installation progress of the laser-based synchronization ......mar 04, 2010 · free-space optics...

Installation Progress of the Laser-based SynchronizationSystem at FLASH.

Overview, Experiences, Performance and Outlook

Sebastian Schulz1,2 on behalf of the FLASH LbSyn Team

1Institute of Experimental PhysicsHamburg University, Germany

2Deutsches Elektronen-Synchrotron (DESY), Hamburg, Germany

48th ICFA Advanced Beam Dynamics Workshop on Future Light SourcesMarch 1 – 5, 2010

S. Schulz (Hamburg U, DESY) FLASH Optical Synchronization FLS Workshop 2010 1 / 25

Outline.

1 Introduction and Overview

2 Progress and Results of Selected SubsystemsMaster Laser OscillatorFiber LinksLaser-to-Laser SynchronizationInfrastructure and Electronics

3 Latest DevelopmentPhotoinjector Laser Synchronization

4 Summary and Outlook


Introduction and Overview

Optical Synchronization Systems.Overview

• projected point-to-point stability: 10 fs• enable the implementation of a longitudinal feedback system



The Laser-based Synchronization System at FLASH.Layout, Implementation and Upgrades

RF gun

L2RF

ACC1 ACC39L2RF

L2RF

BAM EBPM

ACC2 ACC3L2RF

EO/THzlaser

EO photon beam

target

ACC4 ACC6 ACC7ACC5L2RF

injectorlaser 1

HHG

seedlaser

BAM

BAMEO/THz

BAMseedingundulators SASE undulators

bypass

OXC

OXCOXC

OXC

16-port FSD unitEDFAs

master laseroscillator EDFLFL

RL

L2RF

FL

master laseroscillator SESAM

dump

BAMEBPMBCM BCM

Yb berlaser

FL

FL

FL

RL

FL

pump-probelaser

OXCFLadd. diag FL

FL

FL

FL

FL

• last user run (till September 2009)

→ MLO with distribution and 4 fiber links, 3 BAMs, 1 EBPM (3 front-ends),→ standard BCMs, EO with Ti:sapphire laser




RF gun(exchanged)

L2RF

ACC1 ACC39L2RF

L2RF

BAM EBPM

ACC2 ACC3L2RF

EO/THzlaser

EO photon beam

target


injectorlaser 2

HHG

seedlaser

BAM

BAMEO/THz


bypass

OXC

OXCOXC

OXC



RL

L2RF

FL


dump

BAMEBPM

L2RF

BCM

Yb FLOrange

EO

EOBCM

Yb berlaser

FL

FL

FL

FL

FL

FLRL

FL

pump-probelaser

OXC L2RF

FLadd. diag FLFL

• last user run (till September 2009)→ MLO with distribution and 4 fiber links, 3 BAMs, 1 EBPM (3 front-ends),

→ standard BCMs, EO with Ti:sapphire laser

• after the FLASH upgrade (just finished, now in commissioning phase). . .




RF gun(exchanged)

L2RF

ACC1 ACC39L2RF

L2RF

BAM EBPM

ACC2 ACC3L2RF

EO/THzlaser

EO photon beam

target


injectorlaser 2

HHG

seedlaser

BAM

BAMEO/THz


bypass

OXC

OXCOXC

OXC



RL

L2RF

FL


dump

BAMEBPM

L2RF

BCM

Yb FLOrange

EO

EOBCM

Yb berlaser

FL

FL

FL

FL

FL

FLRL

FL

pump-probelaser

OXC L2RF

FLadd. diag FLFL

• last user run (till September 2009)→ MLO with distribution and 4 fiber links, 3 BAMs, 1 EBPM (3 front-ends),

→ standard BCMs, EO with Ti:sapphire laser

• upgraded MLO and distribution, 3 more fiber links, 1 new BAM, 1 newEBPM, new BCM setups, new EO station, new OXC-IL2, improvedinfrastructure, and more


Progress and Results of Selected Subsystems Master Laser Oscillator

Master Laser Oscillator I.

Overview and Current Status

Specifications/Requirements

• topology: EDFL in σ-configuration

• repetition rate: 216.66MHz

• average power: ≈ 100mW

• pulse duration < 100 fs (rms)

• integrated timing jitter < 15 fsin the interval [1 kHz, 10MHz]

• mechanically robust, easy to maintain

⇒ established additional diagnostics to ensure⇒ single-pulse operation

L/4

L/2

L/2coll.

collimator

WDM

L/4

isolator

PBC

L/4

piezo mirror on

motorized stage

Er-doped

fiber

L/2

PBC

PBCL/2

photo diode

to free-space

distribution

pump

combiner

980 nm

pump

diodes

autocorrelator

OSA,

amplitude mon.

• back to breadboard design (severe issues with engineered versions)

• prepared for further automatization, exception handling

• in operation for > 6 months without major problems



Master Laser Oscillator I.

Overview and Current Status

Specifications/Requirements

• topology: EDFL in σ-configuration


• average power: ≈ 100mW

• pulse duration < 100 fs (rms)


• mechanically robust, easy to maintain

⇒ established additional diagnostics to ensure⇒ single-pulse operation

• back to breadboard design (severe issues with engineered versions)

• prepared for further automatization, exception handling

• in operation for > 6 months without major problems



Master Laser Oscillator II.

Some Recent Results

−180

−160

−140

−120

−100

−80

−60

−40

SSB

phas

e no

ise

L ! (

dBc/

Hz)

102 103 104 105 106 107 1080

2

4

6

8

10

12

14

inte

grat

ed ti

min

g jit

ter

(fs)

offset frequency (Hz)

free−running, P = 10.8 mW −> u2t photo diode @ monitor port, Vbias = 7.00 V, thick RF [email protected] GHz = 10.4 dBm, att. = 10 dB, 1000 corr., no avg., open housing

integr. timing jitter = 11.5 fs

• integrated timing jitter 11.5 fsin the interval [1 kHz, 10MHz]




Some Recent Results

50 100 150 2000.99

0.995

1

1.005

1.01

time (h)

MLO

2 re

lativ

e op

tical

pow

er

peak−to−peak = 0.0180, standard deviation = 0.0039


• amplitude drift over 240 h:< 2% (peak-to-peak),∼ 0.4% (rms)→ no active stabilization




Some Recent Results

−800 −600 −400 −200 0 200 400 600

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

delay (fs)

auto

−co

rrel

atio

n si

gnal

(a.

u.)

measured datataup = 87.3 fs


• amplitude drift over 240 h:< 2% (peak-to-peak),∼ 0.4% (rms)→ no active stabilization

• pulse duration τp = 87 fs, fromsech2(t/τp)-fit→ important for fiber link→ dispersion compensation



Master Laser Oscillator III.

Alternative Concept: Investigation of a Commercial Laser System

Promising: OneFive ORIGAMI-15

• topology: unknown (SESAM, soliton)


• average power: > 100mW

• pulse duration: τp < 150 fs


• mechanically robust, easy to maintain(sealed housing, one button)

⇒ ordered custom system⇒ delivery/installation: April 2010

⇒ PSI enabled progress on direct conversion⇒ November 2009 – . . .

⇒ open questions: long-term stability, life-time

−170

−160

−150

−140

−130

−120

SSB PhaseNoise [dBc/Hz]

1k 10k 100k 1M 10M0

1

2

3

4

5

Timing Jitter [fs]

Frequency [Hz]

Origami / Direct−Conversion − RF Phase Noise: 4.3 fs

2.6 0.5 1 3.2

6 V, 13 mV, 1.5 GHz, 4.2 dBm

integrated timing jitter: 4.3 fs


Progress and Results of Selected Subsystems Fiber Links

Actively Length-Stabilized Fiber Links I.

Experiences with the Engineered Version

Design: 3 Layers

• free-space optics

• fiber installations

• electronics compartment

⇒ solved, taken into account for in first redesign (manufacturing now)

⇒ easy: dispersion compensated EDFA (between FSD and link box)⇒ advantageous: new pre-spliced DCF (lower insertion loss)

⇒ major redesign considerations in progress → end-of-year



Actively Length-Stabilized Fiber Links I.

Experiences with the Engineered Version

Design: 3 Layers

• free-space optics

• fiber installations

• electronics compartment

Problems

• motorized delay stage

• telescope design→ incoupling efficiency

• optical isolation of EDFA

• ∼ 20 minor issues

⇒ solved, taken into account for in first redesign (manufacturing now)

⇒ easy: dispersion compensated EDFA (between FSD and link box)⇒ advantageous: new pre-spliced DCF (lower insertion loss)

⇒ major redesign considerations in progress → end-of-year



Actively Length-Stabilized Fiber Links II.

Performance Example: Long-Term Drift

50 100 150 200

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

time (h)

fiber

link

in−

loop

tim

ing

jitte

r (f

s)

LINK09 in−loop timing jitter, mean = 0.91 fs, std = 0.08 fsLINK09 120 minute moving averageLINK15 in−loop, mean1 = 2.68 fs, std1 = 0.83 fs, mean2 = 1.11 fs, std2 = 0.28 fsLINK15 120 minute moving average

LINK09 → BAM UBC2

• ≈ 165m

• FRM link-end

• returning pulses τp = 115 fs

LINK15 → OXC EO

• ≈ 440m

• loop-mirror link-end

• returning pulses τp = 200 fs

• engineered fiber link boxes ensure reliable operation

• timing distribution to the femtosecond level over long periods

• out-of-loop measurement setup currently under consideration


Progress and Results of Selected Subsystems Laser-to-Laser Synchronization

Laser-to-Laser Synchronization I.

Idea and First Implementation

RF Synchronization

• based on a RF down-mixing scheme

• timing jitter & 35 fs

Optical Synchronization

• basis: non-linear optics⇒ more precise measurements⇒ timing jitter � 10 fs

• issues:→ two individual oscillators→ different repetition rates→ different wavelengths

Characterization & Drift Measurements

• only possible with out-of-loopmeasurement⇒ requires second OXC



Laser-to-Laser Synchronization II.

Single-Crystal Two-Color Balanced Optical Cross-Correlator

[email protected]&1550nmHT@

HT@HR@

527.7nm800&1550nm

2

1

BBO GDD

• sum frequency generation (SFG) in type-I(−) phase-matched BBO crystal

• group delay, then again SFG

• measurement of SFG intensities⇒ difference signal highly sensitive to timing changes⇒ (nearly) independent on amplitude noise of the lasers (“balanced”)⇒ input signal for PLL



Laser-to-Laser Synchronization III.

Performance of the Optical Lock of the EO experiment’s Ti:Sapphire Laser

−600 −400 −200 0 200 400 600 800

−1

−0.75

−0.5

−0.25

0

0.25

0.5

0.75

1

relative time delay ∆t (fs)

OX

C2 d

iffere

nce

sig

nal (

V)

raw data

Gaussian fit

slope = 29.5 mV/fsfit

• used to lock, i.e. firstcross-correlator

• measured by scanning an opticaldelay stage

0 50 100 150 200 250 300 350 400−60

−40

−20

0

20

40

60

time (s)

rela

tive

timin

g ch

ange

(fs

)

RF lock, jitter = 19.3 fs (rms over 400 s)optical lock, jitter = 7.9 fs (rms over 400 s)

• out-of-loop signal

• timing jitter over 400 s: 7.9 fs(rms), practically no drift

• RF lock over 400 s: 19.3 fs (rms)

→ challenging task to reproduce routinely/automatically


Progress and Results of Selected Subsystems Infrastructure and Electronics

Infrastructure and Electronics.Some Important and Critical Points

Infrastructure

• new climatization in the synchronization hutch

• new power supplies with battery backup for optical table and criticalrack-chassis

• new power supplies for BAM/EBPM installations in the tunnel

• improved BCM, VME & cabling installations in the tunnel

Electronics & Software

• addressed all issues of the fast ADC for BAM and EBPM readout

• improved VME laser diode drivers

• extension of the RF-lock server (exception handling!)

• polarization/amplitude feedback for the fiber links

• better failure detection and remote control

⇒ important for a robust and reliable system

⇒ time-consuming and expensive


Latest Development Photoinjector Laser Synchronization

Photoinjector Laser Synchronization I.

Motivation

New BAM upstream of First Chicane

• timing change:

δtbeam = Ggun δtgun +Glaser δtlaser

with

δtgun =δφgun

ωRFand δtlaser =

δφlaser

ωRF

• variation of gun and laser phase:

⇒ δtgun + δtlaser = 2.12 ps/deg

⇒ (32% + 68%)

• variation of RF-gun phase slope:

⇒ δtmax = 0.52 ps

⇒ RF-gun phase feed-forward or feedback⇒ requires arrival-time information of⇒ photoinjector laser pulses on cathode!

→ optical cross-correlator

−0.3 −0.2 −0.1 0 0.1 0.2 0.3−0.5

0

0.5

Change of Gun or Laser phase [deg]

Varia

tion

of a

rriva

l tim

e [p

s]

Laserdt/d! = 1.449 ps/degGundt/d! = 0.675 ps/deg

0 5 10 15 20 25 30−0.8

−0.6

−0.4

−0.2

0

0.2

0.4

Bunch number in macro pulse (10µs bunch spacing) [No #]

Aver

aged

arri

val t

ime

of b

unch

es [p

s]3.6 deg/ms4.2 deg/ms4.8 deg/ms5.4 deg/ms6.0 deg/ms



Photoinjector Laser Synchronization II.

Prototype Implementation at FLASH

diode amplifier & pulse picker

flashlamps

LBOBBO

e

toroidRF gun

Nd:YLF osc.

manual shifter

photoinjector laser 2

1047 nm~ 15 nJ27 MHz

AD

C

OpticalX-Correlator

AD

CA

DC

DACDAC

IQ

13

.5 M

Hz

10

8 M

Hz

1.3

GH

z

MO

LO generation

controllerADCADC

MLO

20m SMF, PSOF, orRF-based FiberLink

synch. hutch

fR

LDD

EDFA

AD

C

controller

dispersion comp.

MD

R

L4,L2

delay line

MD

R

VME + Beckhoff or uTCA system

EOM AOM AOM

• MLO delivers precise timing information over an optical fiber• measuring timing jitter between PTO and reference on O(10 fs) level with

the optical cross-correlator• stabilize 1.3 GHz phase of the PTO’s EOM by a feed-forward algorithm or a

control loopS. Schulz (Hamburg U, DESY) FLASH Optical Synchronization FLS Workshop 2010 15 / 25


Photoinjector Laser Synchronization III.


HR@628nmHT@1047&1550nm

HR@628nmHT@1047&1550nm

BBO crystalL = 5 mm

input 1550 nm216 MHz200 fs FWHM

input 1047 nm27 MHz (10 Hz burst)12 ps FWHM

"2" fast detectors "1"

filters @ 628 nmHR@1047nmHT@1550nm

fixedend mirror

adjustableend mirror

HR@1047nmHT@1550nm

• one short pulse (126 fs) and one long pulse (∼ 4 ps)

• sum frequency generation (SFG) in type-I(−) phase-matched BBO crystal

• separation of pulses, delay with mirror, then again SFG

• measure sum frequency intensity shot-by-shot with fast detectors and ADCs(νrep = 27MHz)



Photoinjector Laser Synchronization III.




Photoinjector Laser Synchronization IV.

First Results

PRELIMINARY

• scan delay between MLO and PTO pulse → SFG in both directions of OXC

• measure for PTO pulse duration ⇒ 4.9 ps (rms)

• delay with mirror for second SFG not adjusted correctly


Summary and Outlook

Summary and Outlook.Current Status of Upgrades to the Optical Synchronization System

1 many new components and subsystems

2 considerable progress towards a robust, reliable and engineered system

3 implementation of a longitudinal feedback


Acknowledgements

Acknowledgements.

Thank you for your attention!

The FLASH LbSyn Team

M. K. Bock, M. Felber, P. Gessler, K. E. Hacker, F. Ludwig, H. Schlarb,B. Schmidt, S. Schulz, L. G. Wissmann

and its former membersV. Arsov, F. Loehl, A. Winter, J. Zemella

with its many collaborators, technicians and people of other groups


Additional Slides: RF, L2L-Testsetup, FiberLink Principle... RF-Lock Electronics for Ti:Sapphire Lasers

RF-Lock Electronics for Ti:Sapphire Lasers.Extended RF Circuit with 81 MHz, 1.3 GHz and 9.1 GHz Phase-Detectors


Additional Slides: RF, L2L-Testsetup, FiberLink Principle... Laser-to-Laser Synchronization

Laser-to-Laser Synchronization A-I.

Idea and First Implementation

test reference laserEDFL (self-built)

• λ1 = 1550 nm, ∆λ1 = 100 nm

• τFWHM,1 = 120 fs

• frep,1 = 40.625 MHz

experiment’s laserTi:Sapphire oscillator (Coherent Micra-5)

• λ1 = 800 nm, ∆λ1 = 65 nm

• τFWHM,1 = 35 fs

• frep,2 = 81.25 MHz

RF Synchronization

• based on a RF down-mixing scheme

• timing jitter & 35 fs

Optical Synchronization

• basis: non-linear optics⇒ more precise measurements⇒ timing jitter � 10 fs

• issues:→ two individual oscillators→ different repetition rates→ different wavelengths

Characterization & Drift Measurements

• only possible with out-of-loopmeasurement⇒ requires second OXC


Additional Slides: RF, L2L-Testsetup, FiberLink Principle... Laser-to-Laser Synchronization

Laser-to-Laser Synchronization A-II.

Implementation using the FLASH EO Ti:Sapphire Laser System – Layout


Additional Slides: RF, L2L-Testsetup, FiberLink Principle... Fiber Links

Actively Length-Stabilized Fiber Links A-I.

Principle of Operation


Additional Slides: RF, L2L-Testsetup, FiberLink Principle... Fiber Links

Actively Length-Stabilized Fiber Links A-I.

Principle of Operation

• sum frequency generation (SFG) of reference and reflected pulses in type-IIphase-matched PPKTP crystal, delay, then again SFG

• measurement of SFG intensities⇒ difference signal highly sensitive to timing changes⇒ (nearly) independent on amplitude noise of the lasers (“balanced”)⇒ input signal for PLL (acting on piezo stretcher and delay stage)


Additional Slides: RF, L2L-Testsetup, FiberLink Principle... Free-Space Distribution

Free-Space Distribution A-I.

Schematics and Implementation

• designed for redundant two laser operation

• allows for setting optical power per device

• no dispersive pulse broadening


Additional Slides: RF, L2L-Testsetup, FiberLink Principle... Free-Space Distribution

Free-Space Distribution A-II.

Measurements

0 5 10 15 2090

91

92

93

94

95

96

97

98

99

100

beam cube number

tran

smis

sion

Tp, r

efle

ctio

n R

s (%

)

Transmission and Reflection of Newport 05FC16BC.9 at 1550 nm

specification: Tp > 90 %, R

s > 99.5% (w/o surfaces)

Tp meas. 1, rms deviation = 0.96 %Tp mean 1 = 96.51 %Tp meas. 2, rms deviation = 1.08 %Tp mean 2 = 96.23 %Rs meas. 1, rms deviation = 0.39 %Rs mean 1 = 98.27 %Rs meas. 2, rms deviation = 0.66 %Rs mean 2 = 98.14 %

• beam cubes do not meet specs(to a small amount)

• use only the best (→ QA)

Optical Power Drift

• 0.3% (peak-to-peak) over 14 h

• but: cannot distinguish betweentemperature and pointingstability

• can it be done better?

⇒ consider also alternative schemes:⇒ → dispersion compensated high-power EDFA⇒ → hybrid design: free-space + fiber splitter(s)


installation progress of the laser-based synchronization ......mar 04, 2010 · free-space optics...

Documents