TRD Status Report

About This Presentation

Title:

TRD Status Report

Description:

TRD Status Report – PowerPoint PPT presentation

Number of Views:109

Avg rating:3.0/5.0

Slides: 75

Provided by: han62

Category:

more less

Transcript and Presenter's Notes

Title: TRD Status Report

1
TRD Status Report
LHCC Comprehensive Review, 7-8 March 2005,
Johannes P. Wessels Univ. of Muenster

brief overview
super module
insertion tests
design
integration of services
chamber production
status
tests
services
testbeam
stack test

electronics
analog pre-amplifier shaper
digital ADC development, tracklet processor
system integration MCM readout boards
detector control system

summary
milestones

2
Transition Radiation Detector (TRD)

Purpose
electron ID in central barrel pgt1 GeV/c
fast trigger for high momentum particles
Parameters
540 modules -gt 760m2 (50 funded)
length 7m
anticipated X/X0 15
28 m3 Xe/CO2 (8515)
1.2 million channels
Institutions
Athens, Bucharest, Darmstadt, JINR,
Frankfurt, GSI, PI Heidelberg, KIP Heidelberg,
Cologne, Karlsruhe, Kaiserslautern, Münster, Worms

3
General Status Aims

Bucharest, JINR, GSI, Heidelberg in series
production mode Frankfurt soon ready
incurred additional delay
modified design of super module - new layer 0
delay in chip production, redesign of ROB
still open funding issues
Japanese proposal for funding resubmitted
apply to BMBF for funding period 2006-2009
reschedule with the aim to have
1 super module ready in 2005 - installation
Phase 2
4 completely equipped super modules -
installation Phase 4

4
SM Insertion Tests at CERN
full scaledummy super module insertion
belowTOF dummy August 2004
TOF
TRD
5
SM Insertion Tests at CERN
August 2004
6
SM Insertion Tests at CERN
Nov./Dec. 2004
Modified Mounting Tool
7
SM Insertion Tests at CERN
Nov./Dec. 2004
8
Nominal Dimensions of 1 Space Frame Segment
TOF
TRD
9
RB 26
BOTTOM / TOP 6 cross bars each LEFT / RIGHT 10
radial bars each
TOP
6
5
4
10
3
9
8
RIGHT
7
2
6
1
5
4
TOP left
TOP right
3
RIGHT top
LEFT top
2
LEFT
TRD
1
RIGHT bottom
LEFT bottom
BOTTOM left
BOTTOM right
BOTTOM
SF-numbering scheme for TRD insertion
test Counting starts on RB 24 side
RB 24
10
Nominal Distances/Dimensions between Supermodule
and Space Frame
In the above table the values for LEFT and RIGHT
are valid for segment 04 and 13, assuming the
supe rmodule is centered in the space frame
sector. for segment 05 to 12 LEFT(top and bottom)
0 and RIGHT(top and bottom) 12.46 for segment
14 to 03 LEFT(top and bottom) 12.46 and
RIGHT(top and bottom) 0
11
Sector 15, corrected data, ?H9mm
(-?H)
(?H)
(-2?H sin 10)
12
Summary of Insertion Tests

-
-
-
-

-
-
-
fits
- does not fit

-

-
-
-
-
-
-
13
Solution to the Dilemma
After considering all options

TOF unchanged
SF unchanged
? modify TRD (shrink by 4)
do not build L6
move L1-L5 up by one layer
introduce new layer (L0) at the bottom
gain 20 mm on either side

14
Space Frame TOF TRD TRD modified
TOF
L5
L4
L3
L2
L1
L0
15
Chamber Stack L0 L5
Bernd Windelband, Uni Heidelberg
16
Super Module Design
after decision to introduce L0 detailed design
work on the super module has started
17
Status of ServicesBus Bar / Cooling Channel
original version
18
Status of ServicesBus Bar Version 1
copper bus bars
cooling
Bernd Windelband, Uni Heidelberg
19
Status of ServicesBus Bar Version 2
20
Servicesnext steps

final design and construction of new bus bar /
cooling channel

design of manifolds (gas and cooling)

fully equipped ROB with cooling and all cables

21
Status HV
Athens

prototype development progressing - expect 1st
card in summer
working on
fault protection
current measurement
layout

22
Radiator
Muenster

CF-laminated quilt structure of
thin CF laminate
Rohacell HF71
polypropylene fibers from Freudenberg LRP375BK -
completely cut for 100 of TRD
status production completed in July 04
layer 0 - try modification of L6
order extra panels

23
QA - Radiators
Muenster

systematic check of X-ray absorption of HF71/CF
laminate
Test device
10 kV X-ray source from Dubna
source up to 50 keV
Si/Li detector
measurement of individual componenets and whole
radiators

24
Radiator X-ray absorption
Muenster

quantitative understanding of absorption
properties of all materials over large energy
range
parameterizationOrlic et al. NIM B74(93)352

25
Chamber Design

HEXCELL/CF backing (FACC)
finished - layer 0 ordered
pad plane - (Optiprint) finished
layer 0 delivery in April
amplification region
drift region
radiator

26
Chamber Assembly
Heidelberg
27
TRD chamber production
Heidelberg
preparation of side profiles
preparation for high voltage and gas tightness
test
control of wire grid
28
Quality Assurance

common set of tools developed and adopted by
all production sites
standardized assembly procedure with sign-off
sheet training in Heidelberg
gas tightness better than 1 mbar l/h
(lt0.2 mbar l/h in preproduction)
dark current lt 1 nA _at_1.5 kV
gain uniformity lt 15
wire tension 5

29
Chamber Assembly
Heidelberg
30
Chamber Testing
Heidelberg
2d - gain map
JINR X-ray source
phi
z
test of leak tightness
10
O2 level (ppm)
anode current vs. position
time
31
Wire Tension Position Device
contact-free wire tension and position measurement
step motor
bar code head
chamber
32
Wire Tension Position Device
as delivered to Heidelberg and Bucharest
33
Wire Tension Measurement
Tension
Tension
45.9 0.1 cN nominal 45 cN
76.4 0.05 cN nominal 80 cN
34
Wire separation
Anode
Cathode
5 mm 28 µm
2.5 mm 46 µm
35
Clean Rooms at JINR

36
Wire Frame Transfer
JINR
Several microscopes are used to define a right
wire position relatively to reference hole ,
precision 50 mkm
TRD status meeting, 3-5 January 2005, GSI

37
Gain Uniformity Measurement
JINR
38
Status JINR

13 chambers ready and tested
gain uniformity about 14
2 chambers under transferring of anode planes
4 boxes and 19 pad plane panels are ready
will finally need 8 days for 2 chambers
(as of February 4)

cosmic trigger for super module test under
preparation
39
Status of Bucharest Lab
laboratory fully operational
40
(No Transcript)
41
Heidelberg Status

manpower
2 technicians (chamber building)
1 student (pad plane QA, chamber production)
3 technicians (front- L-profiles production)
1 technician from Bucharest (L-profiles
production)
2 physicists
production rate 1 chamber/week
(w/o complications)

9xL1C1 finished
2xL1C1 being wired
3xL1C0
3xL2C0
1xL1C0 not usable? (badly glued pad plane repair
may be possible)

42
Layer 0 Implications

pad plane delivery starting April 05
back panel delivery starting June 05
radiators - modify L6, order additional material
few months delay - implented in schedule

current distribution of types
Dubna L1-6C0
Bucharest L2C1, L3C1
GSI L5C1, L6C1
HD L1C1(part), L4C1
Frankfurt L1C1 (rest), ?

proposed change
Dubna L0-5C0
Bucharest L2C1, L3C1
GSI L4C1, L5C1
HD L1C1(part), L0C1
Frankfurt L1C1(rest), ?

43
Chambers until End of 2005

Hd
7 L1C0
6 L2C0
24 L1C1
28 L0C1
Dubna
12 L0C0
7 L1C0
6 L2C0
10 L3C0
8 L4C0
12 L5C0

Bucharest
2 L1C0
2 L2C0
20 L2C1
20 L3C1
GSI
20 L4C1
24 L5C1
Frankfurt
20 L1C1

more than 50 by the end of 2005
44
Chambers Ready for Installation in SMs

- assumption funding commitment for 100
detector Spring 2006
all estimates based on 1 chamber/week average
production rate
-gt chamber production not limiting super module
assembly

45
Gas System Status

all features tested, including (CERN 04)
membranes for filling with Xe
flush with CO2
inject Xe while pumping out CO2 from membrane
double regulation system (3rd rack)
CERN 04 test beam a satisfactory experience
work needed to automate procedures

46
(No Transcript)
47
Gas System Schedule

to be revisited with GWG
design change in the distribution racks in
preparation
construction schedule to match installation

48
Beam Time October 2004
49
(No Transcript)
50
Test Beam - Event Display
51
Test Beam - Charge Spectra
52
Test Beam - dE/dx, e/p
quantitative understanding of deposited charge
final electronics on stacksignificantly larger
noise
53
Test Beam - Angle Reconstruction
envisaged TRD performance achieved - noise will
be improved detailed understanding in simulation
(space charge, isochronicity) NIM A 540 (2005)
140-157 neural networks for PID, NIM submitted
54
TRD Electronics Status and Planning

Chip production and testing
PASA
TRAP
OASE
MCM
Readout Board
DCS
Integration
October tests
Recent measurements
Planning

Volker Lindenstruth Chair of Computer Science
Kirchhoff Institute for PhysicsUniversity
Heidelberg, Germany Phone 49 6221 54
9801 Fax 49 6221 54 9809 Email ti_at_kip.uni-heide
lberg.de WWW www.ti.uni-hd.de
55
PASA

full production ordered and delivered in 2004
all tests performed since then were using
untested chips from production rundue to high
production yield no significant loss
PASA wafer tester produced
PASA wafer testing to start in March 2004

56
Tracklet Processor

Features
deep submicron UMC180nm
integration of 22 ADCs, 4 RISC CPUs, 4x2.5 GBit
Network IFDCS infrastructure, Temperature
sensor,
concurrent operation of processor and ADCs
Status
pilot run launched in April 04
chips returned working with one minor bug
disabling the redundant DCS interface
ADC performance over all 9.5 ENOB
M2/M3 patched to enable redundant DCS interface
reprocessed, backordered 4 wafers received Feb
25, 05

57
OASE

OASE (Optical Advanced Serializer/dEserializer)ch
ips on TRAP3 wafer received back DOA (error in
mask fabrication, affecting only TWELL mask)
UMC finally agreed in October to repair mask and
reprocess 4 wafers from scratch
4 new wafers with OASE chips received Feb 25,
2005(wafers were processed with corrected TRAP
masks, yielding additional 4 wafers for
production)
OASE backup (FPGAglue logic) designed in order
to avoid project delays (OASE serializer is
mounted on mezzanine card)

58
MCM Production
FZ Karlsruhe

fully automatic MCM tester completed being
phased into production
electronics identical for TRAP wafer tester
(needle card under development)
first automatic MCM testing starting at KIP
several batches of 16x MCM lots produced at FZK

59
Readout Board

first readout boards (V1) produced in September
04
readout board V1 fully opeational during test
beam
production process requires further optimization
of yield
current profiles measured on V1 and filter caps
optimized readout board redesigned with improved
PWR and GND routing ? V2
first equipped module V2 received 2/05 testing
started
additional respin anticipated
use of very low dropout regulators to save power
adapt to layer 0

60
DCS Single Board Linux Computer

successfully operated in test beam, October 2004
working fully to specification
full software stack available and functional
two minor features / inelegant usage issues found
and repaired
layout patch completed
second preproduction run ordered (25 units)
production run of 200 units to follow 5/05 as
needed
no delay for either TPC or TRD as DCS board is
not in critical path
production tester in production

embedded single-board linux computer
radiation tested
ethernet as field bus operational in strong
magnetic fields
very light weight ethernet interface
redundant, in situ reconfigurable
emerging ALICE standard
reconfiguration master

61
Test Beam Readout Scheme
1
2
3
4
5
6
0
7
RB
RB
RB
RB
RB
RB
RB
RB
All RoBs V1
TRAP 3 Merger
TRAP 3 Merger
TRAP 3 Merger
FER

ACEX card last
minute solution for FER
FER converts data to DDL

DDL
LVDS
Adapter
HLT PubSub
HLT- RORC
ACEX
62
DAQ System setup
TRD LDC
FER
TRD stack
TRAP 3 Merger
DDL
GDC
VME LDC
Switch
Monitoring detectors
PVSS
8 DCS boards
Temp

beam area
counting room
DCS
DCS
DCS
DCS
DCS
DCS
DCS
DCS
63
Trigger System Setup
TRD stack
8 DCS boards
Trigger signal
128 TRAP3
DCS
DCS
ev
DCS
DCS
TTCvi
DCS
DCS
DCS
DCS
FER
Trigger fan out
Monitoring detectors
TRD LDC
ev
CORBO
VME LCD

busy
GDC
? entire trigger/clock distribution system
operated
64
First Online Tracks (1 of 2)
lt?Ygt 524 ?m
?single track reconstructed by the tracklet
processor on the MCM
65
First Online Tracks (2 of 2)
lt?Ygt 233 ?m
lt?Ygt 329 ?m
lt?Ygt 836 ?m
?multiple tracks reconstructed by the tracklet
processor on the MCM
66
Test Beam Summary

first fully operational TRD Stack
all on-detector electronics components
implemented and tested
tracklet processor and readout working without
any problems
complete readout tree exercised under realistic
experiment conditions (however, at reduced rate
but full bandwidth due to ACEX readout mode)
full DCS functionality with 10 DCS boards working
fine
complete TTC Trigger System implemented
TRD not used as trigger but verified in data
600k events recorded in 7 days, trigger rate 25
Hz
noise in test beam area very high due to various
possible reasons (very preliminary grounding, 80
unused pads, first version of complete
electronics chain, uncontrolled current returns)
noise about 6000 e-

67
Readout Board Topology
V1
68
Optimization of RoB (V1)

one readout board V1.0 on Stand alone chamber
connected to stack power
HV unconnected
gas system unconnected

69
RoB V2.0 in Lab
70
Integration Summary

electronics noise of stand-alone MCM 600e- RMS
electronics noise of stand-alone RoB V1 700e-
RMS
electronics noise of stand-alone chamber 1200e-
RMS
electronics noise of chamber in stack 1500e- RMS
various issues identified and being repaired
improvement of grounding (e.g. ethernet, readout,
HV, chamber)
improvement of systematic cross talk (trigger
related cross talk, most likely via digital
power)
improvement of power filtering
better understanding of RLC properties of power
distribution

71
Power

power measured in system
power increase as compared to original estimate
some heat dissipation compensated with very LDO
regulators in latest RoB design
cable and rack allocation being reworked

PASA 3.3V Heat per SM W Heat W Vdrop V Heat W
Analog 281.77 20723.47 0.40 2511.94
TRAP 3.3V
20.97 5653.14 0.40 685.23
TRAP 1.8V
Analog 475.19 14679.45 0.60 4893.15
TRAP 1.8V
Digital 414.43 13708.82 0.30 2284.80
DCS 3.3V
Digital 237.60 0.40 28.80
Sum
1192.35 55002.48 10403.9176
Total Heat W
66598.75
72
Detector Summary

construction of TRD moving steadily ahead
5 sites working on chamber production
space constraints with space frame resolved
introduction of new layer 0
slight increase in cost
major goal - finish first super module this year
funding commitment for 100 desirable inearly
2006

73
Electronics Summary

significant progress on electronics in 2004
design of all electronics completed, tested and
verified
8 wafers of digital chips deliveredproduction
launch of digital chip wafers during March/April
05
production test infrastructure approaching
completion
complete electronics, DCS, and trigger
infrastructure successfully operated during beam
time
noise and performance of stand-alone readout
board satisfactory
still some yield issues with glob top of MCMs
readout board production not yet running smoothly
remaining issues
noise performance of electronics on chamber still
not optimal
full integration of a super module