam nominal

ultimate beam

ns lhc beambut

lhc proton injector