1

2

Alternative Options in the Injectors – Preliminary Summary

H. Damerau

LIU-TM#8

18 October 2013

Many thanks for discussions and input to T. Argyropoulos, M. Benedikt, H. Bartosik, T. Bohl, C. Carli, R. Garoby, S. Gilardoni, B. Goddard, S.

Hancock, W. Herr, B. Mikulec, Y. Papaphilippou, G. Rumolo. E. Shaposhnikova, H. Timko, R. Tomas and many others

3(Alternative) upgrade option overview

Linac4

PSB

PS

SPS

Basic choices Additional benefit/margin

• Faster recombination kickers PSB-PS (with 1.4 GeV)

• Double-batch or h=5 single batch injection

• 3-split, BCMS, BCS or PBC (pure batch comp.)

• Mini-microbatch, 8b+4e together with 3-split or BCMS

• Resonance compensation• Special injection optics• Long. flat or hollow

bunches

• 2.0 GeV at PSBPS transfer

• SPS RF upgrade: 43+24• More RF power plants:

42+43 or 102• Relaxed el with 200 MHz in

LHC• LHC ACN cavities in SPS?

• Split tunes optics• Special injection optics

• 28 GeV at PSSPS transfer

+5 %

+? %

+15 %

+25 %

+? %

+? %

+? %

• Vertical painting Linac4

+25 %• Long. flat or hollow

bunches

4PS-SPS space charge limit, standard

3-sp

lit, 1

.4 G

eV BC

MS

BC

S

3-sp

lit, 2

.0 G

eV

PBC

3-sp

lit, 1

.4 G

eV3-

split

, 2.0

GeV

BCS

BCMS

PBC

BCS Batch compression + split

h = 9 10 20 21

BCMS

Batch comp. + merge + split

h = 9 10 11 12 13 14 7 21

PBC Pure batch compression h = 9 10 … 20 21 1.4 GeV,PS: -DQy = 0.31

2.0 GeV,PS: -DQy = 0.31

• With PS at 1.4 GeV, pure batch compression reaches brightness as 2 GeV transfer

® But ~15% less bunches in LHC and ~twice longer filling, but squeezed to limit

(at

SP

S e

xtr

acti

on

, n

um

ber

of

bu

n-c

hes

per

LH

C r

ing

n

ot

con

sid

ere

d)

5

PS-SPS space charge, alternatives

3-sp

lit, 1

.4 G

eVh5

SB

3-sp

lit, 2

.0 G

eV

8b+4e

+BCMS

3-sp

lit, 1

.4 G

eV3-

split

, 2.0

GeV

h5SB Single-batch h = 5 injection

h = 5 10 20 21

8b+4e Double split + empty bucket

h = 7 21 (2b+1e … 8b+4e)

8b+4e+BCMS

Batch comp. + batch comp.

h = 9 … 14 21 1.4 GeV,PS: -DQy = 0.31

• h5SP attractive only together with Linac4 and PSB-PS transfer at 1.4 GeV

• 8b+4e schemes approach or push brightness beyond SPS space charge limit

8b+

4e8b+4e

8b+4e+BCMS

2.0 GeV,PS: -DQy = 0.31

(at

SP

S e

xtr

acti

on

, n

um

ber

of

bu

n-c

hes

per

LH

C r

ing

n

ot

con

sid

ere

d)

6

Larger bunch intensity from SPS?

t = co

nst, LD+PWD

24+25

43+2443+42

102

• Baseline upgrade: shorter cavities and 21.6 MW RF power• Nb ≈ 2 · 1011 ppb without degradation, 2.5 · 1011 ppb for 10%

longer bunches

® Even shorter cavities and more RF power?

E. Shaposhnikova et al., LIU-TM#2

LD: Loss of Landau damp.PWD: Potential well dist.

Slope?+0.51 · 1011

+0.29 · 1011

+0.18 · 1011

• Additional step of adding 21.6 MW: intensity gain ≈ half of first step• No gain at low intensity MWs of RF only heating the cavities

® Significant uncertainty in LD+PWD line, sensitive as LD proportional to l

5/2

® Additional emittance constraints on bunch length during acceleration

Pold = 1.05 MWPnew = 1.6 MW

RF voltage at transfer to LHC RF voltage at transfer to LHC

Analysis still in progress

7

8b+4e scheme or 200 MHz in LHC?

t = const, LD+PWD

VRF for 3 MV at 200 MHz in LHC

8b+4e scheme (SPS ejection):• Reduced line density due to less

bunches• Single bunch effects LD+PWD

unchanged

LD: Loss of Landau damp.PWD: Potential well dist.

200 MHz in LHC ~ 2 · l ≈ 1.1 eVs• No issue with LD or PWD• Beam loading and matching with

LHC24+25

43+2443+42

102

24+25

43+2443+42

102

Analysis still in progress

• Significant gain from larger longitudinal emittance at transfer to LHC (assumes blow-up in SPS, to avoid bottleneck around start of acceleration)

• Revisit 200 MHz capture system in the LHC?• Install additional 200 MHz standing wave cavities (those for LHC?)

in SPS?

t = co

nst, L

D+PWD

8

Preliminary summary and remarks

• No magic alternative to Linac4 + 2.0 GeV + SPS RF upgrade

• Large number of schemes to increase intensity and brightness from injectors® Linac4+PSB+PS may push SPS to space charge limit

• Longitudinally larger bunches in SPS would help a lot • Limited reach of brute-force approach for even more RF

power

• Interesting alternatives can be studied in injectors after LS1® PSB: Hollow bunches® PS: Flat or hollow bunches, special flat-bottom optics, pure

batch compression, 8b+4e schemes, higher PS-SPS transfer energy

® SPS: split tunes optics, higher intensity with slightly longer bunches

• Combinations of alternatives keep flexibility of injector complex to react to requests from LHC

• Numerous alternatives, e.g., H- injection into the PS, 400 MHz or slip stacking in SPS do not appear as studied in literature

99

THANK YOU FOR YOUR ATTENTION!

10

Introduction• Evaluation of alternative schemes based on

® ‘Common references’ document (https://edms.cern.ch/document/1301268/2)

® Assumptions presented in LIU-TM#4 (http://indico.cern.ch/getFile.py/access?contribId=3&resId=1&materialId=slides&confId=266540)

• Parameters for US1/2 with standard schemes now in good agreement with G. Rumolo’s tables (https://edms.cern.ch/document/1296306/1)

• SPS space charge limit to be revised

• Analysis for most of the schemes very superficial and simplified

• Optimization strategy:1. Fix maximum intensity per bunch at SPS ejection2. Calculate minimum transverse emittance backwards

through the chain® Results: Transverse emittance at extraction from SPS

(25 ns by default) and number of bunches per batch

11

Evaluation of injector scenarios

Constraints of the injectors taken into account for evaluation:

• Linac + PSB® Transverse brightness® Maximum longitudinal emittance (PSB-PS transfer)

• PS® Space charge at flat-bottom® Various longitudinal emittance limits (RF manipulations,

transition crossing, final emittance)® Coupled-bunch instability limit (with new feedback)

• SPS® Space charge at flat-bottom® Maximum intensity per bunch, maximum line density