Those are random plots, recording progress on understanding of FGT response

Feb 8 , examples of pules passed by simple filter (Attachement A), latency=87, HV=3600V

The pulses you see in PDF were found with this simple algo:
-ped from DB  (thanks Renee) are subtracted  from 7 ADCs , next 
-applied this 4 cuts: 
     if(minAdc>100) continue;
      if(maxAdc<200) continue;
      if(maxAdc>3000) continue;
      if(iMin>1) continue; // require minimum is in time bin 0 or 1

2 observation:
* this is just 4 minB events and the algo above found ~70 pulses. You can easily convert it to # of pulses per strip per hour of data taking - we will have a lot of data per hour of running. Now it is not based on some simulations - this is what we see in the real collision environment. 

* count how often  a non-physical pulse shape sneaked thrugh: 3 on page 4 and ... that is it. So the signal purity is of 95% - no shape fitting applied yet.  This is really nice.  (if you know what timing to use)

Feb 9, maching of amplitudes for R*P clusters, 3 random events, oct=3AS, attachement B 

Feb 11  , inclusive spectra of highest strip in event as function of HV 3600, 3400

pulse-shape filter:  minAdc<100, maxADC>200, minBin=0,1,or 2

Attachemnts C,D,E show spectra for 1A+B, 3A+B, 6A+B 

Feb 12, fitted slope to a section of inclusive spectra

Attachemnt  F - HV changed 300,3200,3400,3600, disc 1 fit range 500,1200,

Attachement G - gains vs. HV. proposed target gain=200

 Feb 22, att. H, pulses selected with pulseSeedFinderA

Feb 22, Att,: I,J  Pulse yield for 2 quadrants, missing voltage on one section is visible.