Stable engineering fixture · VOF CFD · source-neutral benchmark

Tank Sloshing CFD Validation

A fixed review point for subject-matter experts. Every number below comes from the pinned public release; report pages should link here when they claim analytical agreement, convergence evidence, or numerical qualification.

Engineering disposition

The published solver reproduces the rectangular-tank first-mode frequency within 0.2964–0.3264% across seven accepted free-decay cases. The stationary baseline, refined forced-roll sweep, complete-cycle fields, and sectional reconstruction support comparative screening within the published model and loading envelope. This is verification evidence, not full vessel validation.

Validation fixture at a glance

0.758054 HzAnalytical first-mode target
0.2964–0.3264%Published frequency-error range
7 casesFree-decay mesh and timestep fixture
0.783862%Two-case forced-roll exchange-amplitude mesh change

Review checks and acceptance meaning

Analytical agreement

Measured free-decay frequencies span 0.755580–0.755807 Hz against the 0.758054 Hz linear-wave target. All seven metric records are marked pass.

Timestep sensitivity

The 0.001 s and 0.004 s cases return 0.755717 Hz and 0.755702 Hz. This supports timestep stability for the measured frequency in this benchmark.

Mesh evidence

Five free-decay meshes from 1,600 to 25,600 cells remain close to the analytical target. The release deliberately classifies the convergence study as non_asymptotic_caution.

Forced-roll confirmation

The 20,480-cell and 69,120-cell T = 24 s cases provide a two-level mesh comparison. The published exchange-amplitude change is 0.783862%.

Numerical caveat retained and reduced

The original pressure follow-on configured a maximum Courant number of 0.35 but observed 0.47841479. The new fine rerun tightened the target to 0.25 and reduced the observed maximum to 0.32648391, with 0.00280% volume drift and final-cycle changes below 0.08%. It completed and converged, but still exceeded its configured target; both records therefore retain publish_with_courant_caveat.

Evidence matrix

QuestionPublished fixtureUseBoundary
Is the natural frequency reproduced?Seven free-decay cases vs. 0.758054 Hz analytical targetFirst-mode frequency verificationLinear rectangular-tank theory only
Is the result stable with timestep?0.001 s and 0.004 s free-decay casesFrequency sensitivity checkTwo timestep levels
Is mesh dependence quantified?Five free-decay meshes; two forced-roll meshes at 24 sScreening-grade mesh evidenceNo demonstrated asymptotic convergence order
Is the pressure follow-on numerically qualified?Original and tighter-Courant completion, conservation, cycle-change, and Courant auditsQualified comparative useConfigured Courant target remains exceeded
Are sectional loads internally consistent?24 s left/right wall-patch histories reconstruct the 9.677 MN·m aggregate harmonic amplitudeSectional load review within the disclosed patch splitNot a structural stress or local impact-pressure validation

What this fixture does not validate

It does not establish full-scale vessel behavior, structural stress, air-entrainment fidelity, local impact-pressure convergence, response at other fill levels or amplitudes, or anti-roll performance. Those require purpose-specific experimental or field benchmarks and additional numerical qualification.

Traceability

Dataset: aceengineer/digitalmodel-sloshing at revision aa37e6d365f3059674a49dc4bafbe4d64d12e5fd.

Website release digest: a0da3df2d32f03e5dd7014dc81af6aeb6f332a8570738eecca4a3c62c6f9af97. Primary fixture tables: cases.csv, metrics.csv, derived_metrics.csv, studies.csv, and qa_audit.csv.

This page is intentionally static and revision-pinned so its URL remains a consistent review fixture while interactive browsers and case selectors evolve.