[Halld-pid] TOF ADC Issues

[Beni Zihlmann]

Hi All,

I want to touch on two issue regarding the TOF ADCs, which result in
at least two jobs we need to do. (feel free to start working on one or 
both ;-) )
This mail is a little long so bare with me.

1) ADC -> Energy calibration
2) Best determination of ADC Integral

1) ADC-> Energy calibration
     As you can see from the attached plot the peak position of the 
Landau distribution is
     strongly correlated with paddle number (equal to position in the 
TOF plane).
     These peak positions were determined from events hitting the paddle at
     150cm +- 3.25cm from the pmt. The red color is the left/up pmt the 
cyan color is
     the right/down pmt. Note that these data are pedestal subtracted so 
zero is zero.

     The first question is: do we want to adjust the HV such that all 
peaks are at the same value?

     We need to use Monte Carlo data to determine the translation from 
peak position (ADC units)
     to energy (GeV) for each PMT. In other words what is the mean and 
most probable energy
     deposition of 1GeV pions in the paddles.

2) Best determination of the Integral. This is a more subtle question. 
Studies of both the BCAL and
     the FCAL show that the best resolution is not achieved by 
integrating the signal over the full
     range starting from the beginning of the signal (or slightly 
earlier) to the end (far into the tail).
     In fact for both detectors it is found that integrating only from 
the peak for about 15 samples
     which is I think about where the signal reaches half the amplitude 
in the decay gives the best
     resolution.
     We should also do such a study with real data were we select one 
paddle with sufficient hits
     like paddle 15 (not too close to the center) and select hits at a 
fixed distance from the PMTs
     similar to what I did to generate the attached plot. And then use 
different integration windows
     for the Integral and see what that does to the width of the landau 
peak. Of course that means
     we need to do this with data where we have to full wave form, so 
readout mode 8.


cheers,
Beni

-------------- next part --------------
A non-text attachment was scrubbed...
Name: pmt_response.pdf
Type: application\pdf/unknown
Size: 22626 bytes
Desc: not available
URL: <https://mailman.jlab.org/pipermail/halld-pid/attachments/20150813/a7d684a8/attachment-0001.bin>