Status of the Tile HCAL prototype

1
Status of the Tile HCAL prototype

• MiniCal prototype operational in e beam _at_ DESY
• 3 types of photo-detectors tested (PM, SiPM,
  APD)
• preliminary analysis results
• ? beam calibration
• ? cosmic calibration
• ? energy scan results
• Preliminary data/MC comparison

• analysis of all data in progress
• large group effort
• HH-university, DESY, MEPHI, Prague, LPI, ITEP

2
Goal of MiniCal
First working prototype of Analogue HCAL Study
of energy resolution and shower shape Control
calibration and monitoring Compare with MC
prediction ? tune MC
Study various photo-detectors
against tuned MC Saturation effects in the range
0.5 7 GeV ? dynamic range ?
linearity
get detector hardware under control !!!
Get ready for studies on Physics Prototype
? see V. Korbel talk
3
The MiniCal Structure
e 1-6 GeV
0.1 cm Ø WLF
97 Shower contained

0.5 cm active
2 cm steel
4
The Cassette structure
3x3 tiles / layer
Tile size 5x5x0.5 cm3
e
1-loop fiber placed in groove (not glued) Single
tiles covered by 3M reflector
1 cell 3 tiles combined in depth (for PM/APD)
5
Test of 3 types of Photo-Detector

• MA-PM 16 channels
• best photo-detector
• cannot be operated in magnetic field
• cell read out 3 tiles connected to one channel

Only for reference

• Avalanche photo-diode (APD)
• different from those used by CERN experiments
• 3x3mm2 low capacity
• gain 500, quantum eff. 75
• cell read out 3 tiles

6
Cosmic calibration of PM
2 possibilities to calibrate MiniCal
- Cosmics run (vertical position) - Beam w/o
absorbers (horizontal)

• 4 calibrations over 2 months
• large shifts in calib. factors
• due to PM voltage change
• LED system used to monitor
• one channel of each PM

7
LED monitoring system
RUN 366
PIN diode
PIN diode stability 1
PM 2 550V
LED position corrected for ped. and PIN
diode fluctuations stability 1
PM 3 500V
time sec

• possibility to correct PM response to get
  stability of 1
• one LED fiber to one channel of each PM
• system available tested on PM
• very stable monitoring system crucial for APD
  operation

8
Calibration correction reproducibility
RUN 366 / RUN 378

• 10-15 shift observed on the 1 channel monitored
  per PM
• ?corrected on all PM channels

PIN diode
PM 2 550V
PM 3 500V
9
PM Calibration with beam

• Comparison of 4 beam calibration ?
• - individual tiles
• - cell of 3 tiles
• Good reproducibility (2)
• (?) Dependence on beam hit position on tile

• - 108 tileWLFPM
• ? 16 spread
• of which ? 7 spread from tileWLF system

10
PM Calibration comparison
Cell of 3 tiles
single tile
Ped 21.3 , s 0.9 MIP 37.7 ltRMSgt 8.4
/- 1.3
Ped 21.6 , s 1.1 MIP 81.1 ltRMSgt 7.7 /-
4.9
Ped 21.2 , s 1.0 MIP 36.6 ltRMSgt 12.8
/- 1.8
Ped 20.5 , s 1.5 MIP 76.3 ltRMSgt 10.5
/- 6.2
Analysis from Evgeni Devitsin, LPI
11
PM Calibration comparison

• average of cosmic calibration 1 shifted wrt e
  one
• calibration spread 6

12
SiPM Calibration
- Cosmic and beam calibration of all tiles w/o
pre-amplifier ? calibration analysis (MEPHI)
? reproducibility studies (LPI) - Single
photoelectron peak visible with fast
pre-amplifier ? for calibration only
One photoelectron peak
MIP peak
With low intensity LED
d
pedestal
Without LED light 1 phe noise visible
From MEPHI group ?see B. Dolgoshein talk
1 MIP 25 phe.
13
SiPM noise rate
Old meas. With 70ns gate 10Hz noise
14
SiPM noise rate
Fit to noise in all channels
Rate of noise _at_ ½ MIP

• 108 SiPM measured
• ltnoise/eventgt 410-3
• ?ltnoise rategt 24 Hz
• on each SiPM
• (24 kHz on total physics prototype)
• Establish requirements
• for pre-selection

Bad SiPM
½ MIP
15
Shower development with SiPM
Single layer readout 1 MIP peak visible in
first layer
Tile 2
Tile 1
Tile 4
Tile 3
Tile 6
Tile 5
Tile 8
Tile 7
97 shower contained for 3GeV beam
Tile 10
Tile 9
16
Energy Resolution for PM readout
Energy Sum

• different calibrations lead to same s(E)/E
• 5 systematic uncertainty

1 GeV
2 GeV
N MIP
N MIP
4 GeV
3 GeV
N MIP
N MIP
6 GeV
5 GeV
N MIP
N MIP
17
Energy Resolution for SiPM

• - saturation correction not applied
• 5 systematic uncertainty
• ? work needed on MC simulation

MC simulation of SiPM dynamic range Saturation
at N(ph.e) N(pixel)
Preliminary
18
Result Comparison

• good agreement PM/SiPM
• systematic uncertainty still to be
• calculated (fix 5)
• SiPM w/o saturation correction
• Fit function
• Fit values for PM / MC
• a 0.1 ? 0.2 / 0.4 ? 0.1
• b 21.0 ? 0.4 / 17. 1 ? 0.1

Preliminary
19
MC studies
Energy sum comparison GEANT4/data ? large
disagreement
GEANT4
data

• s(E)/E
• 10.46 ? 0.076

• s(E)/E
• 13.95 ? 0.077

mean 90 MeV
mean 110 MeV
Improve MC simulation including detector
properties ?
20
MC studies
Take into account technical features of
detectors SiPM poissonian smearing according
to of ph.e/MIP
?see R. Poeschl talk
21
Avalange Photo-Diode (APD)

• Characteristics of 40 APD tested
• _at_ CERN (CMS test room)
• quantum efficiency 77.5 (500nm)
• ? stable at 1 level
• capacitance 30pF at 400V
• gain vs voltage
• ? 50 gain variation at V 415V
• ? reduced with APD pre-selection
• ? no single APD voltage adjustment
• on board required

• required high voltage stability
• ? observed 1/GdG/dV 6 (1/V) _at_ gain 400
• required high temperature stability
• ? observed variation 1/GdG/dT 0.7 (1/oC) _at_
  gain 50
• ? can be measured corrected by LED monitoring
  system

22
APD Homogeneity test
U405V, M 200, T290C
U410V, M130 , T29.50C

• 12 APD tested
• ?homogeneity 6
• 1 APD used to
• read out 3-4 tiles

23
Pre-amp Boards (beam test)
Voltage sensitive preamp Prague Design
Charge sensitive preamp LAL/ECAL chip Larger
gain
APD
pedestal
s/MIP 0.21
s/MIP0.19
Pre-amp. gain required APD gain gt200 12
mV/7.2 fC APD gain lt100 12 mV/1.8 fC
Beam MIP
s/MIP0.39
s/MIP0.48
24
Pre-amp Boards (source test)
Charge sensitive Minsk preamp
Voltage sensitive preamp Prague Design
ped
Sr MIP
Sr MIP
ped
LED
LED

• - gate adjustment 90 signal contained
• 300 ns
  120 ns
• noise comparison s(ped)/(MIP-ped)
• 5.0/85 0.06
  12.4/82 0.15
• MIP resolution s(MIP)/(MIP-ped)
• 36.4/85 0.42
  42.3/82 0.52

25
Conclusion Outlook

• First, very promising results on SiPM
• single tile readout possible, calibration
  under control
• good agreement with PM energy resolution
  results
• detailed photo-detector simulation needed
• APD operation w/o single voltage adjustment
  possible
• promising results from APD preamp. boards
• Comparison of various APD readout boards
• Finish tests for all photo-detectors
• Finalize data analysis MC comparison
• ? LCIO format for data acquisition
• fast comparison to MC
• fast integration of other detectors
• fast exchange of data for detector
  comparison

? 2004 construction and commissioning of the
Physics Prototype