Timing Counter

1
Timing Counter

• Report of Feb 20th, 2008
• F.Gatti

2
Final Construction Phase of TC
TC with fibers exposed
TC upside down for Fiber APD gluing
High reflectance polymer foil coating
TC before insertion in COBRA
Cables and pipes of TC in COBRA before final
positioning
3
TC/Bag final positioning
DCH supports
Bag SS inner cap
COBA inner step
Bag SS inner edge
Bag EVAL inner side backside illuminated
PM at the final position inside bag
4
Fibers detector

• Fibers detector turned on
• The analog output achieved a good S/N level that
  has been tested channel by channel.
• A selected event of a CR that hits at least 3
  fibers is shown as example

8 Channels analog sum
8 Channels analog sum
8 Channels analog sum
Signal of 88 Interleaved fibers
5
Timing Analysis

• Use positrons runs with TRIGGER on TC only (2 or
  3 contiguous bars in coincidence).
• Reference data 250k triggers belonging to
  8378-8408 runs
• Measure bar average time Tk
• First method s (Tk-Tk1) s (s/cdT)
• Second method s ?(TkTk2 )/2 - Tk1? s (0DT)

s
6
No cuts timing analysis 3 bars example
Correction for the first two bars used
Higher order effects
7
No cuts timing analysis all bars
Different chip
upstream
downstream
Bar2Bar1
Bar1Bar0
8
Searching the intrinsec timing resolutions

• The methods give z-dependent results? low z cuts
  give better resolution? not fully explained by MC
  and presently under study
• Time reconstruction from digitized pulse are
  affected by the electronics/algorithm intrinsic
  jitter? evaluation of this time jitter has been
  done in a dedicated run.
• Double Threshold Discriminator pulse is
  available to overcome the elctr./alg. jitter

9
Z-cuts Timing Analysis with DTD and PMT pulses
DTD Time resolution s 52 ps
DTD
Bar3 - Bar2
Bar3 - Bar2
PMT
Bar3 - Bar2
PMT Time resolution s 62 ps

• Cuts 0.1Vltpulse heightlt0.3V (Landau peak)
• Reconstructed z on first bar lt 55 cm

10
Intrinsic electronics/algorithm jitter in PMT
signal

• Dedicated positron runs one PM signal feeds
  the two input channels of the corresponding bar
• ?same signal to the two electronics chains of the
  bar.
• ?Time distribution only affected by
  electronics/algorithm jitter
• It has been measured
• s(T1-T2) 2 s ((T1T2/2))
• Average jitter value 54 ps
• Final PMT timing resolution after elect/algor.
  jitter subtraction ? 55 ps rms ? 129 ps FWHM
• DTD analysis gives 120 ps FWM without jitter
  correction.
• Latest preliminary results on DTD after jitter
  correction 108 ps FWHM

Runs 9362-9363
11
First comments on time analysis

• DTD pulse analysis method shows better
  performance in the first analysis (not all
  correction applied as in PMT signal analysis).
• PMT and DTD give us good redundancy for the
  timing of a single event.
• Corrections for the time-walk, hit position,
  electr./algorithm jitter, systematic of the
  timing resolution analysis methods, are under
  study.
• Z dependence of T resolution under study (MC vs
  data comparison) the effect suggests a
  dependence of the chosen timing method on track
  inclination (low Z ?low impulse z projection).
• Finally the TC intrinsic time resolution in
  operating conditions is now evaluated to be 120
  ps FWMH (20 excess respect to the proposal),
• Latest preliminary results show 108 ps FWHM
  jitter corrected DTD signal.
• We are confident that there is room for the fine
  tuning of the timing analysis and a further
  assessment of the of the timing resolution

12
PMT equalization ( CR and CW data)
13
Charge Analysis

• Veff
• measurement of positron impact time on the bar,
  T
• measurement of z can be made, in principle, from
  the width of the t1-t0 distribution
• Measurement of z with the charge, once that ?eff
  is known
• ?need ?(veff)/veff 1 to achieve the desired T
  resolution
• ?eff
• independent measurement of z can be measured
  from
• - ln(Q1/Q0) vs z determined with time once that
  veff is known
• - ln(Q1/Q0) using z measurement from fibers

14
l eff. and Veff.

• Significant differences among bars, consistent
  with results at BTF.
• Data suggests not uniform losses in the internal
  reflection
• - residual surface roughness (measured value
  lower than 0.2 mm RA)
• - plastic enclosure residual reflectivity
• Effective velocity 14.5/-0.2 cm/n

l eff
15
Laser for 532 and 266 nm monitoring pulse

• The system delivered during the run. Now under
  test with the optical fiber distributor
• Power stability at 48 MHz, 1064 nm, within 0.3
  over a week of monitoring
• Timing pulse distributed via optical fiber and
  detected on TC DS at 50 Hz free running
  repetition rate

To opt. fiber and fast APD for trig. Out.
To fast APD for power and int.pulse sync
16
Laser Status

• Laser assembly and table under integration before
  transportation at PSI

17
Preparation for the run08

• Completion of the commissioning of the digital
  hit map for the fibers with CR (Mostly done last
  week)
• Works on TC change of PMTs with unexplained low
  gain, close light leak in the APD fibers
  detector, change not working APD boards, improve
  S/N and noise immunity of the APD analog output
• Improvement of the S/N will allow to decrease the
  threshold level in trigger algorithm ?recover
  some delay (20-30 ns ?) in the trigger latency
• Rebuild N2 Bags
• Integrate the Laser for TC and XEC

18
Question of shaping time of APD electronics

• The 160 ns shaping time (10-90) the latency of
  the trigger APD algorithm prevented the use of
  the in the on line trigger selection in the run07
  
• An anticipation of about 60 ns of the formed
  trigger signal from APD should be enough ton
  solve the problem
• Improvement of the S/N will allow to decrease the
  threshold ? recover some delay (20-30 ns?)
• From the trigger side it is possible to reduce
  the processing time of the algorithm.
• ? APD online in the trigger for Run08
• Alternative possibility of reducing shaping time
  for 512 channels is not a trivial task and
  requires 7 weeks of with 6 people. This could be
  done eventually at the end of the run 08

19
Last on PMT life

• The measurement lasted 290 days for a total
  current of 2130 Coulomb on a new PMT
• The average current produced by the PMT over the
  period has been 88,2 microA
• Data presented has been already corrected by the
  laser power variation with the photocell values

20
Previous measurements on old PM
Drift region of the old 1-1/2 PM
1
2
3
5
21
New measurements 6.6 x life without changes
1
2
3
4
5
6
22
Schedule
23
End of slides
24
backup
DT-data
DT-MC
25
backup
26
Backup

• T1 t0 x/c
• T2 t0 (L-x)/c
• T1T2 2 t0 L/c
• TA(T1T2)/2 t0 L/2c
• TBt0s/c L/2c
• TA-TB s/c s/2
• TATB2t0s/c(L/c)
• TCt0s/2cL/2c
• (TATB)/2-TC t0s/2cL/2c-t0-s/2cL/2c0