October 18 th, 2007 LNGSC Yves Déclais 1 OPERA/CNGS1 O scillation P roject with E mulsion-t R...
-
Upload
peter-blair -
Category
Documents
-
view
214 -
download
0
Transcript of October 18 th, 2007 LNGSC Yves Déclais 1 OPERA/CNGS1 O scillation P roject with E mulsion-t R...
October 18th, 2007 LNGSC Yves Déclais 1
OPERA/CNGS1Oscillation Project with Emulsion-tRacking Apparatus
1. Status of the installation of the electronic detectors
2. Results of the commissioning of the electronic detectors & DAQ
3. Brick production and detector filling: status
4. CNGS first Physics run5. Event selection & Brick finding6. Brick processing & Emulsion
analysis7. Looking at CNGS events in
OPERA bricks8. Conclusions
Outline:
October 18th, 2007 LNGSC Yves Déclais 3
Electronic detectors installation & commissioning completed
SM1
Veto
BMS
Target tracker
SM2
Spectrometer:XPC, HPT, RPC, magnet
October 18th, 2007 LNGSC Yves Déclais 4
Completion of HPT installation
-March 2007: HPT 9&11
-May 2007: end of mass production
-June 2007: HPT 10&12• gas system SM2 completed July 07• cabling SM2 completed June 07• all TDC inside the main DAQ• all cables calibrated
Main remaining problem:Pickup noise from magnet coil for HPT inside the magnet:• filter the data onto the pulse width basis• filter the supply of the front-end board with ferrite
October 18th, 2007 LNGSC Yves Déclais 5
VETO installation
The VETO consists of 2 planes of 32 Glass Resistive Plate Chambers. Each plane is read-out by 2 planes of orthogonal copper strips (X,Y).
October 18th, 2007 LNGSC Yves Déclais 6
VETO commissioning
Noise map:• All the GRPCs are working, but the pick-up has to be optimized• A detailed efficiency study using cosmic & beam events is underway
Flow rate monitor:• important for humidity control• working continuously since 210 days
Current monitor:
Water vapor
gasflow
October 18th, 2007 LNGSC Yves Déclais 7
DAQ computing architecture
Eventbuilder
Root •Alignment•Data analysis•Brick finding
Gateway
DB
CNGS DB(Oracle)
BMM
TTdaq RPCdaq HPTdaq•Majority trigger•Same DAQ forRPC-XPC-VETO•54 sensors
•X-Y coincidence•L-R coincidence•N planes coinc?•992 sensors
•Trigger cross-check?•Tests procedure•105 sensors
… TTsensors … RPCsensors … HPTsensors
ASCII
• Time coincidence between subdetectors (merge sort)• Event classification wrt the beam UTC• Event header production
Raw data DB•postGres (DAQ)•Oracle
Event DB
(Oracle)
DAQmanager
Beamsyncronizer
•Decodes CNGSearly warning•Flag the DAQ cyclewrt the beam
CNGS Early Warning
•Connects theDAQs•Send flaggedcycles•Writes GPSdates
ASCII ASCII
ASCII-2-Root
October 18th, 2007 LNGSC Yves Déclais 8
Currently in Hall B:
• DAQ servers
• Gateway server
• DAQ-MMS Database
Computer room inside the HallB Control room
DAQ DB: migration to Oracle completed
DAQ live time > 95%During the CNGS run
October 18th, 2007 LNGSC Yves Déclais 9
DAQ: event building concept
Event building :by sorting in time individual hit from sub-detectors:• for each sub-detector in the dedicated sub-detector DAQ machine + filtering algorithms • between sub-detectors in the event builder machine + filtering algorithms
HPT
RPC
TT
Δt between consecutive hits10 nsec binning
100 nsec 600 nsec
Trigger for sub-detectors :• free running mode at the hardware threshold for TT, RPC & Veto• Common stop formed by using RPC signals (dedicated electronics) for the HPT
October 18th, 2007 LNGSC Yves Déclais 10
Time stamp resolution
Without fiber transmission correction
With position correctionSigma : 9 nsec
down going muon : side view
Tim
e
diff
ere
nce
Str
ip
nu
mb
er
October 18th, 2007 LNGSC Yves Déclais 11
Up-going muon selection using the time stamp
2006 data:Cosmics and beam data
Only cosmics data:• 44 days of data• expected 2• observed 1
Beam events
Down goingUp going
Up going muonfrom atmospheric ν
v/c
v/c
October 18th, 2007 LNGSC Yves Déclais 12
2. Results from the commissioning of the electronic detectors & DAQ
October 18th, 2007 LNGSC Yves Déclais 13
VETO commissioning
CNGS rock muon
Cosmic muon
Veto 1 resolution Veto 2 resolution Y Veto 1 vs Y TT Y Veto 2 vs Y TT
~2 cm ~4 cm
October 18th, 2007 LNGSC Yves Déclais 14
Target Tracker efficiency
using cosmic ray tracks
Trigger rate:• 1 p.e. 20 cps per pixel• 2 p.e. 10 cps per pixel
1 p.e.
2 p.e.
3 p.e.
99%
October 18th, 2007 LNGSC Yves Déclais 15
Mgnetic field influence on TT response 10<B<40<G
auss inside the TT
gain ratio Npe ratio
rock
sid
e
corr
idor
sid
e
1.15
0.85
1.00
R=
ON
/OFF
ch #
October 18th, 2007 LNGSC Yves Déclais 16
RPC efficiencies using cosmic ray tracks
All problems have been cured
Inefficiencies mostly located in first SM and due to HV/gas distribution failures:
maintenance scheduled at the end of brick insertion on SM1 to avoid interference with BMS
RPC XPC
SM1 SM2 SM1 SM2
October 18th, 2007 LNGSC Yves Déclais 17
RPC spatial resolution
Tracking resolution ~ 1.3 cm,
Misalignments < 1 cm
using cosmic ray tracks
Cluster size
October 18th, 2007 LNGSC Yves Déclais 18
RPC/XPC timing system for HPT
σRPC = 3.5 nsec
σXPC = 5.3 nsec
Efficiency vs time
XPC1XPC2RPC1
RPC11RPC12RPC13
RPC20RPC21RPC22
October 18th, 2007 LNGSC Yves Déclais 19
Status of reconstruction for HPT
- Reconstruction running stable for data and MC events
- Timing corrections implemented
- Alignment ongoing, starting from TDA measurements
- Resolution of ~500m in single planes for cosmic data (see plots below)
- Alignment will bring improvements (300m prototype measurements)
- 80% track efficiency for cosmic data Trigger efficiency vs track angle
October 18th, 2007 LNGSC Yves Déclais 21
Brick components
Emulsions: delivered in GS material for 154000 bricks (including CSd) : 75% no room for finding new funds (Funding Agency Meeting, March
23, 2007)Lead:• present production rate is 32000 pcs/day (~ 2.5 drum/day). • the production rate has been increased following the stability of brick production (BAM) and detector filling (BMS) rate.• Lead mechanical quality is now good and stable. Air flow is still needed to secure single sheet pick up in BAM piling stations.• Lead radioactivity
0
2
4
6
8
10
12
14
16
18
March-2006average
cts/d/cm2
July - 2006average
cts/d/cm2
September-2006 average
cts/d/cm2
November-2006average
cts/d/cm2
January-2007average
cts/d/cm2
total 4,5 12 14,3 14,3 17,9
5-5.5 MeV 1,3 1,6 1,6 1 2,3
March-2006 average cts/d/cm2
July - 2006 average cts/d/cm2
September-2006 average cts/d/cm2
November-2006 average cts/d/cm2
January-2007 average cts/d/cm2
+ no surface migration observed+ family equilibrium restored following the expected T1/2 ~4.6 months
October 18th, 2007 LNGSC Yves Déclais 22
BAM
Achieved milestones:• 10 Kbricks by last SPSC• 2 drums/day in May 20 Kbricks by end of may• 2.5 drums/day in June• 3 drums/day from oct 1st
• all technical aspects under control• but resource dependent !
Drum loader(1 drum = 234 bricks)
BAM in Dark5 Piling/Pressing stations1 Wrapping station
Lead Loading
Running with 2 shifts/day
October 18th, 2007 LNGSC Yves Déclais 23
Brick production summary
produced bricks
0
10000
20000
30000
40000
50000
60000
70000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
weeks
bri
cks
produced bricks
0
500
1000
1500
2000
2500
3000
3500
1 4 7 10 13 16 19 22 25 28 31 34 37 40
2007 (weeks)
bric
ks
weekly production
~60000 bricks inside the detector
October 18th, 2007 LNGSC Yves Déclais 24
Completion of the Detector filling
0
20000
40000
60000
80000
100000
120000
140000
160000
180000
1 8 15 22 29 36 43 50 57 64 71 78 85 92
weeks
bri
cks
2 drums/day
4 drums/day
2.5 drums/day
3 drums/day
real
End of April 2008 June 10th July 25th
October 24th
2008 SPS schedule for physics:May 10th November 10th
October 18th, 2007 LNGSC Yves Déclais 25
Brick Manipulator System
Loading station and brick drum Draw bridge for
brick insertion
Storage Carrousel
Digital camera for positioning
Vacuum sucker vehicle to retrieve bricks
Electric jack to insert bricks
Portico
Manipulating platform
Spatial precision o(0.1 mm) on ranges o(10 m) using digital camera vision
October 18th, 2007 LNGSC Yves Déclais 26
BMS : Insertion performances
Verified performance (without failures or waiting time in drum exchange): insertion of 500 bricks in one shift
Running staff: 1 expert and
1 operator per shift
Daily rate from 21/10/06 to 21/06/07
- Max daily rate 815 -
Running with 2 shifts per day
October 18th, 2007 LNGSC Yves Déclais 28
CNGS run in 2007
Bricks at the start of the run
Bricks at the end of the run
Integrated p.o.t Bricks with interactions
58749505 tons
71687616 tons
0.43 1019 pot 180 (~6/day)10 charm decay
Hypothesis:• 3 weeks of CNGS commissioning: 38 (half intensity, 1 CNGS cycle) , 39, 41• 3 additional weeks of physics run : 40, 42, 43• super cycle with 3 CNGS : 39.6 sec• intensity as in the 2006 run : 1.7 1013 pot/extraction (70% of nominal high intensity)• 70% efficiency of the machine complex
• 2006 CNGS run : ~ .1 1019 pot• nominal CNGS year : 4.5 1019 pot
OPERA request to the SPSCJune 2007
October 18th, 2007 LNGSC Yves Déclais 29
SPS & CNGS commissioning
Repairing completed week 38• Horn & reflector cooling system refurbishedCommissioning:• week 39: low intensity CNGS alignment• week 40: intensity increase up to 3.6 1013 pot/cycle
fix for strip lines SPS commissioning for 3 CNGS cycles
• Friday, October 5th, at 5:30 pm start physics beam
Beam very stableBeam stopped on October 13th for alarm on the control system for the ventilation work going for recabling a new control system, beam should be back not later Friday October 19th
CNGS/OPERAAllocated timeResearch Board
July 2007
October 18th, 2007 LNGSC Yves Déclais 30
CNGS performance
MoTue Fri 5/10
Commissioning Physics
Extraction 1
Integrated flux:6.72 1017 pot
Extraction 1
1.94 1013 pot
Fri 5/10, 18:00 Tue 9/10, 18:00
Nominal high intensity2.4 1013 pot
October 18th, 2007 LNGSC Yves Déclais 31
CNGS event selection by using the GPS Timing Information
Searching events in O(ms) windows just yields a narrow peak of the order of the spill width (10.5 us) with practically no CR background: O(10-4) CR background
TSPS TOPERATOF=2.44 ms
TOPERA – (TSPS + TOF) < Tgate
GPS time stamp intercalibration ~100 ns Tgate slightly larger than the extraction length (10.5 s)
+- 52 ms
Spill length
Beam time structure shown with an enlarged windows around the first extraction
October 18th, 2007 LNGSC Yves Déclais 32
Absolute Time Comparison
New calibration campaign 2007:
July 2007 with a portable atomic Cs clock capable of providing a 1PPS signal synchronized with UTC and stable at the level of few ns over several days Measure the timing signals at various points of the chain at CERN and LNGS compared to the 1PPS of the Cs clock absolute calibration transported in all the single points, check the calibrations performed last year with independent method Precise timing: implement a fast digitizer recording the proton current at the level of the BCT
2007 data:• 393 CNGS events (all categories)• using the same calibrations as last year
Cross check ongoing @CERN
Status for this run:same hardware than in 2006 (Kicker synchro)
Latest news:CERN will provide a new GPS receiver, an atomic Cs clock and the hardware/software for tracking the LNGS GPS system with a resolution of 1nsecto be installed at LNGS
October 18th, 2007 LNGSC Yves Déclais 34
CNGS event selection
Data processing:• data are extracted from the DAQ data base every 12 hours• all events are processed through the OPREC reconstruction package• In time events are selected and stored on a dedicated file
Brick finding:• all in time events are scanned visually and sorted by categories:
in target event CC or NC like (according to the muon id) magnet interaction rock muon
• in target events are analyzed through the brick finding package each step of the reconstruction for the muon is checked manually iterative process including;
neural network for the wall finding hadronic shower parametrization probability estimate for the vertex location in each brick exact configuration and alignment of the target at the time of the event (BMM data base)
at the end the list of bricks is provided for each event
Accumulated statistics for 6.72 1017 pot • 24 CC like events• 8 NC like event• 3 events in the scintillator
October 18th, 2007 LNGSC Yves Déclais 35
CNGS events outside the target
Event vertex location defined as the first hit of the 3D track
Fit: 3.4 ± 0.2 degree
Muon candidate energy spectrum
Vertical beam angle
Data = 331MC = 303.5
Magnet 1 Magnet 2
October 18th, 2007 LNGSC Yves Déclais 36
CC like event « Opera baby » Pμ : 7.87 GeV/c
Xtalk in MAPMT:• ~60% of the nb of hits• ~3% of the nb of PE
October 18th, 2007 LNGSC Yves Déclais 37
Hadronic shower parametrization 11451 p.e. tot
1579 p.e. muon
Wall 5 Tray 24 Cell 6 prob= 0.9Wall 6 Tray 24 Cell 6 prob= 0.09
Fitted muon trackPredicted brick
October 18th, 2007 LNGSC Yves Déclais 38
NC like event
Pwall >93% Bricks:wall 3 tray 28 cell 7 prob= 0.67wall 3 tray 28 cell 6 prob= 0.13wall 3 tray 27 cell7 prob= 0.08wall 3 tray 29 cell 7 prob= 0.05
October 18th, 2007 LNGSC Yves Déclais 41
X ray marks
8 cm
7,5
cm
1 shot Xray
12
3 4
Brick films alignment
• 6 lines for alignment (crossing point)• 1 line (inclined) for plate numbering
X-ray machine
CS-doublet
brick
CSd alignment & CSd brick connection
October 18th, 2007 LNGSC Yves Déclais 42
Chemical plant
11
22
33
44
5566
Almost readyAlmost ready
• Main control unit powered, working
• Demi water ready, 20° C ok
• Filtered water available, 20° C ok
• Collection tanks secured
• A nice chemical Lab set aside (small operations, & quality check! : thanks to LNGS Chemical Service)
• Big preparation tanks
11
22
33
44
55
66
October 18th, 2007 LNGSC Yves Déclais 43
Brick processing lab
Progress report:Progress report:
• Climatization resumed, ok!
• Safelight darkroom ok
• 5 proc. lines cabled to front-end (6th suspended)
• Lines # 1,2,3 tested, ready and operational in automated mode
Lines # 3,4,5 to follow soon; some refinement suggested by tests
October 18th, 2007 LNGSC Yves Déclais 44
European Scanning System
• R&D on 20 cm2/h European Scanning System completed successfully (end 2004)
• 30 ESS distributed among 11 labs ( 22 ESS in 8 Italian labs ): Bari, Bern, Bologna, Lngs, Lyon, Napoli, Salerno + Neuchatel, Roma, LNF, Padova
• Integrated in a online computers network :including DB, manager , on line computation , plate changer …..
October 18th, 2007 LNGSC Yves Déclais 45
Scanning in Japan
The 1st S-UTS (20cm2/h)
SUTS status:• first unit ready : 20 cm²/h• R&D in progress in order to reach 60 cm²/h
X50 lensOptimized FPGA code
• 5 units in preparation 300 cm²/h
CSD scanning
Brick scanning
Vertex finding performances:• Single track scan back :• Vertex finding:
• νμCC : 15’ for vertex finding nx10’ for vertex studies• νμNC : 2-3 x CC vertex finding time
October 18th, 2007 LNGSC Yves Déclais 46
Automated Track Reconstruction and Vertex Search
position residuals between base-tracks and fitted volume tracks
angular residuals between base-tracks and fitted volume tracks
(vertical tracks)
scan-back efficiency 93.1%
volume scan
scan-backtracks followed backward
film by film
track search in the CS doubletFollowing the electronic detectors predictions
vertex reconstruction
October 18th, 2007 LNGSC Yves Déclais 47
CSd & Brick : track location rehearsal
Using nearly horizontal cosmic muons:
Transverse position difference
(found – pred)
Longitudinal position difference
(found – pred)
Transverse slope difference
(found – pred)
Longitudinal slope difference
(found – pred)
Difference wrt TT prediction for tracks located in the CSd
σ= 5 μm σ= 10 μm
σ= 3 mrad σ= 11 mrad
Track following inside the brick
October 18th, 2007 LNGSC Yves Déclais 49
Looking at CNGS events First located vertex
Interesting di-muon event
CS film development Wed., Oct. 10th
Brick film development Thurs., Oct. 11th
CS scanning Fri., Oct. 12th
Brick transportation LNGS BARI Sat., Oct. 13th, 1.30 p.m.CS scanning predictions LNGS BARI Sat., Oct. 13th , 2.52 p.m.SB started Sun., Oct. 14th, 11.50 a.m.Track disappearance Sun., Oct. 14th, 5.25 p.m.Quasi-online reconstruction Sun., Oct. 14th, 5.42 p.m. and event confirmation
October 18th, 2007 LNGSC Yves Déclais 50
Muon track following
4 μm 2 mrad
Mean value: 22.5RMS : 3.3
Grain density
Fill factor:• 82% for base tracks• 100% including μ-tracks
Difference wrt the CSd prediction x=27 μm, y=6 μm
muon candidate followed back from film 57 to 30track disappearance in film 29stop confirmed in 4 consecutive films
(last measured film 26)
October 18th, 2007 LNGSC Yves Déclais 51
Vertex parameters
VTX ID 17Downstream Tracks 8Upstream Tracks 0Avg. Track Distance 6.0 micronX 100932.9Y 21913.3Z -36190.8
in lead between films 29 and 30
depth ~ 40 micron
Could be a Charm decay candidate !
October 18th, 2007 LNGSC Yves Déclais 52
Tracks association wrt CSD tracks
track 689film 44: 0.0324 0.1715
compatible with track found in CS1-CS2 ?
CS2: 0.0270 0.2027
SB track
found also in CS1-CS2!
CS2: -0.0075 -0.0804
film 44: -0.0099 -0.0944
track 71
October 18th, 2007 LNGSC Yves Déclais 53
Lookig at CNGS events
• CSd prediction received @Naples on Monday morning (high fog CSd)• track found in the brick : Δx = 56 μm , Δy = 25 μm (wrt CSd predictions)
Δθx= 9 mrad, Δθy= 10 mrad• track followed until plate #37• fill factor with base-tracks : 100%
2nd located vertexNice quasi-elastic event
October 18th, 2007 LNGSC Yves Déclais 54
Vertex found: Quasi elastic event
Nuclear fragment
Nuclear fragment
Nuclear fragment
Muon track
October 18th, 2007 LNGSC Yves Déclais 55
Looking at CNGS event 3rd located vertex @Nagoya
• films shipped on October 12th, delivered on October 16th @Nagoya• CSd scanning on october 16th (2h per CC like event, 12h per NC like)
event 179264151 CSd track: difference wrt TT prediction Δx= 17.3 mm , Δy= -2.1 mm scan-back on October 17th : first vertex found and confirmed
Event : 179264151
October 18th, 2007 LNGSC Yves Déclais 57
8. Conclusions
The concept of the OPERA detector is now fully validatedfrom the electronic detectors emulsion analysis
The τ hunting is starting from now on and for ~5 years !
The brick production is launched and the detector could be filled before the physics run in 2008Caveat: the schedule for the detector filling is tight
and the resources limited any problem will now induce delays impossible to
recover
The OPERA collaboration is grateful towards the staff members & the management of the LNGS for their constant support