Transcript R&D towards a Laser Based Beam Size Monitor for the FLC

The Laser System at PETRA Wire
G. A. Blair, Royal Holloway Univ. London
ACFA Workshop, Mumbai
16th December 2003
Accelerator-Related Session
• Motivation for the project
• Laserwire at PETRA
- Environment at PETRA
- Installation of Hardware
- First measurements
• Conclusions and Outlook
16th December 2003
G. Blair, RHUL
1
Motivation
• Maximise Luminosity performance of Linear Collider
• Control of transverse beam size and emittance in the Beam
Delivery System (BDS) and at the Interaction Point (IP)
• Conventional techniques (wirescanner) at their operational limit
• Development of standard diagnostic tool for LC and LC Test
Facility operation based on optical scattering structures 
Laserwire, Laser-Interferometer
• Features
- Resolution error smaller than 10%
nb N e2 f rep
- Fast (intra-train) scanning
L
 HD
* *
- Non-destructive for electron beam
4 x y
- Resistant to high power electron beam
16th December 2003
G. Blair, RHUL
2
Trans.+Long. Profiles
16th December 2003
Trans: 10-100 m
Long: ~200
m
G. Blair, RHUL
Trans:
10-100 nm
3
LC Layout and Parameters
CLIC
NLC/GLC
TESLA
BDS
x/ m
y/ m
3.4 to 15
0.35 to 2.6
7 to 50
1 to 5
20 to 150
1 to 25
IP
x/ nm
y/ nm
196
4.5
335
4.5
535
5
16th December 2003
G. Blair, RHUL
4
Optical Scattering Structures
• Scanning of finely focused laser beam through electron beam
• Detection of Compton photons (or degraded electrons) as function
of relative laser beam position
• Challenges
- Produce scattering structure smaller than beam size
- Provide fast scanning mechanism
- Achieve efficient signal detection / background suppression
16th December 2003
G. Blair, RHUL
5
Laserwire for PETRA
• Positron Electron Tandem
Ring Accelerator
• Injector for HERA, upgrade to
synchrotron light source
 Long free straight section
 Easy installation of hardware
due to existing access pipe
and hut outside tunnel area
• Q-switch Nd:YAG with SHG
• From CERN LEP polarimeter
 Trans Mode: large M2 ~9
 Long Mode: stability ± 20%,
beating  ps substructure
 Homegrown timing unit for
external triggering
16th December 2003
PETRA parameter
Energy
Bunch Length
Charge/bunch
Hor. beam size
Ver. beam size
E/GeV
z/ps
nC
x/m
y/m
4.5 to 12
~100
1 to 3
500 to 100
~100
Laser parameter
Wavelength
Energy
Pulselength
Reprate
Beam size
Divergence
G. Blair, RHUL
l/nm
E/mJ
dt/ns
frep/Hz
x,y/mm
q/mrad
1064/532
250/90
10
30
~7
0.7
6
Laserwire for PETRA
16th December 2003
G. Blair, RHUL
7
Signal and Backgrounds
• Signal: Compton scattering
• Background sources:
- Synchrotron radiation
- Cosmic rays
- Bremsstrahlung
• Simulation with Geant4 plus
tool kits with realistic setup
16th December 2003
G. Blair, RHUL
8
Setup at PETRA
BPM
COMPTON
PHOTONS
e+
CAL
IP
PD@IP
TUNNELAREA
HALLEAST
EMBLHUT
SCOPE
PIT.E
IN
LASER
Q-SWITCH
IN
OUT
PIT.Y
PD
TRIGGER BOX
OUT
16th December 2003
PETRA
TIMING
OUT
G. Blair, RHUL
IN
9
Installation at PETRA
16th December 2003
G. Blair, RHUL
10
Installation at PETRA
16th December 2003
G. Blair, RHUL
11
Lab Measurements at RHUL
16th December 2003
G. Blair, RHUL
12
Installation at PETRA
16th December 2003
G. Blair, RHUL
13
Detector
• Requirements for detector material
- short decay time (avoid pile up)
- short radiation length
- small Moliere radius
• Cuboid detector crystals made of PbWO4
• 3x3 matrix of 18x18x150 mm crystals
• Energy resolution
better than 5%
16th December 2003
G. Blair, RHUL
14
Detector Calibration
• Detector studies with DESY II testbeam
• Beamline with electrons with energy
from 450 MeV to 6 GeV
• Ten detector crystals were calibrated
using a single PMT
• Combination of nine crystals in matrix
• Resolution
- High intrinsic resolution
- Full matrix less good
16th December 2003
G. Blair, RHUL
15
First Photons 31.07.03
Laser on
Laser off
Calorimeter
Q-switch
Photodiode at IP
16th December 2003
G. Blair, RHUL
16
First Beam Profile Scans
•
•
•
•
Positron beam in PETRA
Beam energy: 7 GeV
Bunch pattern: 14 x 1 bunch evenly filled
Average current: 12 mA
- Bunch charge = avg. current / (reprate * Nbunches) = 6.5 nC
• Laser energy measured: 40 mJ (specs 90 mJ), PL= 4 MW
• Optimization: qswitch delay, timing of ADC sample point
• Vertical and horizontal orbit bumps to steer positron beam
- Closed symmetric bumps using four steerers
- Bump length: 50 m, max offset: 10 mm
• Operation of fast piezo scanner
16th December 2003
G. Blair, RHUL
17
The Laser
•
•
•
•
•
•
The laser has been given to us by B. Dehning from CERN.
It has been used at LEP to measure beam polarization
It’s a Nd:YAG Q-switched system, running with 30 Hz
pulse energy measured: 40 mJ, power: 4 MW
synchronization to PETRA beam by triggering the Q-switch
Pockels-cell
transverse beam quality is modest (multimode)
measured spot size at IP: σL = (80 ± 10) μm
16th December 2003
G. Blair, RHUL
18
Measurement of the
longitudinal Profile
•
•
•
•
The longitudinal profile has been measured with a streak
camera: FESCA 200 from Hamamatsu
largest window of the camera: 500 ps with a resolution of
5 ps (fwhh)
The camera was triggered with the laser via a fast photo
diode
Problem:
stability of the trigger probably not better than 0.5 ns
16th December 2003
G. Blair, RHUL
19
Averaged Profile
•
Measured averaged profile:
fits to gaussian with a width of 12.5 ns (as expected)
16th December 2003
G. Blair, RHUL
20
Structure in the Longitudinal
Profile
•
Example of a single shot measurement of the profile
500 ps window, resolution 5 ps
66 ps
60 ps
16th December 2003
G. Blair, RHUL
21
Unfortunately, the structure is
not stable
•
•
The longitudinal structure is due to longitudinal mode
beating – this was expected
The beating changes from shot to shot
79 ps
30 ps
16th December 2003
G. Blair, RHUL
22
Laser
Transverse
Profile
Units – number of CCD pixels
16th December 2003
G. Blair, RHUL
23
Laser Summary
•
•
•
•
•
As expected for a this type of laser, the longitudinal
profile shows substructure due to mode beating
The spikes have a width of 30 to 60 ps and a distance of
60 to 80 ps
Unfortunately, the structure is not stable and changes
from shot to shot
To overcome this, the laser has to be equipped with a
frequency stabilized seed laser or eventually with an
Etalon
Hot spots a problem
16th December 2003
G. Blair, RHUL
24
Orbit Scan
• First scan with signal on scope
• Then sampling of peak using ADC
• Moving beam orbit up and down
with vertical orbit bump
• 5k counts at each orbit position
• 3 min for each spectrum
• 40 min for complete scan
• Background with 20k counts
- Mainly synchrotron radiation
and bremsstrahlung
- Rate changed by factor 10
• Signal rate expected at peak
- 200 γs x 380 MeV avg Energy
16th December 2003
G. Blair, RHUL
25
Result Orbit Scan
• Gaussian approximation of beam shape
σm = (0.175 ± 0.020stat ± 0.038sys) mm
• Vertical beam size
σe = sqrt(σm - σL )
laser σL = (40 ± 10) μm
σe = (170 ± 23 ± 37) μm
• Result of fit sensitive to
background modelling
• Systematic error dominated
by vertical orbit jitter
• More measurements and
understaning of bkg sources
necessary
16th December 2003
G. Blair, RHUL
26
Fast Scanner Operation
• Next scan with remote
controlled fast scanner
• Orbit position stable
• Scan range: ± 2.5 mrad
- Scan line = range * flens=
0.625 mm (± 20%)
• Change amplitude of scanner
power supply (1-100V)
• Take 5k counts
• Record laser IP image with
CCD
• Move laser beam
• Take 5k counts ...
16th December 2003
TO
DUMP
CCD
VIEWPORT
MIRROR
G. Blair, RHUL
SPLITTER
BEAM
LENS
125 mm
IMAGING
LENS
SCANNER
27
Data and Analysis
•
•
•
•
•
•
•
•
•
Seven scan points recorded
5 min / point
40 min for full scan
Positron beam position stable
within ± 40 μm
Moving low energy pedestal
No background model
Orbit stable  bkg const.
Simple pedestal cut instead
Sufficient background
rejection
16th December 2003
G. Blair, RHUL
28
New Setting 5.12.03
•
•
•
•
•
Positron beam in PETRA
Beam energy: 7 GeV
Posittron beam optics not as in October scans!
Bunch pattern: 14 x 1 bunch evenly filled
Low current: 7.1 mA, first bunch 0.458 mA
- Bunch charge = avg. current / (reprate * Nbunches) = 3.9 nC
• High current: 40.5 mA, first bunch 2.686 mA
- Bunch charge = 22.3 nC
• Vertical and horizontal orbit bumps to steer positron beam
into laser beam
- Closed symmetric bumps using four steerers
• Scanning of laser beam using the fast piezo scanner
16th December 2003
G. Blair, RHUL
29
Results 04.12.03 Data
• Slopy Gaussian approximation of beam shape
σm =(68 ± 3 ± 20) μm at low current
σm =(80 ± 6 ± 20) μm at high current
16th December 2003
G. Blair, RHUL
30
Conclusions and Outlook
• Laserwire at PETRA produced first compton photons and
measure vertical beam size Next steps:
• Full characterisation of laser: beam size, divergence, and
power (stability) with slot scans and imaging techniques
• Update all readout software, merge BPM and PMT software
• Do more systematic scans with the fast scanner
• Go to smaller spot sizes and reduce error bars
• Build second dimension scanner.
• Start designing a complete laser-wire emittance
measurement system for the LC BDS.
16th December 2003
G. Blair, RHUL
31
Collaborators
•
•
•
•
DESY
BESSY (Thanks to T. Kamps for many of these slides)
UK: RHUL, UCL, RAL, (Oxford).
CERN: (Laser, plus collaboration)
Close contact with:
• SLAC
• KEK
16th December 2003
G. Blair, RHUL
32
People
K Balewski, G Blair, S Boogert, G Boorman, J
Bosser, J Carter, J Frisch, Y Honda,
S
Hutchins, T Kamps, T Lefevre, H C Lewin, F
Poirier, I N Ross, M Ross, H Sakai,
N
Sasao, P Schmüser, S Schreiber,
J
Urukawa, M Wendt, K Wittenburg,
Thanks to PETRA and BKR shift crews !
16th December 2003
G. Blair, RHUL
33