Characterization of Fast Orbit Feedback System Om Singh, APS, ANL NSLS-2 Beam Stability Workshop...

Characterization of

Fast Orbit Feedback System

Om Singh, APS, ANL

NSLS-2 Beam Stability Workshop

BNL, April 18-20, 2007

2Characterization of Fast Orbit Feedback System – APS, ANL Om Singh NSLS-2 Beam Stability Workshop, April 18-20, 2007

Outline

Bpm & Magnet Layout – One SR Sector

AC Beam Stability

Goals & Present Performance

Orbit Feedback Sub-systems

Resolution & Responses

Digital Process Rate & Orbit Correction Configuration

Summary & Upgrade Plans

Photo


27.6 Meters

Number of sectors = 40 Circumference = 1104 m

Transverse Tunes νx= 36.2, vy = 19.27 Energy = 7 GeV

Beam Current = 102 mA RF Frequency = 352.194 MHz

Revolution Frequency = 271.554 KHz Harmonic number = 1296

One Sector of the Advanced Photon Source Storage Ring


Beta Functions


X-Ray Beam Position Monitors

Photo -emission sensory blades Excellent thermal insulation

and vibration damping Beam Position Measurement BW

up to Tens of KHz

One Sector - Cartoon

Bending Magnet

SlowCorrectionMagnet

FastCorrectionMagnet

Monopulse BPM – turn by turn (AM/PM)

Up to 8000 turn-by-turn samples of beam history - @ ~ 271 KHz Averages to 50 Hz, 2 Hz, 0.03 Hz BW

Sector N

Narrowband (Switching) BPM

Small Bunch Pattern & Intensity Dependence (vs Mp bpm) High Reliability and easy maintainability Beam Position Measurements BW up to ~ 5KHz

EllipticalChamber PUEs

ID ChamberPUEs

ID Device

Beam Position Monitors and Dipole Magnets


AC Beam Stability Goals

SR Emittance εe= 3.1 nm; coupling = 0.8%

1. Presently σx = 280 μm; σy = 10 μm; σ’x = 12 μrad; σ’y = 3 μrad

Beam stability goal and present performance

1. Stability goal* (AC) 5% of the APS beam size/ divergence

Goal / Present Performance

x, y (μm rms ) x’, y’ (μrad rms )

Goal (0.017 – 200 Hz) 14.0, 0.5 0.60, 0.22**

Present (0.017 – 200 Hz) 6.0, 2.0 0.26, 0.34

Broadcast (0.017 – 30 Hz) 0.8, 0.6

*G. Decker presented to 2005 DOE-BES Review at APS** Includes photon divergence contribution, 7th harmonics, APS undulator A


AC Stability @ ID’s Sources

4um

1.5um

Correction BW


AC Pointing Stability @ ID’s SourcesPower Spectral Density Sqrt[Integ[PSD]] Sqrt[ReverseInteg[PSD]]

rad2/Hz

rad2/Hz

Horz.

Vert.

radrms

radrms

radrms

radrms

with Feedback

Frequency (Hz)

without Feedback

Frequency (Hz)

Frequency (Hz) Frequency (Hz)Frequency (Hz)

Frequency (Hz)

* G. Decker – DOE-BES Review 2005

220 nrad

Spec


Global 1.5 KHz Processing Architecture

20 Double Sector Processors

1 Master Processor

1 IOC Processor(Datapool) Reflective

MemoryNetwork


Orbit Feedback - Subsystems

Subsystems & Performance Limiting Factors

1. BPM/ Corrector - Resolution 2. BPM/ Corrector - Responses

3. DSP Processing Power - Process Rate & Orbit Correction Configuration


Beam Motion @ ID Source vs Bpm Resolution (Noise)

Cumul RMS 30 Hz band (H & V)

Beam Motion = 0.8 & 0.6 micron RfBpm Noise = 0.3 & 0.35 micron Faster digitization should lower noise floorXBpm Noise = 0.04 & 0.04 micron Low noise level, but so far only in DC orbit feedbackDigitizer Noise = 0.01 & 0.01 micron

30 Hz Band 30 Hz Band


BPM / Fast Corrector – Responses

BPM Response

One pole low pass filter @ 2KHz

Fast Corrector Response

One pole low pass filter @ 1.5 KHzDelay ζ = 0.2 ms

BPM & Fast Corrector Combined Phase Contribution

@ 100 Hz = 15o

@ 200 Hz = 27@ 300 Hz = 45@ 400 Hz = 57@ 500 Hz = 70


Orbit Feedback - Limitations

Digital Signal Processor -

Present hardware allows to process up to 1.5 KHz sample rate – limits orbit correction BW to 50-90 Hz

Up to 4 Bpms per sector are included due to processing time constraint

Study in progress to include Xbpms

Corrector

Only one corrector (A3) available with fast response – limits orbit correction

A second fast corrector is required for optimal orbit correction – simulation shows further noise reduction by ~ 2.5

RF

Bp

ms

XB

pm

s

Fast Corrector


Optimized Overall O. L. Responses (Vert) – 1.5 KHz vs 15 KHz

Optimized parameters @ 1.5 KHz s.r.

Digital Process Rate = 1.5 KHz Digitization Phase = (f/fs) * (360deg)

Regulator 1 HPF @ 0.07 Hz 1 LPF @ 25 Hz Gain = 4

Orbit Correction Bandwidth 100 Hz

Optimized Parameters @ 15 KHz s.r.

Digital Process Rate = 15.0 KHz Digitization Phase= (f/fs) * (360deg)

Regulator 1 HPF @ 0.105 Hz 1 LPF @ 75 Hz Gain = 6

Orbit Correction Bandwidth 400 Hz

o oUnit gain line


Summary – Stability Goals, Status & Plans

Beam Size/ Divergence

StabilityGoals

Stability Status

Meet Goals (yes/no)

x µm 14 6.0 yes

y µm 0.5 2.0 no

x’ µrad 0.6 0.25 yes

y’ µrad 0.22 0.35 no

Proposed Upgrade Plans

Add 2nd fast corrector per sector - simulation shows noise reduction factor is 2.5 Increase process rate ten-fold > 15 KHz - AC obit correction BW to 400 Hz Improve RfBpm noise floor – increase digitization rate Include Xbpms in fast orbit feedback

AC Beam Stability (0.017 – 200 Hz)

- Horizontal beam stability goals are met- Vertical beam stability requires a factor of ~ 4 improvement


Bpm/ Corrector In-Use Status – Orbit Feedback

RF

BPMs

Xbpms

Fast Corr

Configuration Layout – L. Erwin

HorizontalVertical


Photo


In Tunnel BPM Hardware (Rf Bpms)

Elliptical Chamber Buttons & Matching Networks

Filter & Comparator

Small Gap Chamber Buttons

Photo – M. Hahne


In Tunnel BPM Hardware (Xbpms)

* Deming Shu -APS

Xbpm Main Assembly

X-YTranslation

Stages

Mounting Stand

Xbpm Blades Assembly


BPM Front-End Electronics and Processors – Two Sectors

Patch Panel

Pre-Amps

Filters &Controls

Narrowband RFbpms

AM/PM RFBpms

FDBK Processors

BPM IOC

16 Bits ADCs

X-ray BPMs Interface

RF Bpms & Processors

Photos by M. Hahne

Outside Tunnel Hardware


Acknowledgement

To

All APS members for contribution

to support ongoing Beam Stability work


BPM Hardware Development Highlight RF Bpms

1. Mpbpm data acquisition* upgrade in progress, using fast 88 MHz A/D and FPGA

technology; can provide bunch by bunch position data

2. Evaluation of yet another kind BPM in progress from Instrumentation Technology

(Libera) – ALL digital bpm with direct sampling of each button RF signals. Two

units are in hand & evaluation with beam to follow

Photoemission type - Xbpm

1. BM Xbpm – extremely reliable; used in DC Orbit control routinely

2. ID Xbpm – “Decker distortion” has been completed redirecting unwanted stray

radiations away from ID radiations. Provides beam measurement to sub-micron

level at fix gap. When gap varies, photon beam measurement held to tens of

microns with feedforward algorithm

Xbpm – hard X-ray type Bpm under development; to resolve to sub-micron level even

with ID gap change; has potential for beamline alignment gold standard


ID photons

77 mrad

1 mrad

Stray radiation from upstream dipole, quadrupoles, sextupoles and correctors

Stray radiation from down-stream dipole, quadrupoles, sextupoles and correctors

Re-direction of Stray Photons by Girder Alignment*

Phys. Rev. ST Accel. Beams 2, 112801 (1999)* (G. Decker)


Beam-definingAperture

Water-cooledMoveable Scrapers

Shielded X-ray Detectors (4),Beryllium Filter

(Target: Cu or W)

X-rayFluorescence

Plan View of Hard X-ray Beam Position Monitor Concept

White ID X-ray Beam

InsertableFilter Array(C)(Fixed)

Above and Below Plane of Beam

* G. Rosenbaum & G. Decker

*

Characterization of Fast Orbit Feedback System Om Singh, APS, ANL NSLS-2 Beam Stability Workshop...

Documents

Transcript of Characterization of Fast Orbit Feedback System Om Singh, APS, ANL NSLS-2 Beam Stability Workshop...