Transcript DHCAL_GEM_niu

Digital HCAL using GEM
J. Yu*
Univ. of Texas at Arlington
Nov. 7 - 9, 2002
NIU/NICADD
•Introduction
•Digital Hadron Calorimeter Requirements
•GEM in the sensitive gap
•UTA GEM DHCAL Prototype Status
•Simulation Status
•Summary
(*on behalf of the UTA team; A. Brandt, K. De, S. Habib, V. Kaushik, J. Li, M. Sosebee, A. White)
Introduction
• LC physics topics
– Distinguish W from Z in two jet final states  Good jet mass resolution
– Higher Jet energy resolution;  E ~ 30% E
– Excellent jet angular resolution
Jet
Jet
• Energy flow algorithm is one of the solutions
– Replace charged track energy with momentum measured in the tracking
system
• Requires efficient removal of associated energy cluster
• Higher calorimeter granularity
– Use calorimeter only for neutral particle energies
– Best known method for jet energy resolution improvement
• Large number of readout channel will drive up the cost for
analogue style energy measurement  Digital HCAL
• Tracking calorimeter with high gain sensitive gap
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
2
DHCAL Requirements for EFA
• Small cell size for higher position resolution for good
multiple track shower separation
• High efficiency for MiPs in a cell for effective shower
particle counting and MiP tracking
• Possibility for Multiple thresholds
• Dense and compact design for quick shower development
to minimize confusion and resolution degradation
• Large tracking radius with optimized magnetic field for
sufficient separation between tracks for shower isolation
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
3
Goals for UTA DHCAL Development
• Develop digital hadron calorimetry for use with EFA
– Aim for cost effectiveness and high granularity
– Look for a good tracking device for the sensitive gap
•
•
•
•
•
•
Develop GEM cell(s) and prototype
Develop module/stack design for EFA optimization
Simulate GEM behavior in calorimeter
Implement GEM readout structure into simulation
Develop EF and calorimeter tracking algorithms
Cost effective, large scale GEM DHCAL
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
4
Why GEM?
• GEM developed by F. Sauli (CERN) for use as pre-amplification
stage for MSGC’s
• Allow flexible and geometrical design, using printed circuit
readout  Can be as fine a readout as GEM tracking chamber!!
• High gains, above 104,with spark probabilities per incident  less
than 10-10
• Fast response
– 40ns drift time for 3mm gap with ArCO2
• Relative low HV
– A few 100V per each GEM gap compared to 10-16kV for RPC
• Rather reasonable cost
– Foils are basically copper-clad kapton
– ~$400 for a specially prepared and framed 10cmx10cm foil
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
5
Large
amplification
140mm
70mm
CERN-open-2000-344, A. Sharma
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
6
GEM gains
High gain
Low voltage
differential!!
CERN GDD group
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
7
Double GEM
DHCAL Design
Embeded
onboard
readout
Ground to
avoid
cross-talk
Anode
pad
Ground
AMP
DISC
Thr.
AMP
REG
Nov. 7, 2002 Digital/serial
output
DISC
REG
Thr.
Preliminary readout design
Jae Yu: GEM Based DHCAL
8
Double GEM test chamber
•Sufficient space for foil
manipulation
•Readout feed-through,
retaining large space for
ease of connection
•Clear cover to allow easy
monitoring
•Readout pads connection
at the bottom
2cmx2cm pad design
J. Li, UTA
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
9
UTA GEM Test Chamber HV layout
2.1kV
Drift gap
Transfer gap
HV fed from one
supply but
individually
adjusted 
Good to prevent
HV damage on
the foils
Induction gap
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
10
UTA GEM
Prototype Status
• Readout circuit board
(2cmx2cm pads)
constructed
• HV Connection
implemented
• Two GEM foils in the
UTA Nano fabrication
facility cleanroom
• Preamp in hand and
characterization
completed (LeCroy
HQV800)
Nov. 7, 2002
Amplification
factor of 300
for GEM size
signal
(LeCroy
HQV800 )
Jae Yu: GEM Based DHCAL
11
Single GEM gain/discharge probability
A.Bressan et al, NIM A424, 321 (1998)
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
Simulation study in
progress using single
pions before multijet
events
•Determine Maximum
total charge deposit in
a cell of various sizes
and gains
•Study fake signal
from spiraling charged
particle in the gap
12
UTA Simulation Status
• Two masters students have been working on this project
– Mokka Geometry database downloaded and installed at UTA
– Preliminary mixture GEM geometry implemented
– Completed single pion studies using default geometry
• Reproduced expected response
• Energy resolution seems to be reasonable also
– Single pion study with mixture GEM begun
• Root macro and JAS based analysis packages developed
• Proceed with more detailed GEM geometry
implementation
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
13
Single Pion Studies w/ Default TESLA Geometry
• 1000 single pion events using Mokka particle gun
command.
– Incident energy range: 5 – 200GeV
– kinematics information on primary particles in the files
• Developed an analysis program to read total energies
deposited per pion for each incident energy.
– Mean Energy vs Incident pion energies
– Energy conversion from the slope of the straight line
– Conversion factor is 3.54% and agrees with the computed
sampling fraction
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
14
TESLA TDR
Geometry
Ecal – Electromagnetic Calorimeter
Material: W/G10/Si/G10 plates (in
yellow)
•1mm W absorber plates
•0.5 mm thick Si, embeded 2 G10
plates of 0.8 mm each
Hcal – Hadronic Calorimeter
Material:
•18 mm of Fe
•6.5 mm of Polystyrene
scintillator (in green)
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
15
TESLA TDR detector live
energy deposit for single pions
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
16
TESLA TDR Elive vs E
%
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
17
TESLA TDR CAL Single Pion
Resolution
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
18
GEM Simulation Status
• Mokka Geometry database downloaded and installed at UTA
• New Geometry driver written  Mixture GEM geometry
implemented Need to use ArCO2 only
• Single pion study begun for discharge probability
– Initial study shows that the number of electron, ion pair with gain of 104
will be on the order of 107 for single 200GeV pions
– Getting pretty close to the 108 from other studies  Might get worse for
jets from W pairs, due to fluctuation
– Need more studies to compute the discharge probability.
• Cell energy deposit being investigated to determine optimal
threshold based on cell energy  Proceed to energy resolution
studies
• Determine optimal gain using live energy deposit vs incident
energy
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
19
GEM Prototype Geometry
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
20
GEM Geometry Implementation
Mechanics in Mokka
TDR / Hcal02 Model chosen
for modification
Fe-GEM sub-detector
instead of the existing FeScintillator
New driver for the HCal02
sub-detector module
Local database connectivity
for HCal02  Database
downloaded and implemented
at UTA
Courtesy: Paulo deFrietas
Nov.Venkat,
7, 2002 TSAPS Meet Oct 10 - 12, 2002
Jae Yu: GEM Based DHCAL
21
Single Pion Cell Energy Deposit in
GEM HCal
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
22
Single pion Energy with GEM
50GeV 
ELive
15GeV 
EMeas
10.6MeV
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
23
GEM Sampling Weight
Live Energy Deposit (MeV)
300
Sampling: 2~4x10-3
250
200
150
Statistics too low to produce
Series1
reliable
gaussian fit  This
depends
heavily
on EM section
Linear
(Series1)
without proper GEM gain factor
taken into account.
100
50
0
0
10
20
30
40
50
60
Incident Pion Energy (GeV)
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
24
Summary
• Hardware prototype making significant progress
–
–
–
–
–
–
GEM foils delivered and are in the clean room for safe keeping
Preamp and Discriminator in hands  Preamp characterized
HV System implemented
Readout Pad implemented
Almost ready to put GEM foils in the prototype box
GEM foil mass production being looked into by 3MSimulation effort made a
marked progress
• Simulation effort made a marked progress
– Single pion study of Mokka default TESLA TDR geometry complete
• Analysis tools in place
• The resolution seems to be reasonable
– Preliminary GEM Mixture geometry implemented
• First results seems to be a bit confusing
– Initial estimate of e+Ion pair seems to be at about 107 for 200GeV pions
– Local Geometry database implemented
– Optimal threshold for digitization and gain factor will come soon
– Will soon move onto realistic events, WW, ZZ, or t`t  jets
Nov. 7, 2002
Jae Yu: GEM Based DHCAL
25