GBS Elektronik MCA-527 - Manual (Page 46)

MCA527
The stabilization area defines the minimum counts in the area of the stabilization peak
collected in a stabilization cycle before the fine gain is readjusted. The optimum value
depends on the peak FWHM, peak count rate and peak drift rate. If the drift rate is high,
then a rather low stabilization area is good to adjust the fine gain frequently. If drift is
rather low a high stabilization area is better as then the centroid is calculated more
accurate. Practical values are between 1000 and 25000. A formula to estimate the
optimum area N
opt
setting is given here 8 :
where
FWHM = expected width of the stabilization peak in keV
˙n = peak area count rate
E = energy of stabilization peak in keV
 = expected drift rate for the energy calibration.
Example:
A big NaI detector shall be stabilized with the 1460keV peak from the K40 background.
Within a 300 seconds measurement the 1460keV peak has an area of 3721 counts and a
FWHM of 72keV. The possible detector drift is expected to be in the order of 5 ·10
-6 s -1 . The
optimum area setting then calculates as:
Here, setting the stabilization area to 1500 should yield the best results.
3.5 Direct Input Pulse Height Analysis
This mode is useful when dealing with Gaussian pulses which may be the output of a
shaping amplifier or detector signals with a preamplifier decay time constant <40µs.
However, this cannot be used with high count rates as the fast rise of a preamplifier signal
is missing and pile up rejection does not work.
Furthermore, a significant part of the signal needs to be the base line, as this is required
to determine peak height.
Function of this is simple. Measurement is triggered if the input signal deviates some 6σ of
the base line noise. Then the algorithm checks if the signal reaches a maximum. If the
maximum is reached, the 4 highest values around are averaged, subtracted by the base
line average and this, multiplied by fine gain setting, is the value to be determined. After
that, it waits until the base line is reached again, and the algorithm continues.
In contrary to MCA166, coarse gain and fine gain settings work here and affect the
outcome of the spectrum.
8 Jörg Brutscher, “Behavior of the MCA 166 at different Temperatures and Gain settings and limits of
centroid accuracy”, internal report 2001
46
N
opt
=
 FWHM⋅˙n
2⋅E⋅
 2
3
N
opt
=

72keV⋅3721
300s
2⋅1460keV⋅5⋅10 − 6 s − 1
 2
3
= 1552