SPS scrubbing run in 2014 - PowerPoint Presentation

Slide1

SPS scrubbing run in 2014

H. Bartosik, G. Iadarola, G. Rumolo

LHC Performance Workshop (Chamonix 2014), 22/9/2014

Many thanks to:

G.

Arduini

, T

.

Argyropoulos

, T.

Bohl

, K.

Cornelis

, H

.

Damerau

,

J

. Esteban

Müller, B. Goddard, S

. Hancock, W.

Höfle

,

L.

Kopylov

,

H

.

Neupert

,

Y

.

Papaphilippou

, G.

Papotti

, E. 

Shaposhnikova

, M.

Taborelli

and the SPS operator crewSlide2

Outline

Introduction

The 2014 SPS scrubbing run

Possible cycles

Scrubbing stages

Possible

supercycle composition

The doublet scrubbing beam

Motivation

Production scheme

Experience in 2012-13 MDs

Scrubbing preparation

Beams from the PS

SPS setup

MeasurementsSlide3

Outline

Introduction

The 2014 SPS scrubbing run

Possible cycles

Scrubbing stages

Possible

supercycle composition

The doublet scrubbing beam

Motivation

Production scheme

Experience in 2012-13 MDs

Scrubbing preparation

Beams from the PS

SPS setup

MeasurementsSlide4

Electron cloud in the SPS and scrubbing

Strong limitation due to e-cloud in the past

Instabilities at injection + incoherent effects

Emittance

blow-up along the batch

High chromaticity needed for beam stability

Pressure rise around the machine

Situation improved gradually due to scrubbing

Requires days of dedicated running in high electron cloud

conditions

Secondary

Electron Yield reduction by the e-cloud

itself

Scrubbing runs since 2002

Performed at 26

GeV

in cycling mode (~40 s cycle length)

Typically limited by heating and/or outgassing

~1-2 weeks periods

SPS scrubbing history

43.2 s

2002

(14d)

2003

(8d)

2004

(10d)

2006

(5d)

2007

(7d)

2008

(2.5d)

2012

(5d)

2009

(1.5d)

Shutdown

400%

2000

(48 b. - 0.8x10

11

p/b

@inj.)

3.5

μmSlide5

SPS scrubbing run 2014

Goals for 2014:

Q

ualify the loss of conditioning

due to LS1

Recover 2012 performance

with 25 ns beamsQ

uantify

amount of beam/time

needed

Test “doublet” scrubbing beam

to be used in the LHC in 2015

Qualification criterion



beam quality measurements

Ideally,

achieve by the end of the allocated scrubbing time:

25 ns, 4 batches, up to 1.3e11ppb,

emittances

below nominal, no blowup along the

train

as in 2012

basis for LIU strategy on e-cloud mitigation – coating vs scrubbingSlide6

From 2014 injector schedule (current version)Slide7

From 2014 injector schedule (current version)

S

crubbing originally foreseen in two consecutive weeks (W39-40), before the start-up of the NA physics. Then split between Weeks 39 and 45

 F

inally

spread over Week 45 plus an additional two-day mini-block in Week 50

Several reasons for splitting the scrubbing run into two blocks

(requested by LIU-SPS)

:

Gives

time to

analyze

the first

block’s

results

and

adapt accordingly

Untangling scrubbing from the

machine commissioning, NA setup

and

vacuum

conditioning of all the newly-installed or vented equipment

Allows

setting up scrubbing beams before

the 2nd scrubbing

block

the “doublet” beam

– its potential

to scrub the SPS can be explored already in 2014 (also in view of LHC scrubbing in 2015)Slide8

Pre-scrubbing cycles

Over weeks 41 – 44

some experience will be already gained during the available parallel and dedicated MD time

Both single bunch and 25 ns beam (1 batch) on short

f

lat

bottom cycle (6BP) should

be

already set up

Work on

recovery performance for nominal 25 ns beam

might have already started

(

with the consequent scrubbing

)

6 BPs (7.2 s)

Inj.

DumpSlide9

Scrubbing cycles

We will need to accumulate dose and monitor the evolution of beam parameters for both coherent and incoherent effects



4 or more batches circulating in the

machine at 26 GeV



Acceleration to 450 GeV should

be

fully set up and used for scrubbing qualification

18 BPs (21.6 s)

33 BPs (39.6 s)

Inj.

Inj.

Inj.

Inj.

Dump

Inj.

Inj.

Inj.

Inj.

Inj.

Dump

Inj.

Inj.

Inj.

Inj.

19

BPs

(22.8

s)Slide10

Planning (to be steered on the fly)

Monday

Tuesday

Wednesday

Thursday

Friday

Saturday

Sunday

Intensity ramp up at 26

GeV

on intermediate flat bottom cycle (21.6 s)

First

scrubbing

block

Intensity

ramp up

at 26

GeV

with 25 ns beams (ideally up to

5

injections –

try to push bunch intensity up

to 1.5x10

11

p/b?)

Inj.

Inj.

Inj.

Inj.

Dump

Possible

supercycle

(to be coordinated with

physics in the PS complex)

Inj.

Week 45Slide11

Planning (to be steered on the fly)

Monday

Tuesday

Wednesday

Thursday

Friday

Saturday

Sunday

Intensity ramp up at 26

GeV

on intermediate flat

bottom cycle (21.6

s)

First

scrubbing

block

Intensity

ramp up

at 26

GeV

with 25 ns beams (ideally up to 5 injections – try to push bunch intensity up to 1.5x10

11

p/b?)

Study

residual electron cloud effects on beam lifetime and quality

for nominal beam (e.g.

emittance

growth, bunch shortening over long flat bottom

) while gradually lowering vertical chromaticity setting for

stability

Inj.

Dump

Inj.

Inj.

Inj.

Inj.

Possible

supercycle

Studies on long flat

bottom cycle (39.6 s)

Week 45Slide12

Planning (to be steered on the fly)

Monday

Tuesday

Wednesday

Thursday

Friday

Saturday

Sunday

Intensity ramp up at 26

GeV

on intermediate flat

bottom cycle (21.6

s)

First

scrubbing

block

Intensity

ramp up

at 26

GeV

with 25 ns beams (ideally up to 5 injections – try to push bunch intensity up to 1.5x10

11

p/b?)

Study

residual electron cloud effects on beam lifetime and quality

for nominal beam (e.g.

emittance

growth, bunch shortening over long flat bottom) while gradually lowering vertical chromaticity setting for

stability

Acceleration

and

scrubbing qualification

Studies on long flat

bottom cycle (39.6 s)

Scrubbing qualification

(25 ns

std

and BCMS, up to 450

GeV

)

Possible

supercycle

for scrubbing + qualification

Week 45Slide13

Planning (to be steered on the fly)

MondayTuesday

Second

scrubbing

block

At this stage

doublet beam could be ready including acceleration

The goal is to

accumulate the largest possible electron dose on the beam chambers

The results of the tests with the doublet beam will be important for the LHC scrubbing in 2015

Wednesday

Week 50

Possible

supercycle

Scrubbing with doublet

(possibly with acceleration)

Dedicated MD (scrubbing)

18 BPs (21.6 s)

Inj.

Inj.

Inj.

Inj.

DumpSlide14

Outline

Introduction

The 2014 SPS scrubbing run

Possible cycles

Scrubbing stages

Possible

supercycle composition

The doublet scrubbing beam

Motivation

Production scheme

Experience in 2012-13 MDs

Scrubbing preparation

Beams from the PS

SPS setup

MeasurementsSlide15

“Doublet” scrubbing beam: introduction

Scrubbing with 25 ns

beam allowed to lower the SEY of the dipole chambers well below the multipacting threshold for 50 ns

 e-cloud free operation with 50 ns beams

 Can we go to lower bunch spacing to

scrub for 25 ns operation?

Due to RF limitations in the PS it is impossible to inject bunch-to-bucket into the SPS with spacing shorter than 25 ns

An alternative is to inject

long bunches into

the SPS

and capturing each bunch in two neighboring buckets

obtaining a

(5+20

)

ns “hybrid” spacing

Non adiabatic splitting at SPS injectionSlide16

“Doublet” scrubbing

beam: introduction

Non adiabatic splitting at SPS injection

20 ns

5 ns

Scrubbing with 25 ns

beam allowed to lower the SEY of the dipole chambers

well below the

multipacting

threshold for 50 ns



e-cloud free operation with 50 ns beams

 Can we go to lower bunch spacing

to

scrub for 25 ns operation?

Due to RF limitations in the PS it is impossible to inject bunch-to-bucket into the SPS with spacing shorter than 25 ns

An alternative is to inject

long bunches into

the SPS

and capturing each bunch in two neighboring buckets

obtaining a

(5+20

)

ns “hybrid” spacingSlide17

“Doublet” scrubbing beam: introduction

Close to the threshold

all the electrons produced after a bunch passage are absorbed before the next one



 small accumulation

over subsequent bunch passages

PyECLOUD simulation

Std

25 ns beam

e-cloud enhancement

mechanism:Slide18

More e

- production and shorter e- decay

 accumulation possible

PyECLOUD simulation

Std

25 ns beam

Doublet beam

e-cloud enhancement

mechanism:

“Doublet” scrubbing

beam: introductionSlide19

“Doublet” scrubbing

beam:

SPS simulation resultsMBA dipole magnet

MBB dipole magnet

Significantly lower

multipacting

threshold

for large enough intensity per doubletSlide20

“Doublet” scrubbing

beam:

SPS simulation resultsSignificantly lower

multipacting

threshold for large enough intensity per doublet

Beam orbit modulation

needs to be applied to condition a wide enough area of the chamber

MBB - 26GeV

Intensity per bunch of the doublet (

b.l

. 4 ns

)

(

b.l

. 3 ns)Slide21

First machine tests

have been conducted at the SPS at the end of 2012-13 run in order to validate the production scheme and obtain first indications about the e-cloud enhancementThe production scheme has been successfully tested for a train of (2x)72 bunches with

1.7e11 p per doublet

4

2

6

6

0

1

2

3

Time [

ms

]

200 MHz RF Voltage [MV]

4

-1

0

1

st

inj.

Measurements by to

T.

Argyropoulos

and

J. Esteban Muller

“Doublet” scrubbing

beam:

first test at the SPSSlide22

“Doublet” scrubbing

beam:

first test at the SPS

4

2

3604

3600

3602

3598

3596

3594

6

6

3592

3590

0

1

2

3

Time [

ms

]

200 MHz RF Voltage [MV]

4

-1

0

1

st

inj.

2

nd

inj.

Profile of the first doublet

First machine tests

have been conducted at the SPS at the end of 2012-13 run in order to

validate the production scheme

and obtain first indications about the e-cloud enhancement

The production scheme has been

successfully tested

for a train of (2x)72 bunches with

1.7e11 p per doublet

The

possibility of injecting a second batch

without degrading the circulating

beam

has also been shownSlide23

“Doublet” scrubbing

beam:

first test at the SPS

Measurement

Simulation

First results looked very encouraging

,

e-cloud enhancement confirmed by:

Measurements

with the

electron cloud detectors

(agreement with measured cloud profile gives an

important validation for our simulation model

and code)Slide24

25ns std. (1.6e11p/bunch

)

(

1.6e11p/doublet)

25ns “doublet”

“Doublet” scrubbing

beam:

first test at the SPS

First results looked very encouraging

,

e-cloud enhancement confirmed by:

Measurements

with the

electron cloud detectors

(agreement with measured cloud profile gives an

important validation for our simulation model

and code)

Dynamic pressure measurements

observed in the SPS arcsSlide25

First results looked very encouraging

,

e-cloud enhancement confirmed by:

Measurements

with the

electron cloud detectors

(agreement with measured cloud profile gives an

important validation for our simulation model

and code)

Dynamic pressure measurements

observed in the SPS arcs

“Doublet” scrubbing

beam:

first test at the SPS

Provided that we can produce and preserve a good quality (multiple batches, large bunch intensity),

this beam will be used during the two-day mini-scrubbing run at the end of the 2014 run



Acquired

experience

will be very important for the definition of

the LIU-SPS strategy

with respect to e-cloud and scrubbing and

for

the LHC scrubbing in 2015

To be noted

Need to commission the new transverse damper for doublets at injection

SPS BQM software was updated for doublet beamsSlide26

Outline

Introduction

The 2014 SPS scrubbing run

Possible cycles

Scrubbing stages

Possible

supercycle composition

The doublet scrubbing beam

Motivation

Production scheme

Experience in 2012-13 MDs

Scrubbing preparation

Beams from the PS

SPS setup

MeasurementsSlide27

Beam requirements

From the PS:

Before first scrubbing week (W45):

25 ns beam

(std. production scheme and BCMS, 72 b., up to 1.5e11 ppb)

50 ns beam (std. production scheme, 36 b., up to 1.7e11 ppb

) – as backupIn addition, before second scrubbing block

(

W50):

25 ns

beam

for doublet production

(>1.5e11 ppb, long bunches at extraction)

From earlier SPS setup and MDs:

Before

first scrubbing week (

W45):

Basic setup

(injection, orbit, working point, RF, damper) of

26

GeV

flat bottom cycle

(with 25 or 50 ns beams, Q20)

Setup

of

25 ns

LHC filling

cycle

Before second block (W50):

Setup of “doublet” beam at 26GeV

(capture of multiple batches, orbit, working point, RF, damper) and possibly accelerationSlide28

Measurements

Tests which compromise the scrubbing efficiency should be kept to the minimum possible

Collect as much data as possible to learn about

ecloud

effects and scrubbing in the SPS

BCT/FBCT (to estimate beam dose)

Longitudinal parameters (BQM,

mountain range, faraday cage scope)

Beam transverse oscillations

BBQ, LHC BPMs,

Headtail

monitor, fast pickup from HBWD

feedback setup, new digitizers on BPW exponential pickups

Beam transverse size

Wirescanners

(bunch by bunch), BGI

(?)

Pressure

along the ring (1 Hz

rate

)

Special attention to the a-C coated magnets

Dedicated e-cloud equipment

Electron cloud monitors (MBA

StSt

, MBB

StSt

, MBB a-C, MBB copper)Shielded pickup

In situ SEY measurement (if available)

Removable

StSt sample (for lab SEY measurement)

COLDEX Slide29

Summary and conclusions

In the past, SPS was strongly limited by e-cloud

Scrubbing proved to be an effective

mitigation for 25 ns beams up

to nominal intensity

25 ns

beams delivered to LHC in

2012 were well

within design report

specs

Scrubbing

run 2014

To recondition SPS after LS1, since

large

parts

of

the SPS

were vented

1

st

block: 7 days (week 45)

Qualify the loss of conditioning

due to LS1

Recover 2012 performance

with 25 ns beams

2

nd

block: 2(3) days (week 50)Test scrubbing with doublet (also in view of LHC in 2015)

Experience gained will be

needed for LIU decision about SPS coating Need to prepare beams from the PS and test instrumentation in SPSSlide30

Thanks for your attention!Slide31

No additional impedance heating

is expected with the doublet beam (same total intensity)Beam power spectrum is modulated with cos

2 functionLines in the spectrum can only be weakened by the modulation

“Doublet” beam: beam induced heating

Thanks to

C.

Zannini