View
215
Download
2
Category
Preview:
Citation preview
U N C L A S S I F I E D
U N C L A S S I F I E D
LA-UR-06-7034
Current Status of theNJOY Nuclear Data Processing Code System
and Initial ENDF/B-VII Data Testing Results
Presented at the
Workshop on Nuclear Data Evaluation for Reactor Applications (WONDER2006)
Cadarache, FranceOctober 9 – 11, 2006
A. C. (Skip) Kahler & R. E. MacFarlaneLos Alamos National Laboratory
U N C L A S S I F I E D
U N C L A S S I F I E DLA-UR-06-7034
NJOY Code Status• Bob MacFarlane may be retired, but is alive & well and
remains actively involved with NJOY development and data testing .
• LANL is maintaining and updating NJOY99 as needed to process ENDF/B-VII.
• An f95 version has also been developed and we expect to release this version by the end of 2006.– The absence, until this summer, of a code sharing agreement
between RSICC and the NEA has delayed release of this new version.
U N C L A S S I F I E D
U N C L A S S I F I E DLA-UR-06-7034
NJOY Code Status• NJOY99 is distributed through the Oak Ridge (RSICC) or
NEA code centers.– Updates are available from LANL at http://t2.lanl.gov.
• NJOY99.112 was the latest version available prior to the start of ENDF/B-VII “beta” testing.– Version 99.125 was released to the National Nuclear Data Center
(NNDC) at Brookhaven National Laboratory (BNL) to support ENDF/B-VIIβ2 testing (June, 2006).– This unofficial update is available at
http://www.nndc.bnl.gov/exfor2/4web/upn125.txt.– Version 99.161 was released to the NNDC at BNL to support
release of ENDF/B-VIIβ3 in late September.
U N C L A S S I F I E D
U N C L A S S I F I E DLA-UR-06-7034
NJOY Code Status• Some of the more significant changes since NJOY99.112
include:– Recognize β-delayed photon (MF=1, 12 & 14, MT=460) data.
– For this MT only, no file 3 data are required.– These data will be present in ENDF/B-VII 235U and 239Pu files.
– ACER “TYR” fix for fissile nuclides using File 6.– Identified at the June CSEWG data validation meeting and a
corrective patch has been provided by Andre Trkov.– INT=22 (unit base interpolation) for File 5.
– A long-standing feature in many JENDL fission product evaluations that will appear in the corresponding ENDF/B-VII evaluation.
– Continuous energy/angle sampling of thermal kernel data.– MCNP, version 6, will be able to use these data.
U N C L A S S I F I E D
U N C L A S S I F I E DLA-UR-06-7034
NJOY Code Status• Some of the most significant changes since NJOY99.112
include (con’t):– Coding to read the compact covariance format in mf32.
– Provided by Andre Trkov.– Corrected energy dependent fission pseudo-Q formula.
– Previously used formula did not reflect the energy dependency equations specified in the ENDF format manual for mf1, mt458.– This change made in update #136.– Calculated fission energy release changes by <1% at 1 MeV incident
neutron energy, and less with decreasing incident neutron energy.
– Miscellaneous explicit variable initialization, external subroutine declarations, increased array sizes in various modules.– This resolves various compiler warnings, but we are not aware of any
changes in answers resulting from these improvements.
U N C L A S S I F I E D
U N C L A S S I F I E DLA-UR-06-7034
NJOY Code Status• In-house testing with different compilers and different
optimization levels continues.– No differences were seen when the standard suite of test
problems was run on a LINUX PC and NJOY99.161 was compiled with either the Intel 9.0 compiler or the g77 compiler.
– Four reconr output files (from 393 ENDF/B-VIIβ2 input files) differ when NJOY99.161 was run on a Windows PC and compiled with the Intel 9.0 compiler and either O0 or O3.– All energy mesh were identical, but in each of these files one to three
cross sections differed in the least significant digit.– User’s compiling their own versions should continue to use O0.
• Final updates will be posted to the t2.lanl.gov website prior to the official release of ENDF/B-VII.
U N C L A S S I F I E D
U N C L A S S I F I E DLA-UR-06-7034
ENDF/B-VII Data Testing• Criticality Testing Using Benchmarks Defined in the
ICSBEP Handbook.– Testing for HEU, LEU, Pu and 233U Benchmarks– Spectra are “FAST”, “INTERmediate” or “THERMal”.– Materials are “METal”, “COMPound” and “SOLution”
• More than 1000 critical configurations have been calculated.– >700 by S. van der Marck.– Several dozen by J. C. Sublet.– Several dozen at Bettis, KAPL and ANL.– Several hundred at LANL.
• Representative LANL results follow.
U N C L A S S I F I E D
U N C L A S S I F I E DLA-UR-06-7034
ENDF/B-VII Data TestingCalculated Eigenvalues for LANL HEU, Pu and 233U Unmoderated Benchmarks
with ENDF/B-VI.8 and ENDF/B-VIIβ2 Cross Section Data Sets
0.9850
0.9900
0.9950
1.0000
1.0050
1.0100
1.0150
No
rmal
ized
Eig
en
valu
e
HMF-1(Godiva)
HMF-28(Flattop-25)
IMF7s(Big-10)
PMF-1(Jezebel)
PMF-2(Jezebel-240)
PMF-6(Flattop/Pu)
PMF-8c(Thor)
UMF-1(Jezebel-23)
UMF-6(Flattop-23)
Open squares are ENDF/B-VI.8.Closed squares are ENDF/B-VIIβ2.Error bars indicate the ICSBEP estimated 1σ experimental uncertainty.
U N C L A S S I F I E D
U N C L A S S I F I E DLA-UR-06-7034
ENDF/B-VII Data TestingCalculated Eigenvalues for LEU-COMP-THERM Benchmarkswith ENDF/B-VI.8 and ENDF/B-VIIβ2 Cross Section Data Sets
0.9850
0.9900
0.9950
1.0000
1.0050
1.0100
1.0150
No
rmal
ized
Eig
en
valu
e
LCT1 (2.35 w/o, USA - PNL)
LCT2 (4.31 w/o, USA - PNL)
LCT6 (2.6 w/o, Japan)
LCT7 (4.74 w/o, France - Valduc)
LCT22 (9.8 w/o, Russia - Kurchatov)
LCT24 (9.8 w/o, Russia - Kurchatov)
LCT39 (4.74 w/o, France - Valduc)
Open squares denote normalized eigenvalues with ENDF/B-VI.8 cross sections.Closed squares denote normalized eigenvalues with ENDF/B-VIIβ2 cross sections.Error bars indicate the ICSBEP estimated 1σ experimental uncertainty.
U N C L A S S I F I E D
U N C L A S S I F I E DLA-UR-06-7034
ENDF/B-VII Data TestingCalculated Eigenvalues for HEU (or LEU)-SOL-THERM
Benchmarks with ENDF/B-VIIβ2 Cross Sections
0.9750
0.9800
0.9850
0.9900
0.9950
1.0000
1.0050
1.0100
1.0150
1.0200
1.0250
0.00 0.10 0.20 0.30 0.40 0.50 0.60
Above-Thermal Leakage Fraction (ATLF)
No
rmal
ized
Eig
enva
lue
HST1 HST9
HST10 HST11
HST12 HST13
HST32 HST42
HST43 HST50
E7β2 Fit Various LST
ENDF/B-VI.8: kcalc(ATLF) = 1.0009(31) - 0.0020(83) * ATLF;ENDF/B-VIIβ2: kcalc(ATLF) = 1.0008(32) - 0.0011(85) * ATLF.Values in parenthesis represent 95% confidence intervals on the preceding regression coefficient.
The Monte Carlo eigenvalue uncertainty is typically less than 15 pcm.
Error bars represent ICSBEP 1σ experimental uncertainty estimates translated into reactivity.
U N C L A S S I F I E D
U N C L A S S I F I E DLA-UR-06-7034
ENDF/B-VII Data TestingCalculated Eigenvalues for PU-SOL-THERM Benchmarks
with ENDF/B-VIIβ2 Cross Section Data Sets
0.9750
0.9800
0.9850
0.9900
0.9950
1.0000
1.0050
1.0100
1.0150
1.0200
1.0250
0.00 50.00 100.00 150.00 200.00 250.00 300.00
Pu Concentration in Solution (grams/liter)
No
rmal
ized
Eig
en
valu
e
PST1 PST2
PST3 PST4
PST5 PST6
PST7 PST9
PST10 PST11
PST12 PST18
PST22 PST28
PST32
U N C L A S S I F I E D
U N C L A S S I F I E DLA-UR-06-7034
ENDF/B-VII Data Testing• Comparisons with LANL and Europe (Petten) for ENDF/B-VIIβ2.
– NJOY/MCNP used (independently) at both institutions.– 218 ICSBEP benchmarks in common.
– Petten/LANL eigenvalue ratio varies from 0.9964 to 1.0024.– 2/3rds of these ratios fall between 0.9990 and 1.0010.
• Comparisons with LANL and Europe (France) for ENDF/B-VIIβ2.– NJOY/MCNP used at LANL; NJOY(CALENDF)/Tripoli used at Cadarache.– 43 ICSBEP benchmarks in common.
– Tripoli/MCNP ratio varies from 0.9996 to 1.0010.– Common benchmarks include HEU-MET-FAST, IEU-MET-FAST, PU-MET-
FAST, HEU-SOL-THERM and LEU-COMP-THERM.
• Comparison with LANL (MCNP) and Naval Reactors (RCP01/RACER).– Excellent agreement in calculated eigenvalues has been observed for
many years (mostly HEU-SOL-THERM).
U N C L A S S I F I E D
U N C L A S S I F I E DLA-UR-06-7034
ENDF/B-VII Data Testing• Observations
– Important contributions have been received world-wide and are incorporated into the ENDF/B-VII neutron library.– new actinide evaluations
– new fission product evaluations
– new light isotope evaluations
– new scattering kernels
– Significant criticality testing has been performed world-wide.– various US Laboratories (LANL, Bettis, KAPL, ANL)
– van der Marck (Petten)
– Sublet (Cadarache)
U N C L A S S I F I E D
U N C L A S S I F I E DLA-UR-06-7034
ENDF/B-VII Data Testing• Observations
– xxx-MET-FAST calculated eigenvalues are significantly improved.
– LEU-COMP-THERM calculated eigenvalues are significantly improved.
– HEU (or LEU)-SOL-THERM calculated eigenvalues remain close to unity.
– PU-SOL-THERM calculated eigenvalues are little changed and remain biased high by ~0.4% ( approximately 0.1% higher than with ENDF/B-VI).
• The final ENDF/B-VII neutron cross section data files represent a significant improvement over the corres-ponding ENDF/B-VI data files.
Recommended