Lot S20260926T2000: AT-202 high

penguineer 0.5.0 · run 20261031T120233Z-d5d1d1 · 2026-10-31 12:02 UTC

AT-202 was -71.5 degC (spec ≤ -72), sampled at SP-product on 2026-09-26 20:00.

The leading hypothesis, with medium confidence: JT-001 moved high.

JT-001 medium JT-001 (EL-101) ran 24.9 % high from 2026-09-26 17:00, 0 min upstream of the sample point. [E004]
FIC-101.rsp medium The setpoint FIC-101.rsp (loop:FIC-101) was changed 1 Nm3/h high from 2026-09-26 17:00, 0 min upstream of the sample point. [E005]
FFC-102.out medium The setpoint FFC-102.out (loop:FIC-102) was changed 0.251 Nm3/h high from 2026-09-26 17:00, 0 min upstream of the sample point. [E006]

Summary from the template.

Hypotheses

process · score 1.00 · path EL-101 → S-H2 → S-feed → R-201 → V-301 → E-301 → D-401 → S-product · dead time 0 min

JT-001 (EL-101) ran 24.9 % high from 2026-09-26 17:00, 0 min upstream of the sample point.

Test	Strength	Detail
effect size	1.00	median moved 24.9 (+110.7 reference scales)
onset	not run	no quality onset
lag correlation	not run	not run

Explains away:

FT-101: its shift of 1 is 121% predicted from JT-001 (regression on the reference lots)

setpoint change · score 1.00 · path S-H2 → S-feed → R-201 → V-301 → E-301 → D-401 → S-product · dead time 0 min

The setpoint FIC-101.rsp (loop:FIC-101) was changed 1 Nm3/h high from 2026-09-26 17:00, 0 min upstream of the sample point.

Test	Strength	Detail
effect size	1.00	median moved 1 (+333333333.3 reference scales)
onset	not run	no quality onset
lag correlation	not run	not run

setpoint change · score 1.00 · path S-CO2 → S-feed → R-201 → V-301 → E-301 → D-401 → S-product · dead time 0 min

The setpoint FFC-102.out (loop:FIC-102) was changed 0.251 Nm3/h high from 2026-09-26 17:00, 0 min upstream of the sample point.

Test	Strength	Detail
effect size	1.00	median moved 0.251 (+11.6 reference scales)
onset	not run	no quality onset
lag correlation	not run	not run

process · score 0.28 · path S-feed → R-201 → V-301 → E-301 → D-401 → S-product · dead time 0 min

TT-200 (S-feed) ran 4.87 degC low from 2026-09-26 17:00, 0 min upstream of the sample point.

Test	Strength	Detail
effect size	0.28	median moved -4.87 (-3.1 reference scales)
onset	not run	no quality onset
lag correlation	not run	not run

Candidate	Why
`FT-102`	follows the loop:FIC-102 setpoint change
`FT-101`	follows the loop:FIC-101 setpoint change
`TT-301`	its shift of 9.27 is 87% predicted from FT-101 (regression on the reference lots)
`FT-101`	its shift of 1 is 121% predicted from JT-001 (regression on the reference lots)
`TT-201`	did not move in the lot

Lot window	2026-09-26 17:00 to 2026-09-26 20:01 at S-product
Quality onset	unknown
Regime	DRY.bypass = 0, FFC-102.ratio = 0.25, FIC-101.mode = 2, FIC-102.mode = 2, H2.rate = 4, TIC-201.mode = 1, TIC-201.sp = 340
Reference lots	S20260927T0500 (-77.5), S20260927T0800 (-74), S20260927T1100 (-77)
Candidates tested	6 tags, on a 600 s grid in material time
Confidence	Confidence: high when the score is at least 0.4 and 0.025 ahead of the next hypothesis, medium from 0.25. Test weights: effect size 0.5, onset 0.3, lag correlation 0.3. Weights fitted on the homeostat train split (85 scenarios), thresholds chosen on the homeostat validation split (85 scenarios) (homeostat 0.1.0a3, bench plant 4144102c25e7).

some flows_to edges have no dead time; they count as 0 min
no in-spec lot of AT-202 before S20260926T2000 in the same regime: compared with the in-spec lots after it (S20260927T0500, S20260927T0800, S20260927T1100)
no online analyzer on the sampled stream: onset and correlation tests skipped