How to run mcfgen

We shall give most of our attention to mcfgen because that is where the data processing is done. One may run mcfgen alone, but because it is just one step in a chain, one should also examine the script

$$\tt /nds/bin/build.mcf.general$$

for the entire process, including creation of the `library' files described below. For mcfbin there are two scripts,

$$\tt /nds/processing/scripts/tar.mcfbin$$

and

$$\tt /nds/processing/scripts/build.mcfbin$$

The script tar.mcfbin assembles the source code for mcfbin into one subdirectory along with the `mcf.asc' files, and it produces a compressed file mcfbin.tar.gz for transfer to a Cray computer. The script build.mcfbin uncompresses this file, makes the code mcfbin, and produces the `mcf.bin' files, renaming them mcf1, ..., mcf7.

If one wants to run the code mcfgen alone, the current directory must contain the following files or links to them.

`bdfls':

This is an ASCII file containing group boundaries, flux weightings, atomic masses, half-lives, physical constants, temperature sets, and atomic subshell designators. The standard instance of this file on the Nuclear Data Group's computer network is

$$\tt /nds/processing/constants/physcons$$

`library' or `epdl.asc':

For incident neutrons `library' is a binary file of ENDL data, and for incident charged particles it is ECPL data. This file is made by running the codes create and endlret. For incident photons the `library' file is replaced by the ASCII file `epdl.asc' of EPDL data made by running epdlmcf.pl. See the script build.mcf.general.

`mcfgen.input':

A file in which the user may specify various data, the most important of which are the identity of the incident particle and the range of targets. The standard `mcfgen.input' files are located on the Nuclear Data Group's computer network as

$$\tt /nds/processing/inputs/mcfgen.input.N$$

where the N is one of the values 1 (neutron), 2 (proton), 3 (deuteron), 4 (triton), 5 (³He), 6 (alpha), or 7 (photon). These files are very rigid in format, and the spacing must be exact. For example, the standard input file for an incident proton is

     0  2  2 71  1 33  2     0  0                 00
  1001  8016

Here marks a space. The significance of these numbers is explained in Section 3.1, so we point out only the most important ones here. The first 0 indicates that the file has a second line containing the range of targets. The first 2 specifies that the incident particle is a proton. The last zero before the 00 specifies that, except for the angular distributions and the energy distributions, the code mcfgen is to compute averages over energy groups. The case of an incident photon has a 1 in that location, indicating interpolaton onto an energy grid. The second line identifies the targets in terms of atomic number Z and mass number A using the code 1000*Z + A. Thus, for this computer run the targets range from isotopes ¹H through ¹⁶O.

The standard way to run the code is simply to type

$$\tt mcfgen$$

in a directory containing soft links for the files `bdfls', `mcfgen.input', and `library' or `epdl.asc'. There is a variant to permit the user to process the data for a single target without changing the `mcfgen.input' file. Specifically, one may use the command line to override the range of targets given in the `mcfgen.input' file by typing

$$\tt mcfgen$$ $$\it nnza$$

Here, nnza is an integer of the form 1000*Z + A, coding the atomic number and mass number of the (single) target. Thus, to process only the data for protons incident on deuterium with the `mcfgen.input' file as above, one types

$$\tt mcfgen\ 1002$$