GEMDetector Projects GEMTPC Planar GEMTrackers Bernd Voss Helmholtzzentrum
GEM-Detector Projects (GEM-TPC & Planar GEM-Trackers) Bernd Voss Helmholtzzentrum für Schwerionenforschung Gmb. H (GSI)
Outline The Experiment The Detectors Simulations PANDA@FAIR Tasks, Requirements & Set up Basic design & Detector assembly Particle flux & Tracks Pad Plane Front End Electronic Higher Level Readout Design & Consequences The XYTER Family Sys. Core / Exploder First Results Noise figures The Future Towards a FAIR-XYTER Detector Control System Detector Mounting ‘Joint. GEM’ Integration Road Map B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 2
PANDA@FAIR The site Facility for Antiproton and Ion Research Darmstadt, Germany HESR High-Energy Storage Ring p production target PANDA B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 3
PANDA@FAIR Setup Overview Target Central Tracker GEM-TPC p Forward Tracker Planar-GEMs cting u d n rco e p u T) S 2 ( d i o solen eturn yoke pe S ~12 m d r orwa F eter r + iron r te e m o ctr om r t c e t Sp e Targ B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 4
GEM-Detectors@PANDA Basic assumptions Figures of Merit: Particle identification Momentum resolution p/p (p, θ, z, r) particle momentum p • scattering angle θ • vertex coordinates z, r • Basic assumptions: GEM-TPC (‘short’ version: 1, 5 m 1, 2 m) GEM-Tracker, 3. . 4 stations Maximize shape-conformity Full angular coverage (φ) not possible Radiation hardness 100 krad Target spectrometer@PANDA ‘V 833’ B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 5
GEM-Detectors@PANDA Figures of merit Requirements Planar-GEM Trackers GEM-TPC Momentum resolution + Particle identification σrΦ ≈ 100 µm ≈ 10 mm, 5° Position Resolutions Double track Secondary vertex σrΦ ≈ 150 µm σz ≈ 1 mm Momentum p/p ≈ 1 % Solid angle Material budget 5. . 18° (2. . 26°) 0. 5 % X 0 Features Challenges B. Voss et. al. near 4π 1% X 0 Operation in magnetic field (2 T) high granularity, fast signal multi-track resolution suppressed ion-feedback, reduced E x B effect High count rate Low B-Field Low curvature Continuous operation: Space charge (5 charged tracks/event) Event mixing GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 6
GEM-Detectors@PANDA Feature GEM-TPC …in numbers GEM-T 1 GEM-T 2 GEM-T 3 GEM-T 4 1530 1890 Length (active) mm 1500 Position from target mm -400. . 1100 Radius inner mm 150 (active) outer mm 420 450 560 740 mm 2 0. 5 0. 6 1 1. 7 3 2 x 3, Single foil 300 384 Area GEM foils No. of GEM sectors Pad planes 1, single sided 2 Projections Readout geometry Hexagonal pads (x, y) Structure size/pitch (resolution driven) 4 mm 2 Max. channel no. Weight kg 60 810 1170 25. . 50 2 x 3, Patched or large-area 600 1, double sided 4, 2 tracklets Cartesian (x, y) Concentric circles (φ) Radial strips (r) Tilted strips (+60°, -60°) radial concentric 100. . 600 µm 400 µm 80 k 20 k 32 k 45 k 50 20 30 40 Ne/CO 2 (90/10) Ar/CO 2 (80/20) Drift time / -velocity 50 µs, 2. 8 cm/µs n x 10 ns 10000 B. Voss et. al. mm 2 50 mm < r < 150 mm < r < 450 mm Gas Cabling 1044 GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 7
Prototype GEM-TPC Detector assembly Prototype L: 650 mm D: 300 mm Modular Under construction at GSI-DL To be tested 2009 at: Staggered strip-type field system CB-ELSA FOPI Several hundred pieces B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 8
Planar GEM-Trackers Detector assembly Possible cabling hooks: central bar & circumference Window GEM stack Drift electrode B. Voss et. al. Cooling Pad Plane Support & LV & - Distribution Media. Distribution Front-End Electronic Shielding Cover & Read-out Plane GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 9
Layer stacking 10 B. Voss et. al. 10 2 2 2 active Planar GEM-Trackers 2 2 2 10 10 GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 mm 10
Planar GEM-Trackers For the set of four (in%) Active Absorber Weight Contribution 0, 5 4 Radiation Length design goal 0, 093 1, 405 Radiation Length status quo technique *) …in numbers 0, 093 (no backing) 3, 485 Supply Support Front-End ≈ 33 ≈ 34 ≈ 28 n. ev. *) ‚n. ev‘ = not evaluated so far Requested 0, 5% X 0 per detector achievable B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 11
Detector Structure Basic design concepts Meshing Maximum deformation (mm) Ux Uy Uz Ures ± 1, 5 ± 0, 24 12, 7 Design input: ! Minimize material budget No mechanical backing structure Symmetric frame layout FEM simulation: Results: Planar stretching forces by foils Much better than half-disk layout Deformations manageable Still needs some support e. g. point-like on outer rim Open questions: Sagging of foils Spacers (e. g. fishing lines opposing foil layer) B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 12
GEM-Foil Sectioning Sector side Discharge prevention Triple GEM setup (per side) Asymmetric gain sharing Segmented foils Requirements: ! ! Avoid dead areas gluing to copper ! Avoid ! B. Voss et. al. Max. 104 mm 2/Sector Easy contact scenario including series resistors Layout: 2 x 5 µm Copper on 50µm Kapton Inter-sector distance 0, 1 mm Combine ¼-circles & strips Contact from window frame (surface) via trough-hole SMD connectors GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 13
Planar GEM-Trackers Beam pipe… Simulation results@15 Ge. V/c Inner | Outer area… Mean 1, 5 mm ‘Standard’ physics run 2· 10 7 annihilations/s Mean 0, 15 ° Structure size Ralf Kliemt (TU-Dresden) confirmed by Radoslaw Karabowicz (GSI) B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 14
Pad. Plane Basic design concepts Simulation results: HIT-rate 5. . 50(140)k particles/cm 2/s (r) Track lengths: Cluster sizes <1 mm (single-cluster HIT to be avoided) radial 0, 5. . 8 mm angular 0. . 0, 8° Purely resolution driven (mean 1, 3. . 1, 6 mm) (mean 0, 15°) Patterning under investigation structures 2 D, pitch 400µm, 2 combined on every side Strips 1. Circular/polar + Radial 2. Circular/polar + 60° Tilted 3. Cartesian 1. Pixels – regular polygon shapes 2. Hybrid readout structures B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 15
Pad. Plane Pattern design Circular GEM 1/2 3 Radial ‚acentric‘ ‚isocentric‘ 5888 (1571) 9600 (1571) 4 60°Tilted Cartesian 5880 (2096) 9600 (2096) 13440 (1571) 13400 (2096) Required for constant resolution, Assumed in simulations Problems with Patching (GEM 4) Patching process Signal routing B. Voss et. al. Dead areas & Increased material budget (Support & Additional routing layers) Cut channels unequal length, non-uniformity in response GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 16
Front-End Electronic Pad Planes General ideas Structure size (3. . 4 GEM-Ts) Resolution evolution 0. 3. . 0. 5 mm 1 mm decreasing 20. . 26 kch constant 96. . 116 kch constant 80. . 95 kch Channel no. FEE system (to be used with TPC & Trackers) ‘Division bar’ not sufficient nor feasible Circumferential arrangement FEE-Ring ≈ 7. . 11% of total detector area width ≤ 50 mm (n-)XYTER-based FEB cards 180. . 900 cards (2 ASICs à 128 channels each, 100 x 65 mm 2) è overall operating power 21 m. W/channel 2, 7 W/chip ! è≈ 1. . 5 k. W power/cooling requirements, Axial cooling structure, ≈ 30% of weight B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 17
Front-End Electronics • AMS CMOS 0. 35 mm technology • 128 channels / chip • Charge-sensitive preamp (n-)XYTER • Time-stamping with 1 ns resolution • Data driven, autonomous hit detection • Token ring readout @ 32 MHz de -randomizing, sparcifying • Fast (30 ns) & Slow (150 ns) Shaper • Expected noise: • Peak detector 370 e-@10 p. F, 550 e-@20 p. F B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 18
Front-End Electronics XYTER testing board Simple hybrid PCB with signal fan-in ADC interconnect to DAQ chain (Sys. Core / KIP, Exploder / GSI) CBM beam time September 2008: whole signal chain operative ‘Chip-In-Board’ avoids space eating vias allows pitch adaptation: B. Voss et. al. 50, 7 μm on chip 101, 4 μm on PCB (two levels) 10 Rev. C boards, fully functional GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 19
Front-End Electronics XYTER TPC Prototype board Gold plated edges B. Voss et. al. 2 XYTER chips / card chip-in board glued to aluminium backing step-down & staggered bonding ~350 2 XYTERs 4 -layer PCBs for PANDA TPC prototype 2 x 128 300 ADC 42 channels / card pins / connector Watt power dissipation passive heat pipe system cooling liquid HFE-7100 (3 M) PCB lab-tested awaiting in-beam test (@CB-ELSA) Higher Power regulators integration level (1: 1): Sys. Core Boards (KIP Heidelberg) Exploder+… Others …? (GSI-EE, MBS) GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 20
Experimental data Energy-Output (preliminary) Sr 90 Energy [ADC digits] 60 ke. V Am 241 ~2/3 MIP Photo escape? Pedestal ? B. Voss et. al. Si Detector 300μm strip-pitch DC coupled 40 V ‘as is, on the table’ Pulse-height spectra on one strip Enhanced low-energy part: -electrons & other scattering GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 21
Experimental data Energy-Output (preliminary) Sr 90 Pulse-height spectra Energy [ADC digits] 60 ke. V ~2/3 MIP Am 241 Pedestal ? Photo escape? B. Voss et. al. with internal test pulse Cdet not determined (5 -10 p. F) Peak-to-peak distance ~5560 e- σ ~425 e- (370 e-expected) Analog read out chain operative Temperature coefficient spoils the effort GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 22
Front-End Electronics n-XYTER Engineering Run Preparations by Hans Kristian Soltveit Several thousand chips@110 k€ Issues addressed: Temperature coefficients on three amplifiers Pad arrangement, input-ESD-pads, LVDS in/out arrangements Shielding (in particular mono stable cross-talk) Choice of process Epoch marker output (Physikalisches Institut Heidelberg) Time line (03/2009): Current: End of April: Finalize schematics & Review on modifications May & June: Layout modifications End of June: Submission readiness review & Successive submission to AMS B. Voss et. al. Extensive corner analysis on new schematics GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 23
The Future: FAIR-XYTER …for gaseous detectors GEM-Tracker Channels/chip GEM-TPC 64 -128 32 -128 Power limit/channel 10 m. W Noise limit 500 e-@5 p. F Max. rad dose 7 p. F Max. det. capacity Ampl. 7 p. F 4. . 11/200 k. Hz 250 k. Hz 25 ke > 25 ke 94 ke Measured feature Max. possible signal Spatial resolution + Hit-time 30. . 40 ns 200 ke = average*2 750 ke=average*30 1000 ke, 3000 ke, Dual range 7 bit, 5 ke 8 -9 bit linear, 4 ke 10 ns/hit 4 ns Spark protection + baseline restoration +forced neighbor readout ADC res. /ampl. B. Voss et. al. Landau Signal shape Special tasks 7. . 150 p. F 2. . . 150 p. F Signal distribution Time resolution 500 e-@5 p. F, 900 e-@25 p. F 100 krad Avg. det. capacity Rate/channel Others GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 24
Detector Control & Monitoring Structure Current concept: Modular, expandable, flexible Same structure for all GEM-Detectors Hierarchical interlocking Failsafe operation Hard- & software limits Integration into PANDA-DCS database started B. Voss et. al. Identified 45 parameters so far GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 25
Detector-Mounting B. Voss et. al. ‘Riddle’ design Optimize position determination Maximize inter-detector distances Equalize weighting of information Balance inter-detector distances Maximize shape-conformity Symmetrise the loads, moments and thicknesses Minimize mounting and adjustment efforts Ì Simple & stiff support structures Multi-conical shape of the supports (‘Matroschka’ principle) Shell structure (0. 5. . 1 mm CRP) 50µm sag GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 26
GEM-Projects Road map FP 7 -JRA WP 24 (605 k€ granted) 2009 GEM-TPC, GEM-Tracker Tasks / TPC-GEM Task 1 2 3 2010 4 1 2 3 2011 4 1 2 3 2012 4 1 2 3 4 Precursor-Prototype design / fabrication / test Development of thin large area GEM foils Development of large read-out structures ASIC (and FEE) adaptation and optimization Quality control and calibration Material research TPC field-cage study and optimization Light-weight frames and support structures Read-out and DAQ Detector assembly and integration PANDA Time horizon: 2016. . 2018 B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 27
GEM-Projects: TPC 1 & Tracker 2 GSI crew-members & tasks Jörg Hehner 1 Andreas Heinz 1 Markus Henske 1 Radoslaw Karabowicz 2 Volker Kleipa 1 Jochen Kunkel 1, 2 Rafal Lalik 1 Christian Schmidt 1 Sandra Schwab 1 Daniel Soyk 1 Eduard Traut 1 Ufuk Tuey 1 Bernd Voss 1, 2 Jan Voss 1 Joachim Weinert 1 Aging tests Pad. Plane, GEMs, sensors, Web. Info Material tests, sensors, infrastructure, purchase Root-Simulations Front-End Electronics (XYTER) Mechanics, drawings, simulations, assembly Front-End Electronics (XYTER) Part production, tooling , FOPI environment FEM-Simulations GEM generals General mechanics, drawings ‚All & nothing‘, ideas & concepts, project & logistics General mechanics, material tests Part production, tooling
Backup slides
PANDA@FAIR Physics goal Precision spectroscopy: mass, width, branching ratios Charmonium Gluonic excitations above 2 Ge. V/c 2 Charmed meson properties in nuclear medium Double hypernuclei production and spectroscopy Electromagnetic processes electromagnetic form factors Time-like Generalized parton distributions Transverse spin distribution in nucleons Open charm physics Spectroscopy Rare decays CP violation in charm sector B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 30
TPC Layout @ PANDA B. Voss et. al. Basic Properties GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 32
Event Mixing • 2· 107 annihilations / s • 5 charged tracks / event • v. D=3 cm/ms, L=150 cm t. Dmax=50 ms 5000 tracks superimposed in one TPC “picture” tracks mixed in time B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 33
PID Performance PID Processes: • Cherenkov radiation: above 1 Ge. V radiators: quartz, aerogel, C 4 F 10 • Specific energy loss: below 1 Ge. V TPC B. Voss et. al. DIRC best accuracy with TPC • Time of flight • EMC for e and γ GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 34
Cooling Motivation Backside view Simulations Parameters: NXYTER(2) + ADC + voltage regulators on board FR 4 + Alu backing, sided mounting 10°C cooling media temperature 25°C backflow temperature Results: B. Voss et. al. single Chips reasonably cool GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 35
Analogue Pulses, Peaking Time, Front-End Noise FAST channel SLOW channel ENC 26. 9 e/p. F + 200 e 12. 7 e/p. F + 233 e peaking timea (1% to 99%) 18. 5 ns 139 ns Engineered for 30 p. F, giving 1000 e 600 e power consumption: 12. 8 m. W per channel; OK for neutrons! B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 37
System Integration Issue Sys. Core generates 48 to 64 bit data elements: n-XYTER provides 14 bit time stamp @ 1 ns resolution 7 bit channel number FEB provides 10 bit peak height Sys. Core adds Global Timing chip number Relate local time stamps to global events: Feed a global epoch marker or event strobe into Sys. Core epoch counter some diagnostic Sys. Core provides data FIFO The serialized event number with successive time stamp could be fed into auxiliary inputs. This event header would be added to the data elements Such synch scheme should be fine up to about 50 k. Hz event rate B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 39
Test Setup Complex Multi-layer Data Chain Detector n-XYTER-FEB n-XYTER FEB and Bonding Technology Documentation (Manual) ADC Interconnect Sys. Core / Exploder Firmware Ongoing: Embedded software Soft configuration Realize software design freeze to be packaged Ethernet-Interconnect Still missing: PC and DAQ Diagnostic toolbox for system analysis to make successful deployment in other labs feasible KNUT GUI DAQ System B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 40
The ‘Riddle’ Deformation under load Conservative assumptions worst-case material constants Max. deformation quite small Local reinforcement still possible Reasonable concept B. Voss et. al. GEM Detector Projects@GSI-DL RD 51 mini week, CERN April, 28 th 2009 41
- Slides: 38