Nuclear Emulsion Readout Techniques Developed for the CHORUS Experiment

1
Nuclear Emulsion Readout Techniques Developed for
the CHORUS Experiment
Ioannis M. Papadopoulos CERN EP-CHE division

• or
• ...how we managed in CHORUS to convert nuclear
  emulsion into an electronic detector

2
Emulsion highlights

• 3-d spatial resolution better than 1 ?m
• no dead time
• permanent storage device
• cheapest detector
• large discovery potential

3
Emulsion readoutuntil recently
W.H.Barkas, Nuclear research emulsions,1963 1973
4
CHORUS physics

• Search for ????? oscillation
• SBL appearance
• Clean signal

NOMAD
CHORUS
E531
CDHS
CCFR
gct 1 mm qkink 100 mrad
CHORUS (Phase II)
X
n?
?n?
n?
?-
Phys. Lett. B 424 (1998) 202-212 Phys. Lett. B
434 (1998) 205-213 hep-ex/9907015
qkink
m-
5
CHORUS detector
Nucl. Instr. Meth. A 401 (1997) 7-44
? -
T5
Calorimeter
h-

Muon spectrometer
770 kg emulsion target and scintillating fibre
tracker
Air core spectrometer
and emulsion tracker
Veto plane
6
Inside a vertex plate
n beam
Focal depth 3 mm Red frame 30x40 mm2
7
Automatic microscopes in Nagoya
http//flab.phys.nagoya-u.ac.jp/flab.html
8
Automatic scanning
140x160 ?m view, 120 Hz frame rate
track
microscope stroke
CCD camera
X50 magnification 3?m focal depth
350 ?m (175 ?m) emulsion sheet
90 ?m plastic backing
emulsion plate
350 ?m (175 ?m) emulsion sheet
9
Track Selector algorithm
track
track
grain finding, binarizing expanding
shifting
track
found track
summing
10
Track Selector implementation
During 1 camera cycle (8 ms) digitizes image
(FADC) finds pixels of grains (hardware FIR
filter) expands grains (3?3 expander)
stores pixels in FIFO buffers After 16 images
are taken tries out a number of shifts IN
PARALLEL (to cover angular acceptance)
for every shift tried performs
summing (FPGA technology) clustering
on the sum image (Pentium) finds
tracks stage has moved already to next
position
Current performance 3 Hz (for all tracks with
?z lt 400 mrad)
T.Nakano, Ph.D. Thesis, Nagoya Univ., 1997
11
CERN/NIKHEF microscopes
Larger stage (MICOS) to handle the
whole emulsion sheet Larger digital
camera (Thomson Mpixel CCD) to analyze
larger emulsion volume per view using DSP
cards (Blue Wave) based on the TI-C80 and TI-C620
processors for flexible data treatment Advanced
optics from Jenoptik ? FOV0.5
mm DOF1.2?m shutter up to 120 Hz
350 x 350 ?m2/view
http//choruswww.cern.ch/CERN-Microscope
12
Microscope layout
Data (LVDS)
FastEthernet
I/Odriver
Interfacecard
Network
controls (RS422)
DSP C80, C620 (Image filtering)
CCDMpixel camera
Matrox Pulsar (Emulsion display, graphical tools)
Slow Control Temp.,Hum.,Oil
Matrox Mil. (primary display)
Objective, shutter, light source
DAC/ADCIO
Clustering, tracking, controlling stage and
light (2?Pentium)
RS232
Motor control
RS232
COM1
PCI
COM2
10 m
Clean room
Operator room
13
Data flow
2 ? Pentium II (C)
2?C620 (Assembly)
prediction tracking
IIR filter
Clustering (grain finding)
128 kb pixels
PCI
16-25?30kb grain data
JPEG compression
C80 (C, Assembly)
acquisition control
20-250?30kb grain data
250?50kb image data
tracks grains
1 Mb raw data
FastEthernet card
input _at_30 Hz
Ethernet
Objectivity DB
tracks
vertices
kinks
grains (emulsion hits)
...
14
We want to find many of such events fully
automatically
...and analyze the data pretending they were
taken using an electronic tracker with a 3-D
spatial resolution of 1 µm.
Phys. Lett. B 435 (1998) 458-464