The MHTB Primary Is a TVS Performance Study, Not a Self-Pressurization Characterization

What this note is: a reading of the primary document beneath the MHTB validation citations examined in Note 001 (doi:10.5281/zenodo.21895605), applied against the six referent-quality criteria of the series. Every factual claim about a specific document is traceable to a passage in that document, listed in Section A of the sources, with page numbers. (“Audit” here means documentary audit, as defined in Note 000.)

What it is not: a criticism of the test team — the reading finds the opposite, again — and not a claim that any downstream paper misused this document. The claim is narrower: the public record beneath a standard validation citation is a different kind of document than the citation suggests, and a modeler needs to know that before building on it.

I. The debt from Note 001

Note 001 examined a 2015 NASA paper that validates CFD models of tank self-pressurization against data “obtained from the MHTB self-pressurization experiment,” and scoped every MHTB claim carefully, because the primary test report had not yet been obtained. That was the series’ declared weak point: judging a referent through the papers that cite it is exactly the failure mode this project exists to name.

The audit behind this series — Technical Note 000 — located and audited the principal public document behind that citation, and in doing so found that the gap moves rather than closes: the underlying tabulated record for the long-lockup tests remains unresolved (§V). The document is NASA/TM—2003–212926: Spray Bar Zero-Gravity Vent System for On-Orbit Liquid Hydrogen Storage, by Hastings, Flachbart, Martin, Hedayat, Fazah, Lak, Nguyen and Bailey, October 2003, 161 pages. Eight authors across four organizations — Alpha Technology, NASA Marshall, Boeing, and Sverdrup Technology; the lead author sat at a contractor, not at NASA.

The chain of custody is short and verifiable. The 2015 paper’s reference list contains exactly two MHTB documents: the multilayer-insulation test report (NASA/TM—2001–211089) and this one. Within the 2015 paper’s published reference list, this is the MHTB document that contains the self-pressurization behavior. What follows is what a modeler finds on opening it.

II. What the document actually is

Here is the finding that reframes everything else in this note:

The primary public document beneath the MHTB self-pressurization citations is not a self-pressurization characterization report. It is a performance report for a pressure-control device.

The Multipurpose Hydrogen Test Bed itself is exactly what the modelling literature says it is: a flight-representative liquid hydrogen tank — aluminum 5083, cylindrical, 3.05 m diameter by 3.05 m with 2:1 elliptical domes, 18.09 m³, insulated with a foam substrate and a 45-layer multilayer insulation blanket, tested at 90, 50 and 25 percent fill under two heat-leak series. A serious, large-scale, well-instrumented test article.

But the test this TM reports is a thermodynamic vent system test: a spray-bar mixing and venting device whose job is to keep tank pressure inside a control band. Self-pressurization appears as the phase between the device’s interventions:

“…the tank was locked up and allowed to self-pressurize until the ullage pressure attained the maximum tank pressure setpoint of 138 kPa (20 psia). Upon reaching this pressure, the recirculation pump was turned on and mixing continued until the ulllage [sic] pressure reached 131 kPa (19 psia), the minimum setpoint.” (p. 63; the triple-l is the original’s, verified against the page image)

Two consequences follow for anyone using this as a self-pressurization referent.

First, the clean free-rise data are the lockup phases — bounded to a 7 kPa band and terminated by active intervention. The tabulated self-pressurization content of the report is Table 5: saturation-pressure rise rates after lockup, by fill level and heat-leak series (the high-heat-leak rates ran 2.9 and 3.3 times per its Table 5 (the TM’s own prose prints “2.9 and 3.1 times” on p. 68 — a prose-vs-table inconsistency internal to the source, reported here per Erratum 26’s policy) the low-heat-leak rates at 90 and 50 percent fill).

Second — and within this TM the numerical record bottoms out here — the long self-pressurization tests that the modelling literature actually wants, the 1996 and 1998 test series appear in this TM only as model-comparison figures — P263968E/F (1996) and P263981D (1998) at 90 percent fill, P263981T at 50 percent, per the TM’s own figure captions (Figs. 54, 71, 75). One association must be stated as inferred: the 2015 modelling paper prints no test identifiers at all — it describes its cases only as the 50 % and 90 % fill levels — so the link from its comparisons to these segments rests on fill level and campaign description, not on a printed ID. Not as data tables. The figures exist to show how the authors’ own analytical model tracks the tests; the underlying time histories are not tabulated anywhere in the document. A modeler wanting those tests as numbers must digitize curves — from figures that were themselves drawn to compare a model.

Note 001 found that experimental context did not travel into the papers it examined. This is something different and prior, and it does not depend on how any downstream paper behaves: beneath this citation, the tabulated record of the headline use-case does not appear in the public primary at all.

A companion paper, and what it does not close

There is a second public document by four of the same authors — Hedayat, Bailey, Hastings and Flachbart, “Test Data Analysis of a Spray Bar Zero-Gravity Liquid Hydrogen Vent System for Upper Stages” (NTRS 20030106051, 8 pp.) — which reports directly on the 1996 and 1998 campaigns. I found it after drafting this note, and it is worth reporting exactly, because it changes the picture in both directions.

What it adds is real. Its Table 1 tabulates seven test segments, more than the TM discusses: fill level, ambient heat leak in watts, year, test segment number, operation mode, and two derived comparison metrics — a model/test pressure-rise ratio and a cycle-rate error in percent. It names segments this note had not seen, including P263968G, P263968K, P263968L, P263981D and P263981X. Anyone auditing MHTB should have this document; the audit behind this series did not, and that is a gap in our search rather than in the record.

What it does not add is the thing a modeler needs. The tabulated quantities are comparisons, not measurements: ratios and percentage errors against the authors’ own analytical model. The pressure and temperature time histories remain, here as in the TM, in figures. The action item at the end of this note stands unchanged.

And the paper continues the pattern of §III below. It states plainly why its model diverges during long lockups — the analysis “assumes that the liquid and ullage are each represented by a single node, did not accurately simulate the complex energy exchange that actually occurred at the liquid-vapor interface.” The sentence appears twice, in near-identical form, in eight pages: it is the authors’ central explanation, not an aside. And it rhymes with, without repeating, what the K-Site authors found in 1992 — a single-node treatment cannot represent the interface, just as a single vertical sensor rake could not resolve the stratification it produced. The two cases are not the same failure: K-Site’s is a measurement-representativeness caveat (a single rake cannot see radial nonuniformity), the companion’s a model-abstraction one (a single liquid node cannot resolve interface exchange). What they share is the deeper point: collapsing spatially distributed liquid–ullage–interface behavior into an under-resolved representation breaks an energy interpretation — on the measurement side at K-Site, on the model side here.

The paper contains no statement of measurement uncertainty at all.

III. The candor, again

For MHTB, as for K-Site, the original team turns out to have been more forthcoming than the downstream paper examined. Four practices deserve naming.

The instrument accuracies are stated. Ullage pressure P4 is an MKS Baratron 0–666 kPa absolute transducer “with an accuracy of ±0.02 percent”; the pressure-control loop used a second Baratron (0–133 kPa, ±0.02 percent) plus a 1-torr-head differential transducer at ±0.04 percent (P4 on p. 6; the control-loop pair on p. 60). During insulation-performance determination the ullage pressure was held in a control band of ±9×10⁻⁴ kPa about the setpoint. This is precise pressure work, stated as such — and, as Note 001 documented, none of it appears in the downstream validation paper. [Corrected at v1.0, Erratum 26: the TM prints this band as ±9×10⁻⁴ kPa on p. 65 but as ±0.001 psi (≈0.0069 kPa) on pp. 60, 62, 74 and 96 — a factor-7.7 internal inconsistency this audit initially copied without detecting. The conservative ±0.0069 kPa governs any use of this number.] [Scoped at v1.0, Erratum 26: the specification does not state percent-of-reading versus percent-of-full-scale — at 133 kPa, 0.02 % of reading would be ±0.027 kPa, larger than the control band — an ambiguity that must be resolved before this figure can serve as a u_D contribution.]

A closure failure is reported and worked. The TVS heat-extraction rate derived from the vent measurements does not close against the measured tank heat leak. The report investigates, and rules instrumentation out, in its own words — a claimed exclusion the report supports arithmetically —:

“One source considered, but ruled out, was instrumentation uncertainties. The error in measured quantities would have to have been much larger than the instrumentation uncertainties to yield the additional enthalpy necessary…” (p. 71)

It then names the likely culprit — the method of calculating vent mass flow rate, an extremely small quantity (averaging ≈0.0035 kg/s in the low-heat-leak series, where a gain of only ≈0.0009 kg/s would close the gap). Readers of Note 001 will recognize the structure exactly: it is the K-Site energy-balance episode of 1992, repeated a decade later on a different tank at a different NASA centre — quantified instruments, an honest non-closure, the instrument contributions priced against their own error bars, a physical hypothesis named.

That this recurs across two centres is the most interesting thing in this note. Whether it amounts to a pattern of practice in the field, rather than a habit of two organizations, is more than two instances can establish; §VI returns to the limits of that claim.

Hardware misbehavior is declared. Part of the high-heat-leak series is qualified: “believed to be due to sometimes erratic J-T valve operation coupled with limited test durations” (p. 66). The less-consistent data are flagged as less consistent, with a hardware reason.

The model comparisons disclose their own adjustment — and show the before picture. To correlate the mixing/vent cycling against the 1998 test, “the model start time with venting, was shifted by 60,350 s (to 187,460 s) to better align with the test data” (p. 89). That is a 16.8-hour shift. And the report plots the unshifted comparison too (Figure 75), so the reader sees exactly what the adjustment did. For the report’s own purpose — correlating cycle characteristics of a pressure-control device — this is a declared, legitimate alignment. But the declaration lives here, on page 89 of a TVS report. Anyone downstream who cites those figures as evidence of predictive skill, without carrying the shift, would be treating a post-hoc aligned comparison as though it were an unadjusted prediction. Whether anyone has done so is a question for the audit, not an accusation of this note.

IV. The three things a modeler must know

1. Bulk liquid temperature is one diode. The quantity used to monitor bulk liquid state is silicon diode TD23, at the 11.5-percent fill level, 53.3 cm above the tank bottom, which “is considered to be representative of the bulk liquid temperature” (p. 6). One point, low in the tank, standing for 18 cubic meters of liquid. Note 001 quoted the K-Site authors questioning — from their own energy balance — this class of assumption at their conditions: their measured liquid temperatures appeared higher than a bulk-consistent temperature, the hedge theirs and preserved. Here, the same class of assumption is the basis of the reported saturation conditions, and there is no equivalent published closure analysis of the liquid to test it during lockup. It may be adequate — a mixed tank after spray-bar operation is a better-stirred tank than K-Site’s quiescent one. But adequate is a finding someone would have to establish; the report asserts representativeness, reasonably, for its own control purposes.

2. The fluid’s spin composition is unstated — for the third time. The words ortho and para do not appear in the document. Three referents into this series (1992, 2003, 2021), the measured composition of the actual test fluid has yet to appear in a primary record.

3. Within this TM, the published numerical record is eight numbered tables plus figures. Tabulated: pump characteristics (Table 1), measured tank heat leak (Table 2), mixing-cycle durations and intervals (Tables 3–4), post-lockup saturation-rise rates (Table 5), derived heat extraction with its declared discrepancy (Table 6), model/test correlation metrics for seven representative segments (Table 7), and the tanking table (Table 8) — a census of every numbered table caption, verified by machine at v1.0. None is a time-history series: every pressure and temperature time history, including the 1996/1998 self-pressurization tests, is plotted, not tabulated. And I found no statement that the underlying data files are archived anywhere a researcher can reach. Stated carefully, as before: absence of a statement in the document read, not evidence that no archive exists.

The downstream lineage, updated at v1.0

Earlier drafts assessed criterion 6 against the 2015 paper alone. A dated search at v1.0 located more — the expiring-negative discipline of Errata 16–17, applied here before freeze:

Majumdar, Valenzuela, LeClair & Moder, Numerical modeling of self-pressurization and pressure control by a thermodynamic vent system in a cryogenic tank, Cryogenics 74 (2016), doi:10.1016/j.cryogenics.2015.12.001 — located, paywalled, not yet obtained. [Superseded at v1.0, Erratum 21: its public NTRS conference version — same title, same authors — was located, obtained and audited the same day; see below.]
Kassemi & Kartuzova, Effect of interfacial turbulence and accommodation coefficient on CFD predictions of pressurization and pressure control in cryogenic storage tank, Cryogenics 74 (2016), doi:10.1016/j.cryogenics.2015.10.018 — located, paywalled, not yet obtained. [Superseded at v1.0, Erratum 21: its public NTRS conference version — same title, same authors — was located, obtained and audited the same day; see below.]
Patel, Validation of Pressure Control in a Flight-scale Liquid Hydrogen Tank using a Spray Bar, Space Cryogenics Workshop, May 2025, NTRS 20250004637, 17 pp. (presentation), SHA-256 8d9137873c0c2a1bb4c69718c4c8534b12c61a4de5b36617a4cb536c2d5025a6 — obtained and audited: it cites this TM, and carries zero statements of measurement uncertainty, zero ± on measured quantities, and none of the TM’s declared context (energy discrepancy, alignment shift, single-node caveat); its accuracy mentions refer to the model. A presentation is weight-limited, and is scored accordingly — as the ZBOT deck was in Note 000.
Soriano, Kartuzova, Kassemi & Hauser, Modeling Self-Pressurization and Spray Bar Pressure Control of a Cryogenic Storage Tank in Normal Gravity, AIAA SciTech Forum, January 2025, NTRS 20240016269, 29 pp. (presentation), SHA-256 796eb38586fdee3c92fb86b49e06b722d14742e63c22e6c88556116472802190 — obtained and audited: it concludes that a “validated CFD model accurately predicts pressure and temperature evolution” of the MHTB experiment, and carries zero statements of measurement uncertainty, zero ± on measured quantities, and none of the TM’s declared context. This item was located hours after this note’s dated downstream search had missed it — by a third external AI reviewer following a fourth AI reviewer’s garbled citation. The negative-claim discipline exists for exactly this; Erratum 20 records it.
Majumdar, Valenzuela, LeClair & Moder, Numerical Modeling of Self-Pressurization and Pressure Control by Thermodynamic Vent System in a Cryogenic Tank, conference paper, 2015, NTRS 20170006183, 17 pp., SHA-256 407e21d5983fbabe465e5c16c175f8a8a45a20c1a6b9447d1c78d43e5d6b82f6 — obtained and audited: the public conference version of the first Cryogenics 74 (2016) paper above — same title, same authors. GFSSP nodal model against MHTB 50 % fill data; an MLI degradation factor of 4.0 is assumed; observed discrepancies are attributed to “the uncertainties of heat and mass transfer coefficients at ullage-liquid interface” — model-side uncertainty. Zero statements of experimental measurement uncertainty, zero test-series IDs, none of the TM’s declared context.
Kassemi, Kartuzova & Hylton, Effect of Interfacial Turbulence and Accommodation Coefficient on CFD Predictions of Pressurization and Pressure Control in Cryogenic Storage Tank, conference paper, 2015, NTRS 20150021282, 25 pp., SHA-256 900c77631c5d23ef82c1575d46739768d6fdbb54f61740ce1a7f95af046749a9 — obtained and audited: the public conference version of the second Cryogenics 74 (2016) paper above — same title, near-identical author set. CFD pressurization and pressure-control study drawing on both the MHTB and K-Site experiments. Zero statements of experimental measurement uncertainty, zero ± on measured quantities, zero test-series IDs, none of the TM’s declared context.
Soriano, Kartuzova, Kassemi & Hauser, Modeling Self-Pressurization and Spray Bar Pressure Control of a Cryogenic Storage Tank in Normal Gravity, AIAA SciTech Forum, January 2025, full conference paper, NTRS 20250000041, 30 pp., SHA-256 ebc8e048c818151436cd87c29d4fc2b4b6c78c9a7fb67b6bd9957439b6bfaf53 — obtained and audited: the full-paper version of the presentation examined under Erratum 20, public on NTRS while this note was auditing the slides. The initial tank temperature profile is “set to match the experimental data”; the word family accuracy/accurate/accurately appears 18 times (11/3/4), every one describing the model. Zero statements of measurement uncertainty, zero ± on measured quantities, zero test-series IDs, none of the TM’s declared context.
Soriano, Kartuzova, Kassemi & Hauser, Modeling Spray Bar Pressure Control of a Large-Scale Liquid Hydrogen Propellant Tank in Normal Gravity, Space Cryogenics Workshop 2025 manuscript, NTRS 20250005060, 36 pp., SHA-256 87efd7b6efef5130e5842b56c0d4268ea1e31a3ec9dba2cf33d443b7c81562dd — obtained and audited: MHTB 25 % fill spray-bar validation. It names model-side uncertainties expressly (the quarter-tank approximation, the diffuse VOF interface) and records a declared experiment-side gap: detailed spray characterization data were not collected, so “both the experimental and modeling efforts lack sensitivity assessments related to these factors” — more context about the experiment’s own limits than any earlier NASA artifact in this lineage carries. Still: zero statements of MHTB measurement uncertainty, zero test-series IDs.
Wang, Jordan & Xiao (Karlsruhe Institute of Technology), Non-Equilibrium Liquid Hydrogen Tank Modeling When Thermal Insulation Fails, Hydrogen Safety (2025), open access, doi:10.58895/hysafe.35, 19 pp., SHA-256 c94e25a4568a64c5d420c8a26569c8f4568fc35c43ddb766bf2cc614f511c284 — obtained and audited: the first non-NASA downstream use examined in this series. It validates a lumped non-equilibrium model against “the NASA MHTB 50% fill level test (p263981t, 1998)” — the only downstream document examined in this series that prints a test-series ID — cites the TM directly, and declares its initial-condition assumption (saturated initial temperatures) with a pointer back to the TM. The uncertainty it analyzes (evaporated mass, after Saury et al. 2002) is model-side; it carries none of the TM’s measurement uncertainty and none of its declared anomalies. Twenty-two years and one continent from the experiment, the external paper preserves more referent identity than any NASA downstream artifact examined here.

Eight downstream artifacts audited, three research groups, zero carriers of the TM’s declared context — and a gradient worth naming: referent identity travelled furthest in the hands farthest from the source (the external group alone prints the test ID), while measurement uncertainty travelled nowhere. Whether that gradient generalizes is a question for the registry, not for this note.

The pattern of Notes 001 and 002 — data travels, qualifications do not — held across every artifact examined, with one gradient recorded below: the external 2025 paper carries more referent identity than any NASA downstream document — and still no measurement uncertainty.

V. The scorecard

Criterion (as refined in Note 002)	MHTB primary (TM—2003–212926)
1. Machine-readable measurement data?	Partial — result tables numeric (heat leak, cycle timing, rise rates, extraction); all time histories plotted only; the 1996/1998 self-pressurization tests exist here only as model-comparison figures; no public archive statement found
2. Boundary conditions?	Partially documented — tank heat leak measured per series; spatial distribution of that heat leak not characterized; environmental shroud described
3. Measurement uncertainty?	Strong on pressure (±0.02 % stated, control band quantified), by-type elsewhere; no propagation into derived quantities — and the report itself shows why that matters, in its own heat-extraction discrepancy
4. Geometry?	Substantially documented — material, dimensions, dome geometry, insulation build
5. Fluid state?	Absent — ortho/para composition unstated
6. Does the quality information travel into downstream use?	PARTIAL ASSESSMENT — absent from all eight downstream artifacts audited at v1.0 (2015–2025, three research groups; §IV, The downstream lineage). The two 2016 journal versions remain unread, but their public conference versions — same titles, same authors — are audited. One gradient: the external 2025 paper prints the test-series ID (referent identity travels) yet still carries no measurement uncertainty. OPEN for further downstream use classes

Criterion 6 deserves two more paragraphs, one of them uncomfortable.

The comfortable one: in Note 001, the 2015 paper’s five-fold disclaimer looked like a limitation of the modelling work. Read against the primary, it resolves into something cleaner — it is a faithful report of what the experiment could and could not see. The mechanism the models exist to represent was never a measured quantity anywhere in this chain.

The uncomfortable one is about the strength of my own evidence. For MHTB, the audited downstream sample now stands at eight artifacts — the 2015 CFD validation paper; the public 2015 NTRS conference versions of both 2016 journal papers; the Patel and Soriano 2025 presentations; the Soriano 2025 full conference paper and its Space Cryogenics companion manuscript; and one external validation, Wang, Jordan & Xiao 2025 (Karlsruhe Institute of Technology) — across three research groups, two of them NASA lineages sharing authors, one independent. Note 001 examined five documents for K-Site — the 2015 and 2022 computational papers and the three 2025 versions of one study — of which four share both authors and an institution. The related weakness runs one level up: the four papers whose bibliographies define the referent list in Note 000 include three that share an author and an institution. Neither fact was noticed until it was pointed out to me. For this referent specifically, the audited downstream sample stands at the eight artifacts of §IV — none carrying the TM’s declared context. That is enough to say what this paper does with this referent’s caveats, and it is the honest limit of the claim. It is not enough to characterize how the field treats MHTB data, and this note does not.

VI. What three referents now show

The series so far, in one table:

Referent	Primary’s conduct	What the public record cannot support
K-Site (1992)	Instrument accuracies + closure failure + representativeness caveat, all published	The caveat did not travel into the papers examined
SHIIVER (2019–21)	Prospective measurement requirements + redundant instrumentation + declared degradations	Time histories plotted only; no consolidated budget; archive unstated
MHTB (2003)	Instrument accuracies + energy-discrepancy analysis + disclosed alignment	The headline use-case (long self-pressurization) is not tabulated in the public primary at all

The left-hand column is the stronger one, and it is worth separating from the right — but it is weaker than I first wrote, and the correction is worth making in public because of how I got it wrong.

My draft called these “three test teams, three institutions, three decades.” That is false, and the evidence was on the title pages I had already read. NASA Lewis Research Center and NASA Glenn Research Center are the same institution — Lewis was renamed for John Glenn in 1999, a fact the SHIIVER report itself states in passing (“Lewis Research Center (now Glenn Research Center)”). K-Site 1992 gives its authors’ address as Cleveland, Ohio 44135; SHIIVER 2021 gives Glenn Research Center, Cleveland, Ohio. Same centre, thirty years apart. The K-Site facility and SHIIVER’s thermal-vacuum testing were both, moreover, at that centre’s Plum Brook Station in Sandusky.

The defensible count is two NASA centres, three decades: Glenn/Lewis in 1992 and 2021, Marshall and its contractors in 2003. That is still a real finding — two NASA-center-led test teams, with different contractors, publishing limitations that made their own results look worse — but it is half the strength the draft claimed, and it admits a reading I have to name: the candor could be institutional culture at one centre rather than a norm of the field. Among the three ground-test cases this section compares — K-Site, SHIIVER, MHTB — MHTB, from Marshall, is the only evidence against that reading. One data point is not much to hang a professional norm on.

I record the error rather than quietly fixing it because of its timing. This paragraph exists at all because a reviewer showed that my sample of modelling papers was not independent — three of four sharing an author. In the act of correcting that, I asserted independence in the experimental sample without checking it. Same failure, one paragraph later, against documents already open on my desk.

The right-hand column is weaker and differently sourced. Each entry is a fact about a public document, verified. But the inference — that a custodial gap exists, that nobody owns the link between a dataset’s numbers and the context its authors published — rests on a handful of downstream papers, concentrated in two research groups. It is a hypothesis this series has made concrete and checkable, not a conclusion it has established. The audit in Note 000 is what tests it, and it may not survive.

The immediate action item is concrete: locate the primary tabulated record of tests P263968E/F and P263981T, if one exists — a Marshall test report, a data package, an internal archive. The original authors, successor project teams and Marshall’s surviving test records are the logical first places to ask. That letter is the next thing this project sends.

VII. Corrections and additions wanted

If you worked on MHTB, the spray-bar TVS program, or the 1996/1998 test series, three things would be genuinely useful:

where the tabulated time histories of P263968E/F and P263981T live, if anywhere reachable — this note states only that the TM does not contain them;
whether an uncertainty propagation for the derived quantities exists beyond what the TM prints;
cases where I have this wrong.

Corrections will be credited. Where one changes a conclusion, it will be recorded as having changed it.

On verification. Every quotation was checked verbatim against the source PDF, whose SHA-256 hash matches the artifact recorded and audited in Note 000 — an internal cross-check of this same project, not an independent verification; page numbers cite the report’s printed pagination. The title page, authorship and identifiers were verified against page images. Anything found wrong after publication produces a new version carrying a visible erratum — the superseded version remains preserved and citable — including when the correction weakens a conclusion already stated.

What follows this series

These notes are the documentary groundwork — published first so it can be checked first — for a committed quantitative study: reconstruct a defensible result-level experimental uncertainty (u_D) for one referent in this registry, combine it with the numerical and input uncertainties of a published comparison, covariances included, and report whether that comparison’s validation conclusion moves. In either direction: if nothing moves, that result publishes too. Until that study exists, everything here remains what Note 000 §IX declares — documentary findings whose engineering consequence is argued, not demonstrated. The study will appear as a new version under this series’ concept DOI, with these notes as its prior work.

Sources

A. Documents obtained and read for this note

Hastings, L.J. (Alpha Technology, Inc.), Flachbart, R.H., Martin, J.J., Hedayat, A., Fazah, M. (NASA Marshall Space Flight Center), Lak, T., Nguyen, H. (The Boeing Company), Bailey, J.W. (Sverdrup Technology, Inc.) — Spray Bar Zero-Gravity Vent System for On-Orbit Liquid Hydrogen Storage. NASA/TM—2003–212926, October 2003, 161 pp. NTRS 20040000092. Title page, authorship and identifiers verified against page images; SHA-256 of the audited artifact recorded in the series verification log and candidate list (companion dataset, doi:10.5281/zenodo.21895803), and in Note 000 §VIII. — https://ntrs.nasa.gov/citations/20040000092
Hedayat, A. (NASA Marshall Space Flight Center), Bailey, J. W. (Sverdrup Technology, Inc.), Hastings, L. J. (Alpha Technology, Inc.), Flachbart, R. H. (NASA MSFC) — Test Data Analysis of a Spray Bar Zero-Gravity Liquid Hydrogen Vent System for Upper Stages. 8 pp. NTRS 20030106051. Companion analysis of the 1996 and 1998 MHTB campaigns; Table 1 discussed in Section II. — https://ntrs.nasa.gov/citations/20030106051
Kartuzova, O., Kassemi, M., Agui, J., Moder, J. — Self-Pressurization and Spray Cooling Simulations of the Multipurpose Hydrogen Test Bed (MHTB) Ground-Based Experiment. Conference paper, 2015. NTRS 20150000249. Examined in Note 001; its reference list connects it to source 1. — https://ntrs.nasa.gov/citations/20150000249
Patel, C. (Jacobs ESSCA Group) — Validation of Pressure Control in a Flight-scale Liquid Hydrogen Tank using a Spray Bar. Presentation, Space Cryogenics Workshop, May 2025. NTRS 20250004637, 17 pp. SHA-256 8d9137873c0c2a1bb4c69718c4c8534b12c61a4de5b36617a4cb536c2d5025a6. Audited for the downstream-lineage subsection. — https://ntrs.nasa.gov/citations/20250004637
Majumdar, A., Valenzuela, J., LeClair, A., Moder, J. — Numerical Modeling of Self-Pressurization and Pressure Control by Thermodynamic Vent System in a Cryogenic Tank. Conference paper, 2015. NTRS 20170006183, 17 pp. Public conference version of the Cryogenics 74 (2016) paper in C. SHA-256 407e21d5983fbabe465e5c16c175f8a8a45a20c1a6b9447d1c78d43e5d6b82f6.
Kassemi, M., Kartuzova, O., Hylton, S. — Effect of Interfacial Turbulence and Accommodation Coefficient on CFD Predictions of Pressurization and Pressure Control in Cryogenic Storage Tank. Conference paper, 2015. NTRS 20150021282, 25 pp. Public conference version of the Cryogenics 74 (2016) paper in C. SHA-256 900c77631c5d23ef82c1575d46739768d6fdbb54f61740ce1a7f95af046749a9.
Soriano, S., Kartuzova, O., Kassemi, M., Hauser, D. — Modeling Self-Pressurization and Spray Bar Pressure Control of a Cryogenic Storage Tank in Normal Gravity. AIAA SciTech Forum, January 2025, full conference paper. NTRS 20250000041, 30 pp. SHA-256 ebc8e048c818151436cd87c29d4fc2b4b6c78c9a7fb67b6bd9957439b6bfaf53.
Soriano, S. A., Kartuzova, O. V., Kassemi, M., Hauser, D. — Modeling Spray Bar Pressure Control of a Large-Scale Liquid Hydrogen Propellant Tank in Normal Gravity. Space Cryogenics Workshop 2025 manuscript. NTRS 20250005060, 36 pp. SHA-256 87efd7b6efef5130e5842b56c0d4268ea1e31a3ec9dba2cf33d443b7c81562dd.
Wang, F., Jordan, T., Xiao, J. — Non-Equilibrium Liquid Hydrogen Tank Modeling When Thermal Insulation Fails. Hydrogen Safety (2025), open access, doi:10.58895/hysafe.35. 19 pp. SHA-256 c94e25a4568a64c5d420c8a26569c8f4568fc35c43ddb766bf2cc614f511c284.

B. Framework

ElarionX CPMS, Technical Note 000 — The Cryogenic Referent Registry (doi:10.5281/zenodo.21895568): canonical six criteria, verdict scale, candidate list, provenance and errata, including this document’s companion audit record A3 — an internal record of this same project, not an independent review.
ElarionX CPMS, Technical Notes 001–002 — the K-Site and SHIIVER readings referenced throughout; their scoring refinements are incorporated in Note 000.

ElarionX CPMS develops and evaluates engineering models for cryogenic propulsion systems, and publishes what it finds — including about its own work. Competing interest: declared in Note 000 §IX. This note is exploratory: it states a reading and a method, not a validated result. Corrections and correspondence: Luis.emc2@elarionx.com

C. Located, not read — journal versions of documents read in A

These two are the Cryogenics 74 (2016) journal versions of the 2015 conference papers read above; the journal texts themselves remain unread (paywalled). Nothing in this note relies on their content beyond what the conference versions carry.

Majumdar, A., et al. — Cryogenics 74 (2016), doi:10.1016/j.cryogenics.2015.12.001.
Kassemi, M., Kartuzova, O. — Cryogenics 74 (2016), doi:10.1016/j.cryogenics.2015.10.018.