Status of GRETINA

1
Status of GRETINA
David Radford ORNL Physics Division JUSTIPEN
February 2009
2
Status of GRETINA
David Radford ORNL Physics Division JUSTIPEN
February 2009

• Outline
• Brief description of GRETINA
• Status
• Detectors, support, electronics, computing
• Signal decomposition
• Siting plans
• Summary

3
GRETINA

• Gamma-Ray Energy Tracking Array for in-beam
  nuclear structure studies
• 28 highly segmented Ge detectors,
• in seven groups of four
• 36-fold segmentation
• (6 azimuthal, 6 longitudinal)?
• Tapered irregular hexagons
• Total 1? steradians
• Funded by DOE, under construction at LBNL
• Scheduled for completion (CD4) in March 2011

8 cm
9 cm
4
Principles and advantages of gamma-ray tracking

• Efficiency Proper summing of scattered gamma
  rays, no solid angle lost to suppressors
• Peak-to-background Reject Compton events
• Doppler correction - Position of 1st interaction
• Polarization Angular distribution of the 1st
  scattering
• Counting rate - Many segments

3D position sensitive Ge detector
Resolve position and energy of interaction points
Determine scattering sequence
5
GRETINA - People

• Contractor Project Manager I-Yang Lee (LBNL)?
• GRETINA Advisory Committee (GAC)
• Con Beausang (U. of Richmond)?
• Doug Cline (U. of Rochester)?
• Thomas Glasmacher (MSU / NSCL)?
• Kim Lister (ANL)?
• Augusto Macchiavelli (LBNL)
• David Radford (ORNL) (Chair)?
• Mark Riley (Florida State U.)?
• Demetrios Sarantites (Washington U.)?
• Kai Vetter (LLNL)?
• Working Groups and chairs
• Physics M. A. Riley    (FSU)?
• Detectors A. O. Macchiavelli (LBNL)?
• Electronics D. C. Radford (ORNL)?
• Software M. Cromaz  (LBNL)?
• Aux. Detectors D.G. Sarantites (WUSTL)?

6
Highlights of recent achievements

• Received CD2B/3B approval construction of all
  systems started
• Received and tested the first two
  quadruple-detector modules
• Ordered detector modules 3 to 6 from
  Canberra-Eurisys
• New version of signal decomposition program
  signal basis
• Achieved 2mm position resolution
• Understood and eliminated preamplifier crosstalk
  and oscillation
• Completed signal digitizer and trigger modules
• Ordered first set of computers for decomposition
  cluster
• Developed a suggested national lab rotation
  schedule for the first round of experimental
  campaigns

7
First Quadruple Cluster (Q1)?
Delivered Dec 2006
A-type
B-type
8
Second Quadruple Cluster (Q2)?
Delivered Dec 2008
9
Mechanical Support Structure - under construction
Spherical shell Rotation
Axle gear set
One hemisphere machined, tested and accepted
10
Electronics
Designed, fabricated, and tested digitizer
modules (LBNL) and trigger modules (ANL)? -
Worked beautifully together on first try
Digitizer and trigger modules under test
Digitizer module, 150 produced
11
Computing System
Bench-mark CPU-scaling tests for decomposition
algorithm - Almost linear scaling up to dual
quad-core CPUs - A relief had feared that
algorithm may be limited by memory
bandwidth Ordered first batch of 30 nodes for
the decomposition cluster - Dell, dual 2.8 GHz
xeon, quad core, 8BG, 1U - Final system will
have 70 80 nodes
12
Signal Decomposition

• - The heart of gamma-ray tracking.
• Tracking depends on knowing the positions and
  energies of the Compton interactions
• Digital pulse processing of segment data
• Determine number, position, energy of multiple
  ?-ray interactions
• Uses data from both hit segments and image
  charges from neighbours
• Must allow for at least two interactions per hit
  segment
• Uses a set of calculated basis pulse shapes, on a
  predetermined grid
• Ideally suited to parallel processing
• Requires about 90 of CPU cycles used by
  GRETINA
• The major processing bottleneck, was greatest
  technological risk
• Baseline design specifies 20,000 gammas/s
• - allows only 4 ms/crystal/node for
  decomposition

13
Examples of calculated signals Sensitivity to
position
Hit segment
Signals colour-coded for position
Image charge
Image charge
14
Signal Decomposition Algorithm

• Current algorithm is a hybrid AGS SQP
• Pair-wise Adaptive Grid Search
• 200,000 two-par. least-squares fits per segment
  (for energies)? in 2 ms
• Non-linear Least-Squares (a.k.a. SQP)?
• Have also been developing Singular Value
  Decomposition
• Collaboration with Tech-X Corp., funded under DOE
  SBIR
• Developed SVD SQP hybrid algorithm
• - SVD on a coarse grid, with 250 eigenvalues
• - Localize interaction regions
• - Estimate number of interactions in each
  segment
• Approx. 25 faster, but slightly poorer fits
  further work planned
• Also demonstrated speed-up of SVD algorithm by
  factor 30 to 40 using Graphics Processing Units
  (GPUs) rather than CPUs.
• CPU time required AGS O(500n)?
• SQP O(n ?n2)? for n
  interactions
• SVD O(n)?

15
Status of Signal Decomposition Algorithm

• Three orders of magnitude improvement in CPU
  time now meets timing requirements
• Developed new optimized, irregular grid for the
  basis signals
• Developed method to accurately correct calculated
  signals for preamplifier response and for two
  types of cross talk
• Incorporated fitting of signal start time t0
• Much improved fits (?2 values)?
• Can handle any number of hit detector segments,
  each with up to two interactions
• Although some work remains to be done, we have
  demonstrated that the problem of signal
  decomposition for GRETINA is solved

16
Optimized Quasi-Cylindrical Grid

• Spacing arranged such that ?2 between neighbours
  is approximately uniform, i.e. inversely
  proportional to sensitivity
• Optimizes RAM usage and greatly simplifies
  programming of constraints etc.

17
Latest Decomposition Algorithm Data and fits

• Red Two typical multi-segment events measured
  in prototype triplet cluster
• - concatenated signals from 36 segments, 500ns
  time range
• Blue Fits from decomposition algorithm
  (linear combination of basis signals)?
• - includes differential cross talk from
  capacitive coupling between channels

18
Collimated Cs-source test
Pencil beam of 662 keV Distribution of
deduced interactions points throughout the
crystal, from decomposition plus tracking
algorithms
Position resolution sx 1.5 mm sy
1.7 mm
19
Scanning-table coincident-data test
Evaluated positions (red) Collimator position
(blue) Position resolution sx 1.2 mm
sy 0.9 mm Systemic error 1.1 mm 1.5 mm
respectively.
20
Siting

• Workshop in Oct 2007, organized by the GAC
• Plan for optimizing physics impact of GRETINA
  following completion in March 2011
• Participation and presentation by Prof. Shimoura,
  U. of Tokyo expressed interest in hosting
  GRETINA at RIKEN
• Report prepared by GAC, submitted to DOE
• Unanimous agreement on a plan for the first
  physics campaigns
• Commissioning runs at LBNL, starting March 2001,
  coupled to BGS
• Then rotate to other national laboratories, 6
  month campaigns
• Suggested sequence for the first cycle
• MSU - NSCL
• ORNL - HRIBF
• ANL - ATLAS
• We look forward to further discussions with our
  Japanese colleagues and are excited about the
  possibility of future collaborations.
• Unique opportunities at RIKEN Beams, auxiliary
  instruments, expertise

21
Summary

• Construction is proceeding, on schedule, on
  budget
• Received CD2B / 3B approval Oct 2007
• Signal decomposition problem is solved
• Scheduled completion date 2 March 2011 (Wed)?
• We have proposed a plan for the first round of
  physics campaigns
• Next stage Full 4? steradians
• GRETA received strong community support in
  LRP
• construction of GRETA should begin upon
  successful completion of GRETINA

22
Acknowledgements

• LBNL
• I-Yang Lee
• Overall project manager and source of all wisdom!
• A. Machiavelli, M. Cromaz, P. Fallon, M.
  Descovich, J. Pavan, many others
• Detectors, DAQ, in-beam data analysis,
  simulations, electric field calc'ns, etc.
• Sergio Zimmermann, John Joseph, Carl Lionberger,
  many others
• Electronics, engineering, computing, etc.
• GRETINA Advisory Committee and Working Group
  Leaders
• Endless proposal writing, review preparation,
  working group meetings, conference calls, ...
• John Anderson (ANL)?
• Trigger system
• Karin Lagergren (ORNL / UTK)?
• Signal calculation code in C, Optimized
  pseudo-cylindrical grid
• Tech-X Corp, especially Isidoros Doxas
• SVD development

23
Backup Slides
24
GRETINA Collaborating Institutions
Roles are defined by MOUs with LBNL

• Argonne National Laboratory
• Trigger system
• Calibration and online monitoring software
• Michigan State University
• Detector testing
• Oak Ridge National Laboratory
• Liquid nitrogen filling system
• Data processing / signal decomposition software
• Washington University
• Target chamber

25
In-Beam test at LBNL
3-crystal prototype 82Se 12C _at_ 385 MeV v/c
0.09 2055 keV (10?8) in 90Zr
Derived 3D average effective position resolution
s 2.1 mm
26
In-Beam test at MSU
36Ar at 86A MeV, v/c 0.4 Be target, 1mm
thick PIII in coincidence with S800 using
time-stamp technique Obtained effective position
resolution of 2.2 mm Doppler
correction Crystal 10 Segment
5 Tracking 2
2
E g (keV)
27
ORNL Installation

• LBNL railroad car design compatible with ORNL
  site
• No interference w/GRETINA at optimal RMS target
  location

ORNL site CAD integration models
Overview photograph of ORNL site
28
Cross-talk

• Differential cross talk arises from capacitive
  coupling between the inputs to the preamplifiers
  (some due to physical capacitance of the
  detector)
• Can be modelled well in SPICE, but needs to be
  carefully characterized in reality

29
Fitting to Extract Cross-Talk Parameters

• 36 superpulses averaged signals from many
  single-segment events (red)?
• Monte-Carlo simulations used to generate
  corresponding calculated signals (green)?
• 996 parameters fitted (integral and
  differential cross-talk, delays, rise times)
  (blue)?
• Calculated response can then be applied to
  decomposition basis signals

30
Why is it hard?

• Very large parameter space to search
• Average segment 6000 mm3, so for 1 mm
  position sensitivity
• ??? two interactions in one segment 2 x 106
  possible positions
• ?? two interactions in each of two segments
  4 x 1012 positions
• ?? two interactions in each of three segments
  8 x 1018 positions
• PLUS energy sharing, time-zero,
• Underconstrained fits, especially with gt 1
  interaction/segment
• For one segment, the signals provide only
  9 x 40 360 nontrivial numbers
• Strongly-varying, nonlinear sensitivity
• ??2/?(?z) is much larger near segment boundaries

31
Comparison Old Basis and Code vs. New
Distribution of deduced interactions points
throughout the crystal
Old
New
32
Signal Decomposition
GEANT simulations 1 MeV gamma into GRETA Most
hit crystals have one or two hit segments Most
hit segments have one or two interactions
33
Evolution of g-ray Detectors
The calculated resolving power is a measure of
the ability to observe faint emissions from rare
and exotic nuclear states. Taken from the NSAC
2002 Long Range Plan document, page 132.
34
Singular Value Decomposition

• Very roughly
• The full signal -vs.- grid position matrix can be
  decomposed into the product of three matrices,
  one of which contains the correlations
  (eigenvalues).

MxN MxN NxN NxN
M interaction sites

A UWVT
N voltages
35
Singular Value Decomposition

• Very roughly
• The full signal -vs.- grid position matrix can be
  decomposed into the product of three matrices,
  one of which contains the correlations
  (eigenvalues).
• By neglecting the small eigenvalues, the length
  of the signal vectors (and hence computation with
  them) can be greatly reduced.

MxN MxN NxN NxN
Mxn nxn nxN
M interaction sites
?

A UWVT
N voltages
36
Singular Value Decomposition

• Very roughly
• The full signal -vs.- grid position matrix can be
  decomposed into the product of three matrices,
  one of which contains the correlations
  (eigenvalues).
• By neglecting the small eigenvalues, the length
  of the signal vectors (and hence computation with
  them) can be greatly reduced.
• The more eigenvalues kept, the higher the quality
  of the fit.

MxN MxN NxN NxN
Mxn nxn nxN
M interaction sites
?

A UWVT
N voltages
37
1? ? 4? coverage, 28 ? 120 detectors
From GRETINA to GRETA

• Greater resolving power by factors of up to 100
• GRETA is the most requested instrument at the
  next generation RIB facility - RIA Facility
  Workshop, March 2004

? GRETA
? GRETINA
? Gammasphere
38
Physics Opportunities

• How do extreme proton-to-neutron asymmetries
  affect nuclear properties, such as shell
  structure and collectivity?
• What are the properties of nuclei at the limits
  of mass and charge?
• What are the properties of nuclei at the limits
  of angular momentum?
• Nuclear astrophysics, fundamental interactions
  and rare processes

39
GRETA in the 2007 NSAC Long Range Plan

• Gamma-Ray Tracking
• The construction of GRETA should begin upon
  successful completion of GRETINA. This gamma-ray
  energy tracking array will enable full
  exploitation of compelling science opportunities
  in nuclear structure, nuclear astrophysics, and
  weak interactions.

40
GRETA Cost and Schedule
Could start in FY08, complete in FY16
Program Starts

• As fast as allowed by detector production
  schedule.
• No gap between GRETINA and GRETA construction
• Total cost is M42
• Physics program to start 2011 with continued
  growth of capabilities.
• Match FRIB schedule GRETA will be ready when
  FRIB starts
• Competing European project 1/3 AGATA plan to be
  completed in 2012