August 23, 2016
The SCALE 6.2.1 update is available for SCALE 6.2.0 to provide enhanced performance and resolve issues in the areas detailed below. This update is recommended for all SCALE 6.2.0 users.
The TRITON T-DEPL features ASSIGN and SWAP were updated for SCALE 6.2.1 to correct issues that could affect the results of SCALE 6.2 calculations. SCALE 6.2 T-DEPL calculations that use ASSIGN and/or SWAP should be reanalyzed with SCALE 6.2.1.
Details of all updates:
A restriction that limited CRAWDAD to treating only up to 100 temperatures has been removed.
A memory defect was resolved that could cause a 'glibc' crash for some cases.
Fulcrum was enhanced with several new features for SCALE 6.2.1.
Fulcrum now supports text block column operations. By using the ALT+Left mouse button drag, a block of text is selected and can be deleted, pasted, edited, etc.
Fulcrum no longer incorporates the file name into the file specific action items (save, save as, etc). When conducting a Save as, the name is pre populated with the current file name.
Fulcrum text panels now display the line number in the same font as selected for text viewing.
Fulcrum geometry view now has a 'Show view origin' button allowing users to place a cross-hair at the origin of the view plane, which can be helpful for zoom to a certain region of the geometry.
Data files that include reaction names now also include the reaction identifier (MT) number in the identifier, which is also shown in the plot legend. For example: u-235 mt=18 fission. This applies to many data formats including multigroup and continuous-energy cross sections, sensitivity data files, covariance data, etc.
Fulcrum now provides a highlighting all instances of selected text within an editor to help users recognize patterns of use. This highlighting looks like a shadow box surrounding each additional occurrence.
Fulcrum now supports in-line mathematical expression evaluation in any text file. Selected functions and expressions can be evaluated by clicking Edit -> Evaluate or by using the CTRL+E key (CMD+E on Mac) combination. In addition to basic arithmetic operators (+, -, ‘*’, /, ^), the following functions are available for evaluation: sqrt, cos, sin, root, abs, min, max, avg, sum, mul, floor, ceil, exp, log, logn, log10, hyp, ifFunction, clamp, inrange, sign, deg2rad, tan, equal, acos, asin, atan, cosh, tanh, sec, csc, cot, sinh, round, roundn, d2g, g2d, r2d
Several issues with the previous version of Fulcrum were also addressed.
With multiple documents open, file actions (Find, Auto-complete, Save As, etc.) now operate on the expected file.
Fulcrum now displays the sensitivity data file (.sdf) energy-integrated k-eff sensitivity.
Fulcrum's geometry XZ and YZ MeshView 2D plot creation has been updated to correctly set the voxel index at which to create the 2D plot. The results of previous calculations are not affected, but visualization of results may result in minor offsets of data that use the Z-plane. Mouse hover operation now incorporates the mesh voxel index in the info label in addition to the absolute (x,y,z) cartesian coordinates, unit, and material.
Fulcrum now provides the user a warning message when closing a message or editor view with a running job attached to it.
An empty cell data block no longer produces a validation error.
SCALE input parsing (the form of the input) and validation (correctness of input parameters) errors have been disambiguated. Calculations cannot be started for inputs with parse errors, but can be started for inputs with validation errors to obtain additional diagnostic data from the computational codes.
A TRITON timetable density specification with 0 nuclides (indicating all nuclides) no longer produces a validation error in Fulcrum.
Fulcrum syntax highlighting was originally only available for inputs with all lower case keywords. Syntax highlighting is now case-insensitive.
Fulcrum will now correctly validate MCDancoff inputs that start particles in holes, geometry ring regions, and also supports FIDO statements.
The mixing table generation capability (Run/Mixing table) now also supports material aliases.
An issue was resolved on Windows operating systems where very large models would fail to run.
An issue was resolved in the temperature output edit for multigroup calculations where the nuclide's temperature was listed incorrectly. The results of calculations are not affected.
An issue was resolved on Windows where the KENO plot .png file was not created.
The prompt or total neutrons from fission for continuous-energy calculations (PNU parameter) now properly functions when requested. Previously, only total neutron emissions (prompt+delayed) were used regardless of the parameter setting.
The KENO mean free path edit previously provided the mean distance travelled before a collision or a boundary crossing. In this way, the mean free path depended in part on how the user modeled the geometry. The mean free path has been updated to correctly represent the mean distance between collisions.
An issue was resolved where incorrect gamma yield data were sometimes used in continuous-energy calculations. Testing shows a small decrease (few percent) in flux and response tallies after this update.
An issue was observed where the input keyword 'fluxplanes' was not supported
Previously, only 'fluxplane', 'fluxplan', and 'flux' were supported in SCALE 6.2.
An issue was resolved when performing restart decay in ORIGAMI when the originating reactor library could not be interpolated to 3.2% enrichment.
The “total” row in an ORIGAMI output of the neutron and gamma emission summary table is now suppressed. In this case, the “total” was of rows of particles/second and percentages, which did not make sense.
Additional backwards compatibility was added for reading SCALE 6.1 and earlier FIDO style input.
The default cutoff criteria for (alpha,n) sources has been changed from 1e-5 to 0 (no cutoff) based on the recommendation and experience of Rick Migliore of AREVA, who noticed that there can be 15% error in some cases in the (alpha,n) component of the total neutron source with the previous default cutoff of 1e-5. With the current implementation, the runtime increase incurred by using a zero cutoff instead of 1e-5 is negligible.
An issue was resolved in ORIGEN's beta principle emitter summary, which in the table formatting portion of the output, accessed an out-of-bounds array element which sometimes caused a crash.
An issue was resolved in calculating the time step when isotopics where loaded from an f71 file and simultaneously the user specified a start time of zero in the time block. The output file does show the actual time steps used in the calculation. The bug was fixed and additional inputs were added to control the material timeline and time specification.
An issue was resolved in Polaris where the use of size=2 for large water holes could sometimes cause the calculation to terminate.
The Polaris MOC transport solver has been enhanced. PN calculations are supported, with P2 treatment the new default scattering order. The accuracy of LWR reflector calculations is significantly improved with the addition of anisotropic scattering. Transport cross section edits have also been improved by computing hydrogen transport cross sections using the neutron-leakage-correction method.
Sampler was updated to enable multi-dimensional parametric studies to study the effects of combinations of variables on various responses. Using the READ PARAMETRIC block, a user may enter one or more variables that have been defined with uniform distributions along with the number of samples of each variable. Each variable is sampled with uniformly spaced values from its minimum to maximum value.
Sampler generates a summary table of the parametric study, including values for which the minimum and maximum of each response occurs. Sampler also generates PTP plot files showing the dependency of each response on each variable.
SCALE Runtime Environment
An issue was resolved where the HTML icon graphics were not being properly copied to the htmd directory.
Standard Composition Processing
SCALE was updated to provide improved warning messages for nuclides that are not present on the specified cross section data library. Several warnings can be observed with models with multiple compositions containing oxygen. Oxygen contains 0.2% at of O-18, which is not available in ENDF/B-VII.0 or ENDF/B-VII.1. The user can now export the environment variable MISSING_NUCLIDE_MSG_LEVEL=QUIET to reduce the amount of information printed for nuclides that are missing from the cross section library. The calculation continues with the nuclide removed from the model which will produce identical radiation transport and very similar activation/depletion results as using a cross section of 0.0 for each of the missing nuclide(s).
The material processor now supports the legacy SOLN formatted solution composition input that was supported in SCALE 6.1.
Two issues were resolved for T-DEPL calculations using the ASSIGN function. The ASSIGN function now updates the material concentrations correctly, and mass and volume edits of these materials in the system mass table are now correctly computed. All SCALE 6.2 T-DEPL calculations using ASSIGN should be reanalyzed with the SCALE 6.2.1.
Two issues were resolved in T-DEPL calculations using the new timetable SWAP feature. The timetable swap logic introduced time steps too close to one another causing some jobs to fail. The power normalization logic did not account for the swapped materials. All SCALE 6.2 T-DEPL calculations using SWAP should be reanalyzed with the SCALE 6.2.1.
An issue was resolved in multigroup depletion calculations where ADDNUX nuclides were missing their titles in output edits. The results of calculations are not effected.
An issue was resolved in KENO-based TRITON calculations where the geometry volume was being recalculated for each depletion step. The results are not affected, but the calculation efficiency is improved.
ALIAS expansion is now available in KENO-based TRITON calculations.
The job information data imbedded in sensitivity data files (.sdf) from TSUNAMI-3D continuous-energy calculations has been updated to provide the expected data.
SAMS mixture numbers that are longer than 4 digits are now printed in various sensitivity edits. This issue was a holdover from previous SCALE releases where mixture numbers could not exceed 2147.
Total Sensitivity Coefficients by Nuclide and Total Sensitivity Coefficients by Mixture are now displayed for all mixtures.
An issue was resolved were temperature interpolation could sometimes generate a negative cross section for Th-232.
A cell data issue was addressed where changing the origin of the zone definition in a multiregion model caused the calculation to fail.
An issue was resolved where using upper-case CE library name caused the problem to not run.
The default settings for legacy standalone (FIDO input) CENTRM calculations have been updated for consistency with XSProc-based CENTRM.
An issue was resolved where Double Heterogeneity calculations failed when CELLMIX was used.
An issue was resolved in CENTRM when using npxs=0 and nfst=6 where CENTRM would fail.
An issue was resolved in depletion calculations where memory grew per depletion step when running CENTRM with a user-input Dancoff factor specified.
The CENTRM "iterp" temperature interpolation message is now correctly identified as a warning (not an error) and limited to one occurrence per calculation.
Multigroup calculations involving multiregion cells with invalid zone radii (equal radii for more than one zone) produced a misleading 'lbar is zero' error. This message has been updated to indicate the line and column of both offending cell radii.
An issue was resolved with the thermal calculation routine using the CENTRM 2D MOC solver option (npxs=6). The impact on criticality calculations for SCALE sample input problems was very small. Calculations with other npxs values are not affected.
The MOC solver was also updated to use equal-volume radial mesh for discretizing each material zone. Previous versions of the MOC solver used equal-radii radial mesh. Although the equal-volume radial mesh is more appropriate, this code change has little impact on the accuracy of self-shielded calculations.
The AMPX code PUFF for generating multigroup cross section covariance data has been modernized and enhanced. PUFF has been rewritten in C++ to take advantage of the new C++ reading routines for ENDF (which will also support the new GND format), the new in-memory input/output resource for the COVERX file format, and the new resonance processing application program interface (API). In addition the new PUFF version adds covariance matrices for redundant reaction if not given by the evaluator (e.g. absorption reaction when fission and capture are provided). The processing for fission spectrum (chi) covariance matrices, previously available in a different module has been incorporated into PUFF.
Errors found in the old version of PUFF that affected some test libraries that where not distributed with SCALE have been corrected and results from previous and updated codes have been compared and found to be in good agreement except in cases where erroneous covariance matrices were previously computed.
A new key-word based input with more options has also been added, however, the old FIDO-style input is still supported.
Error reporting has been enhanced to report inconsistencies in the data.
A memory defect was resolved in the AMPX Y12 module which would sometimes cause calculations to fail.
SCALE compilation requires CMAKE 3.3 or higher. See ReadMe file for details.