From tigwiki

[edit] Germanium, Silicon and BaFF Detectors

[edit] Gain Calibration of Online Spectra

Some energy signals are simultaneously histogrammed both as a raw ADC number x and as calibrated energy E in keV, where E=gx+o, g is gain and o is offset.  For example, the 20 HPGe raw ADC channel data are histogramed as EGe00 to 19,  while the calibrated energy are histogrammed in EnGe00 to 19.  The g and o coefficients are stored in the ODB under /Analyzer/Parameters/fifohist/Ge/Gain and /Analyzer/Parameters/fifohist/Ge/Offset, respectively.  These parameters are hot-linked, so online, any changes take effect immediately.  Similar structures apply to the PACES Si and could be implemented in the future for the plastics and BaF2s following the same conventions.

[edit] Automatic Calculation of HPGe Gain Coeffficients

Chris has written some nice scripts for automatic gain matching.   You can find them on isdaq08 in /home/8pi/online/analyzer/gainmatch. Type gainmatch -h to get help.  A typical command line would look like:

Mon> gainmatch isdaq08:9090 2 y88

This will calculate gain coefficients from the online raw ADC spectra,  assuming that the amplifiers are already set to give you roughly 2 ADc channels per keV (0.5 keV/channel), and assuming a 88Y source has been counted for a while.  At present, the sources that can be specified are y88, co60, cs137, eu152, na22, and bkg (for the 1460 and 2614 keV background lines)

[edit] Manually Changing Gain Coefficients

Since the Gain and Offset are simply ODB entries, they can be modified either through the web interface, Starting the Web-Based Run Control Interface, or the odbedit program just like any other ODB entry.  They are hot-linked, which means that as soon as they are changed, they take effect.

[edit] Evaluating Peak-To-Total Ratios

If you specify a negative number for the dispersion, then the peak-to-total ratios for the source will be calculated:

Mon> gainmatch isdaq08:9090 -2 co60 <br>

This is also useful to see how much the peaks might have drifted.

[edit] Hints

The automatic procedure assumes that there is one and only one peaks in a window around target channel for the specified source and dispersion (i.e, around channels 1796 and 3672 in the y88 example above). By default that window is +/- 25 keV (in this case, +/- 50 channels).  It is recommended that the potentiometers on the ADCs be adjusted so the peaks are within about 7 keV of the nominal position before running the gain-matching procedure.

[edit] Beta counters (Sceptar)

Gain is set by changing the voltage on the PMTs. Offset is done by pedestal subtraction.

Per Paul Garrett:

SCEPTAR pedestals can be adjusted through the "ODB" "Equipment" 
"Camac" "Settings" "SCfera" page.  However, I have found many 
times that its essentially useless to adjust with a source - 
for some reason I'm never fully understood, the pedestals change 
in-beam vs source.  Having a few pedestals low is, however, not 
a real problem - it just means some useless data is brought across 
in every event.  Small increment in the dead time, and small 
increment in the file size, are the only consequence.