F9 weekly Ljubljana Slovenia 07 12 2018 Bojan Hiti Jožef Stefan Institute Ljubljana Slovenia October 2018 Test beam campaign Test beam at CERN SPS H6 10 24 October 120 GeV hadrons ID: 1039771
Download Presentation The PPT/PDF document "Beam test results with BCM' and TowerJaz..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
1. Beam test results with BCM' and TowerJazz CMOSF9 weekly, Ljubljana, Slovenia, 07. 12. 2018Bojan Hiti (Jožef Stefan Institute, Ljubljana, Slovenia)
2. October 2018 Test beam campaignTest beam at CERN SPS H6 10 – 24 October, +120 GeV hadronsTwo separate projects:PART 1: BCM' prototype readout chipPART 2: TowerJazz CMOS investigatorKartel beam telescope 6 mimosa planes + 1 FEI4 module< 5 μm tracking resolutionDRS4 readout of DUTs, few channels7. 12. 2018F9 weekly2
3. CERN H6 test beam10 – 17 October, CERN SPS H6 test beam: 120 GeV hadronsMimosa beam telescope6 mimosa planes for tracking1 FEI4 module for timing and region of interest (ROI) triggerTracking resolution ≈ 3 μmDRS 4 analog readout3 GS/s acquisition rate, 700 MHz bandwidthRecorded one channel per chip per run – two channels per runDataset:All channels measured (separate runs)200k telescope triggers per channel 60k tracksAll tests done with one version of front end configuration+ 1000 V bias voltage7. 12. 2018F9 weekly3DUT3 planes + timing + ROI3 planes
4. 7. 12. 2018F9 weekly4PART 1: BCM'
5. BCM' chipBCM': Readout ASIC prototype for ATLAS Beam Condition Monitor (BCM) upgrade for HL-LHCAnalog front end with 4 readout channelsCurrent amplifier, fast pulses (baseline restored after < 25 ns)Two BCM' wire bonded to a specially segmented diamond sensor Large pitch bump bonding also considered to reduce capacitanceChip configuration with external DACs via a Labview interface7. 12. 2018F9 weekly5[1, 4][1, 1][1, 2][1, 3][2, 1][2, 4][2, 2][2, 3]Chip 1Chip 2
6. Charge measurementAverage waveform of 1000 acquisitonsSelect an integration window around the peak [tmin, tmax] (typically [– 6 ns, 6 ns])Waveform subtraction to mitigate pick up – w.r.t. track positionSignal = Integral around the peak Hit criterium: signal > 4σ above noise7. 12. 2018F9 weekly6Noise spectrumbaseline offset+4σSingle event waveformcharge 20 mV(Landau tail)U (V)
7. Chip 2, channel 17. 12. 2018F9 weekly7
8. Channel 27. 12. 2018F9 weekly8
9. Channel 37. 12. 2018F9 weekly9zoom on next slide
10. Chip 2 channel 37. 12. 2018F9 weekly10
11. ConclusionsDebugged the software, now nice test beam resultsFinal analysis code uploaded to Git:https://github.com/bhiti/testbeam_analysis_drsThe plan is to create a Git repository with the "full" test beam software collected in one placeWorking on obtaining an eos partition to save the BCM' data7. 12. 2018F9 weekly11
12. 7. 12. 2018F9 weekly12PART 2: TowerJazz Investigator CMOS
13. TowerJazz Investigator7. 12. 2018F9 weekly138 x 8 active pixels10 x 10 pixel arraydummy pixelsCMOS prototype based on small collection electrode architectureOld (2015) analog only designBased on this prototype fully monolithic chips TJ MonoPix and TJ MALTA have been producedFully monolithic samples have reduced efficiency in pixel corners after irradiationAgenda: Measure efficiency with Investigator and correlate with E-TCTCaveat: can read out only one pixel per channelreadout 2 x 2 pixels (50 x 50 um2 pixel size) – low rate!
14. Test beamSample:1e15 neutron irradiated ROI: 1250 x 400 um2 (DUT 100 x 100 um2 – cca. 0.5 % area)Charge sensitive amplifier – relatively complicated charge calculationPick up was also presentBeam intensity was HUGE. Still working on reconstructing tracksFor now only spectra7. 12. 2018F9 weekly14Waveforms of 4 channelsSignal wf, 2 references, discardedCharge = signal – reference
15. Charge spectra7. 12. 2018F9 weekly15110k entriesexpect 130 full wfs per channeltracking t.b.d.
16. 7. 12. 2018F9 weekly16BACKUP
17. TriggerAcquisition triggered by the FEI4 moduleROI to trigger only on tracks through the active areaROI determined in a test run by self triggering on the DUT7. 12. 2018F9 weekly17Example ROI:(chip 1, ch. 4) + (chip 2, ch. 4)FEI4 ROITelescopeDRS4HitORROIread out FEI4
18. Waveform subtractionPick up has the same shape in all the channelsSubtracting Ch. 2 from Ch. 1 reduces "noise" by a factor of ≈ 2Additional selection has to be introduced for events when signal only in Ch. 27. 12. 2018F9 weekly18U (V)Noise before subtractionNoise after subtraction
19. Waveform subtraction7. 12. 2018F9 weekly19Blue wf = Black - redFine tuning of the phase between black and red might reduce the noise