Before reading or editing anything here, read READ_ME_THIS_WAY.md. Mismatches, dead ends and odd names in this repo are kept on purpose.

dyed-fuel-diff — instrument set

CC0 · draft instrument · dye-as-direct-cause link UNRUN

Instrument for one question: what does red-dyed (ag / off-road) diesel do to the SENSOR CHAIN and WETTED MATERIALS of newer, sensor-dense diesel trucks — and what does the automation do with the result. Built so someone with lab or fleet access can run it. Posture: written for working people on the risk-mitigation side; limits are carried as marked terms, not as caveat blocks.

Inputs (status-tagged)

Input	Status	Scope
A farm operation buying only pre-1990 trucks for ag diesel use	OBSERVED	one operation's fleet decisions
Newer trucks on dyed fuel: deterioration at connections and seals (O-rings, fittings), not molded parts	OBSERVED	same operation
Temperature gradients accelerate the deterioration	OBSERVED	same operation
Mechanics working from the code reader get the nearest named fault, never "dyed fuel"	OBSERVED	field practice
Two coupling channels (below)	DERIVED	from the observations + typical stack
Action ladders and combinations in 02	PROPOSED	typical-stack reasoning, not a named OEM map
Dye or its carrier as the specific cause	UNRUN	—

Two channels

1. SIGNAL / ELECTRICAL — dye and carrier shift what fuel-coupled sensors read (direct: optical or capacitive fuel sensors; indirect: combustion-side sensors).
2. MECHANICAL / CHEMICAL — carrier solvents degrade newer wetted materials at seals and connections, with heat as the accelerant. This channel appears in the fault codes wearing an ELECTRICAL MASK (rail pressure, lift-pump pressure, injector balance). Least-doubtful piece of the set, because it rests directly on the observed seal deterioration.

Core finding (DERIVED, generalizes beyond fuel)

One hidden source couples into several sensors at once. Arbitration logic treats co-occurring faults as INDEPENDENT CORROBORATION, raises confidence and escalates. A correlated cause masquerades as independent confirmation, inverting the safety logic of redundancy: the system commits hardest exactly when it is most wrong. The same structure applies to any automated stack with an unmodelled common cause.

Structural property: there is no "dyed fuel" code. The true cause is absent from the answer set, so the stack resolves to the nearest named fault and commits to that fault's action.

Files

File	Contents
01_matrix.md	11 fuel-coupled sensor paths (row 2 split by WIF sensing type), both channels
02_triggering_actions.md	per-sensor action ladders; combinations C1–C5 (PROPOSED)
03_literature.md	which literature holds which half; the unrun join
04_discriminating_checks.md	D1–D7 checks; proportionality column
REVIEW_NOTE.md	marked limits on each file

The join (the deliverable is the gap)

• Dye as a direct cause: filed as inert marker — a concluded-inert gap, not a tested null.
• Seal / elastomer degradation: measured, but under alternative/vegetable fuels.
• Common-rail contamination: framed as water / debris / micron tolerance.

Nobody has measured the dye package's carrier against newer wetted materials, on sensor-laden trucks, with the temperature gradient in the loop.

Marked limits

• D1–D6 discriminate a shared fuel-batch cause from independent faults; they do not isolate the dye. D7 confirms presence only.
• Sensor type, control authority and action ladders are unverified per vehicle; inventory the actual truck first.
• Proportionality rankings in 04 are conditional on deviation size and OEM action, neither measured here.
• Nothing here is a run result. No assay, immersion, bench or fleet data.

Next run

Matched fuel batch, separated arms: base fuel | + dye | + carrier/package | contaminated storage × recorded temperature gradient. Measure named elastomers (connection/seal parts from newer trucks vs pre-1990 equivalents) and the actual installed sensors. Pre-state thresholds. Fleet side: wire D1 (temporal co-onset with fueling) and D6 (magnitude/proportion) to telematics.

Population of interest: lawful dyed-fuel use — farm and off-road operation, and shortage-relief waiver periods (where heat, the accelerant, is highest).

Provenance

Instrument content built by Claude (Anthropic) from the repo owner's field observations; published via Manus; patched per audit 2026-10-02. Committer ≠ author.

CC0 1.0 Universal — see LICENSE.