Document audit reveals inconsistent experimental uncertainty reporting across ten cryogenic tank benchmarks, highlighting vulnerabilities in computational model validation.
This is the foundation document of the ElarionX CPMS Technical Note series. It fixes the six criteria every audit note applies, defines the verdict scale, lists the candidate experiments, records the verdict for each one audited so far, and carries the SHA-256 provenance of every source artifact. Computational models of cryogenic propellant tank self-pressurization and boil-off are widely described as validated. A validation claim compares a simulated value S with a reference value D, and the comparison error E = S - D only carries meaning if the uncertainty attached to D is known. Whether that uncertainty is available, and whether it travels from the original experimental reports into the modelling literature, is an empirical question. This registry answers it with documents rather than opinion. Ten referents are audited: K-Site (NASA TM-103804 and TM-105411, with the Stochl and Knoll boundary-condition report), MHTB (NASA TM-2003-212926 and the variable-density MLI report), SHIIVER (NASA/TP-20205008233), liquid nitrogen zero boil-off testing (NASA/TP-2017-219389), Saturn AS-203 (NASA CR-94045, 1967), the Tank Pressure Control Experiment and the Zero Boil-Off Tank experiment. Twenty-eight source and downstream-use artifacts — fourteen registry sources and fourteen downstream documents examined by the notes — are identified in manifest.csv by NTRS accession number or DOI and SHA-256 hash so any reader can confirm they hold the byte-identical document that was audited. An errata table in the note records every defect found to date, including defects found in this work by cross-checking and by a commissioned adversarial audit. Limitations are stated at length: this is exploratory (A0) work by one auditor with AI assistance, with no independent human review; eight candidate sources remain unobtained; and the candidate list derives from the bibliographies of four modelling papers, three of which share an author and an institution.
No takes yet. Share an insight, caveat, or question.
Luis Johan Elizondo Arias (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: