CSC Validation Update Woochun Park (Univ of South Carolina) @USC - - PowerPoint PPT Presentation

1

CSC Validation Update

Woochun Park (Univ of South Carolina) @USC ATLAS Meeting 8/30/2007

2

Low Efficiency in Endcap Region

• David Adams found efficiency loss in endcap region.
• New MC simulation is available in release 13.0.20.
• Related to CSC detector, gas gain is reduced by factor of two to have

more realistic baseline value.

• We want to make sure that CSC reconstruction doesn’t have

problem for this.

R12 simul. with R13 recon. R13 simul. with R13 recon.

3

Gas Gain

• Gas gain in MC simulation was different from that in cosmic data.
• The lower gain it becomes, the worse resolution we expect.
• Charge threshold in the algorithm should be lowered.
• Error of charge measurement in strip should be increased.

Qpeak Qright Qleft Charge

4

• Residual increased about 50% (41 microns 62 microns) which was

expected.

• With 10k electrons, CSC segment efficiency is recovered.
• With previous charge threshold 20k electrons, it becomes lowed by 1.5%.
• But, spoiled fraction is increased compared to the previous MC.
• Anyway, it’s proved that CSC is not show-stopper.
• New charge error is 4510e and non-normal error is 0.552mm
• Previously, 4300e and 0.57mm.

5

TGC problem

6

To do list

• 13.0.20 MC should be validated further.
• Struggling to analyze cosmic data.
• Analysis code should be written for physics analysis in AOD/ESD

level (such as WH;Wµν; Hbjets)

7

Local workstation

• Bee is used for ATLAS machine successfully.

With SLC4.5, it has 12.0.6/13.0.10/13.0.20. All the releases are successfully installed and being used for

CSC reconstruction analysis as a major machine.

Ryan plans to move his ntuples into higgs. (275Gb) Then, beeraid1 will be 783Gb available. Arjun and I will make Bar another ATLAS machine soon. Bar has

340 Gb. Higgs has 660Gb (will be 385Gb with Ryan’s).

Strategy: To be cost effective, purchase new machine as late as

possible??