R&D Towards a Linear Collider Detector

1. R&D Towards a Linear Collider Detector • DOE Site Visit Wednesday July 27, 2011 • Senior: Fadeyev, Schumm, Spencer • Students: Bogert, Carman**, Chappelletvolpini*, Cunnington, Gomez, Key, Khan*, Maduzia, Mallory, McFadden, Michlin, Mistry, Moreno, Newmiller, Ramirez, Schier**, Taylor*, Thompson* • * Senior thesis completed • ** Campus award for senior thesis

2. Areas of Activity • Generic Studies • Charge division • Long-ladder noise • Electronics Development • LSTFE readout • KPIX readout • Radiation Damage • High-dose electromagnetic irradiation • Simulation • Beamline calorimeter reconstruction • Non-prompt track reconstruction

3. Charge Division Can a longitudinal coordinate be measured with microstrip sensors? Explore with PC-board microstrip mock-up and PSpice simulation

7. Areas of Activity

8. Long-Ladder Readout Noise Probe conventional notions about dependence of readout noise on distributed capacitance and series resistance

9. Standard Form for Readout Noise (Spieler) Series Resistance Parallel Resistance Amplifier Noise (parallel) Amplifier Noise (series) Dominant term for long ladders (grows as L3/2) Fi , Fv are signal shape parameters that can be determined from average scope traces.

10. Expected Noise vs. Ladder Length “Lumped element” Load Series noise expected to dominate for narrow (50 m) pitch sensors above ~25 cm long

11. Sensor “Snake” Sensor “Snake”: Read out up to 13 daisy-chained 5cm sensors (with LSTFE-1 ASIC) LSTFE1 chip on Readout Board Can read out from end, or from middle of chain (“center-tap”)

12. Naïve Prediction vs. Observation “Lumped” Load Observed

13. Exploring Long-Ladder Noise Results To explore/understand difference between expected and observed, a full network simulation was developed in SPICE by Aaron Taylor ( UNM physics Ph.D. program) and Khilesh Mistry

14. Comparison with Full Network Model Full network simulation

15. Further Reduction: “Center-Tapping” Center-Tap Observed Center-Tap Simulated  NIM paper in preparation

16. The LSTFE Microstrip Readout ASIC Designed at SCIPP by Spencer, Schumm et al.

17. Time-Over-Threshold (TOT) Readout: the LSTFE • Pulse-development simulation  no loss of accuracy for TOT readout (relative to direct ADC conversion) • Targets low-complexity, long-ladder tracking solution • Real-time readout stream favorable for forward tracking also • LSTFE-I prototype relatively successful; LSTFE II under testing. Upgrades relative to LSTFE-I include • Improved environmental isolation • Additional amplification stage to improve S/N, control of shaping time, and channel-to-channel matching • Improved control of return-to-baseline for < 4 mip signals (time-over-threshold resolution) • 128 Channels (256 comparators) read out at 3 MHz, multiplexed onto 8 LVDS outputs

18. Li Hi Li+1 Hi+1 Li+2 Hi+2 Li+3 Hi+3 Li+4 Hi+4 Li+5 Hi+5 Li+6 Hi+6 Proposed LSTFE Back-End Architecture Low Comparator Leading-Edge-Enable Domain 8:1 Multi-plexing (clock = 50 ns) FIFO (Leading and trailing transitions) Event Time Clock Period  = 400 nsec

19. Some early results: TOT respone Time over Threshold (s) Time over Threshold (s) Very uniform response for large pulses; increased sensitivity in min-i region Minimum ionizing region

20. More early results: Noise v. Capacitive Load Result at 100 pF (optimization point): 1375 electrons noise, but without detector resistance (distributed RC network)

21. Use of the SLAC KPiX Chip for Tracking SCIPP charged with looking at KPiX in the one-few fC range (tracking regime)

22. 0-Charge Input Offset (mV) by Channel KPiX 7 Four mis-behaving channels (x10) Offset in mV for no input charge

23. Number of channels with occupancy greater than 0.1% Number of channels with efficiency less than 99.9% KPiX 7 Use SCIPP pulse-development simulation to assess impact of offset variation • Window of operating thresholds • Very narrow! Need to consider for further KPiX versions…

24. The ILC BeamCal: Radiation Damage Studies Reconstruction Studies

25. The Issue: ILC BeamCal Radiation Exposure ILC BeamCal: Covers between 5 and 40 miliradians Radiation doses up to 100 MRad per year Radiation initiated by electromagnetic particles (most extant studies for hadron –induced) EM particles do little damage; might damage be come from small hadronic component of shower? 25

26. Irradiation Plan • Use existing Micron sensors from ATLAS R&D • n-type and p-type • Standard float-zone and Magentic Czochralski • Runs of 0.1, 0.3, and 1 GRad for each sample • Runs with samples far from radiator (no hadronic effects) • Total integrated dose of ~10 Grad • Will assess the bulk damage effects and charge collection efficiency degradation. Sensors Sensor + FE ASIC DAQ FPGA with Ethernet 26

27. Charge-Collection Modularization Activity • Can connect multiplicity of Sensor Board modules to PMFE w/out wire-bonding • Requires development of 6 PCB/litho components (Donish Khan; accepted to Stanford Ph.D. program)

28. BCAL Simulation • Alex Bogert: • Geometry and virtual segmentation • Overlay • Mean pair background subtraction • Developing high-energy electron pattern recognition

29. Average Deposited Energy per Layer Background Signal

30. Pair Background Energy Fluctuations

31. Non-Prompt Tracking with the SiD Explore performance via explicit signature: Metastable stau NLSP (Gauge-Mediated SUSY)

32. Reconstructing Metastable Staus w/ SiD • Gauge-Mediated SUSY • Large tract of parameters space as stau NLSP • Metastable (cstau ~ centimeters) is in cosmologically preferred region • Process is • with

33. Reconstructing Metastable Staus w/ SiD Started with: 5+1 layers for inside track 4 layers for outside track New result: Include VTX-only inside track

34. Measuring Staus with the SID • Stau sample: • 11.1 fb-1 of e+e- stau pairs with • mstau = 75 GeV • Ecm = 500;  = 90 fb • c = 23 cm • Background sample: • 5.3 fb-1 combined SM background

35. Reconstructing Metastable Staus w/ SiD Focus initially on rdecay = 22-47 cm… Reconstruct decays by requiring:     - Outer hit of inner trk on last VXD or 1st tracker layer     -  1 missing layer between inner & non-prompt trks    - Both tracks on the same side of the Barrel (in z)     - Tracks have a geometric intersection in the x-y plane And: When inside track has  1 Central Tracker Hit     - The sign of the track curvatures match     - Non-prompt track curvature larger than the primary Of 897 staus with 6cm < rdec < 47cm, 642 staus are reconstructed, of which 592 truth-match

36. Stau Reconstruction Efficiency Truth-Matched Staus

37. Signal to Background for 10 fb-1

38. Signal to Background (10 fb-1) pT of inside track #Prompt Tracks/event Good separation between signal and background for #prompt tracks/event and inside track pt  Require, e.g., fewer than three prompt tracks

39. Reconstructing Metastable Staus w/ SiD Started with: 5+1 layers for inside track; 4 layers for outside track New result: Include VTX-only inside track Next step: Try to use cal-assisted tracking to get three-hit kinks (dedicated recon-struction by Mallory, Michlin, Bogert)

40. Wrap - Up • Many projects underway: • Charge-division study published • Ladder noise study in preparation for publication • Electronics development proceeding • Simulation studies bearing fruit • New BeamCal contribution in full swing • Meaningful research experience for many undergraduates; 8 theses in past three years, including two campus-wide thesis awards

41. Backup Slides

43. Areas of Activity