1. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori1 HB TB2002 Some Results Shuichi Kunori Jordan Damgov 05-Nov-2002

2. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori2 HB Testbeam 2002 Dates: June 26-July 1 “ECAL” // July 24-July 31 HF // Aug. 01- Sep. 18 HB Goals (HB): Demonstrate 144ch working Demonstrate DCS going Source data vs GeV/ADC (beam: e /  /  / p) Muon signal in HO for muon trigger Eta dependence (attenuation) Eta dependence (timing) Pulse shape (needs TDC) Weight in Layer 0  start construction of Calibration Database Additional Goals (left over from 1999TB) Crack between wedges e/pi (resolution and linearity) Cerenkov light in clear fibers

3. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori3 HCAL+”ECAL” Layout Calibrate 4 wedges ’02. Check HO response as tail catcher and as muon trigger element. In ’03 use PPP to study 40 MHz beam and HE/HB transition region.

4. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori4 Testbeam Layout H2 beam ECAL (49 crystals) HB Wedges (2) (144 channels) HO (16 channels) Moving Table

5. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori5 “ECAL”“ECAL” Have a 7 x 7 crystal array in front of HCAL which can be moved for eta scan. Online get a usable resolution of 2.3 %.

6. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori6 HB 2 Wedges - 16  x 8  300GeV  -

7. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori7 Prototype HTR, DCC

8. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori8 Test Beam Results: Data  - 225GeV e - 100, 20,30,50GeV  - 300, 20,30,50GeV

9. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori9 Data and Software Data Files ROOT format All files (~300GB) are in the tape robots at CERN and Fermilab. Software for Analyses HTBDAQ_data http://flywheel.princeton.edu/~jmmans/HTBDAQ_data/ Provides data access methods to DAQ root files Simple C++/root scripts to analyze data. e.g. wire source analysis H2Reco http://home.fnal.gov/~jdamgov/h2reco/ Reads DAQ root file using HTBDAQ_data. Construct higher level objects, e.g. 1x1, 3x3, 5x5 ECAL or HCAL clusters in addition to raw ECAL and HCAL data, etc. Writes output root files V11 was used for quick data validation during September runs. Many data sets have been processed through v11d at Fermilab. V12 will be released soon. It includes calibration constants from the wire source data for September data runs.

10. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori10 Pedestals

11. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori11 Noise Level Pedestal RMS distribution In ADC count

12. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori12 Wire Source Scan Source scan uses QIE at 25 nsec to sum up to a D.C. current. Signal to noise is good (3 mCu). Assignment of calibration constant to tile is simple.

13. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori13 Source Histograms by HTR a bad case ADC count (in 3x mode)

14. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori14 Wire Source Some channels are noisier than others.

15. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori15 Gaussian Fit to y projection Fitted values are less sensitive to pedestal width.

16. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori16 Source vs Beam Data 5% e-e- -- Some muon points in  =3 were bad. Besides the bad points, the muon data show better agreement with the source data.

17. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori17 ADC  GeV

18. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori18 7 bits ADC ADC Codes 00 - 31 (0 - 20GeV) 32 - 63 (17 - 119GeV) 64 - 95 (104 - 614GeV) 96 - 127 (539- 3089GeV)

19. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori19 Dynamic Range Original Requirement: Jets: 15-20GeV for top reconstruction/rejection >3TeV for compositness & QCD Readout: Et Threshold 500MeV Max. E = 3TeV Noise < 200MeV/time slice ( <0.66 LSB) QIE(ADC): 7 bits (128 integer codes in non-linear scale ) 1-10000 counts equivalent in linear scale. 300MeV LSB 3TeV at the max scale Pedestal: Higher statistics for better mean value  before/after physics runs.  during abort gaps. Energy extraction:  Optical sum of layer 0 signal with others.  1 or 2 time slices for signal extraction. Resolution Energy offset Occupancy 0.9 LSB (TB2002)

20. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori20 Effect of Threshold on low E T jet and MET Electronics noise and occupancy define the threshold. >> aim at 0.5GeV/tower @ 10E34 Lower threshold is better! (S.Nikitenko) 20GeV parton jet @ 10E34MET (I.Vardanian) (SK Dec 2001)

21. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori21 MET for Signal Events with Pile-up and Tower Threshold RMS 35 GeV 37 GeV 42 GeV 44 GeV >.5 GeV > 1 GeV > 2 GeV > 3 GeV 36 GeV > 1GeV With 17.3 min-bias events No min-bias Tower = Ecal+Hcal GeV >> Not much pile-up effect with this resolution! >> Resolution gets worse as threshold increase. (SK June 2000)

22. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori22 E in single HCAL readout >3TeV jets 2.9fb  290 events/year at 10E34. Fraction of events above E threshold (%) QCD binsMC evts 1.5TeV2.0TeV2.5TeV3.0TeV3.5TeV4.0TeV 80-12010001.400.600.300.00 2600- 3000 200016.653.550.700.200.050.00 3000- 3500 200028.857.401.750.300.100.05 3500- 4000 200046.0518.155.700.900.150.00 4500- 5500 5064.046.0020.006.004.000.00 Need to cover up to 3TeV? YES. (J.Damgov)

23. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori23 Signal in Time Slices 20 time slices at TB. (5~10 at CMS) No synchronization between beam arrival time and QIE clock at TB. (Rohf)

24. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori24 GeV/ADCGeV/ADC 60MeV ~ 240MeV / 1 ADC count  0.6TeV ~ 2.4TeV / Full range (1 time slice)  1.2TeV ~ 4.8TeV / Full range (2 time slices)

25. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori25 HPD Quantum Efficiency x x x xx HPD QE520

26. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori26 Adjust HPD Gain For 3TeV 1 time slice : 300MeV/ADC, noise=270MeV 2 time slices: 150MeV/ADC, noise=212MeV 200 280 250 350 1.Change HV 2.Lower gain HPD in HE

27. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori27 Eta dependence attenuation

28. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori28 Variation in eta Eta 1/attenuation1 20.947 30.947 40.918 50.904 60.887 70.847 80.862 90.831 100.825 110.786 120.756 130.769 140.730 150.618 160.605 2.5TeV ….. 1.8TeV 3.0TeV ….. 2.2TeV 4.1TeV ….. 3.0TeV

29. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori29 Pulse Shape

30. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori30 Pulse Shape Simulation (S.Abdullin)

31. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori31 Scope Measurement (Elias/Rohf)

32. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori32  225 E i /E all t i -t ave  300  20 e100 e20 Need correction for 25ns/30ns! (Damgov)

33. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori33 Eta dependence timing Coming soon.

34. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori34 HO for Muon Trigger Analysis in progress at Tata.

35. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori35 Layer 0

36. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori36 Layer 0 Goal: 1.Overweight 1.5 or 3.0? 2.Demonstrate how to implement. Wt ~ 1.0 Implemented! (de Barbaro) TB 1996 and MC indicated ~1.5! Need real ECAL super module to determine final weight.

37. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori37 Crack Scan (between wedges)

38. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori38 Wire Chamber Wire chamber data useful after Run 2698. (R.Vidal)

39. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori39 W1-W2W1-W2 xx yy  ~ 0.6mm  ~ 0.5mm Ready for gap study!

40. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori40 Crack/Tower Scans for HCAL

41. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori41 ConclusionConclusion QIE dynamic range. Not much head room for different input (HPD) gain. Need to control input gian by changing HPD HV? QIE calibration Need to calibrate pedestal and gain (full range) for each CapID. Source calibration Need to calculate constants for each CapID (and then average over 4 CapID) May require higher statistics (because of 4 caps and noise). TB data analysis. Extract results for all defined goals. Finish by March 2003, and publish. TB 2003 New goals are suggested.

42. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori42 Additional

43. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori43 Wire Source Data

44. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori44 Peds R3378 18Sep02  mean = mean(4)-mean(cap) rms (cap) mean(4)  = 1 2 3 4 5 6 7 8 (J.Dagmov)

45. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori45 Peds R806 9Aug02

46. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori46 Source/LED-Laser/BeamSource/LED-Laser/Beam

47. HCAL TB2002 Resuts @ CPT Week, CERN, 05-Nov-2002, S.Kunori47 HB TB 2003 (my suggestion / PRS) Goal: Local synchronization. (time in all channels.) Operation of full calibration system, including laser. Emulation of pedestal estimates (4 caps) in the LHC environment. 1) Before and after physics runs. 2) During abort gaps. Remote monitoring and analysis. Preparation: (addition to tasks obvious to get the goal) Before the TB period HPD calibration (all) – QE QIE calibration (all) – gain in full range for each CapID Source calibration E-log book.