Wolfram KORTEN, JRA-02 AGATAEURONS PCC-Meeting, Mainz (Germany), 6.-7. April 2006 Status of the AGATA project Recent developments Milestones and deliverables.

EURONS PCC-Meeting, Mainz (Germany), 6.-7. April 2006Wolfram KORTEN, JRA-02 AGATA

Status of the AGATA project

• Recent developments• Milestones and deliverables for 2006• Plans for 2007/08

Main features of AGATA

Efficiency: 40% (M =1) 25% (M =30)today’s arrays ~10% (gain ~4) 5% (gain ~1000)

Peak/Total: 55% (M=1) 45%

(M=30)today ~55% 40%

Angular Resolution: ~1º FWHM (1 MeV, v/c=50%) ~ 6 keV !!!today ~40 keV

Rates: 3 MHz (M=1) 300 kHz (M

=30)today 1 MHz 20 kHz

• 180 large volume 36-fold segmented Ge crystals in 60 triple-clusters • Digital electronics and sophisticated Pulse Shape Analysis algorithms • Operation of Ge detectors in position sensitive mode -ray tracking


Ingredients of -ray Tracking

Pulse Shape Analysisto decompose

recorded waves

Highly segmented

HPGe detectors

·

·

··

Identified interaction

points(x,y,z,E,t)i

Reconstruction of “tracks”

e.g. by evaluation of permutations

of interaction points

Digital electronicsto record and

process segment signals

1

2 3

4

Spectroscopic studies


AGATA symmetric prototypes


Acceptance tests on prototypes

0,00

0,50

1,00

1,50

2,00

2,50

Reihe1

Reihe2

FWHM at 1.3MeV

FWHM at 60keV

Mean(1.3MeV)=1.99keV

Mean(60keV)=1.14keV

IKP Cologne 001 Core FWHM: at 1.3MeV : 2.10keVat 122keV : 1.20keV



0,00

0,50

1,00

1,50

2,00

2,50

Reihe1

Reihe2

FWHM at 1.3MeV

FWHM at 60keV


Mean(60keV)=1.07keV

Core FWHM: at 1.3MeV : 2.08keVat 122keV : 1.19keV

GSI 002



0,00

0,50

1,00

1,50

2,00

2,50

Reihe1

Reihe2

FWHM at 1.3MeV

FWHM at 60keV


Mean(60keV)=1.03keV

Core FWHM: at 1.3MeV : 2.13keVat 122keV : 1.10keV

INFN Padova 003


First AGATA triple module

Univ. Cologne August 2005

Goal: Validation of pulse-shape analysis codes under realistic experimental conditions


Planning for AGATA detectors in 2006

• Three symmetric detectors are being mounted in test cryostats for characterisation (3D scanning):

cryostat 1: scanning in Liverpool ongoing cryostat 2: delivery to Orsay in April cryostat 3: delivery to GSI in June• Two first asymmetric Ge crystals do not meet

specifications and were returned to Canberra-Eurisys for repair.

• First detector from “new batch” expected in April.• If problems have been solved by Canberra-Eurysis,

first asymmetric triple ready by the end of 2006


Electronics and DAQ developments

Up to 180detectors

Synchronous

Buffered

Analogue

Detectorpreamp.

Digitisers

TrackingControl,Storage…

EventBuilder

Clock100 MHzT-Stamp

PreprocessingPSACore +36 seg.

Ancillary 1.

Ancillary 2.

Handling of ancillary detectors1. interface to GTS via mezzanine 2. merge time-stamped data into event builder3. prompt local trigger available from digitisers

Ancillary 3.

GL Trigger


Digitiser module 36+1 channels, 100 MhZ, 14 bits

(Strasbourg - Daresbury – Liverpool)

• Mounted close to the Detector 5-10 m

• Power Dissipation around400W

• Water Cooling

Prototype Segment Board2 boxes per crystal


Pre-processing modules (E,T, hits, …) (Orsay – Daresbury)

1 ATCA crate for 2 clusters(6 Ge crystals, 222 channels)

ATCA standard :“full mesh” communicationwith Gbit Ethernet or PCIexpress switches

166 8 1012144

1,5U

8U

4,5U

21’’

11 9 7 5 313

14U

2115

1 2 3 4 5 6 7 8 9 10 11 12 13 1514 16

3 X 2 Carriers1 cluster

3 X 2 Carriers1 cluster

Segment preprocessing mezzanine for 6 channels


Global Trigger System (GTS) (Padova – Legnaro)

GTS Pre-Proc. mezzanineOne per detector

(also for ancillaries)Trigger processor layoutFor up to 12 detectors


Status of electronics and DAQ prototypes

• Digitisers Q1 2006• Pre-processing mezzanines Q1 2006• ATCA carrier card Q1 2006• GTS mezzanine 1 (Virtex2) Q2 2005• “ mezzanine 2 (Virtex4) Q2 2006

• All prototypes tested by Summer 2006• Full processing chain, including

prototype of DAQ tested by Autumn 2006• Any needed reprocessing Autumn 2006• Production and start delivery Q4 2006• Test of a triple-cluster Q1-Q2 2007

• Ready for Demonstrator by fall 2007


Pulse shape analysis developments

• Determine the number of interactions– Neural Net Algorithm (IPN Orsay)

• Recognizes ~86% of single hits• Close to 100% of double hits with distance > ~10 mm

Use specific PSA algorithm depending on number of interactions

• Several PSA algorithms under development:– Wavelet based grid search (GSI)– Matrix inversion algorithm (IPN Orsay)

• Only code to treat full crystal “at once”• Very good results for simple events (1-3mm, typical time 1ms)• Still problematic for multiple adjacent interactions

– Particle Swarm search (TU München)• So far only single hits per segment• Hit in single segment: ~300 s, mean distance 1.7 mm (front), 3.5 mm (back)• Hits in two segments: <3 ms, distance 3 – 4.6 mm

PSA has been solved for certain event classes,e.g. single and multiple hits in the detector, but not in the same segment

5 mm average position resolution for 600keV -rays

• Implementation and testing in “real” hardware is ongoing


Status of Pulse shape analysis

M. Schlarb et al. (TUM, 2006)

• Example of the mean distance (in mm) of reconstructed position from real position

• in front segment• simulated signals• Similar results from

other PSA codes


Procedure of Gamma tracking

~ 100 keV ~1 MeV ~ 10 MeV -ray energy

Isolated hits Angle/Energy Pattern of hits

Photoelectric Compton Scattering Pair Production

Probability of E1st = E– 2 mc2

interaction depth

E' E

1E

m0c2 1 cos


Algorithm development and optimisation

Simulation of interactionsof M photons in Agata(with Agata Geant4 code)

Simulation of PSA: packing, smearing,energy threshold

Tracking code

(ei,xi,yi,zi)

(e’k,e’k x’k,x’k y’k,y’k z’k,z’k)

(Ej,j,j) : incident photon energies and emission directions

(and scattering directions for polarisation measurements)

Data analysis


• Efficient forward tracking algorithms are ready:

• Effect of neutrons has been investigatedph reduced by ~1%/neutron

• Codes have been optimised for speed andspecifications for the demonstrator have been met :0.1 – 3.5 ms/evt for M=1 -30 (on 1 Opteron 1.7 GHz CPU)

• FT code currently being implemented into the DAQ

Status of gamma-ray tracking

E=1MeV, Mult=1 (30)

Mgt (Padova)

FT (Orsay)

Efficiency (%) 43 (28) 37 (24)

P/T (%) 58 (49) 67 (51)


Status of the AGATA project

• Recent developments– General discussion at the next AGATA week: Liverpool, June

6-9– Characterising the first prototype Ge detectors– Testing the first electronics and DAQ prototype boards

• Milestones and deliverables for 2006– Ge detector prototype characterized (06/06)

• o.k., but will use symmetric detector due to delay in delivery– Pulse-shape analysis algorithms optimised (06/06)

• Benchmark will be achieved, but will be ongoing– Gamma-ray tracking algorithms optimised (06/06)

• Benchmark will be achieved, but will be ongoing– Electronics and DAQ prototypes (06/06)

• Prototype local level processing electronics and specific data acquisition hardware: Delivery on schedule

• Ready for Demonstrator by fall 2007


Planning (EURONS and beyond)

• Demonstrator commissioning at LNL: second half of 2007

• First physics campaign at LNL in 2008further campaigns (from 2009) at GANIL, GSI, ILL, …

• LoI for construction phase signed in 2005to allow bids for new funds from 2006 (D, …)

• MoU for AGATA construction ready in 2007• Start construction in 2008 1 in possible in 2011• Support in FP7 as an independent RI (I3-TNA, CNI, …)

Wolfram KORTEN, JRA-02 AGATAEURONS PCC-Meeting, Mainz (Germany), 6.-7. April 2006 Status of the AGATA project Recent developments Milestones and deliverables.

Documents