Time resolution measurement with Madison testDAQ

1
Time resolution measurement with Madison test-DAQ
Dmitry Chirkin, Spencer Klein, LBNL
IceCube meeting in Uppsala, 2004
2
Dataset

• laser run
• 1.5 kHz
• 75 ps wide
• attenuated by beam splitters optical occupancy
  set at 0.1
• 46211 DOMS
• 1 SYNC board connected to laser (here DOM 37)
  175 ns early
• 1 REF PMT (room T) (here DOM 62)
• (info by Kael Hanson)

files from
IceCube/2004/09/15/UWDAQ1_run0002141_TimeResolutio
n-ATWD0
3
DOM time calibration concept

• Calibration event time
• Tc_dorTc_domdelta
• Tc_dor
• dor_txdor_rx-d(wf_dor)/2
• Tc_dom
• dom_txdom_rx-d(wf_dom)/2

Roundtrip time TdTu dor_rx-d(wf_dor)-dor_tx-
dom_tx-dom_rx-d(wf_dom)
Time delta delta(Tc_dom)Tc_dor-Tc_dom
4
DOM/DOR received waveforms

• COMM Waveform fit algorithms
• centroid
• crossover
• leading edge intercept
• leading edge threshold

laser SYNC board
5
Roundtrip time
6
Time delta measurement
7
ATWD feature extraction
Bin size to time conversion is done for each
OM using info from the monitoring
files frequency 20 MHz (slope850intercept)
300 MHz, e.g., bin size 3.3 ns. (info by
John Kelly)
time direction
8
Laser-correlated signal
9
Laser-correlated signal
crossover
centroid
le threshold
le intercept
10
Time resolution results
11
Conclusions

• Time calibration precision is 0.5-1.5 ns for
  centroid and crossover, and 1-2.5 ns for leading
  edge intercept and threshold. However, centroid
  and le threshold give higher values for 30 of
  DOMs (could be improved by modifying the
  algorithm?)
• Although round trip time depends on the
  algorithm used to fit the COMM waveforms, time
  correction does not (with differences between
  methods 1-3 ns)

12
Conclusions

• clock stability is lt 0.5 ns between calibrations
• time resolution obtained with the laser method
  does not depend on the algorithm used to fit the
  COMM Waveforms and is 1-3.5 ns for the majority
  of tested DOMs
• Laser time resolution setup can be used for DOM
  verification tests?