If the weighting function is very well known, you can do good
transport calculations with very few groups, but with modern
computers, the trend has been to use more and more groups.
GROUPR contains a number of builtin options for group structures
(some of them of mostly historical interest), and the user can
enter any arbitrary structure that matches his or her needs.
Reactor physicists have always been fond the the variable called
"lethargy," with is defined as follows:
Lethargy increases as neutrons slow down. Note that in the
slowing-down region, where the flux varies like 1/E,
groups with constant lethargy width all contain the same
portion of the flux. In elastic scattering, neutrons loose
a fraction of their energy with each scattering; thus, the
lethargy increases by a fixed amount with each scattering.
Because lethargy is such a natural variable for neutron
slowing-down problems, many group structures are based on
groups with certain lethargy widths. For example,
- the LANL 30-group structure uses "one lethargy" widths
in the resonance range,
- the ANL 27-group structure uses even half-lethargy steps, and
- the GAM-I 68-group structure uses even quarter-lethargy steps.
Some group structures start with even lethargy groups and
then subdivide some groups that contain especially interesting
structure, such as the Fe-56 resonance at 27 keV. In other
cases, the group structure starts with even lethargy groups
and then moves some bounds to try to center important
resonances (e.g., the WIMS 69-group structure).
Another factor when choosing a group structure is its range.
Many group structures intended for reactor analysis stop at
10 MeV. They are clearly not suitable for fusion problems.
Other group structures only provide a few groups in the
thermal range, and they are most suitable for fast-reactor
and fusion problems.
The neutron group structures currently available in GROUPR are as
- IGN=2, CSEWG 240-group structure, a "supergroup" structure
containing bounds from many other structures. Not widely
used these days.
- IGN=3, LANL 30-group structure, used for fusion problems.
Still in active use.
- IGN=4, ANL 27-group structure, even half-lethargy groups used
for fast-reactor analysis after collapse from more detailed
- IGN=5, RRD 50-group structure, a standard fast-reactor set
that is now obsolete.
- IGN=6, GAM-I 68-group structure, even quarter-lethargy set
used for the epithermal part of thermal reactor codes (most
- IGN=7, GAM-II 100-group structure, an extension to higher
energies of GAM-I. Note widely used these days.
- IGN=8, LASER-THERMOS 35-group structure, a thermal structure
for the low energy part of power-reactor codes (most recently
- IGN=9, EPRI-CPM 69-group structure, the WIMS structure, still
widely used all over the world for reactor calculations.
- IGN=10, LANL 187-group structure, a single set of fine-lethargy
bounds suitable for thermal reactors, fast reactors, and fusion
- IGN=11, LANL 70-group structure, another standard fast-reactor
set, now supplanted by the 80-group structure.
- IGN=12, SAND-II 620-group structure, very fine group structure
used for flux unfolding applications. Mostly supplanted by the
- IGN=13, LANL 80-group structure, optimized for fast-reactor and
fusion systems. Still in use through MATXS files.
- IGN=14, EURLIB 100-group structure, similar to GAM-I except
for additional high-energy groups and groups near the iron
resonance. Used for a number of NEA exercises.
- IGN=15, SAND-II 640-group structure, used for dosimetry work
with flux unfolding codes. It extends the 620-group structure
to higher energies.
- IGN=16, VITAMIN-E 174-group structure, used for the widely
distributed fine-group library from Oak Ridge.
- IGN=17, VITAMIN-J 175-group structure, an extension of the
174-group structure. This structure is currently widely
used in the fusion community, e.g., ITER design.