What’s new in 1.5
ACTINV 1.5.0 is mostly about work that used to leave ACTINV for another tool: reading a FISPACT-II deck, printing FISPACT-style tables, estimating dose at a distance, tracing production routes, and checking an inventory against a transport or clearance table. This page is a tour by task. Versions and releases has the version table, and the 1.5.0 release notes have the full list.
Nearly everything below is opt-in. A problem that does not name a new section computes what it computed before; the changes to default results are collected in Results that may change. The nuclear-data catalog stays at 1.1.0. The desktop is 0.4.0-preview.1, and the desktop release page describes it.
Start from a FISPACT-II deck
actinv import-fispact converts a FISPACT-II input deck into an ACTINV problem and writes a report that marks every keyword as mapped, approximated, noted or unsupported:
actinv import-fispact DECK.i OUT.json [--fluxes FLUXES] [--files FILES] [--library LIB.npz] [--decay DECAY] [--covariance COV.npz] [--strict]
All 132 public FNS decks import. POWER, WALL, PARTITION, OVER, GRPCONVERT, DOSE 2, GROUP, BREMSSTRAHLUNG, PATH/ROUTES and ATWO are mapped, and the importer accepts three-letter keyword abbreviations such as TOL. Several keywords still need you to add a section by hand, and a few have no equivalent; Move a FISPACT-II deck to ACTINV has the full keyword map, including the new Coming from FISPACT-II table.
The deck gives ACTINV the problem, not the data. ACTINV uses its own activation and decay libraries, so results differ from a FISPACT-II run by the difference between the data sets as well as by the solvers.
The same importer is in the desktop and the browser workbench as Import FISPACT-II deck (Ctrl/Cmd+I in the desktop).
Reproduce FISPACT-II features
These are the FISPACT-II capabilities added in 1.5. Each is an optional section of the problem specification.
-
Spectrum conversion.
spectrum.rebinconverts a spectrum between group structures inside the run, and every FISPACT-II neutron structure fromfispact-66tofispact-1102(to 1 GeV) is available. See Spectrum conversion and Group structures. -
Schedule controls. Steps can be driven by a wall loading, a power density or a specific power (with a groupwise kerma table), and a step can apply an instantaneous elemental partition. Power is renormalized on sub-intervals within a step;
max_substeps: 1gives FISPACT-II’s single normalization. See Kerma heating and power-driven steps. -
Overrides and decay-only runs. Per-nuclide, per-reaction cross-section, half-life and uncertainty
overrides, and cooling runs from decay data alone. See Named data overrides and Decay-only runs. -
Fission. Spontaneous-fission products, the cumulative fission count, burn-up and the spontaneous-fission neutron source. See Fission count, burn-up and spontaneous-fission neutrons.
-
Projectiles and libraries. Triton and helion projectiles, and switching the activation library between steps. Neither triton nor helion has a library in the standard catalog, and neither has been tried on real data, so treat them as untested beyond synthetic fixtures.
-
Query data without a run.
actinv inspectprints decay data, group and spectrum-collapsed cross sections, fission yields and group boundaries:actinv inspect decay DECAY_PATH [NUCLIDE...] actinv inspect xs LIBRARY.npz [--target NUCLIDE] [--mt MT] [--spectrum SPEC.json]See Query data files.
Get new outputs
Tables and a step summary
The optional summary section adds the full inventory with grams, activity, alpha/beta/gamma heat and half-life per nuclide, step totals (including activity and heat without tritium), an element table, dominant lists with cumulative percentages, and MIND/NOSTABLE-style filters. A summary.depletion table adds, for each initial nuclide, the depleted fraction and the removal rate split between decay and each reaction. See Step summary and Depletion table.
actinv report renders an existing result as FISPACT-style tables (TAB1 to TAB4, step totals, dominant and element tables, a plot-ready time series) in csv, tsv, markdown or text. It runs no solver:
actinv report RESULT.json [--tables LIST] [--format csv|tsv|markdown|text]
[--out DIR] [--step N|all] [--nuclides LIST]
See Render result tables.
Photons and dose
- The 22-group Steiner photon structure alongside the 24-group structure. See Steiner 22-group structure.
photon.point_sourcegives the gamma air-kerma rate at a distance from the sample (at least 0.3 m), like FISPACT-II’sDOSE 2. See Point-source dose.photon.bremsstrahlungreports the bremsstrahlung photons and their dose as a separate block; it is not added to the decay-photon dose. See Bremsstrahlung.
The dose values are null when the photon response has no air attenuation curve or when nuclides have decay-record gamma power with no photon spectrum, rather than a number that looks complete. The shipped response has no air attenuation curve, so a point-source dose needs a rebuilt response or air_attenuation: false; the specification explains both. Approximate gamma spectra for such nuclides (SPEK) are not estimated; the power that has no spectrum is reported instead.
Production pathways and routes
The optional pathways section traces multi-link routes from a source nuclide to a target, including loops, and integrates each route exactly through every schedule step. Each target reports a closure: the total production, the sum over the listed routes, and what the routes leave unexplained. See Read the pathway routes and Pathways and routes analysis.
Regulatory tables
A problem can now compare an inventory with bundled public tables: the A1 and A2 values of 49 CFR 173.435, the exempt values of 49 CFR 173.436, and the clearance and exemption levels of Euratom Directive 2013/59 Annex VII. They are opt-in (radiological.table.path set to bundled:NAME). ACTINV chooses no jurisdiction for you, and whether a table applies to your material and use is your decision.
actinv report has a table for them. It is built only when you name it:
actinv report RESULT.json --tables regulatory
It lists, per step, each clearance, waste or A2 response with its value and the share of the activity the table covers. Some rows have no finite limit, and progeny can be included in a parent’s value; the tables record both, so require_complete can be satisfied by an activation inventory. See Bundled regulatory tables.
ICRP 119 dose coefficients are not bundled because of their copyright. If you hold the publication, scripts/regulatory/fetch_icrp119.py builds a table for your own use; see Building the ICRP 119 dose-coefficient table locally.
Treat self-shielding by geometry
self_shielding.geometry applies a resolved-resonance self-shielding correction for a foil, wire, sphere or cylinder using the universal sigmoid curve, as FISPACT-II’s SSFGEOMETRY does. It reads ENDF resonance files you supply, including R-matrix-limited (LRF=7) evaluations such as ENDF/B-VIII and FENDL-3.2c. self_shielding.dilution_overrides sets the background dilution for chosen groups when you use a Bondarenko table. See Geometry self-shielding and Dilution overrides.
A resonance is used only inside the energy range its evaluation resolves; see Results that may change.
Understand the uncertainty
study.robustness.statisticsadds, for each response, the sample mean, standard deviation, skewness, exact percentiles and a histogram, the input-output Pearson and Spearman correlations by sampled parameter, a response-response correlation matrix, and bootstrap errors next to the first-order standard deviation. It is the counterpart to FISPACT-II’sSENSITIVITYandMCSAMPLEoutput, and it is absent unless you ask for it. See Monte Carlo sampling statistics.- A single sample has no standard deviation, correlations and bootstrap errors need at least three samples, and
first_orderreadsnot_comparablewhen channels beyondcross_section_mf33are sampled. - The removal-rate uncertainty in the depletion table uses the same MF=33 covariance as the rest of the run.
Use the 0.4 interface
Desktop 0.4.0-preview.1 and the browser workbench now have thirteen pages in three groups, with the same look as before. The browser still never runs the solver.
- Outputs (under setup) is one place to enable the step summary and depletion, decay photons and dose, production pathways and routes, regulatory tables, geometry self-shielding, uncertainty with each of its channels (including flux tally statistics) and gas production. A status line on each card gives the solver’s own message when two choices conflict.
- Irradiation & cooling has a Drive for each step (flux multiplier, wall loading, power in W/cm3 or W/g), and Material has an optional density.
- Charts, Tables, Photons & dose, Pathways and Ledger & certificate share one result bar and one chosen step. Tables exports CSV that is byte-identical to
actinv report. - Import FISPACT-II deck converts a deck with the importer’s per-keyword report. Unlike the CLI, the desktop converts a deck that irradiates but has no fluxes file by inserting a flagged placeholder spectrum; replace it before running. The CLI refuses such a deck.
See Use the desktop and Use the browser workbench.
Results that may change
Most 1.5 changes are opt-in. These are not, or they correct a result you may already have.
Defaults that moved
- A fission after another decay transition (beta-delayed fission, RTYP 2.6) was counted as spontaneous fission of the parent. It is now counted in the new ledger key
unmodelled_decay_branches_to_leakage, sospontaneous_fission_branches_to_leakagefalls from 129 to 126 on the shipped data. Inventories do not change. The new key appears only when it is nonzero. - Trace-mode pathway attribution now includes production fed by the decay of a radioactive bulk nuclide, and
pathway_closureis 1 when a nuclide in the inventory has no chain at all. Inventories do not change. In SS316 the leading Ru-100 route is Mo-100 double-beta decay; in tungsten, W-186 alpha decay feeds Hf-182 (the closure of the shipped shielding example falls from 0.98 to 5e-16); and the natural-uranium closure falls from 1.0 to 4e-15. - Trace mode no longer holds radioactive initial material constant.
autonow selects the coupled solver when any initial nuclide decays by 1e-6 or more over the schedule, and an explicit trace request is refused there. Restart or cooling runs from an activated inventory that did not setcoupledshould be rerun. - Trace-mode pathway attribution used to ignore the step’s flux multiplier and spectrum override, and lost production after a cooling gap. It now uses the solve’s own matrices, and
pathway_closureshows the difference.
Opt-in features that were corrected
- Bremsstrahlung used the decay record’s average electron energy per decay of the parent as the energy of one electron per decay. It now uses the beta-minus branching and the matching energy per electron. Partial beta emitters such as Cu-64 change (about 2.6 times more energy); pure beta-minus emitters do not.
- The bundled regulatory tables were regenerated, so their SHA-256 values changed. A spec that declares a bundled table’s hash must use the new one;
actinvfills a missing hash itself. - Geometry self-shielding uses a resolved resonance only inside the range [EL, EH] that its evaluation resolves. TENDL-2025 W-186 is resolved to EH = 4341 eV and loses its factors above that energy; ENDF/B-VIII.1 W-186 loses one group above 10 keV. Whether FISPACT-II clips resonances at the same energy is not verified here, so a W-186 comparison with FISPACT-II may differ for this reason. A
geometry.density_g_cm3that disagrees with the material density is refused.
The changelog lists the remaining fixes (schedule refusals, depletion and sampling statistics, importer robustness).