Closer to Reality From Last Talk 7 Jan06

Closer to Reality From Last Talk ( 7 -Jan-06) – things seemed Great! However. . . 1) In estimating the SRD I use the normalized (true) Extra Galactic Diff Rate instead of our absolute(meas. ) rate. . . Iissue: the acceptance (Aeff) falls off at lower energies (there's almost a factor of 3 here) 2) Our intend operational mode is "Rocking" exposing our backside Issue: Large gaps between Cal. Modules (there's another factor of ~ 1. 3 here) This analysis: Input: 1) First day of 5 B Bkg. Run: dataset v 7 r 3 p 5 (Includes rocking) 2) Extra Galactic Diff: dataset v 7 r 3 p 5 3) All Gamma rep-V 7 r 3 p 4 dataset, runs 401 -800 All these datasets are "CTnew" mean application of the cuts shown by Julie prior to download Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 1 GLAST

Input: CTnew High Energy e. Filter FT 1 Zenith. Angle > 100 Heavy Ion Filter Scrambled Tkr Filter see next chart for Filter Definitions Background and Diffuse Integral Rates 00 CTBCORE >. 1 && CTBBest. Energy. Prob. 1 1 E min 0 20 . 3. 4 500 . 5 0 2000 0 1000 5000 1: 1 SRD . 75 . 65 . 0 AM G B T C Base Class 1: CTBGAM >. 35 Base Class 3: CTBGAM >. 55 Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 2 GLAST

Hi Energy e- Filter: ((min(abs(Tkr 1 XDir), abs(Tkr 1 YDir)) <. 01 & Tkr 1 Die. Edge < 10 & Acd. Active. Dist 3 D > 0 & Acd. Act. Dist. Tile. Energy >. 2) | (Tkr 1 SSDVeto < 7 & Acd. Active. Dist 3 D > -3 & Acd. Act. Dist. Tile. Energy >. 15) | ( Acd. Active. Dist 3 D >(-30 + 30*(Tkr 1 First. Layer-2))) ) & CTBBest. Energy > 5000 Heavy Ion Filter: CTBBest. Energy > 5000 & ((Cal. Trans. Rms - 1. 5*Tkr 1 To. TTr. Ave) < 5) Scrambled Track Filter: (Tkr 1 First. Layer - Tkr 2 First. Layer) < 0 & Tkr 2 First. Layer > 4 & Tkr 2 Tkr. HDoca > 100 Cosmic Proton: Tkr 1 First. Layer < 6 & Acd. Active. Dist 3 D > -80 & ((Acd. Act. Dist. Tile. Energy + Acd. Active. Dist 3 D/100) > 1) Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 3 GLAST

Base Class 3: CTBCORE >. 35 && CTBBest. Energy. Prob >. 35 Same Contours as before Base Class 3: CTBGAM >. 50 Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 4 GLAST

Base Class 3 Cuts On Axis THIS ONE Aeff x Fo. V Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 5 GLAST

Base Class 3 Cuts Diffuse Bkg. Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 6 GLAST

Residual Background Broken out by Sources and CTBBest. Energy Hadronic Res. Bkg. 2500 -12500 Me. V > 12500 Me. V 100 -500 Me. V 500 -2500 Me. V Cosmic Protons – ACD Leakage? Correlated a's! Same energy bins as above QED Res. Bkg. Cosmic (Hi E) e- Irreducible: e+ Conversions Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 7 GLAST

ACD Leakage Correlated Events: Mc. Dir. Err <. 1 and not e+ Mc. Id != -1 Side Ribbons don't go to top! 72 events Require events start in Top Layer Tkr 1 Z 0 > 590 Estimate 47 events are leaked in this way Composition: These are the Top Corner Leakers found in the hand-scan Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 8 GLAST

Project Tracks to + Const. X Sides and fold to YSide > 0: YSide = min(abs(Tkr 1 Y 0 - abs( (832 -abs(Tkr 1 X 0))/Tkr 1 XDir) * Tkr 1 YDir), 742) ZSide = min(abs(Tkr 1 Z 0 - abs( (832 -abs(Tkr 1 X 0))/Tkr 1 XDir) * Tkr 1 ZDir), 800) All Data Inside Holes All Gamma Ineff. = 90/105090 =. 1% Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 9 GLAST

Suspect the same is true on Const. Y Sides – Evidence poor – Const. X Sides Holes Removed Both Sides Holes Removed Difference Combined: BKG: 209 DIFF: 63 GAM: 220 Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 10 GLAST

Remaining Base Class 3 Residuals QED Hadronic Next step: separate out the CTBBest. Energy > 1 Ge. V and hand scan 69 Events after X, Y Side Holes Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 11 GLAST

Uncorrelated Events P Primary 12. 17 Ge. V Tracking error in Y e+ Back-Entering 1. 71 Ge. V Shower Stub in Tracker Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 12 GLAST

Correlated Events P Primary 32. 55 Ge. V Acd Tile Mc. Pos. Hit 49 Ke. V Ribbon Slot (10 of these) P Primary 4. . 494 Ge. V Acd Tile Mc. Pos. Hit 1. 57 Me. V No ACD Digi Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 13 GLAST

e- cosmic 6. 65 Ge. V Ribbon Ineff. And so on. . . Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 14 GLAST

Speculation: We may be able to kill most(all) of the Correlated Bkg. > 1 Ge. V If so where will we be - Present Status Corr. Bkgs. Removed Goal D = 35 events Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 15 GLAST

And. . . 45% Already There: 16 of these events are CTBBest. Energy > 1 Ge. V Top Ribbon Ineff. Req. CTBBest. Z 0 > 590 69 Events Total (out of 365) (Starts at the Top) & Mc. Dir. Err <. 1 (Correlated Event) This will be an issue on the sides as well Need DOCA to nearest Ribbon whether or not it was hit. . Bill Atwood, SCIPP/UCSC, Jan. 30, 2006 16 GLAST