TPC/HBD R&D C.Woody, N.Smirnov, B.Azmoun, M.Sivertz, B.Yu, R.Majka, J. Mitchell, V.Rykov, M.Purschke, C.- Y.Chi PHENIX Collaboration Meeting Nashville, TN June 11, 2003
TPC/HBD R&D
C.Woody, N.Smirnov, B.Azmoun, M.Sivertz, B.Yu, R.Majka, J. Mitchell, V.Rykov, M.Purschke, C.-Y.Chi
PHENIX Collaboration MeetingNashville, TN
June 11, 2003
C.Woody, Nashville Collaboration Meeting, 6/11/03 2
List of Topics
GEM measurements with Ar/CO2 and CF4
Status of new GEM foils Gas transmission measurements TPC field cage design Tracking Issues with TPC and VTX Electronics for the HBD and TPC R&D plan for next year
C.Woody, Nashville Collaboration Meeting, 6/11/03 3
GEM configuration
Initial tests with GEMs using CF4 did not produce usable signals
Problem was with electric field in transfer gaps
Smaller gap spacing and higher fields (> 2 kV/cm) give much higher gains
C.Woody, Nashville Collaboration Meeting, 6/11/03 4
GEM Measurements with Ar/CO2
Small-Triple GEM Gain in Ar-CO2 deltaV(Foil) = deltaV(TG)
y = 2E-08e0.0704x
y = 2E-08e0.0763x
100
1000
10000
100000
315 335 355 375 395
deltaV(Foil) [V]
Gain
Small-Triple GEM Gain in Ar-CO2 deltaV(TG) = 360V, invariant
y = 1E-07e0.0709x
1000
10000
100000
320 330 340 350 360 370 380 390
deltaV(Foil) [V]
Gain
Small-Triple GEM Gain in Ar-CO2 deltaV(Foil) = 360V, invariant
0
5000
10000
15000
20000
25000
30000
0 100 200 300 400 500 600
deltaV(TG1+TG2+1/2TG3) [V]
Gain
BNL
WIS
Resistive divider Fix VTG
Vary VF
Fix VF
Vary VTG
• We seem to get higher gain at the same voltage than WIS
• At fixed VTG the gain curve is nicely exponential as expected up to ~ 105 where discharging starts to occur
• Varying VTG at fixed VF has a much weaker effect on gain, but shows no plateau up to the maximum V
C.Woody, Nashville Collaboration Meeting, 6/11/03 5
GEM Measurements with CF4
Small-Triple GEM Gain in CF4deltaV(Foil) = 525 V, invariant
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
200 250 300 350 400 450 500 550 600 650
deltaV(TG1 + TG2 + 1/2IG)/3 [V]
Ga
in
Small-Triple GEM Gain in CF4deltaV(TG)=500V, invarint
y = 4E-07e0.0449x
100
1000
10000
100000
1000000
450 475 500 525 550 575 600 625 650
deltaV(Foil) [V]
Ga
in
Small-Triple GEM Gain in CF4deltaV(Foil) = deltaV(Trans. Gap)
y = 2E-07e0.0467x
y = 3E-09e0.0534x
100
1000
10000
100000
1000000
450 470 490 510 530 550 570 590
deltaV(Foil) [V]
Gai
n
Resistive divider
Fix VTG
Vary VF
Fix VF
Vary VTG
• We again seem to get higher gain at the same voltage than WIS
• Both BNL and WIS see saturation in CF4 at gain ~ 105 (before breakdown point)
• VTG again has a much weaker effect on gain and shows no plateau
BNL
WIS
C.Woody, Nashville Collaboration Meeting, 6/11/03 6
New GEM Foils from 3M
Expecting delivery of 50 new 10x10 cm foils at BNL in ~ 2 weeks
Roll-to-roll process Limited to 12” width
3M Microinterconnect Systems Division, Austin, TXIn collaboration with Univ. Chicago (J.Collar)
hep-ex / 0304013 (April 2003)
C.Woody, Nashville Collaboration Meeting, 6/11/03 7
Testing Small 3M Foils
3 stage GEM with three 1” dia. 3M foils
80 mm (55 mm) holes spaced in a hexagonal pattern with 140 mm pitch
Visually looks to be excellent quality
Will compare with CERN foils in Ar/CO2 & CF4
C.Woody, Nashville Collaboration Meeting, 6/11/03 8
VUV Transparency of Potential Radiator Gases
VUV transparency is extremely sensitive to O2 and H2O content
80
85
90
95
100
105
110
1150 1250 1350 1450 1550 1650 1750
Wavelength [Angstroms]Tra
nsm
itta
nc
e [
%]
Ar, [O2]=5.6ppm,[H2O]=2.3ppm
CF4, [O2]=0.6ppm,[H2O]=19.0ppm
Tranmittance of Argon and CF4
Gas system being developed andtested in collaboration with LEGS
B.Azmoun
C.Woody, Nashville Collaboration Meeting, 6/11/03 9
Low O2 and H2O levels are critical for good VUV transparency
Absorbance: Ar
0
0.1
0.2
0.3
0.4
0.5
0.6
1100 1200 1300 1400 1500 1600 1700 1800
Wavelength [Angstroms]
Ab
so
rba
nc
e
Static, 0 hr.
Static, 1 hr.
Flow, 0 hr., gauge valveopen
Flow, 20 min., gaugevalve closed
Flow, 1 hr.
Flow, 24 hrs.
Flow, 65 hrs.
Flow, 65 hrs., inc. rate
Flow, 65 hrs., flow meterremoved
Flow, 70hrs., f-mremoved
Static, following flow
(Simple gas flow system)
Absorbance of Ar ( P~700 Torr)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1100 1200 1300 1400 1500 1600 1700 1800
Wavelength [Angstroms]
Ab
so
rba
nc
e
Static (Brfore Flowing), Wait= 0min.
Flow, 0 Hr.
Flow, 1 Hr.
Flow, 2 Hr.
Flow, 3 Hrs.
Static (After Flowing), Wait = 0min.
Older Measurement: No Flowsystem: Static, Wait = 0 min.
Flow Rate: ~100 cc/min
Ref.: Zaidel’ & Shreider, VUV Spectroscopy
Water Vapor. Oxygen.
C.Woody, Nashville Collaboration Meeting, 6/11/03 10
Absolute Quantum Efficiency of CsI photocathodes
0.1
1
10
100
1140 1200 1260 1320 1380 1440 1500 1560 1620 1680 1740 1800
Wavelength [Angstroms]Q
ua
ntu
m E
ffic
ien
cy
(%)
Smpl. 1: Al, 0.54 microns
Smpl. 2
Smpl. 1: Cu+Ni+Au, 0.66 microns
Smpl. 2
CsI PMT
Comparison of our CsI photocathodes with a calibrated CsI PMT
Good quality CsI photocathodesare now being made at Stony Brook
Stack with Au coated GEM foil for depositing
CsI photocathode
VUV Spectrometer
C.Woody, Nashville Collaboration Meeting, 6/11/03 11
TPC Field Cage Design
Question: How uniform can we make the drift field and how large are the electric field distortions near the edges and
in the vicinity of the HBD ?
Autocad geometry MAXWELLNot so easy….
Also try 2D calculation using GARFIELD
N.Smirnov
C.Woody, Nashville Collaboration Meeting, 6/11/03 12
Drift Field Distortions
N.SmirnovGARFIELD MAXWELL
EZ / <EZ> ER / <ER>
R
Drift field distortions are generally < 1%
C.Woody, Nashville Collaboration Meeting, 6/11/03 13
Electric Field Distortions Near the HBD
30 cm
mini TPC field cage wires
3 mm 5 cm
-36 kV 0 V
0 V
+ 75 V
wire plane with 5 mm step
N.Smirnov
Initial indications are that the electric field distortions in the region of the
HBD are small, but we need to do more calculations
ED, V/cm
ER, V/cm
C.Woody, Nashville Collaboration Meeting, 6/11/03 14
Tracking Issues
Need to begin serious simulation work on pattern recognition, track finding
and track fitting using tracking information from all PHENIX detectors
(present and future - VTX and TPC)
V. Rykov
++ Configuration B = 9 KG
Momentum resolution for various combinations of the VTX and TPC
TPC would be used for tracking not only in the electron pair measurement in conjunction with the HBD, but also as an additional tracking detector in
normal field running.
Helps in rejecting false high PT tracks and provides good stand alone
tracking for tracks presently outside the PHENIX acceptance (useful for jet
measurements)
C.Woody, Nashville Collaboration Meeting, 6/11/03 15
Is this PHENIX or STAR ?
J.Mitchell’s event display of the PHENIX TPC for a central Au-Au event
(PISA simulation)
Investigating TPC pattern recognition
software from LBL (LC) and
STAR (Level 3)
C.Woody, Nashville Collaboration Meeting, 6/11/03 16
Readout Electronics
Currently reading out GEM TPC with 100 MHz FADC(SIS 3300 8 channel, 12 bit VME module)
pDAQ courtesy of M.Purschke
FADC readout (10 ns bins) Digital scope trace
55Fe source with 10x10 cm2 triple GEM
C.Woody, Nashville Collaboration Meeting, 6/11/03 17
Future HBD/TPC readout electronics developmentTry and utilize existing or soon to be available components
HBD needs only modest (~ few ns) time resolutionPerhaps can use FADC to provide time information
32 ch preamp/shaper for APD readout0.18 mm CMOS, 125 mW per chip
Final ASIC 4.3 x 1.6 mm2
65 MHz FADC available soon
18
TPC and HBD electronics
R&D:
1) can use the same type of electronics for both HBD and TPC ???
(by fitting ADC samples to get T0)
2) TPC FEE issue:
Heat load and high channels count
=> custom ASIC or commercial ADC+ FPGA
In the next few months we will
construct a possible commercial ADC + FPGA for the TPC readout solution base on Analog Device 4 channel 8 bit ADC (chips scale BGA serial ADC)
This is to understand power + readout + packing density issues
This will also serve as preliminary study for the HBD readout…
TPC FEE: Samples signal every 25ns for 140 samples. need L1 delay buffer (160 samples) +
5 events buffers (700 samples)HBD FEE: Need to measure charge and time
(~ several ns) L1 delay buffer + 5 events buffers.
C.-Y.Chi, Nevis
19
For TPC FEE :60K to carry out TPC FEE study + 20 K (M&S)
1) study commercial IC solution2) study the issues regard to custom ASIC
a) follow ALICE path – buy ADC core + custom Logicsb) fully custom ASIC ( need to use commercial library)
3) building prototype modules ( commercial solution) or custom ASIC prototype
For HBD FEE:40 K on salary + 20K (M&S) - Part of works on done from TPC FEE R&D
1) prototype readout For DCM:
80K on salary + 20K (M&S)R&D on DCMwork out readout method for prototype detectors readout
Assumptions:
we need to 1) readout HBD prototype end of next year
2) readout various prototype detector 1.5 years from now
R&D Requests From Nevis
C.-Y.Chi, Nevis
C.Woody, Nashville Collaboration Meeting, 6/11/03 20
TPC Prototype - Starting Next Year
• Full size module• Tests field cage design• Tests materials to be used in actual construction • Allows testing with cosmic rays• Provides structure on which to design and study readout plane• Provides excellent test bench for readout electronics
C.Woody, Nashville Collaboration Meeting, 6/11/03 21
R&D Plan
FY2003• Complete TPC drift cell and begin comprehensive gas studies • Test new 3M GEM foils• Continue to improve TPC field cage calculation• Begin TPC prototype design• Measure CsI quantum efficiencies w/GEMs + radiator gases• Measure CF4 Cherenkov efficiency + scintillation (NSLS)• Begin design of HBD & TPC readout electronics• Funds: $25K labor + $30K equipment
FY2004• Build and test TPC & HBD prototypes • Hopefully complete design of HBD readout electronics and continue design of TPC electronics.• Start engineering design of TPC/HBD detector system• Funds: $200K HBD + $200K TPC
C.Woody, Nashville Collaboration Meeting, 6/11/03 22
Extra Slides
C.Woody, Nashville Collaboration Meeting, 6/11/03 23
HBD Budget
HBD Construction Costs Total Total + C FY04 FY04+C FY05 FY05+C FY06 FY06+C
Working Prototype 200 200 200 200
UV detector 250 375 100 150 150 225Design (6 Man Months) 25CsI photocathodes 5016 mesh frames 15GEMs 16x3 13016 PCB on honeycomb 20Connectors, resistors capacitors 10
Radiator and support structure 150 225 50 75 100 150
Gas system 150 225 50 75 100 150
Electronics 500 750 150 225 350 525Engineering 250Prototype chip run 100Final chip run 100FEMs + DCMs 50
Total detector costs 1250 1775 200 200 350 525 700 1050Contingency 50 %
C.Woody, Nashville Collaboration Meeting, 6/11/03 24
R&D Proposal Budget RequestCategory Description FY03 ($K) FY04($K) FY05($K)
Salaries Post Doc 45 45 45(incl. fringe) Electrical Engineer (1.0-1.25 FTE) 100 125 125
Electrical Tech (0.25 FTE) 20 20 20Mechanical Engineer (0.25 FTE) 25 25 25Mechanical Designer (0.25 FTE) 20 20 20Mechanical Tech (0.25 FTE) 20 20 20
Supplies Lab equipment 30 20 15Electronics ASIC fabrication 30 60 75
Test equipment 15 25 15Total 305 360 360Total (incl 40% overhead) 427 504 504
PHENIX Institutional Involvement
• HBD BNL, WIS, Stony Brook• TPC BNL (Physics, Instrumentation), Tokyo, FIT• Electronics BNL Instrumentation, Columbia (Nevis)
Related R&D Efforts
• STAR (joint effort)• LEGS TPC• TPC w/GEM readout for NLC/TESLA
C.Woody, Nashville Collaboration Meeting, 6/11/03 25
Upgrade Budget and Timeline
C.Woody, Nashville Collaboration Meeting, 6/11/03 26
Possible Support Structure for HBD/TPC/VTX
H.Van Hecke