TRD Status Report

1
TRD Status Report
LHCC Comprehensive Review, 20-21 March 2006,
Johannes P. Wessels Univ. of Muenster
Project leader Johanna Stachel,
HeidelbergDeputy project leader Peter
Braun-Munzinger, GSITechnical Coordinator
Johannes P. Wessels, Münster

• brief overview
• chamber production
• status
• super module
• insertion tests
• design
• integration of services
• services

• electronics
• PASA - pre-amplifier
• TRAP - digital chip
• 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
Chamber Production

• stable production at 4 sites (Bucharest, Dubna,
  GSI, Heidelberg)
• Frankfurt ramping up production - started in fall
  2005
• since decision for Layer 0 swift procurement and
  production (completed for first super module)
• noticed problem with insufficient surface
  resistivity of AW106 glue after curing (large
  dark current)traced back to change in the
  manufacturing process of the gluecorrective
  action additional cleaning step with ethanol
  after curing

4
Leakage Current Path
backpanel
pad
anode plane
cathode plane
drift space
glue
TRD radiator
5
Surface Resistivity
Chr. Schmidt, GSI
6
ROC production status
C. Adler, Heidelberg
Status 25.2.06 Bucharest Dubna Frankfurt GSI Heidelberg Sum
taped   2 2 5 5 14
tested taped 2     13 2 17
glued (under testing) 1 2   1 2 6
tested glued 36 33   40 21 130
to be repaired   5   3 2 10
Sum 167(10) 62(66) of TRD
since Christmas 40 additional chambers
completed global production rate 6 chambers/week
7
TRD rail system in space frame

• self-locking rail system installed, aligned
  in SF
• different types of rails depending on
  position
• clearances checked
• final adjustment after SF installation in
  magnet (April)

8
(No Transcript)
9
Super Module Production Status
V. Chepurnov, JINR B. Windelband, HD, W.
Verhoeven, MS
-gt all pieces finished for first super module -gt
first assembly of super module mechanics
successful in February 06
10
First Super Module Assembly
11
Mounting L0 Chamber
12
Full Length Assembly
13
Super Module Service Sides
A-side
gas cooling
C-side
LV power detector control high voltage
readout fiberspre-trigger boxDCS power
14
Super Module Installation
15
SM and Manipulator
16
SM Insertion into Manipulator
Manipulator adaptedto final super
moduledimensions
Insertion test at CERN scheduled for April/May
06progress according to schedule
17
Bus Bar Layout
Chr. Lippmann, GSI K. Oyama, HD

• capacitors added
• design successfully tested
• producer identified
• order for first 2 SM placed

18
Cooling
A. Marin, GSI B. Windelband, HD

• 65x10x2mm manifolds under construction for first
  layers
• directly mounted on bus bars
• delivery of all pipes for 50 TRD, 65x8x1mm end
  of April

• cooling plant construction well under way
• commissioning will follow installation of first
  super module(September)

Cooling pipes
19
Layout of cooling meanders
N. Heine, MS A. Marin, GSI
readout chamber
readout board
cooling pathon MCMs maximized 90mm
DCS
ORI
detectorcontrolsystem
optical readoutinterface
20
Production of cooling meanders

• thermal contact epoxy loaded with Al powder
  (plate-meander)
• Thermaglue - heat conducting silicone between
  plate and MCM
• tools ready for production (8/RoB)
• material in hand for 1st super module
• 20 of meanders finished
• manufacturer identified
• material on order for 50 TRD

21
Temperature Distributions
?T20 K
temperature profile for differentload
conditions on 2 ROBs all sensors exhibit
similarprofile
22
Gas System
C. Garabatos GSI R. Stadler, HD

• production and commissioning schedule fixed with
  gas working group (contingent on delays in other
  systems)
• on-detector pipe layout finalized
• small reliable bubbler designed and tested
• Sep/Oct 06 - commissioning with Ar
• Nov 06 - commissioning of purifier and membranes
• Dec 06 - closed-loop operation possible
• April 07 - full commissioning with Xe (analysis,
  recovery)

23
Recovery Plant / Monitoring

• ALEPH recovery plant relocated, needs to be
  re-commissioned
• compressor unit for recovered gas still to be
  designed and produced

drift velocity monitor under test
24
PASA Overview
H.-K. Soltveit, HD
PASA productionfinished125,000 chipstested
(almosttwice as many asneeded) yield 99
Microscope
Control area
Chuck, needle card, needle card adapter and wafer
Main PASA tester
25
TRAP Overview

• wafers from 2 engineering runs with good yield
  ( 80)
• first production run poor yield ( 30) and
  difficult to bond
• corner run proposed by foundry
• received back in February ( 50)
• have enough good chips for first super module
• two further production slots reserved in 2006
  (prices dropping significantly)

26
TRAP Tests (Wafer/MCM)
V. Angelov, F. Ferner, D. Gottschalk, HD

• Power - reject chips with high currents
• Digital part
• Connectivity test using JTAG (MCM only)
• Contact resistances of some signals (wafer only)
• Internal tests using internal CPUs instruction
  memories, data memory, configuration
  registers, CPU registers, look up tables, digital
  filters etc.
• Test the readout-tree functionality

• Analog part
• internal pulser of PASA used to test ADCs (MCM
  only)
• Sine-wave test of 50 of ADCs on wafer (wafer
  only)
• Reliability test using different digital supply
  voltages

MCM tester
needle card of wafer tester
27
TRAP Performance
M. Gutfleisch, HD

• performs angle reconstruction
• good agreement with offline tracking

28
MCM Production
Th. Blank, FZK

• originally FZ Karlsruhe as sole producer
  anticipated
• encountered bonding problems (bonder reliability
  issues, contacts)
• found alternative producer - attractive pricing
  (MSC) (ROB producer)
• optimized bonding procedure and balling of MCMs
• split contract between FZ Karlsruhe and MSC (50
  TRD)
• agreement to build 15000 modules at FZK
• 25000 modules at MSC (delivery schedule to be
  settled end of March)
• current production rate at FZK 800 MCM/wk (gt1
  layer/wk) - yield 90
• FZK will complete 25 by April
• MCM production rate sufficient for ROB and SM
  production

29
ROB Test Setup _at_ HD
J. Mercado. HD

• ROB tests
• power consumption
• SCSN functionality
• internal TRAP
• tests
• IMEM, DBANK, DMEM, DIV, CJP,
• CST, PG, GIO, IRQ, LUT, EBF, Filters, . . .
• network interface test (read-out tree) via SCSN

30
ROB Production/Test Status
I. Rusanov. HD Th. Blank, FZK
design finished - Nov 052 qualified offers - Dec
05delivery - Jan/Feb 06contract to be awarded
end of March
Origin Delivery ROB types Items Tested Good
FZK 12/2005 3A, 3B, 4A, 4B 19 all 18
FZK 19.01.2006 1B, 2B, 3A 18 all 16
Dommel 20.01.2006 2B 9 all 9
FZK 27.01.2006 1A 13 all 11
MSC 06.02.2006 1B 10 all 9
FZK 10.02.2006 all types 30 all 24
ROBs ready and tested as of Feb, 28 76 (2
layers) Yield 86
31
Electronics Integration
Chr. Lippmann, GSI K. Oyama, D. Emschermann. HD
L0C1 003

• ROCs in various stages of completion
• assembly procedure documented
• knowledge base of errors (missing grounds, etc.)
• workload 2 person days / chamber
• one layer finished

in super module
32
Noise Performance
Chr. Lippmann, GSI K. Oyama, I. Rusanov, HD

• Latest measurement ltRMSgt 1.17 LSB1230 e
• stable over many days
• with
• Wiener P.S. for ROBs.
• CERES P.S. for DCS.
• Power bus bar (C.L. type).

33
High Voltage Master/Slave System
Athens University
Controls
6-ch. card / top viewauxiliary power supply box
34
HV System Status
First system test at HD (Feb 23 - 25)- anode
voltages (2000 V) - 6 channels successfully
tested- stability 50mV- full command access
through CAN (and serial interface)- first PVSS
FSM functions error handling available -
primary HV supplies available for 12 super
modules To do - negative voltages (-3000 V)-
software regulation w/ 16 bit ADC _at_10-15Hz ?
5mV Goal- full test of 1st SM equipment at
HD/KIP (end of May)
35
Detector Control System (DCS)
D. Gottschalk, K. Schweda, J. Mercado, HD M.
Stockmeier, R. Bramm, GSI
DCS Hardware Link http//www.kip.uni-heidelber
g.de/ti/DCS-Board/current/

• Each DCS board controls eight readout boards
• Monitoring functions
• temperature of MCMs
• currents of MCMs
• Control functions
• voltage regulator shutdown
• MCM configuration
• Clock and Trigger Distribution

• Optical Clock Receiver (TTCrx)
• Ethernet link
• JTAG in and JTAG out
• 8 analog inputs 16 Bit / 10Sps
• 8 Pseudo LVDS Inputs
• 8 Pseudo LVDS Outputs
• 24 Switching Lines
• two 8 Bit Ports (optionally)
• I²C Bus
• two-wire UART
• autonomous power supply

DCS interface functionality
36
DCS Status
200 TPC DCS boards delivered by MSC 215 TRD DCS
boards expected end of March rest in April 120
TRD DCS boards programmed and successfully
tested PVSS control chain tested Database access
being worked on
37
Optical Sender Status
V. Angelov, HD

• Test setup
• Read-out board
• ? Optical Sender Board
• ? 60m long fiber
• ? dedicated test receiver board
• ? ACEX PCI card software
• Latest test results
• Bit error rates below 10-12 (upper limit)
• Current stock
• 12 Optical Sender Boards built and tested
• components radiation tested - 2 weeks ago

ORI on ROB 3A/B
Optical receiver on the ACEX board
38
TRD Trigger
J. DeCuveland, HD
Global Tracking Inside GTU (Global Tracking
Unit) Objective find high momentum
tracks Search for tracklets belonging
together Combine tracklets from all six
layers Reconstruct pt, compare to threshold and
generate trigger Constraintonly approx. 1.5
µs processing time
Charge Cluster to Tracklet Local tracking units
on detector perform linear fits and reject
uninteresting data
Tracklet(32-Bit word) y position
(origin) slope (deflection) z position
(pad-row number) charge

• Trigger decision after 6 µs
• Charge drift and data pre-processing uses most of
  the time

39
TMU Board Prototype Status

• basic PCB functionality successfully tested
• all supply voltages, oscillators
• configuration flash PROM, FPGA configuration via
  JTAG
• system monitoring and user I/O via I2C
• SFP module controlling via I2C
• multi-gigabit transceiver (MGT) operation shown
  in principle
• transmission between two GTU boards working (with
  limitations)
• known bug in Xilinx Virtex-4 FX60 engineering
  sample-1 chips PLLs lose lock, need prototype
  with ES4 chips
• next to be tested
• SRAM
• CompactPCI interface
• LVDS to backplane

40
TMU Design / Simulation
Reconstruction Precision
pt reconstruction precision 2.1
TMU Processing Time
Detection Efficiency
Simulation with Altera FPGA at 40 MHz Final
design using Xilinx Virtex-4 FX at gt 60 MHz
Simulations with AliRoot data, electrons with
pt gt 3 GeV/c
Results depend on multiplicity density dNch/d?
and number of tracklets delivered by detector
41
TRD Pre-Trigger System
K. Oyama, R. Schicker, HD

• TRD pre-trigger wakeup signal sent to TRD
  MCMs
• generated by T0, V0 and TOF
• T0 Timing detector and M.B. trigger
• V0 Vertex detection and pp trigger
• TOF cosmic trigger and low multiplicity
  (ultra-peripheral)
• timing crucial ? system placed inside L3 magnet
• duplicated for fault tolerance
• functionality of pre-trigger system
• generates coincidence signal of T0, V0 and TOF
  detectors
• distribution of TTC signal from CTP and into
  TRD TTC stream
• provides TTC optical signals to 18 TRD
  super-modules
• generates control signals for FSM in MCMs

42
Pre-Trigger Schematics
K. Oyama, HD
43
Pre-Trigger Timing/Data Flow
44
T0/V0 Pre-Amplifier Module
K. Oyama, PI Electronics Shop, HD
x10 output
x1 feed-through output
final design 55mm x 55mm 3 SMA connectors P0.5
W/ch
power
differential output to TRD pre-trigger system
input from the PMT
30 ps resolution achieved in test beam
experiment high ground and power stability
required circuit contains inductance - still
operational at B0.5 T
45
Location of Pre-Trigger Components
inner bar (7m)
A
C
TRD
FEB
FEB
FEB
FEB
FEB
FEB
FEB
FEB
FEB
FEB
T0-C
T0-A
CB-C
CB-A
CB-TOF
TOF signals
46
Prototype Control Board
K. Oyama, PI Electronics Shop, HD
all components will be locatedinside L3 volume
(back frame,baby space frame)
DCS board
47
Detector Summary

• construction of TRD moving steadily ahead
• 5 sites working on chamber production
• excellent understanding of detector performance
  from test beam data (6 NIM publications last
  year)
• rail system installed in space frame
• final alignment and insertion test (March-May)
• first assembly of super module structure
  successful
• integration ongoing
• first super module ready for installation in July
• looking forward to funding commitment for 100
  inearly 2006

48
Electronics Summary

• significant progress in electronics in 2005
• MCM production with good yield
• ROB production with good yield
• industrial producers identified
• full integration on chambers successful

49
Milestones

• readout board start of production 07/05 done
• digital chip complete 07/05 production in
  batches
• MCM complete 11/06 following chip delivery
• readout board complete 12/06 as above
• GTU start production 06/06
• GTU complete 01/07 completed as needed
• chambers 50 complete 12/05 done
• chambers 100 complete 10/06
• start stacking super modules 11/05 done
• ready to install 1 super module 07/06
• start installation 1 super module 08/06
• installation of 2 super modules 03/07