K. Akiba on behalf of the VELO and UT groups

Size: px

Start display at page:

Download "K. Akiba on behalf of the VELO and UT groups"

Kathryn West
6 years ago
Views:

epoch of incredulity, it was the season of Light, it was the season of Darkness, it was the spring

1 K. Akiba on behalf of the VELO and UT groups It was the best of times, it was the worst of times, it was the age of wisdom, it was the age of foolishness, it was the epoch of belief, it was the epoch of incredulity, it was the season of Light, it was the season of Darkness, it was the spring of hope, it was the winter of despair, we had everything before us, we had nothing before us C. Dickens

2 LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 2

3 HCAL ECAL RICH2 Outer Magnet Tracker TT VELO Muon + Trigger Hard & Soft LHCb Upgrade: Strips and Pixels -- ECFA 2016 Inner Tracker Kazu Akiba RICH1 3

4 TT VELO LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 4

5 Quick Reminder LHCb Upgrade is planned for LS2. The statistical reach is limited by the current readout scheme. New front-end would make a more efficient trigger which can also cope with higher instantaneous luminosity 4x10 32 cm -2 s -1 à 2x10 33 cm -2 s -1 The LHCb Upgrade increases the luminosity by a factor 5, but the readout by a factor 40! Survive 50 fb -1 LHCb upgrade status and outlook Chris Parkes 3/10 morning session. Heavy flavour physics at high luminosity LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba Niels Tuning 4/10 morning session 5

6 Current detectors

7 The current Vertex Locator (VELO): Operates in vacuum CO2 Evaporation of CO2 keeps the temperature stable. 88 silicon sensors R-φ design 300 μm, n+n Si 2048 strips ractive = 8.2mm pitchmin = 40 μm Moves away every fill and centers around the beam with self measured vertices Separated from primary vacuum by thin RF foil with complex shape Protection from beam pickup Kazu Akiba Main task: find Primary (PV) and secondary Vertices. LHCb Upgrade: Strips and Pixels -- ECFA

Current Tracker Turicensis (TT) xuvx station layout: ± 5 2

Expected < 5x10 14 1 MeV n eq max hv 500V 8.

Also retractable For installation and servicing during

8 Current Tracker Turicensis (TT) xuvx station layout: ± 5 2 stations 50 um hit resolution. Expected < 5x MeV n eq max hv 500V 8.4 m2, 144 k channels. Also retractable For installation and servicing during shutdown 500 µm thick p-in-n silicon 512 strips 10 cm long 183 µm pitch Bonded strips for longer detectors LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 8

9 Upgrade strategy: Closer, lighter, faster, colder and more segmented Going closer to the beam axis improves extrapolations and increases coverage. The higher occupancies require higher granularity to keep the performance. VELO Upgrade and Upstream Tracker (UT) will have several changes in: Sensors è Pixels (VELO), strips with custom shapes (UT) New ASICsè Velopix and SALT (Silicon ASIC for LHCb Tracking) Module constructions è Double sided modules (VELO) and staves (UT) Readout and signal chain Active Cooling for sensors and FE è Evaporative CO 2 for both (shared plant) in encapsulated pipes (UT) microchannels (VELO) LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 9

10 VELO System overview Current detector Half Upgrade detector Half 44 C shaped double sided modules (R/φ) è88 sensors. 52 L shaped pixel modules è208 sensors. Keep vacuum and motion features High speed kapton cables and feed throughs LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 10

UT system overview Silicon strip sensors on

Beam pipe high segmentation at low θ, cutout

the sensor transmit zero-suppressed digital

11 UT system overview Silicon strip sensors on double sided staves Staves staggered in Z overlap in X direction Every strip wire-bonded to an ASIC and routed out. Beam pipe high segmentation at low θ, cutout for beam Integrated FE electronics located at the sensor transmit zero-suppressed digital signals LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 11

12 Sensors

13 VELO sensors Planar silicon, n-in-p (n-in-n under consideration) Tile for 3 ASIC chips: ~ 43 x 14 mm, 200 µm thick 55x55 µm 2 pixels, elongated pixels at ASIC boundaries Non homogeneous irradiation sets constraints on guard ring design, HV tolerance Tip close to beam: 8 x n eq, far corner only at.0.8 x n eq guard ring width ~450 µm Bias of 1000V after 50 fb -1 ASIC sensor ~43mm ASIC Longer pixels ASIC ~15mm Sensor tile factor 1/100 factor 1/7 factor 1/20 LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 13

µm 450, 250, 150 µm 39 µm 35 µm 450, 600 µm HPK Sensors Singles : Triples Dashed curve: IRRAD Solid

14 VELO Sensors Micron n-on-n Micron n-on-p Hamamatsu n-on-p HPK: 2 GR design Micron: n-in-n with GR on the back Thickness 150 µm 200 µm 200 µm Implant size 36 µm 36 µm 39 µm Guard ring pixel to edge 450, 250, 150 µm 450, 250, 150 µm 39 µm 35 µm 450, 600 µm HPK Sensors Singles : Triples Dashed curve: IRRAD Solid curve: KIT no breakdown Resolution Charge Collection LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 14

on June 2015 Key technical aspects: x900 x50

Top-side biasing via wirebonds PRR on June

15 UT sensors 320 µm thick, p-in-n, 190 µm pitch 250 µm thick, n-in-p, 80 µm pitch EDR on June 2015 Key technical aspects: x900 x50 x16 Detectors with circular cutout to maximize acceptance near the beam pipe Built-in pitch adapters from 190 µm to 80 µm Top-side biasing via wirebonds PRR on June 2016 LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 15

16 UT Sensor evaluation Type D sensors, S/N: >16 before irrad, >10 at max fluence No inefficiency near cutout region Type A sensors Preliminary results from May 2016 test beam Primarily 320 µm p-in-n sensors, and half width-a Irradiated up to 4x10 13 MeV n eq /cm 2 at CERN IRRAD and MGH S/N ~13, consistent with expectation Half-A p-in-n 320 µm Half-A n-in-p 250 µm LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 16

17 Front-end

Velopixà320 Mtracks/s VELO 900 Mhits/s peak rate Digital processing: SALT: 128 channels: ampàshaperàadcàpedàcmsàzsàtx.

18 Front-end challenges Average rate: multiply by 27 MHz Smart and fast ASICs High radiation enviroment: Same technology (130 nm TSMC) SEU protection: triple redundancy High output bandwidth in the VELO: 20 Gbit/s UT: data through 5 x 320 Mbps e-ports Hottest Velopixà320 Mtracks/s VELO 900 Mhits/s peak rate Digital processing: SALT: 128 channels: ampàshaperàadcàpedàcmsàzsàtx. Velopix: binary, time-stamped, data driven, 2x4 super pixelsà30 % data reduction SALT ASIC Block diagram LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 18

VELOPix ASIC Technology (130 nm TSMC) resistant to required

! Key features Data driven readout: Each hit time-stamped,

Binary readout but optional Amplitude (slow) for calibration

LHCb Upgrade: Strips and Pixels -- ECFA 2016 25 ns Time walk

19 VELOPix ASIC Technology (130 nm TSMC) resistant to required ~400 MRad. Velopix recently produced!! Key features Data driven readout: Each hit time-stamped, sent off chip immediately Fast front-end: time walk < 25 ns Binary readout but optional Amplitude (slow) for calibration Chips will be thinned to 200 µm: è minimize material Velopix LHCb Upgrade: Strips and Pixels -- ECFA ns Time walk measured with Timepix3: lower signals are slower. Velopix has a very similar analog design Kazu Akiba 19

20 Velopix tests (FRESH!) Counts Threshold=0x0F Threshold=0x0F fit Threshold=0x0 Threshold=0x0 fit Threshold equalized DAC Code 5800 Equalisation result DACCode Matrix before equalisation Test pulse pixel gain variation (32 px) Counts DACCode y = x R² = Energy [Ke-] Test pulses in 1 pixel threshold scan results Very preliminary test results look great, more detailed tests ongoing LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 20

21 SALT: Silicon ASIC for LHCb Tracking Each SALT reads 128 strips. More than 5000 are needed. Prototype versions with 8 channels produced in SALT128 submitted. Sensor capacitance 5-20 pf, AC coupled Noise: ~1000e - at 10 pf + 50e - /pf 40 MHz readout: shaper T peak 25 ns, <5% after 2 T peak Power consumption ~768mW/ ASIC LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 21

22 SALT Tests and validation Two 8-channel SALT versions produced Tests of ADC, DSP Noise performance matches expectation Data packet format validated Successful communication with GBT Laser test is a full system test of the salt chip with ECS and with Sensor! Laser pulsed between channels 3 and 4 CM in chip Gain curve is symmetric CM offline LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 22

23 Modules

24 Common conceptual construction The detectors are assembled in a modular way. Electronic signals are sent from and to hybrids built on flex circuits. Off detector electronics communicate to the back-end electronics through optical links. Cooling is tricky, but both use evaporative CO2 LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 24

25 Velo Upgrade Modules Double sided module design with 2 sensor tiles on each side. Electronic hybrid to be attached to cooling substrate. LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 25

26 Velo Upgrade Modules Double sided module design with 2 sensor tiles on each side. Electronic hybrid to be attached to cooling substrate. LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 26

27 Module prototypes First prototype produced May 2016 ~ 1 week production All steps identified, jigs produced LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 27

28 UT Modules Fixtures ready to assemble mockup modules and mount them on instrumented staves LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 28

2 cooling tube Procedures defined for aligning, mounting, and wirebonding

29 UT Staves thermal and structural foam core sandwiched between carbon fiber sheets Each stave supports up to 16 hybrid modules, 4 flex cables, single CO 2 cooling tube Procedures defined for aligning, mounting, and wirebonding hybrids and flex cable LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 29

trimming: target precision of foam element

validate construction process Kazu Akiba

30 UT Stave Construction Foam core pieces glued to backing, cooling tube glued into milled trough in foam core Carbon fiber backing held in vacuum fixture Metrology, trimming: target precision of foam element positions a few hundred µm, currently at 0.5 mm LHCb Upgrade: Strips and Pixels -- ECFA 2016 Two bare staves assembled to validate construction process Kazu Akiba Second backing glued to assembly, aligning vacuum fixtures 30

3 UT Flex Circuits Hybrid: flex circuit, 4 or 8 ASICs Sensor wire bonded to hybrid, wire-bonded to flex cables. data, clock, control, LV, HV, Thermal conduction: 0.

21 24 45 22 48 49 73 52 46 76 50 74 27 27 23 1 5 9 13 47 17 25 29 33 37 51 75 41 56 60 64 68 72 77 81 85 89 2 6 10 14 18 26 30 34 38 42 55 59 63 67 71 78 82 86 90 94 3 BGA

31 3 UT Flex Circuits Hybrid: flex circuit, 4 or 8 ASICs Sensor wire bonded to hybrid, wire-bonded to flex cables. data, clock, control, LV, HV, Thermal conduction: 0.8W/ASIC, 3.2 or 6.4 W/module BGA connectors for output 3 types of flex cable accommodating various sensor configurations LHCb Upgrade: Strips and Pixels -- ECFA Kazu Akiba Signal integrity maintained through flex in realistic signal environment, BER tests ongoing 31

32 Cooling

within silicon substrate High heat transfer coefficients No CTE difference (Si on Si) Excellent uniformity of material in sensitive region Pressure

33 Cooling: micro channel substrate Challenge: Modules produce up to 30W, dominated by ASICs Sensors must be kept < -20 o C to minimize the effects of radiation damage, and to avoid thermal runaway Must avoid material especially at tip of module Chosen Solution: Evaporate CO2 in micro-channels etched within silicon substrate High heat transfer coefficients No CTE difference (Si on Si) Excellent uniformity of material in sensitive region Pressure tolerance of module is important Cross section of the module concept CO2 simulations Race track design Soldered connector LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 33

ΔT and well beyond necessary power Nominal Maximum

34 Microchannels under test Double Micro channel test setup Tested 2 samples Cooling performs with very low ΔT and well beyond necessary power Nominal Maximum Power LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 34

testing, open for picture TRACI pumps CO 2 at 1 g/s Dummy stave cooling tube heat loads, temperature

35 TRACI MulTipurpose Refrigeration Apparatus for CO 2 Investigation Traci-based recirculating prototype system with stave mock-up, nominal power/stave 75 W, at nominal operating conditions Cold box closed for testing, open for picture TRACI pumps CO 2 at 1 g/s Dummy stave cooling tube heat loads, temperature sensors Stave successfully cooled to LHCb Upgrade: Strips and Pixels -- ECFA 2016 (-20.5±0.5) C Kazu Akiba 35

36 Bonus

37 VELO Upgrade RF Foil Specific to the VELO è beam vacuum/em pickup separation. Material and fabrication: Aluminium (AlMgMn) µm thickness: By 5-axis milling of a single homogeneous block Chemical etching IP IP 3D x resolution [µm] [µm] Impact parameter resolution is improved by going closer to the collision point and less material LHCb simulation LHCb simulation 300 µm thick 300/200 µm thick, 44mm wide 300/100 µm thick, 44mm wide no foil /p [GeV c] -1 1/p [GeV c] T T LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 37

38 Upgrade RF Foil Prototypes A B C D A: Milling out of Al Block. B,C: matching of prototypes D: possible masking and etching with NaOH E E: gorgeous latest prototype LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 38

Timeline LOI FW TDR VELO TDR UT TDR Sensor

installation year 2010 2011 2012 2013 2014 2015

s -1 3-4x10 32 cm -2 s -1 4x10 32 cm -2 s -1

3 D -1 5 D -1 50 D -1 Current upgrade plan

39 Timeline LOI FW TDR VELO TDR UT TDR Sensor production module production Assembly installation year beam crossing 50 ns LS1 25 ns TeV Instant Luminosity LS cm -2 s x10 32 cm -2 s -1 4x10 32 cm -2 s -1 2x10 33 cm -2 s ns Integrated Luminosity 3 D -1 5 D D -1 Current upgrade plan still compatible with post LS3 LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 39

Summary We plan to install a fully upgraded detector in the LS2/2019 40 MHz triggerless readout Run @ L = 2 x 10 33 cm -2 s -1 (5x now), but front-end 40 times higher rate.

40 Summary We plan to install a fully upgraded detector in the LS2/ MHz triggerless readout L = 2 x cm -2 s -1 (5x now), but front-end 40 times higher rate. VELO upgrade will consist planar silicon pixel detector Upstream Tracker will be silicon strips. New smart ASICs with zero suppression and high readout rates Evaporative CO 2 cooling as close as possible to the sensors Entering construction phase on schedule. It is a far, far better thing that I do, than I have ever done; it is a far, far better rest that I go to than I have ever known. C. Dickens LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba A tale of two cities 40

41 Back up

42 SALT Test Setup Test pulse can also be injected to laser generator Pulse width ~10 ns, height 1.3V, 1.56V Output width ~ 7 ns, energy deposition ~ 1 & 2 MIPs. LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 42

450 um guard ring Reduced width guard rings (150um) and overlapping guard rings on backside in n-in-n case n Irradiation

43 Silicon sensors Focus on two vendors with proven track record in radiation hard and HV tolerant sensor designs: HPK n n n n-in-p, 200 um thick 450 um guard ring 3x1 and single-asic sensors Micron n n n n-in-n or n-in-p, 150/200 um Baseline of 450 um guard ring Reduced width guard rings (150um) and overlapping guard rings on backside in n-in-n case n Irradiation and test beam program to validate Sensor and ASIC tech at high rates. LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 43

44 Sensor Production Sensor corners to be rounded to optimise foil clearance Implant width: 39 µm chosen better efficiency Y[mm] Intrapixel Efficiencies Micron 36 µm n-on-n 300V: Underdepleted X[mm] Mask detail Under Production at HPK Mask detail Under Design at Micron HPK LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 44

Th t 0 t 1 t 2 Time between hit and track T.

45 Timing Telescope has a very stable and precise time measurement. Single TPX3 measurement resolution measured to 1 ns and telescope down to 0.4 ns. Th t 0 t 1 t 2 Time between hit and track T.Evans Time to threshold Almost all hits below 25 ns Hit time residual amplitude LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 45

The Telescope our infrastructure. The Timepix3 telescope is fast, precise, and easy to combine with Timepix3 DUTs. The software is written in Gaudi architecture: Kepler. ~15k tracks/s.

46 The Telescope our infrastructure. The Timepix3 telescope is fast, precise, and easy to combine with Timepix3 DUTs. The software is written in Gaudi architecture: Kepler. ~15k tracks/s. Charge measurement and clustering give a great pointing resolution èdown to 2 µm. Precise time stamps make it simple and clean for the PatRec. Tracks can be measured with 0.4 ns timing precision We have operated up to 10 Mtracks/s/cm 2. No significant loss Correlated to rate. A.DosilSuarez LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 46

bay Detector Plane Detector Plane Plane ~4k ASICs Cavern PEPI Backplane Interconnections Data Concentrator Boards Master Control (TFC) ECS LV/HV

47 UT DAQ/ECS overview Each PEPI (Peripheral Electronics Processing Interface MCBs: Distribute TFC, ECS, reference clock DCBs: Read out data from and provide reference clock to SALT ASICs Low voltage regulated and distributed to peripheral and FE electronics by dedicated circuits in service bay Detector Plane Detector Plane Plane ~4k ASICs Cavern PEPI Backplane Interconnections Data Concentrator Boards Master Control (TFC) ECS LV/HV Conditioning LV Power Modules Counting room TELL40 TFC SOL40 HV Modules ECS Remote control LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 47

48 VELO DAQ/ECS overview LHCb common DAQ boards (TELL40) 48 copper links from chips. Electrical to optical conversion outside of vacuum tank à 20 Optical links ~5 Gbit/s each 1 FPGA to time order events of 1 module LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 48

49 Physics performance Technology choice made in may/2013. Pixels vs strips Microchanel Vs foam/tpg Simulations support the microchannel case LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 49

Main challenges for the Upgrade Non uniform Radiation exposure 8 x 10 15 n eq /cm 2 @ tip, 0.

50 Main challenges for the Upgrade Non uniform Radiation exposure 8 x n eq /cm tip, 0.1 x n eq /cm outer edge Sensor HV tolerance 1000V after 50 fb -1 Readout data rate Low Temperature operation ASIC power consumption Material budget ~33tracks/Event/module. (LHC: 40 MHz/25 ns) -20 tip close to the beam 3W/ASIC; up to 36 W/module; Good IP and tracking resolution currently: Proper time resolution ~ 50 fs IP resolution ~ 40 mm (p T =1GeV) LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 50

51 Current detectors, common characteristics Silicon micro-strip detectors. Same front-end: LHCb Beetle chip, 0.25 µm IBM technology. Same limitation: 1 MHz Readout. LHCb Upgrade: Strips and Pixels -- ECFA 2016 Kazu Akiba 51

PoS(VERTEX2015)008. The LHCb VELO upgrade. Sophie Elizabeth Richards. University of Bristol

PoS(VERTEX2015)008. The LHCb VELO upgrade. Sophie Elizabeth Richards. University of Bristol University of Bristol E-mail: sophie.richards@bristol.ac.uk The upgrade of the LHCb experiment is planned for beginning of 2019 unitl the end of 2020. It will transform the experiment to a trigger-less