Stopping Muons

1
Stopping Muons

• Jeff Hartnell
• University of Oxford
• Rutherford Appleton Laboratory
• MINOS Meeting,
• January 05

2
Talk Outline

• Track window technique overview
• MEU definition median not mean
• CalDet Analysis Improvements
• New cuts
• New path length correction
• New drift points for T7 2003 ( temperature
  correction)
• Cleaner electron sample to x-check with
• CalDet results
• Light level tuning in MC self-consistency
• Delta rays ( Birk suppression)
• Summary

3
Track Window Technique
CalDet Beam Muon, stops here

• Sum energy deposition in a 49cm (14 planes)
  window located 56cm (16 planes) from the end of
  the track.
• Define a MEU to be the median detector response
  in the track window (per plane) to a
  perpendicular muon.

Dist. to track end (56 cm)
Window (49 cm)
4
MEU Definition Improvement

• Now median response in track window
• More robust (than mean)
• Better for Landau type distribution
• Protects against large anomalous events which
  shift the mean e.g. Brems in FD.
• Changes the value of an MEU by 1
• Same shift in both data and MC.

5
New CalDet cuts

• To remove CalDet specific bkg from PS muons and
  overlapping events.
• Aim ensure MEU can be extrapolated to Near and
  Far detectors.
• Fid. Vol. excl. 3 strips at edge
• Entry point between strip 9-12 (1st 4 pln.)
• Fid time window (-30 -gt 300 ns of TOF)
• Hits per plane whole event (all hits)
• Hits per plane first 10 planes (gt1.5 pe)

6
Effect of Cuts
7
Path length correction

• Aim ensure consistent with Near and Far det.
• Now doing full x,y,z reconstruction
• Calc. path length correction in every plane
  sqrt(dx2dy2dz2)/dz
• Previously assumed muon was perpendicular to
  EVERY plane.
• MC study
• Old MEU value wrong by 0.85 (w/ assumption)
• Path len. cor. OVER corrects by 0.5
• Will run over all the data and check how much
  values change
• Probably assign a systematic error of 0.5

8
Remaining Issue Calibration
9
Cleaner electron sample (for cross-check)

• Use Trishs OLChi2 cut.
• Make fiducial timing cut as well.
• Num hit planes lt 20.
• Correct for energy bias due to energy loss in
  upstream material (20-30 MeV).

10
MEU vs Run

• New drift points for 2003 (and temperature
  correction)
• Can see 2-3 fluctuations in response
• Not sure what the cause is???

11
Calibrated Electron Response
12
Latest CalDet results in figures

• Difference between
• T11 2002 and T7 2002 is 0.2
• T7 2002 and T7 2003 is 1.3
• T11 2002 and T7 2003 is 1.5
• (Assuming electron response is constant)
• Ryan's result in his thesis was that the error
  was 2 between run periods (not detectors).
• Don't know at this level whether the beam
  momentum was off by 1 between run periods or if
  there is a problem with the calibration, or the
  detector, or something?

13
CalDet Summary

• Nearly exhausted my list of potential
  problems/issues.
• MEU values have remained constant to 1.
• Demonstrates robustness of technique.
• The analysis technique is close to delivering a
  2 calibration w.r.t. the other detectors.
• However, problems with the intra-detector
  calibration introduce a systematic of 2.

14
Light-level tuning in the MCself-consistency

• Show that what we get out equals what we put in.
• Use exactly same code on MC and data.
• At CalDet see a 10 discrepancy. Light out too
  low.
• Turned off all detector effects (clipping, birks
  suppression, dynode thresholds, sparsification,
  xtalk, etc) and only a lt1 discrepancy.
• Turned effects on one by one
• Stripend clipping 2 (too long tails in pdf
  now fixed)
• Birks suppression 8 (delta rays)
• FD self-consistent to 5 (preliminary)

15
Delta rays ( Birk suppression)

• Birks Suppression (S) 1 / (1 B dE/dx)
• B 0.133 m/GeV
• Suppression factors
• muon S2.5
• delta rays (10keV cut-off) S35
• delta rays (100keV cut-off) S15
• At CalDet delta rays deposit 30 of the energy gt
  10 suppression of light level
• Birks suppression will affect different particles
  at different levels a way to tune relative
  responses?
• Is Birks suppression of delta rays really 35?

16
An aside Range reconstruction

• How well do we reconstruct the range between
  detectors?
• 2 ? (It has to be at least this good!)
• Problems
• Steel thickness.
• Magnetic field.
• Xtalk patterns.
• 2/36 pre-trigger at FD.

17
Summary

• Relative calibration at CalDet is consistent with
  the beam energy to lt2.
• CalDet MEU analysis is now very close to
  delivering 2 between detectors.
• MC is now self-consistent to a few .
• Interesting Birks suppression effect

18
The End!