Deep Inelastic cross section for the DVCS experiment

  • Slides: 17
Download presentation
Deep Inelastic cross section for the DVCS experiment Maxime DEFURNE CEA-DSM-Irfu-SPh. N 1

Deep Inelastic cross section for the DVCS experiment Maxime DEFURNE CEA-DSM-Irfu-SPh. N 1

Why Deep Inelastic Scattering • During the most part of the DVCS experiment, Trigger

Why Deep Inelastic Scattering • During the most part of the DVCS experiment, Trigger = Cerenkov + Scintillator 2 • Extraction of DIS cross section to check the HRS, its experimental data and globally the quality of the runs. 2

Extraction Correction Estimation of systematic error Cerenkov Efficiency 99. 5% 1% S 2 Efficiency

Extraction Correction Estimation of systematic error Cerenkov Efficiency 99. 5% 1% S 2 Efficiency 99. 9% Negligible Multi-track events Deadtime Charge About 10% Between 7% and 25% 3700 C 0. 9% 3

Deadtime correction • Huge unstability of deadtime! That’s why correction applied run by run.

Deadtime correction • Huge unstability of deadtime! That’s why correction applied run by run. 4

Selection of good events • Simulated acceptance of HRS ->Cuts defined by the match

Selection of good events • Simulated acceptance of HRS ->Cuts defined by the match between experiment and simulation. • Contamination by pions ->Cut on the energy deposit in the pion rejector. 5

6

6

Cut on pion rejector • Necessary to remove contamination from pions which can represent

Cut on pion rejector • Necessary to remove contamination from pions which can represent about 10% of events. 7

Computed cross section • Use the simulation of the HRS acceptance from the experiment

Computed cross section • Use the simulation of the HRS acceptance from the experiment E 97110 with updated transport functions. • Use a program from Alexandre Deur to calculate DIS cross section from structure functions of proton 8

Computed cross section c e E Sp S R L-H E beam E_bs s

Computed cross section c e E Sp S R L-H E beam E_bs s E_p r ato lim Col Scattering Chamber 9

Computed cross section Generate a vertex (x, y, z, phi, theta) External a Generate

Computed cross section Generate a vertex (x, y, z, phi, theta) External a Generate Radiative vertex correction E_bs (x, y, z, phi, theta) External Radiative Correction E_spec Virtual Radiative Corrections Compute Cross section associated to this event Calculate E_ps Reconstruc t the vertex Test Collimator-HRS 10

Unstable deadtime -> Deadtime very unstable -> Cross section quite stable thanks to the

Unstable deadtime -> Deadtime very unstable -> Cross section quite stable thanks to the run by run correction. 11

Miscalculated deadtime => Jump of the cross section sometimes due to miscalculated deadtime. =>

Miscalculated deadtime => Jump of the cross section sometimes due to miscalculated deadtime. => Is there an other way to calculate the deadtime? 12

Results Kinematics Experimental Cross section Computed Cross section Difference Kin 1 low 10. 21

Results Kinematics Experimental Cross section Computed Cross section Difference Kin 1 low 10. 21 9. 81 4. 1% Kin 1 high 61. 44 59. 65 3% Kin 2 low 15. 43 14. 63 5. 5% Kin 2 high 32. 66 31. 53 3. 6% Kin 3 low 8. 2 7. 57 8. 3% Kin 3 high 18. 2 17. 9 1. 7% 13

Kin 3 low • Cerenkov Threshold changed at run 8205. 14

Kin 3 low • Cerenkov Threshold changed at run 8205. 14

Kin 3 low : No collimator • Extracted and computed cross section with cuts

Kin 3 low : No collimator • Extracted and computed cross section with cuts in the center of the acceptance: Computed cross section at 7. 8 nbarn/Ge. V/sr. Experimental cross section at 8. 2 nbarn/Ge. V/sr ->Reduced the difference from 10% to 4. 5 %. 15

Summary • Kin 3 low was running without collimator. • Impact on the cross

Summary • Kin 3 low was running without collimator. • Impact on the cross section of the cerenkov threshold change. • Miscalculation of deadtime using T 5, T 7, T 10 for some specific runs. 16

Perspectives • Determine precisely the consequences of the Cerenkov threshold change. • Use Rfunction

Perspectives • Determine precisely the consequences of the Cerenkov threshold change. • Use Rfunction to apply better cuts on vertex. • Study the multi-tracks events correction. 17