BaBar Drift Chamber performance Markus Cristinziani Stanford Linear - PowerPoint PPT Presentation

Performance Background Upgrade HiLumi Extra BaBar Drift Chamber performance Markus Cristinziani Stanford Linear Accelerator Center Markus Cristinziani BaBar Drift Chamber performance 0 / 28

Performance Background Upgrade HiLumi Extra Outline BaBar Drift Chamber performance 1 Backgrounds 2 Electronics upgrade 3 Higher luminosities 4 Additional material for discussion 5 Markus Cristinziani BaBar Drift Chamber performance 0 / 28

Performance Background Upgrade HiLumi Extra Outline BaBar Drift Chamber performance 1 Backgrounds 2 Electronics upgrade 3 Higher luminosities 4 Additional material for discussion 5 Markus Cristinziani BaBar Drift Chamber performance 0 / 28

Performance Background Upgrade HiLumi Extra Drift Chamber design and operation Main tracking device in BaBar surrounding SVT 2.8m length, 1.6m diameter, 7104 sense wires 40 axial and stereo layers of 1-2cm hex cells Gas: 80:20 He:C 4 H 10 , 4000 ppm H 2 O vapor Operating voltage: 1930V 324 1015 1749 68 469 35 17.19� 202 236 551 973 1618 IP Markus Cristinziani BaBar Drift Chamber performance 1 / 28

Performance Background Upgrade HiLumi Extra Reconstruction Single hit resolution 125 µ m Momentum resolution, p T : 0 . 45% + 0 . 15% p T / ( GeV /c ) Tracking efficiency : > 95% matching with SVT tracks d E/ d x resolution ≈ 7 . 5% Markus Cristinziani BaBar Drift Chamber performance 2 / 28

Performance Background Upgrade HiLumi Extra Particle identification Drift Chamber K/ π Separation Drift Chamber K/ π Separation p < 0.6 GeV/c 50 K π 0 0.6 < p < 0.8 K π 100 B A B AR 0 0.8 < p < 1 K π 100 0 p > 1 GeV/c 1000 K π 0 -600 -400 -200 0 200 (arb. units) DCH dE/dx - dE/dx(K) Good K/π separation up to ≈ 700 MeV /c Provides confirmation of DIRC information Additional coverage outside DIRC acceptance Markus Cristinziani BaBar Drift Chamber performance 3 / 28

Performance Background Upgrade HiLumi Extra Tracking Mass difference as a function of reconstructed distance from IP S → π + π − K 0 Λ → pπ − 0.001 ) 2 M (GeV/c 0.0005 ∆ 0 -0.0005 -0.001 0 10 20 30 40 50 2 2 Decay Radius √ (X +Y ) Vtx Vtx Fits with DCH only are comparable to full tracking “Jumps” are due to material scattering uncertainty go to pseudoefficiency plot Markus Cristinziani BaBar Drift Chamber performance 4 / 28

Performance Background Upgrade HiLumi Extra Outline BaBar Drift Chamber performance 1 Backgrounds 2 Electronics upgrade 3 Higher luminosities 4 Additional material for discussion 5 Markus Cristinziani BaBar Drift Chamber performance 4 / 28

Performance Background Upgrade HiLumi Extra Background characterization – Motivation PEP-II plans to increase luminosity to 2 × 10 34 by 2007 Increasing luminosity means aging go to aging increased read-out time and deadtime larger occupancy potentially affecting tracking A set of background characterization runs were taken to study BaBar behavior as a function of beam currents and luminosity Used to extrapolate to future luminosities Markus Cristinziani BaBar Drift Chamber performance 5 / 28

Performance Background Upgrade HiLumi Extra Background characterization – Method 49 short runs were recorded during 16 hours in January 2004 with varying beam currents and configurations Quantities of interest (Drift Chamber current, occupancy and read-out time) are parametrized with a functional form Bkg = A + B I LER + C I 2 LER + D I HER + E I 2 HER + F L where [ I LER ] = [ I HER ] = A and [ L ] = 10 33 no LER*HER term as no 2-beam, non-collision data taken beam-beam interaction and trickle injection is added linearly in the extrapolations fitting is performed in a 3-step approach fit pedestal runs and extract A fit single-beam runs after pedestal subraction ( → B-E) fit luminosity runs after single-beam subtraction ( → F) 1-step - simultaneous fit to all parameters as cross-check Markus Cristinziani BaBar Drift Chamber performance 6 / 28

Performance Background Upgrade HiLumi Extra Background characterization – Result FCN A B C D E F No beam pedestal 56 0.606 +- 0.033 —0.0— —0.0— —0.0— —0.0— —0.0— Single beam LER 227 —0.606— 0.21 +- 0.07 -0.03 +- 0.05 —0.0— —0.0— —0.0— Single beam HER 117 —0.606— —0.0— —0.0— 4.76 +- 0.21 -0.96 +- 0.25 —0.0— No quadratic terms Single beam LER 227 —0.606— 0.166 +- 0.018 —0.— —0.0— —0.0— —0.0— Single beam HER 131 —0.606— —0.0— —0.0— 3.97 +- 0.06 —0.— —0.0— Lumi HER 2/3, LER scan 36 —0.606— —0.166— —0.00— —3.966— —0.00— 0.460 +- 0.015 Lumi HER max, LER scan 47 —0.606— —0.166— —0.00— —3.966— —0.00— 0.372 +- 0.012 Lumi LER 2/3, HER scan 139 —0.606— —0.166— —0.00— —3.966— —0.00— 0.393 +- 0.015 Lumi LER max, HER scan 116 —0.606— —0.166— —0.00— —3.966— —0.00— 0.525 +- 0.011 Lumi trickle, HER scan 118 —0.606— —0.166— —0.00— —3.966— —0.00— 0.555 +- 0.015 All good Lumi points 348 —0.606— —0.166— —0.00— —3.966— —0.00— 0.420 +- 0.007 Preferred result 0.61 0.17 0. 3.97 0. 0.42 Occupancy(%) = 0 . 61 + 0 . 17 I LER + 3 . 97 I HER + 0 . 42 L +0 . 21 I LER (beam − beam)+0 . 03 L (trickle inj . ) Single beam LER Single beam HER Lumi trickle, HER scan 6 Occupancy - Single Beam contribution (%) Occupancy (%) Occupancy (%) Entries Entries 1865 1865 Entries Entries 2559 2559 6 Entries Entries 2301 2301 Mean x Mean x 1067 1067 Mean x Mean x 588.1 588.1 Mean x Mean x 3523 3523 Mean y Mean y 2.092 2.092 Mean y Mean y 0.7881 0.7881 Mean y Mean y 2.98 2.98 2.5 RMS x RMS x 1905 1905 5 RMS x RMS x 463 463 RMS x RMS x 318.9 318.9 RMS y RMS y 0.9662 0.9662 RMS y RMS y 0.3099 0.3099 RMS y RMS y 1.193 1.193 0 0 1 1 0 0 0 0 1 1 0 0 0 0 0 0 0 0 5 0 0 1864 1864 0 0 0 0 2558 2558 0 0 0 0 2301 2301 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 2 10 10 4 3 10 1.5 2 3 1 1 2 0 0.5 1 -1 1 0 2000 4000 6000 0 1 0 1 Luminosity (1e33) 0 500 1000 1500 2000 0 500 1000 LER current (mA) HER current (mA) Markus Cristinziani BaBar Drift Chamber performance 7 / 28

Performance Background Upgrade HiLumi Extra Babar-Belle Backgrounds Task Force Average Occupancy (%) -- typical run Goal is to understand differences and 15 14 13 commonalities in background patterns 10 12 5 11 For DCH and CDC similar currents but T 10 0 I O different occupancies are observed 9 B -5 8 go to DCH and CDC comparison 7 -10 6 Geographical background distributions -15 5 -10 0 10 0.16 0.09 0.14 0.08 0.12 0.07 Hits/wire (a.u.) 0.06 0.1 0.05 0.08 0.04 0.06 0.03 LER only 0.04 0.02 HER only LER only 0.02 HER only 0.01 Luminosity Luminosity 0 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 SuperLayer phi Markus Cristinziani BaBar Drift Chamber performance 8 / 28

Performance Background Upgrade HiLumi Extra Outline BaBar Drift Chamber performance 1 Backgrounds 2 Electronics upgrade 3 Higher luminosities 4 Additional material for discussion 5 Markus Cristinziani BaBar Drift Chamber performance 8 / 28

Performance Background Upgrade HiLumi Extra Motivation During poor background conditions the dead time is driven by DCH Cause: DCH GLT SVT Serialization and shipping of data EMT 4 from the front-end to the ROM Other Deadtime (%) Increased luminosity L ∼ 9 × 10 33 Increased L1 rate ∼ 2 kHz 2 Problem: Dead time was noticeable in Run 4 Would have been significant in Run 5 0 2 2.1 2.2 2.3 2.4 2.5 with no intervention Hour of March 31 Markus Cristinziani BaBar Drift Chamber performance 9 / 28

Performance Background Upgrade HiLumi Extra Read-out time extrapolations Predictions for future Extrapolations PEP beam configurations Dead time would soon 400 become a serious issue trickle LER beam-beam HER e.g. 30% at 5 kHz in 2007 Lumi constant DCH Read-out time ( µ s) 300 30% deadtime PEP parameters L (10 34 ) year I LER I HER 200 2004 2.45A 1.55A 0.9 2005 3.1A 1.7A 1.1 2006 3.6A 1.8A 1.3 100 2007 4.0A 2.0A 2.0 2008 4.5A 2.2A 2.1 Short-term solution: 0 2004 2005 2006 2007 2008 ship less raw data! → Year Markus Cristinziani BaBar Drift Chamber performance 10 / 28

Performance Background Upgrade HiLumi Extra Implementation: Waveform decimation 120 32 samples of FADC values with bilinear packing; time information of TDC hits override ADC value 100 Algorithm: out of a pair FADC counts 80 keep the TDC hit or ship the second sample 60 This preserves all TDC hits and 40 reduces waveform to 16 bytes TDC Expand decimated waveform to 20 original size interpolating in the ROM after the bottleneck 0 0 5 10 15 20 25 30 Waveform byte Markus Cristinziani BaBar Drift Chamber performance 11 / 28