Run 6 EEMC Gain Calibrations - Towers and Tiles
Goal: Use mips from pp minbias events to set absolute gains for EEMC towers, SMD strips, and all pre- and post-shower tiles for the 2006 run:
- analyze 1.8M pp minbias events from day 89 - runs 7089008, 7089013, 7089014, 7089015, 7089021, 7089022, 7089024, 7089025, 7089091, 7090008
- fit slopes of ungated SMD histograms: 3-4% statistical error in slopes -> sets strip-to-strip relative gains to ~6% accuracy
- mask out dead or "problem" strips
- for SMD 1st pass: estimate absolute SMD gains by global matching to last year -> need ~10% increase in conversion factor from 2005 Cu-Cu to 2006 pp minbias
- identify mips event-by-event via 000xx000 or 000x000 hit signature in both planes -> "mip ID"
- fit Gaussian (Landau) function to towers (pre/post tiles) gated by mip ID, use known geometry / energy loss to set absolute gains
- for SMD 2nd pass: sum ranges of consecutive SMD strips gated by mip ID to set absolute gains, check for eta dependence, compare to 2005 results
- other improvements?
Results: Using isolated SMD hits to gate tower and tile spectra
- Use absolute SMD gains determined in 1st-pass calibration
- For valid hit, require ESMD > 0.4 MeV
- Require at least three `no-hits' on each side of valid hit (1 or 2 strips) in each plane
Some typical performance plots - all taken from Sector 11
Absolute gains used for Sector 11 V plane
Distribution of isolated mips in each plane
Now use SMD UxV tag to gate other detectors. All plots are for Sector 11, stack C07
Comparison of ADC spectra for towers, pre-, and post-shower layers,
with and without SMD tag
Last step: require hits in tower/tiles other than layer to be fit ("hit" -> ADC - ped > 3*sigmaped.
All plots are for Sector 11, stack C07
ADC spectra for towers, pre-, and post-shower layers, with mip tag
Everything seems to be working in this sector!
- Complete Gaussian (Landau) fits to all EEMC mip-gated towers (tiles) to set absolute gains
- Use "best mip" tag to set absolute gains for SMD (2nd pass) after summing over ~10 strips
- Check for consistency / systematic differences with 2005 gains (Cu-Cu minbias), especially eta dependence
- Check for consistency with 2006 pp minbias data taken later in run (not sure we have statistics)
- Consider improvements: taking out z-vertex variations using BBC time differences; others?