Update on SBS Back Tracker GEMs R&D @ UVa

K. Gnanvo, N. Liyanage, V. Nelyubin, C. Gu, K. Saenboonruang, S. Sacher, X. Bai

SBS Collaboration Meeting, June 4 & 5, 2013

Outline

• Design and construction of SBS Back Tracker GEM • Characterization of SBS Back Tracker GEM Proto I • Last round of R&D and Modifications on the design • Other issues related to the GEM for SBS 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 2

Design and construction of SBS Back Tracker GEM

4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 3

SBS Back Tracker GEM Chambers

SBS Back Tracker Chamber Layout of the Super Bigbite Spectrometer (SBS) for GEp(5) Cross section view of Back Tracker 50 x 50 cm 2 Triple GEM module Entrance window foil Drift foil Frame_GEM Frame_GEM GEM foils Readout board Honeycomb support

SBS Back Tracker Module Design

 Triple GEM al a COMPASS • 3,2,2,2 mm for the drift, the transfer and the induction gap • 2D Cartesian readout on flexible Kapton board  Active area 50 × 50 cm • • 2 Replacement of the original 40 × 50cm 2 32 modules needed for 8 Polarimetry chambers  Wide GEM frames (30 mm) and readout board frames (74 mm) along x-axis • Better stretching, GEM HV sectors electrodes • alignment holes away from active area • Room for strips connectors  No protective resistors on the GEM foils • HV sectors accessible through contacts • External board for protective resistances  ZIF connectors replaced by Panasonic connectors • Compatibility with SRS Electronics • More robust and easy to operate 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 5

SBS Back Tracker Module Design

GEM foil (CERN PCB workshop) Support frame with spacers (RESARM Belgium) Flexible 2D readout board (CERN PCB workshop) Honeycomb support board (CERN PCB workshop) 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 6

Storage of the frames

Clean Room & Equipment for the assembly

Large area (3 × 7 m 2 ) Class 1000 Clean Room Storage of the framed foils Frames holder for cleaning in USB Glue dispenser Ultra sonic bath (USB) with demineralized Water 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 Tacky roller  dust removal 7

Construction of the prototype

GEM foil on the mechanical stretcher GEM in N2 box for leakage current test GEM foil glued to the readout board 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 8

Leakage Current Measurement: HV Test in N2

• We use an Iseg EHS 8060x_105 6 kV HV module with a Wiener MPOD high voltage crate. High voltage is controlled using a command line interface through an internet protocol. • For the test we use the fast ramp up mode of the supply with a rate of 1200 V/s. the power • supply is sampling the current on a millisecond scale and the trip occurs within a few milliseconds of current over limit. The leakage current drawn by the GEM foil is measured using a Keithley 6487 picoammeter , at a sampling rate of 120 ms read into the computer through a Labview interface and save the • readings into a text file. A GEM sector has a capacitance of approximately 2 nF; and the resistance engaged in the HV module is ~ 50 M  , once the stable voltage is achieved this resistance is shunted automatically within the supply). As a result, when we are setting the required high voltage of 550 V, the initial current is a couple of m A, then quickly drops and stabilizes to less 1 nA leakage current far better than the 5 nA requirement. 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 9

4/26/2020

Leakage Current measurement: HV Test in N2

• • • • Distribution of leakage current over all the 72 sectors 24 sectors per foil  3 foils  72 sectors 550 V in N2 for naked, framed foils and in chamber Criteria for accepting foil  Leakage current < 5 nA naked GEM foils ~ 0.55 nA Framed GEM foils ~0.68 nA SBS Coll. Meeting, June 4 & 5, 2013 GEM foils in Proto1 ~ 0.72 nA 10

APV25-SRS FE cards

SBS Back Tracker Prototype I

HV divider Spark protection resistors Board 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 11

High voltage divider with the protection resistance board

High voltage divider on prototype I Schematic of the voltage divider = 4100 V • • Standard COMPASS high voltage divider Compact available @ CERN PCB workshop 4.1 to 4.2 kV on the divider for Ar/CO2 (70/30) A more compact HV board for prototype II GEM 1: D V = 410 V GEM 2: D V = 373 V GEM 2: D V = 328 V 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 12

Characterization of SBS Back Tracker GEM proto I

4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 13

APV25-SRS FE cards on the GEM detector

APV25-SRS Electronics

SRS small system 1 ADC/FEC combo CERN ALICE DAQ (DATE/AMORE) ATX PS FEC board ADC board Raw data event on 1 APV25 FE  3 time samples 2D X/Y strips hits event on proto I 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 14

Gain study of APV25 electronics for SRS & MPD

Initialization parameters for the APV25 FE cards I pre = 98, I pcasc = 52, I psf = 34, I sha = 22, I ssf = 34, I psp = 55, I muxin = 16, V fpr = 30, V fs = 60, V psp = 40, MUX gain = 1 (October 2012) Operation of the SBS prototype: Ar/CO2 (70/30) with 4.1 to 4.2 kV on HV divider  Estimated gain 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 15

APV25-SRS Pedestal noise

Pedestal noise vs. number of time sample 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 16

SBS Proto I on Test bench with Cosmic

Hit map Cluster size (X-strips) = 4.6

• • • • • Cosmic data: 1.5M events @ HV = 4.2 kV  Gain ~ 4700 Cluster size = 4.7 Good charge sharing correlation, ratio = 1.3

Efficiency drop at spacers location clearly visible Cluster size (Y-strips) = 4.8

Charge sharing X/Y strips Cluster ADC sum distr. 4/26/2020 X ADC /Y ADC = 1.3

SBS Coll. Meeting, June 4 & 5, 2013 17

SBS Proto I with Cosmic: Gain uniformity

Average ADC counts • • • • Big efficiency drop in the vicinity of the spacers Spacer width of 300 m m leads to a drop over up to 2 mm HV sectors boundaries also leads to efficiency drop  but with lower loss and over a narrower strip Average ADC count per bin over a large number of events = Ʃ N (ADC counts) / N Hits Gain uniformity along X-strips X (mm) Gain uniformity along Y-strips 4/26/2020 X (mm) SBS Coll. Meeting, June 4 & 5, 2013 Y (mm) 18

• • • •

SBS Proto I on Test bench with

90

Sr

Hit map (N Hits per bin) Data taken with the 90 Sr for each of the 24 HV sectors of the GEM Average ADC count per bin = Ʃ N (ADC counts) / N hits , Good gain uniformity for of all sectors ~ 92 to 94 % efficiency for all the 24 HV sectors We should expect about 97% but a few % drops explained by the spacers 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 HV on divider (V) 19

Upgrade of the APV25-SRS Electronics @ UVa

• • • • 80 APV25-SRS FE cards: 10K channels & 6 ADC boards 5 FEC v6: More FPGA resources, DDR memory, dual SFP connector SRU: a 1U x 19" box serves as DTC links of up to 40 FEC cards and multiplexes event data via a single Gb Ethernet link to the DAQ computer. Provides I/O options for user-defined trigger and clock systems Sept 2013: Integration of the SRS Electronics into CODA (Post Doc @ UVa) 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 20

Upgrade of the APV25-MPD Electronics @ UVa

• • • • New version of APV25-MPD cards with Panasonic connectors (P. Musico, INFN Italy) New Back Planes with 5 slots  Compatibility with Back Tracker GEM readout (P. Musico, INFN Italy) We acquired the new system with 5 APV25-MPD cards, on 5-slots Back Plane and Latest version of the MPD VME board Panasonic connectors APV25-MPD FE cards 5-slots back plane 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 21

Last round of R&D and Modifications on the design

4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 22

Plans for Proto II: Reduce number of spacers in the active area

Proto I (February-March 2013) : Current spacer design gas window Drift foil GEM Frame Honeycomb support GEMs R/O board 4/26/2020 Proto II (will be built in June 2013): suggested spacer design gas window Drift foil GEM Frame Honeycomb support SBS Coll. Meeting, June 4 & 5, 2013 23 GEMs R/O board

4/26/2020

Plans for Proto III: New Gas distribution design

Proto I (built February-March 2013) : Current spacer design gas window Drift foil GEM Frame Honeycomb support GEMs R/O board Proto III (will be built in September 2013):

OR

SBS Coll. Meeting, June 4 & 5, 2013 24

On GEM foils: copper free area (1mm) around the edge

Current design • Changes We propose 1 mm clearance between the frame (inner part) and the GEM active area  Kapton foil with holes but no copper Proposed modification 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 25

On GEM foils: Contacts on bottom electrode (needed on GEM3 only) • Changes: Connect the bottom electrode of GEM3 to the ground to both suppress the induced signal (splash events multiplicity with additional contact strips to the bottom electrode of GEM3  Done by KLOEII GEM (INFN-LNF/Italy)

Courtesy D. Domenici INFN/LNF

4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 26

On GEM Foils and Frames: move gas holes away from the inner part Changes: Holes for the gas moved away from the inner part on the frames.

Adjustment done on the GEM foil as well Gas holes on the old frame design 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 Gas holes on the new frame design 27

Other issues related to the GEM for SBS

4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 28

Testbeam @ FermiLab FTBF (October 2013)

• FLYSUB is a consortium consisting of BNL, Florida Tech, Stony Brook University (SBU), University of Virginia (UVa), and Yale University working on the R&D for GEM-based • • • tracking and PID detectors for the EIC We’ve been approved for 3 weeks beam access at the Fermilab Test Beam Facility (FTBT) in October 2013  the consortium will bring more than 20 GEM chambers detectors (tracking, Cerenkov, TPC …) UVa will be bringing: • SoLID / EIC large GEM prototype, one 50 × 50 cm 2 SBS GEM chambers, 3 small (10 × 10 cm 2 ) GEMs + 1 Micromegas for the tracking The main goal for us is to measure position resolution SBS and SoLID GEMs (U/V readout) • and performances at high rate and in magnetic field.

Also an opportunity to test our newly acquired UVa medium size (10K) SRS electronics 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 29

Testbeam @ FermiLab FTBF (October 2013)

http://www-ppd.fnal.gov/FTBF/schedule/MTest14.html

4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 30

Proposal for JLab LDRD to build the Largest GEM chamber (FY14) • • • • CERN - RD51’s goal for 2014: Produce a GEM foil with an 2 × 0.5 m 2 active area Preliminary sketch of the 2D Cartesian readout Submitted a proposal to JLab FY14 LDRD for funding to build the largest (2 × 0.5 m 2 ) triple GEM chamber with 2D readout. (B. Wojtsekhowski & UVa) A lots R&D involved • “Re-Openable” detector • Big challenges for 2D readout with high spatial resolution and low occupancy • Big challenges for the construction of such a big size detector A GEM chamber would be a big breakthrough for tracking system in Hall A https://misportal.jlab.org/mis/apps/physics/ldrd/viewProposals.cfm?ldrdCycleId=1 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 31

Summary

• Design for the SBS Back tracker GEM chamber, successfully build the first prototype.

• Preliminary test in the lab shows expected performances were satisfied.

• A few small modification are been implemented to further improved assembly of the chambers and performances.

• Plan to build 3 to 4 more prototypes in the coming month to finalize the design and conclude the R&D phase.

• Acquisition of a medium size SRS electronics, upgrade our capability to 10K channels.

• Acquisition of new version of the APV25-MPD electronics, will start the test soon.

• Test beam at FermiLab in October 2013 to test SBS Back tracker prototype and rate capabilities of the APV25 electronics 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 32

Backup

HV test of the GEM sectors

• • • • • • We use an Iseg EHS 6 kV HV module in a Wiener crate, HV controlled through an internet protocol. Fast ramp up mode at a rate of 1200 V/s. The leakage current in the GEM is measured using a Keithley 6487 picoammeter, at sampling rate of 120 ms with a Labview interface and saved in txt file. HV GEM sector ~ 2 nF and with a resistance the HV module is ~ 50 M  , (once the voltage is achieved this resistance is shunted automatically within the supply). HV of 550 V, the initial current is a couple of m A, then quickly drops and stabilizes to less 1 nA leakage. We leave the HV for about 2 min and if no spark  sector is good Initial current with the HV ramping up and down

Recovering of a bad HV sector

Excess of glue leaked onto the sector during assembly  sector recovered after curing on N2 or at 50 degree First test after assembly Second test one day later 3 rd test two days later 4 rd test three days later

Cross Talk in APV25 Electronics

5 s cut Current version of the analysis (AMORE) code: Only single cluster event used for 2D hit maps, X-Y charges correlation etc … and minimal cluster size = 2hits For high amplitude hits, cross talk cluster survives a 5 s zero suppression cut  Event seen with 2 clusters and rejected Cross talk is suppressed at 7 s 7 s cut 36

(X Distribution of the relative position cluster1 – X cluster2 ) of two clusters in 2 clusters events Along X-axis Pics at 32, 88 and 120  cross talk effect Distribution of ADC counts ratio (ADC cluster1 / ADC cluster2 ) of two clusters in 2 clusters events Along X-axis Large ratio > 20 correlated to the cross talk pics Along Y-axis 4/26/2020 Along Y-axis Real 2-cluster events SBS Coll. Meeting, June 4 & 5, 2013 37

4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 38

Cost of the SBS Back Tracker GEM components

Frames RESARM (Belgium) GEM foils, R/0 From CERN (Switzerland) Additional drift cathode: $400 (CERN) Mounting of the Panasonic connectors on the readout: ~$600 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 39

Cost of the SRS components (in Swiss Francs 1 CHF ~ $1.1)

4/26/2020 Additional SRU box ~ 4000 CHF to handle up to 40 ADC/FEC cards SBS Coll. Meeting, June 4 & 5, 2013 40

Modification#4: 2D Readout board Honeycomb support Flexible readout board • adjust the length of the readout at the bottom board to the size of the honeycomb board • add alignment holes on the flexible readout board 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 41

Modification#6: Charge sharing and 2D readout strips width old design 40 x 50 cm 2 new design 50 x 50 cm 2 Top strips: 140 mm Bottom strips: 360 mm X/Y ratio = 1.104 

GOOD

4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 Top strips: 90 mm Bottom strips: 360 mm X/Y ratio = 1.34 

BAD

42

Testbeam @ FermiLab FTBF (October 2013)

 Electrons  Energy of beam: 10 GeV, Intensity: Single particles (if possible), Beam spot size: as small as possible  Pions  Energy of beam: > 4 GeV, > 20 GeV, intensity: single particles (if possible); variations 1k – 100k particles/ 4 sec spill, Beam spot size: as small as possible; about 1 cm2  Kaons  Energy of beam: > 13 GeV, Intensity: single particles (if possible)  Protons  Energy of beam: > 27 GeV, Intensity: single particles (if possible), Beam spot size: as small as possible 4/26/2020 SBS Coll. Meeting, June 4 & 5, 2013 43