Here I describe how to adjust the gains for the various 
 spectrometer PMTs:

 Files:

 o pmt_database
   contains all necessary info about the PMTs (for now):
   Index#  Serial#  Date  Tested  LedBox  WeightInfo

 o pmt_map.dat
   contains a map from the pmt# to the NX,NY scheme
   on the detectors. The numbering is:
   BPCn:
   NX1-15 and NY1-16 correspond to  1-31
   BPCs:
   NX1-11 and NY1-16 correspond to  32-58

 o relative_response.dat
   contains the relative response (a factor) with 
   respect to a std PMT at HV=800 LED(V)=4V. The 
   first row is the PMT index number. 
   
 o powerlaw.dat
   contains the powerlaw info for each PMT:
   first row= index#
   secondrow= par(1)
   third row= par(2)
   The power law is given by:
   ADC = (HV/par(1))**(par(2))     (1)

 Actions:

   I chose the mean of the PMT response distribution (look 
   at file relative_response.dat) in order to set all PMTs
   to equally respond. This is factor=1.3

   The user needs to enter the PMT index number in order 
   to get an answer at what value to set the high voltage.

   To do this one has to call "adjust_hv.kumac"

   The macro reads relative_response.dat and powerlaw.dat.
   Then, it divides the response by the mean of all responses.
   If the results is say ... 1.2 (i.e. 20% higher than the 
   mean) then the powerlaw function has to be used to find 
   where to set the HV in order to get 20% reduction in the 
   ADC. From (1)  HV=par(1)*(ADC/1.2)**(1/par(2))
   The ADC used here is for a particular HVmean=800V for 
   example.

 Important:

   To examine how the entries of the powerlaw.dat dbase 
   were calculated look at:

   findpowerlaw.kumac