the lcls bc1 bunch length monitor system

G. TravishG. Travish

FAC MeetingFAC Meeting [email protected]@physics.ucla.edu

28OCT0528OCT05

The LCLSBC1 Bunch Length

Monitor System

Eric Bong1

Mike Dunning2

Paul Emma1

Patrick Krejcik1

Tim Montagne1

Jamie Rosenzweig2

Gil Travish2

Juhao Wu1

1SLAC | 2UCLA



28OCT0528OCT05

Intro: The Problem

The LCLS demands tight beam parametersLongitudinal feedback systems needed (along with other diagnostics and feedback systems)

Bunch length

Energy

This is a critical diagnostic“Machine won’t work without this”

“Has to operate 24-7”

Need to start addressing safety and approval issues

Time is short



28OCT0528OCT05

Intro: The Approach

UCLA to build bunch length monitor systemSystem will consist of two grating polychromators, one at each bunch compressor

Simulations to determine start-end response of monitor.

SLAC to integrate into beamline

Working on testing at and with ANL-APS



28OCT0528OCT05

Introduction



28OCT0528OCT05

1.0 nC 0.2 nCRelevant Parameters

Nominal electron energy, BC1 0.25 0.25 GeV

Nominal electron energy, BC2 4.3 4.3 GeV

Peak current 3400 2500 A

Nominal RMS bunch length, BC1 200 60 µm

Nominal RMS bunch length, BC2 20 8 µm

Nominal RMS bunch duration, BC1 670 200 fs

Nominal RMS bunch duration, BC2 67 27 fs

Max single bunch repetition rate 120 120 Hz



28OCT0528OCT05

Intro: Possible Solutions

Streak Camera

Interferometer

Electro-Optic Techniques

RF Deflecting Cavity

Polychromator (Spectrometer)

______(fill in your favorite method here)



28OCT0528OCT05

* LCLS PRD.

# J. Wu et al., SLAC-PUB-11276, May 2005.

Intro: System Requirements*

Only relative bunch length needed- not absolute bunch length

Need two bunch length monitors- one at each bunch compressor#

Single-shot

Non-invasive

Maintenance free for several days

Maximum repetition rate: 120 Hz

Measurement resolution: 1-2 % of nominal bunch length(sub-femtosecond)

Long term signal drift: <2% over ~24 hours



28OCT0528OCT05

Observables- Bunch length z

- Energy E

Controllables- Linac voltage Vrf

- Linac phase rf

LCLS longitudinal feedback has 2 bunch length loops

Feedback model studied by Wu, et al., SLAC-PUB-11276, May 2005.

Conceptual Schematic of single loop

Intro: Phase Feedback



28OCT0528OCT05

Hamamatsu "FESCA-200”(Femtosecond Streak Camera).

Temporal resolution: 200 fs.

Possible Solutions

Streak Cameras+ Single-shot+ Wide dynamic range- Limited temporal

resolution(~200 fs at best)

- Trigger jitter



28OCT0528OCT05

Possible Solutions

Interferometers+ Can be single-shot? Sufficient temporal

(frequency) resolution

+ Compact- Narrow dynamic

range- Complex



28OCT0528OCT05

Electro-Optic Methods+ Single-shot? Non-invasive? Temporal resolution- Not yet mature- Require expensive femtosecond lasers

Possible Solutions

P. Bolton et al., SLAC-PUB-9529.Transverse probe geometry produces a

spatial image of the bunch. Also see:

http://www.rijnh.nl/users/berden/ebunch.html



28OCT0528OCT05

RF Deflecting Cavities+ Single shot? Temporal resolution - May require separate RF

system- Invasive (destroy

measured shot)

Possible Solutions

The UCLA 9-cell X-band standing wave deflecting cavity. Courtesy Joel

England.



28OCT0528OCT05

Possible Solutions

Polychromators+ Single-shot+ Temporal resolution+ Robust- Require relatively expensive detector &

vacuum system



28OCT0528OCT05

BC1 Single-Shot Spectrometer

After 4th magnet

CSR port

Narrow angle

Magnet justupstream



28OCT0528OCT05

Single-Shot SpectrometerUse CSR/CER from bunch compressor chicane magnets

Vacuum port window Focusing/turning mirror Entrance slit Grating Off-axis parabola (line focus) Multichannel detector

Cryostat & Bolometers

Basic Design

Line focus mirror



28OCT0528OCT05

The Selection: Polychromator

Spectrums are very reliable; no dependence on amplitude (charge) or intricacies of pulse shape.

Speed, bandwidth and phase noise not an issue.

Robust (0-2 moving parts)

Proven

Simple

Easy to model and easy to fix



28OCT0528OCT05

Single-Shot SpectrometerBunch Distributions

BC1 BC2

• Smooth parabolic distribution

+ Simple CSR spectrum

• Wake-induced double-horn

- Complicated CSR spectrum

Cou

rtes

y P

. Em

ma

Cou

rtes

y P

. Em

ma



28OCT0528OCT05

UCLA SimulationsTREDI Post-processor: FieldEye

Calculate far field radiation pattern using Lenard-Wiechert algorithmFourier analysisAngular spectrumDependent on TREDI

TREDI: time intensiveFlexibility in post-processing

Other applicationsCoherent Transition Radiation Coherent Cerenkov radiation

QuickTime™ and aTIFF (LZW) decompressor

are needed to see this picture.

Sample CSR spectrum calculationusing FieldEye



28OCT0528OCT05

Related CER Work (at BNL - ATF)

UCLA built ATF compressor.

Brookhaven Si Bolometerfor CER detection.FieldEye calculated spectrum for 20 micron

long beam, 1000 particles, edge radiation in chicane



28OCT0528OCT05

Challenge: Beamline Integration

Low-loss vacuum port window over desired frequency range (Diamond)

Cryostats: liquid helium & nitrogenHelium hold time (weeks?)

Closed-cycle nitrogen system (Sterling Engine?)

Room temperature alternative?

Windowless enclosure for detector system



28OCT0528OCT05

Central Issue: Flat response curve over THzMaterials:

Z-cut crystalline quartzSiliconHDPEPTX (polymethylpentene)Diamond

References:http://tesla.desy.de/~rasmus/projects/autocorrelator/plan.pdfSLAC-PUB-11249

Vacuum Port WindowMaterials & Issues



28OCT0528OCT05

Diamond Vacuum Port WindowAbsorption Curve



28OCT0528OCT05

Harris International: http://209.123.148.104/windows.asp

Fraunhofer: http://www.iaf.fraunhofer.de/eng/gf/cvd-prod-strah.htm

EOC (Electro Optical Components): http://www.eoc-inc.com/diamond_optics.htm

Diamond Vacuum Port WindowVendors



28OCT0528OCT05

~100 mm

OAP focusing/turning mirror OAP line focus mirror

~100 mm

~250 mm

Mirrors

Conventional: gold coated copper

Here: could be CNC aluminum



28OCT0528OCT05

MirrorsVendors

Kugler of America:

http://www.kuglerofamerica.com/optics.htm

Reynard Corp.:http://www.reynardcorp.com/index.php

Janos Tech:

http://www.janostech.com/index.html

CVI Optical Components:

http://www.ocioptics.com/product.html

Janos Tech

Kugler of America



28OCT0528OCT05

~150 mm

Grating

Machined Copper

Blazed

Groove width ~1 mm

To be optimized



28OCT0528OCT05

BC1• Frequency range:

• 150-500 GHz• ~ 20 channels

• Easy, but big• large vacuum chamber• large optics

• InSb hot electron bolometers

BC2• Frequency range:

• 1-4 THz• ~ 20 channels

• More challenging than BC1• Needs special filtering• Thermal composite bolometers?• Need to research more

Challenge: Detectors



28OCT0528OCT05

+ Speed+ Sensitivity- Cryo-cooling

• Cryostat hold time

- Expensive

VendorsQMC Instruments: http://www.terahertz.co.uk/QMCI/qmc.html

IR Labs: http://www.irlabs.com/irlabs%20pages/irlabs_frameset.html

DetectorsBolometers



28OCT0528OCT05

+ Relatively inexpensive+ Speed- Sensitivity- Resonances

VendorsEOC: http://www.eoc-inc.com/pyroelectric_detectors.htm

Fuji & Co.: http://www.fuji-piezo.com/prodpyro.htm

DetectorsPyroelectric



28OCT0528OCT05

+ Less expensive than bolometers- Speed (25 Hz)- Noise sensitivity- Size

Vendors

QMC Instruments: http://www.terahertz.co.uk/QMCI/qmc.html

Tydex: http://www.tydex.ru/about.html

DetectorsGolay Cells



28OCT0528OCT05

+ Inexpensive

+ Speed- Frequency response- Sensitivity

VendorsVirginia Diodes: http://www.virginiadiodes.com/index.htm

Advanced Control Components: http://www.advanced-control.com/products-detectors.php

DetectorsDiodes



28OCT0528OCT05

20-channel linear array of InSb hot-electron bolometers,courtesy QMC Instruments.

250 mm

• InSb hot-electron bolometers• 10 liter cryostat• Helium hold time: 4-6 weeks!

Detector blocks with

Winston Cones and

Filters

BC1 Detector Assembly



28OCT0528OCT05

Testing

LCLS BC1 is most similar to the APS LEUTL Bunch Compressor.

Propose collaboration with ANL on testing.

Rapid installation of window and turning mirror (site specific) during proximate access.

Installation of polychrometer as assembly.



28OCT0528OCT05

What will we test

Beam energy, energy spread, charge and bunch length can mimic LCLS design.

Simulations can be confirmed for a range of bunch lengths.

Noise, dynamic range in amplitude (charge) and bunch length.

Hard to test: feedback



28OCT0528OCT05

Project Start

Activity NameStart Date

Finish Date Third Quarter Fourth Quarter First Quarter Second Quarter Third Quarter

2004 2005

Third Quarter Fourth Quarter First Quarter Second Quarter Third Quarter

Project History

Initial Proposal 09/01/04 09/16/04

Anticipated Funding Start 10/01/04

MOU Addendum 02/20/05

SOW/MOU approval 03/13/05 03/15/05

Funding ($) arrived 04/18/05

PRD Draft 06/07/05

PRD - Rev 0 09/16/05

CAD Drawings Available 10/28/05



28OCT0528OCT05

Project Overview

Activity NameStart Date

Finish Date Third Quarter Fourth Quarter First Quarter Second Quarter Third Quarter Fourth Quarter

2005 2006

Third Quarter Fourth Quarter First Quarter Second Quarter Third Quarter Fourth Quarter

Project Scoping ... 09/01/0402/20/0503/13/0506/07/0509/16/05

09/16/04

Theory... 10/31/05 01/02/06

Design ... 11/03/05 01/27/06

Acquisition

Detector 01/30/06 06/09/06

Window 01/02/06 04/28/06

Optomechanics04/03/06 05/12/06

Fabrication ... 02/06/06 03/10/06

Assembly ... 06/09/06 07/06/06

Testing ... 06/16/06 09/22/06

Prototype Delivery 11/01/06

LCLS access #1 08/01/06 11/30/06

LCLS access #2 09/01/07 11/30/07

CD-3 review 02/07/06 02/09/06



28OCT0528OCT05

Schedule Issues

Theory: behind on simulations

Acquisition: driven by detector

Testing: can you install and test anything in 2 months?

Delivery: November 2006

Post delivery: beamline integration, safety issues, etc.



28OCT0528OCT05

Conclusions

Simulations behind schedule

Subcomponent selection proceeding well

Challenging schedule can be addresses with APS collaboration, better coordination with SLAC, and timely funding.

Hardware cost and schedule driven by detector

Beamline integration needs to be considered now



28OCT0528OCT05

End



28OCT0528OCT05

Possible SolutionsSummary

Single-shot Non-Invasive

Sufficient Temporal Resolution

Maintenance Free

Streak Camera Y Y N Y

Interferometer Y Y ? N

Electro-Optic Y ? ? ?

RF Deflector Y N ? N

Polychromator Y Y Y Y

PBPL



28OCT0528OCT05

Single-Shot SpectrometerChallenge: BC2 CSR Spectrum

CSR energy spectrum after BC2.Black curve: double-horn distribution

Blue curve: Gaussian distributionRed curve: step function

From J. Wu, et al., SLAC-PUB-11275, May 2005.

• Double-horn distribution complicates CSR spectrum

- Similar to Gaussian below 4 THz

- Wu: Stay below 4 THz



28OCT0528OCT05

WorkplanSimulate CR exiting vacuum ports of BC1, BC2 & arriving at detector

TREDI/FieldEye simulations

Choose detector typeFinalize bolometer evaluations

Continue to study systemFormation length (source to detector distance)Dynamic range (grating, in situ tuning)Calibration methods

Mechanical design & beamline integration with SLACCAD design workFinalized by SLAC

Test system (SPPS or APS Linac)

the lcls bc1 bunch length monitor system

Documents