A-D strip. Sets Pint directly from the Table 3 PPM band and gives a floor estimate for D via titratable acidity. An A-D level is not a diagnosis of vinegar syndrome: the strip reads acetic acid in the air around the cel right now, while onset here is defined by structural damage, D reaching 0.5%. A high reading with low measured D means the acid is coming from somewhere other than this cel — the fix is the enclosure, not the artwork. How you test also decides what the reading means: sealed alone with one cel it reflects that cel; left in a binder or box with others it reflects the whole enclosure.
Production year. Only a way to estimate D when you have no lab value. It applies the handbook's baseline clock of 0.01%/yr under general storage (the 70°F / 40%RH IPI reference), so D ≈ 0.01% × age. That clock is linear — IPI's own assumption that the rate never changes — so treat it as a neutral prior about storage you cannot see, not a framework result.
Measured D. A lab-measured deacetylation value. It overrides the age and A-D estimates entirely and anchors the result at today, since a measured D already accounts for the aging so far. With D in hand the production year no longer drives the Cel Nexus figure — it only anchors the age-only Arrhenius benchmark it is plotted against.
Temperature and humidity. Temperature scales the intrinsic rate through the Arrhenius factor. Relative humidity enters separately as a first-order water-activity term, because deacetylation is a hydrolysis and needs moisture to proceed. Both act on the benchmark rate; neither changes this cel's state penalty, which cancels the environment out.
Leave on Moderate unless you have a reason not to.
Estimated onset window — this cel
Pristine-cel benchmark: R = R0(T) · aw. D=0, Pint=0, no
autocatalytic feedback. Calibrated to reproduce IPI's 50-year guideline exactly at 70°F / 40%RH.
Two readings of the same cel, both counted from today. Arrhenius knows only the calendar. Cel Nexus starts from the cel's measured state. Where they disagree is the useful part: if the state-aware marker sits later, this cel has aged more slowly than its years suggest.
IPI Baseline (benchmark). Arrhenius supplies the temperature dependence only; the water-activity factor aw = RH/100 is a separate first-order moisture term, since deacetylation is hydrolytic. A pristine cel's rate is R = R0(T) · aw with onset at D = 0.5%. R0 = 0.025 %/yr is the Intrinsic Thermal Velocity at the framework reference of 70°F / 40% RH (aw = 0.40): 0.025 × 0.40 = 0.010 %/yr → 0.5 / 0.010 = 50 yrs, reproducing the IPI 50-year guideline exactly. Temperature scales R0 by the Arrhenius factor exp[(−Ea/R)(1/T − 1/T_ref)].
Cel Nexus (relative penalty, Eq. 16). Comparing a degraded cel to the same cel pristine, the environmental terms (β, R0, aw) cancel, leaving a dimensionless state penalty: Relative Penalty = (1 + κ·D) · (1 + Φ·Pint). This cel's effective lifetime = Arrhenius lifetime ÷ penalty. Pristine (D=0, Pint=0) → ×1.00, so Cel Nexus equals IPI Baseline at the reference. The penalty depends only on the cel's state — accumulated damage D and internal vapor Pint — not on storage geometry or scavengers, and not on temperature or RH, which cancel in the ratio.
Scope: pre-autocatalytic only. Everything here models the induction phase, D < 0.5%. At or past the 0.5% brink the tool reports “at/past onset” rather than projecting a date, because post-brink degradation follows a different rate law. Ea = 105 kJ/mol (IPI/Adelstein induction-phase fit) sets the benchmark's temperature sensitivity; κ = 2.0 standard / 1.0 low / 4.0 heavy-paint; Φ = 0.5 PPM⁻¹.
Strip level vs. damage state. Pint (from the strip) and D are different quantities: Pint is free acid present in the enclosure now and is reversible — ventilate or scavenge and it falls. D is accumulated deacetylation and is not. That is why a high strip level on a cel with low measured D still returns healthy remaining life: the model treats the acid as an accelerant acting on the material, not as the damage itself. Onset is defined at D = 0.5%, never at a strip level.
Why the two markers diverge on the same cel. The age clock is linear at 0.01%/yr, which is IPI's assumption that the rate is constant for life. The framework's rate law is not constant: R = R0·aw·(1 + κD) rises as damage accumulates. The two agree exactly at D = 0 and separate from there. Seeding D from the linear clock is deliberate — it is the neutral prior for a cel whose storage history you do not know — and the gap between the markers is the framework's actual claim: once D is high, IPI's constant-rate figure is optimistic.
D is estimated as the larger of: (a) age × 0.01%/yr baseline clock, or (b) the AD strip's titratable-acidity band — unless you supply a measured value directly. Pint comes straight from the AD strip's PPM band (Table 3).