BPM Upgrade Projects Status & Technical Update

BPM Upgrade ProjectsBPM Upgrade ProjectsStatus & Technical UpdateStatus & Technical Update

Bob WebberDOE Review of Tevatron Operations

March 2005

BPM Projects - DoE Tevatron Operations Review - March 2005 - Webber 2

Recent and Active BPM ProjectsRecent and Active BPM Projects

Recycler Ring Completed and operational since early 2004

NuMI Beam Line Completed and operational since first NuMI extracted beam This system reported meaningful beam position data on first pulse !!!

Tevatron Ring (1.4.5.4) (9/2/03 – 5/17/05) ($1081K labor, $1764K m&s) Installed and operational for routine proton measurements in ~25% of

the ring; functioning compatibly with remaining old systems Completion of installation scheduled for end of May 2005

Antiproton Transfer Lines (1.3.6.5.2) (3/26/04 – 7/25/05) ($91K labor, $284K m&s) (details in Elvin Harms’ Rapid Transfer Talk earlier this breakout) In final design stage, manpower assigned, ~80% procurements done Will employ NuMI/Recycler software and Tevatron EchoTek boards Completion expected by ~August ’05

Main Injector Ring (1.1.3.2) (1/3/05 – 6/1/06) ($324K labor, $900K m&s) Requirements being reviewed Awaits manpower currently devoted to Tevatron and Transfer Line

projects All systems are built around a common, commercial digital

receiver board (Echotek)


Tev BPM - General StatusTev BPM - General Status

Joint Computing and Accelerator Division project, CD has provided ~80% of total reported project effort

Essentially all hardware is in hand, including spares Considerable planning and care allow adiabatic

installation without adversely impacting Tevatron operations It is a big deal to seamlessly replace a large operational

system without disrupting Collider operation and luminousity production

7 of 27 systems are installed and routinely operational for proton measurements (functional replacement for old systems)

Antiproton position measurements have been demonstrated but not yet routinely operational

Installation pace is presently limited ~equally by: Opportunity -- we don’t install during a Collider store Debugging –- operational problems must be solved as they

are discovered


TeV BPM - SoftwareTeV BPM - Software

~85% of required VME front-end software functionality has been implemented and debugged

Data is packaged, either in VME front-end or in ACNET BPMLIB library code, to “look like” data from the old BPM system so as to minimize modifications to legacy console applications

Software switch permits transparent operation with mixture of old and new BPM systems in field

Critical legacy console applications are working with new BPM data after little or no modification

Some new applications are being developed to take advantage of the system’s new capabilities

SDA is being updated to use the new BPM data as each new crate is installed


TeV BPM - Colored Pictures!TeV BPM - Colored Pictures!

A3 House Installation BPM Integrity Check


TeV BPM – 36 Bunch Closed Orbit Resolution Upper LimitsTeV BPM – 36 Bunch Closed Orbit Resolution Upper Limits

20 minutes of closed orbit data for A3 BPMs400 microns full vertical scale all plots

Histograms of same data and RMS variation from the mean


TeV BPM – 36 Bunch Closed Orbit ResolutionTeV BPM – 36 Bunch Closed Orbit Resolution

Correlation plot of closed orbit data from two BPMs on opposite sides of ring estimate resolution to be ~5 microns


Tev BPM – Single Proton Bunch TBT Tev BPM – Single Proton Bunch TBT MeasurementsMeasurements

TBT resolution is better than 50 microns RMS


TeV BPM - Antiproton Measurement “Plan A”TeV BPM - Antiproton Measurement “Plan A”

Establish cross-talk (imperfect pickup directionality) coefficients a, b, c and d and compute corrected antiproton signals according to:

Now working to establish unique coefficients for each BPM and automate operational maintenance of time varying values

There is a “Plan B” …

BPM

Ap

Protons Pbars

Apbar

BpbarBp


TeV BPM – Antiproton MeasurementsTeV BPM – Antiproton Measurements

Proton sum signal

Proton position

Pbar positionPbar sum signal

Pbar sum signal during proton loading (ideally zero)

Proton loading

Pbar loading

Ramp


TeV BPM - What’s LeftTeV BPM - What’s Left

Solve currently known bugs Intermittent first-turn and injection turn-by-turn

measurements (phase/timing problem)

Complete hardware installation Implement and test additional required

functionality Robustness of Pbar measurements “Safe” mode (timing robust first turn measurements) Calibration-scaling-offset database

Produce as-built drawings and documentation MOU -> operations Project closeout


Transfer Line BPMTransfer Line BPM

First system required to observe both 53MHz and 2.5MHz beam structures Major impact on design and required dynamic range of

analog circuitry upstream of digital receiver boards Precursor to MI BPM system with similar requirement

Operational scenarios are complicated due to different beam structures through beam lines at different times Recycler/NuMI front-end software that is suited mode

switching and cycle dependent data buffers will be used Design of analog circuit is in final prototype stage All major procurements except analog boards are

complete VME front-end software is ready and waiting Prototype/development system now installed at F23 Schedule calls for completion mid to late summer 2005

– integration will be subject to operational constraints


Main Injector BPM ProjectMain Injector BPM Project

Main Injector BPM project will follow on the heels of the Tevatron and Transfer Line BPMs

Scope Replace signal processing electronics and front-end

data acquisition systems for ~208 beam position monitors located around the Main Injector and in six buildings

Objectives Eliminate obsolete Multibus II hardware and ACNET

communications protocol Accommodate measurement of beam in 2.5MHz RF

buckets Provide improved measurement resolution

Beams Document #471, dated February 2003, specifies system requirements and is now under review by Main Injector Department


Main Injector BPM - Similarities to Tevatron BPMMain Injector BPM - Similarities to Tevatron BPM

General system design and implementation Beam information to be obtained and types of

measurements, i.e. first turn, closed orbit, and turn-by-turn

Anticipated hardware, except more complicated analog circuitry upstream of digitizer

Scope of DAQ, on-line, and off-line software, although different in detail


Main Injector BPM - Differences from Tev -- All in the Main Injector BPM - Differences from Tev -- All in the DetailsDetails

Main Injector operating cycles are more varied in type than in Tevatron and dynamically interwoven in time

Both 2.5MHz and 53MHz signal frequencies must be processed (in different Nyquist bands)

Measurement data for multiple operating cycle types is to be stored separately in parallel front-end data buffers

Both protons and antiprotons must be measured, but Do not circulate simultaneously Pickups are not directional

• no separate p/pbar cables• protons and pbars require different fine timing

Measurement of user selectable segment (batch) of the circulating beam is required


Main Injector BPM - Anticipated ImplementationMain Injector BPM - Anticipated Implementation

Analog front-end electronics will leverage heavily off Transfer Line BPM project (common requirement to deal with 2.5MHz and 53MHz signal frequencies)

VME and VXWORKS front-end DAQ systems as other new systems

Tevatron style EchoTek digital receiver boards Tevatron style digitizer clock and timing

boards Front-end software probably dependent on

what group implements the system (different in detail from both Tevatron and Recycler)


Main Injector BPM - How Project Likely Plays OutMain Injector BPM - How Project Likely Plays Out

Little dedicated effort invested up to now other than clarification and refinement of requirements MI department occupied with NuMI commissioning BPM people occupied with the other new BPM systems

Analog front-end design will flow from Transfer Line project People who provided engineering, software, and technical

support for NuMI BPM will focus on Transfer Line project through ~May ’05

Software support from Tevatron project might become available in ~June ’05

Commodity procurements can begin any time (EchoTek boards already procured and in hand)

A development system initially with either Tevatron or Recycler software should be installed in MI by June 2005

Computing Division will likely play a major role in the project


BPM Projects - SummaryBPM Projects - Summary

We are mid-stream in the effort to re-build all major BPM systems in the accelerator complex except Linac and Booster

Results from completed Recycler and NuMI system and partially installed Tevatron system show excellent performance and operational credibility and utility

Considerable effort remains to: Complete Tevatron system Overcome analog signal challenges for Transfer

Lines and Main Injector Meet the diverse and dynamic functional

requirements demanded in Main Injector


BACKUP SLIDESBACKUP SLIDES

#1577 B3 resolutions 1571 coupling 1581 pbars #1752 for Rob’s most recent very good talk


Antiproton MeasurementsAntiproton Measurements

Present ratio of proton to antiproton intensities combined with directivity of stripline pick-up produces residual proton signals at the antiproton port about 50% the amplitude of antiproton signals

When antiproton intensities increase they will begin to “contaminate” the proton signals also

We opt to not pursue p/pbar separation by precise timing

Typical signal from pbar end of BPM for present bunch intensities

80 nsec/tic

Pbar bunchsignal Undesired Proton

bunch signal


0 5 106

1 105

1.5 105

2 105

0

1

2

protons zz t( )

pbars zz t cptcog( )

t

VA14 Collision

0 5 106

1 105

1.5 105

2 105

0

1

2

protons zz t( )

pbars zz t cptcog( )

t

VA25 Collision

0 5 106

1 105

1.5 105

2 105

0

1

2

protons zz t( )

pbars zz t inj0cog( )

t

VA14 Injection

0 5 106

1 105

1.5 105

2 105

0

1

2

protons zz t( )

pbars zz t inj0cog( )

t

VA25 Injection

Separate p/pbar signals with relaxed timing requiring precision and maintenance of ~50 nsec rather than ~5 nsec

Observe only “isolated” proton or antiproton bunches at ends of 12-bunch trains

No pbar bunches observable One pbar bunch observable

Five bunches observable Two bunches observable

Antiproton “Plan B”Antiproton “Plan B”


Coverage of Ring in Antiproton “Plan B”Coverage of Ring in Antiproton “Plan B”

0 1000 2000 3000 4000 5000 6000 70000

50

100

21

clearz

bpm1

z

After 1st pbar injection nclr 123

0 1000 2000 3000 4000 5000 6000 70000

200

400

21

clearz

bpm1

z

After 4th pbar injection nclr 177

0 1000 2000 3000 4000 5000 6000 70000

200

400

21

clearz

bpm1

z

During acceleration nclr 235

0 1000 2000 3000 4000 5000 6000 70000

200

400

21

clearz

bpm1

z

After 7th pbar injection nclr 226

0 1000 2000 3000 4000 5000 6000 70000

200

400

21

clearz

bpm1

z

At collision nclr 224 Locations with at least one clear pbar bunch (at least 400 nsec from nearest proton bunch) at various times in cycle

x axis is feet around ring from B0, each point is BPM location

y axis is clearance >21 is “clear”


HA15 and VA14 P/Pbar Positions “Plan B”HA15 and VA14 P/Pbar Positions “Plan B”Changing SeparatorsChanging Separators

Horizontal Pbar

Horizontal Proton

Vertical Proton

Vertical Pbar

Position Scales 2mm/box


Horizontal A15 Helix Change and Pbar Loss “Plan Horizontal A15 Helix Change and Pbar Loss “Plan B”B”


Block Diagram Cartoon of Signal PathBlock Diagram Cartoon of Signal Path

Note that proton and pbar signal paths from same half of BPM are directly connected together by the stripline --- reflections from one will feed directly into the signal of the other

attenuator 53MHz BPF Eight

Channel

VME

Digital

Receiver

Proton Inside

attenuator 53MHz BPF

attenuator 53MHz BPFProton Outside

Pbar Outside

attenuator 53MHz BPF

Pbar Inside

VME

CPU

ACNET

Diagnostic

and Calibrate

Signals


Main Injector BPM - Possible Roles for Computing DivisionMain Injector BPM - Possible Roles for Computing Division

Total effort level would be between that provided for Recycler and Tevatron BPM Projects Less engineering design effort than in Tevatron project

Administrative project management Analog circuit board layout and fabrication based on AD

designs Support for procurement, incoming component

inspections and testing, production testing, equipment tracking, etc.

Supply Tevatron BPM style timing boards DAQ, on-line, and off-line software components after

successful completion of Tevatron system Installation manpower To serve all these roles, CD effort estimated as on the

scale of 60 FTE months Learning curve (AD language, machine operational issues,

ACNET) has already been largely traveled by CD personnel on Tevatron project

Details will be different


BPM Accuracy/Resolution SpecsBPM Accuracy/Resolution Specs

This is Table 2 from Requirements document. Table gives the most stringent requirements on the system; for certain types of operation these requirements are relaxed. Note: resolutions are

stated as 3 sigma.

Key Specifications (Protons):Measurement Range: 15mmAbsolute Position Accuracy: < 1.0 mmLong Term Position Stability: < 0.02 mmBest Orbit Position Resolution: < 0.02mm (0.3 sec averaging)Position Linearity: < 1.5%Relative Position Accuracy: < 5%Intensity Stability: < 2%Key Specifications (Pbars)_:Measurement Range: 15mmAbsolute Position Accuracy: < 1.0 mmLong Term Position Stability: < 0.02 mmBest Orbit Position Resolution: < 0.05mm (0.3 sec averaging)Position Linearity: < 1.5%Relative Position Accuracy: < 5%Intensity Stability: < 2%


Demonstrated Beam MeasurementsDemonstrated Beam Measurements

Recycler-type BPM front-end is set up for development and tests in TeV House A1

Connects to both Proton and Pbar signals of one horizontal BPM and one vertical BPM

Interfaced to ACNET with small subset of ultimate required functionality

Closed orbit and turn-by-turn measurement performance have been demonstrated

In use to assess narrowband frequency domain p/pbar signal de-convolution

Will soon demonstrate “wide” time separation of Protons and Pbars (utilize isolated bunches at the ends of the otherwise over-lapping 12-bunch trains); ~50 rather than ~5 nsec timing


Proton Positions in Load of Store #3172: Proton Positions in Load of Store #3172: OldOld vs vs New New

Old: rms error ~140 um for uncoalesced beam, ~70 um for coalesced beam, and 0.6 mm “step” between the two

New: rms error ~25 um for uncoalesced beam, <9 um for coalesced beam, no “step” between the two


Closed Orbit Resolution During Proton LoadingClosed Orbit Resolution During Proton Loading

Green: Vertical Position @ 100 microns/div showing ~10 micron resolution and orbit changes due to leakage fields in ramping injection Lambertson magnet

Red: Beam Intensity showing proton bunches loadingYellow: Time in Supercycle


Upper Limit of Closed Orbit ResolutionUpper Limit of Closed Orbit Resolution

Average of standard deviations for twelve five-minute intervals

First one-hour interval Second one-hour interval

0.0085 ± 0.00061 mm

0.0090 ± 0.00072 mm

Two one-hour periods of 1 KHz bandwidth proton position measurement data, 17 hours apart, in store #3148. (data-logged at

1 Hz) 50 microns / vertical division

Demonstrates upper limit resolution of 9 microns rms in 1 Khz (any real beam motion not excluded) to meet spec of 7 micron 1 sigma

in ~10Hz


Injection Turn-By-Turn --- Un-coalesced ProtonsInjection Turn-By-Turn --- Un-coalesced Protons

0 200 400 600 800 10002.5

2.25

2

1.75

1.5

va14_pos n

n

0 200 400 600 800 10001

1.5

2

2.5

3

ha15_pos n

n

0.5 0.550

0.02

0.04

fftv14p m

1fm

0.5 0.550

0.02

0.04

ffth15p m

1fm

Vertical A14 Horizontal A15

vertical axis units are millimeters in all plots

vertical tune horizontal tune


Turn-by-Turn Long After Injection Single Coalesced Turn-by-Turn Long After Injection Single Coalesced Bunch Bunch

0 200 400 600 800 10002.5

2.38

2.25

2.13

2

va14_pos n

n

0 200 400 600 800 10001

1.5

2

2.5

3

ha15_pos n

n


0.5 0.550

0.02fftv14p m

1fm

0.5 0.550

0.005ffth15p m

1fm

vertical axis units

are millimeters in all plots

Horizontal transverse motion due to persistent synchrotron oscillation


Upper Limits on TBT Resolution --- Vertical A14Upper Limits on TBT Resolution --- Vertical A14

If remove betatron and synchrotron motion by zeroing 17 frequency line amplitudes around betatron frequencies and one at synchrotron frequency, leaving untouched 494 of the original 512 frequencies, TBT resolution is found to be 15 microns

0 500 1000

2.4

2.2

2

va14_pos n

n0 1 10

42 10

40

0.02fftv14p m

fm

Standard deviation of raw data is 34 microns

0 500 1000

2.4

2.2

2

vcooln

n0 1 10

42 10

40

0.02fftv14p m

fm


Same Coalesced Proton Bunch Given Big Vertical Same Coalesced Proton Bunch Given Big Vertical KickKick


0 200 400 600 800 10006

4

2

0

2

va14_pos n

n

0 200 400 600 800 10001

1.5

2

2.5

3

ha15_pos n

n

0.5 0.550

0.5

1

fftv14p m

1fm

0.5 0.550

0.1ffth15p m

1fm

vertical tune horizontal tune vertical tune


Antiproton “Plan A”Antiproton “Plan A”

Model BPM as an 8-port network at processing frequency Measure network transfer functions with beam, e.g. ratio of proton signal

on pbar end to proton signal on proton end Measure signals then correct according to pre-determined transfer

function before computing positions Measurements in process to determine achievable accuracy

LinearSystem

P Upper In

P Lower In

Pbar Upper In

P UpperScope

In Refl

In Refl

InRefl

InRefl

Pbar UpperScope

P LowerScope

Pbar LowerScope

1

2 7 8

3

4

56

BPM

Ap

Protons Pbars

Apbar

BpbarBp

Documents

BPM Upgrade Projects Status & Technical Update