ORTEC DSPEC Pro, Hardware User Manual

Page: 22 / 108

DSPEC Pro
®
Digital Gamma-Ray Spectrometer User’s Manual 794380D / 0914
12
where:
B
1
= sum of background counts for the channels adjacent to the peak start (low-
energy) channel L
B
2
= sum of background counts for the channels adjacent to the peak end (high-
energy) channel H
n
1
= the number of low background points (n
1
= 1, 3, or 5) used
n
2
= the number of high background points (n
2
= 1, 3, or 5) used
The peak area uncertainty is calculated from:
where G is the sum of counts, from the error spectrum, from channels L to H.
!
In our WAN and ISOWAN analysis engines, the peak-fitting routine fits all the library
peaks as singlets to calculate the peak centroid, peak start and end channels, and peak
background. A linear background under the peak is assumed during the peak fitting
process.
!
The error spectrum is always used to calculate the uncertainties of counts whenever
needed. For example, if peak deconvolution is needed, the error spectrum is used to find
the best fit for the peak background.
1.6.4. Choosing a ZDT Mode
When the counting rate is essentially constant during the time required to acquire the spectrum,
the standard mode — ZDT Off — is the preferred mode; only the uncorrected spectrum is col-
lected and stored in the spectrum file. But, if the counting rate varies significantly during the
measurement time, the standard mode will not yield the proper dead-time-corrected counting
rate. This can be most easily understood by noting that the uncorrected mode compensates for
dead-time losses by extending the real counting time. Hence a sample containing both a short-
lived high-activity isotope and a long-lifetime lower-activity isotope will experience very high
dead-time losses during the first few seconds of the measurement, as the short-lifetime isotope
decays rapidly. This high dead time will cause the counting time to be extended after the short-
lived isotope has decayed to zero activity, and the system will count the low-activity isotope for