4. METHOD OF SOLUTION
Considering the three code types in turn:
As the unit of the pulse height distribution must be matched to that of the response matrix, it is necessary to convert from counts per channel to counts per light unit. In this system the light unit used is the Na light unit. Also a new calibration method is used to determine the coefficients of the linear expression that defines the relationship between the channel number and the light unit. This method utilizes the relations between the values of the light unit for the energy of the Compton peak and the corresponding channel number of the gamma ray sources.
The response matrix of the NE213 is obtained using Monte Carlo.
47 neutron groups are used in the test run, with cross sections from ENDF/B-IV. The light output data is that evaluated by Verbinski et al. The response matrix is reduced to 34 groups for the unfolding. The pulse height meshes are arranged at equal log spaces from 0.001 to 10.0 Na light units with 60 mesh points per decade.
For unfolding, two codes, FERDOS and ITEM-II, which are based on different principles are available, so as to check their results. FERDOS employs a constrained least squares unfolding procedure whilst ITEM-II uses an iterative unfolding method.