PMT Calibration

1
PMT Calibration

• R.Sawada
• 7/Jan/2007

2
Time calibration
Method was talked at the previous meeting. The
problems which was shown before were because I
used PMTs in shadow from LED.
3
Time nsec
Time nsec
Distance cm
Distance cm
Time nsec
Time nsec
Distance cm
Distance cm
4
Time offset and DRS channel
According to Matthis, there is similar dependence
in DC data.
Chip0
Chip1
5
Time offset and DRS board
Crate4
Crate5
Slot2
There is small correlation. This was not expected
because time difference due to electronics is
corrected by using clock analysis.
This can be variation of cable length in liquid
xenon. (due to temperature co-efficient)
6
Time distribution before calibration
Crate5,Chip1
Time nsec
Crate4,Chip1
Crate5,Chip0
Crate4,Chip0
Crate5,Slot12
Distance cm
7
Time distribution before calibration
Time nsec
Distance cm
8
Result
One entry is one PMT
Time offset nsec
Time offset uncertainty nsec
This result is available in analysis and
simulation.
This number is uncertainty of measurement of one
LED. Average is 1/sqrt(2)-1/sqrt(4) better.
9
Q.E.
Analyzed by using a task developed by Pisa group,
with my small modification.
10
PMT Nphe fit (Same as LP)
Gaussian when mean of nphe gt 20
Poisson x Gaussian when mean of nphe lt 20 Sigma
of Gaussian is obtained from pedestal events
11
gas alpha
Qsum2 gt 2500
Noise
Outer wall alpha
Wire alpha Top/Bottom wall alpha
12
liquid alpha
13
DataMC
One graph is one PMT. One point is one alpha.
Gas data (1/10 statistics, full statistics
analysis is still running)
14
DataMC
One graph is one PMT. One point is one alpha.
Liquid data (full statistics)
15
Result
Q.E. liquid
Q.E. in liquid
Q.E. in LP or Pisa
Full statistics of gas is running.
16
Summary

• Time calibration was done in 200-500 psec
  uncertainty. Result is available
• Speed of light observed on Inner/Outer and
  lateral is slightly different.
• Q.E. estimation using liquid data was done. Gas
  will be finished soon.
• Result will be available in 1-2 days.
• Uncertainty is currently fitting error of data
  v.s. MC. It looks rather small. I will check the
  code again.
• I used MC result produced by Pisa group. I will
  check it with my MC. (It must be same, because we
  are using the same MC software.)
• PMT calibration was almost finished. I will
  proceed to absorption length estimation, Q.E.
  estimation dependence on geometry (check of MC),
  reconstruction and detector calibration.

1/sqrt(2) 1/sqrt(4)