Beam-Based Alignment Results

1
Beam-Based Alignment Results

• Henrik Loos, for the LCLS Commissioning Team

2
Undulator Trajectory Requirements

• Beam through undulator rms 2 µm per gain length
• Undulator with 33 segments total 100 m
• Not possible with conventional alignment
• Use beam based alignment using set of different
  energies
• RF BPM resolution lt 1 µm

3
Undulator Configuration

• Undulator, Quad, BPM, BFW move with girder
• Beam Finder Wire (BFW) retractable
• Horizontal translation of undulator
• Complete retract (80 mm)
• Undulator K adjustment ( 5 mm)

4
Beam Based Alignment Principle

• BPM offsets unknown
• Magnetic fields (earth, quad kicks, etc.) unknown
• Correct field integrals with quad offsets or
  correctors for dispersion free trajectory at BPM
  position
• Trajectory between BPMs remains unknown
• Measure trajectory at different energies to
  extrapolate to straight line at infinite energy
• Fixed undulator quad fields
• BPM position is BPM offset at infinite energy

5
BBA Measurement Schematic
?b1
?b3
?b2
?b0
?b0
?q1
?q3
?q2
E2
x,x
E1
?y1
?y3
?y2
?y0
?y4
BPM Offsets ?bi Quad Offsets ?qi
E1 lt E2
6
BBA Procedure

• Model beam position (yj) at BPMs as function of
  initial launch at 1st BPM (xi), quad offsets
  (?qi), BPM offsets (?bi)
• y Rx Rq Rb x ?q ?b
• Rxj Rj1,12
• Rqj R1,jend - R1jbeg Riltj,jend - Riltj,j 0
  011
• Rb -I
• Fit solution for y arbitrary to adding linear
  function to quad and BPM offsets
• Add constraint equations for quad or BPM offsets
• 0 Si ?qi and Si zi ?qi for linear quad offset
  constraint
• 0 ?qi for minimum quad offset constraint

7
BBA Implementation

• Setup accelerator for one energy
• Calculate response matrix for this energy
• Measure N orbits at this energy and average
• Repeat for all energies
• Generate final matrix with separate launch
  parameters for each energy and selected
  constraints
• Fit quad and BPM offsets and implement
• Repeat BBA procedure

8
BBA Simulation
Simulation Orbits
Simulation Fit Lin. Quad
9
BBA Results 1st Run

• First test Energy range only 10 13.64 GeV
• 50 orbits each, averaged
• BPMs not well calibrated
• Large oscillation in fit of quad offset, 1mm
  error bar
• Assume BPM offset worse than quad offset
• Apply instead constraint for minimal quad offset
• Initial position rms 300 µm

Measured Orbit
10
BBA Results 1st Run
Fit with Linear Quad Constraint
Fit with Min. Quad Constraint
Applied this to BPM offsets
11
BBA Results 2nd Run

• Energy range now 7 13.64 GeV
• Still large 1 mm oscillation on quad offset fit
• Apply relaxed minimum quad constraint, 100 µm
  error bar
• Orbits very similar after correction
• Position rms 50 µm after

Measured Orbit
12
BBA Results 2nd Run
Fit with Min Quad Scale 20
Measured Orbit after Correction
13
BBA Results 3rd Run
Measured Orbit 4.3 13.64 GeV
Fit with Linear Quad Constraint
14
BBA Results 3rd Run
Measured Orbit 4th Iteration
Fit with Linear Quad Constraint
Position rms 2 10 µm
Offset Error Bar 10 µm
15
BBA Results Best Orbit

• Carefully calibrated BPMs
• Energy range 4.3 13.64 GeV
• 4 different energies
• Undulator launch feedback on
• Average position rms 1 2 µm
• Betatron jitter 20 µm

16
BBA Results Girder Bump Test
58um bump _at_ 13.7 GeV -58um BPM offset
-48um bump _at_ 13.7 GeV
BBA procedure finds both quad offsets and BPM
offsets
17
Typical BBA After Several Months
Observe mostly changes in BPM offsets 10 30um
Some quad BPM offsets in end region of
undulator from incremental orbit corrections
(retracting undulators, changing of taper)
18
Quad Alignment Measurement

• Measure quadrupole offset from beam axis
• Vary quad magnetic field and fit offset to
  trajectory kick
• Verifies earth field compensation from BBA

Z (m)
P. Emma
19
BBA User Interface
Fit Options
Simulation
Measurement
Corrections
20
Fast Linac Energy Change

• User interface to run an automated script
• Block/unblock beam
• Activate saved klystron configuration
• Trim saved magnet configuration
• Toggles feedbacks
• Enables one BBA run in 10 - 15min (at best), 2
  4 h (worst)

21
BBA Undulator Taper

• Orbit effects from undulator motion
• No earth field shielding with retracted undulator
• Undulator translation (80 mm) shifts entire
  girder by 10 - 100 um (quad BPM)
• Undulator field integral depends on taper
• Goal
• Straight trajectory for all undulator
  translations
• Strategy
• Do BBA at design taper, correct quad position
• Compensate field integral change for different
  taper with corrector coil
• Compensate girder shift for retracted undulator
  with corrector coils and BPM offset

22
Undulator Field Integral Measurement
Apply 1st field integral to corrector coil
23
Summary

• Achieved
• BBA procedure successfully implemented
• Converges to 1 µm trajectory rms
• Important to have full energy range
• Errors on fitted quad offsets decreased from 1 mm
  to 10 µm with increasing energy range
• Fast energy switching 15 min BBA possible
• Complemented by measurement of quad offsets by
  varying quad strength
• To Do
• Fully automate energy change (Interface to energy
  management, orbit feedback in linac)
• Study BBA at low charge (lt 250 pC)
• Implement orbit correction from undulator
  translation
• Compare girder position from BBA with alignment
  diagnostic system (ADS)
• Monitor and study BPM offset drifts