May 17, 2026

Generated lab notebook from daily_summary, plan_journal, plan_delivery_log, and setpoint audit data. It is intentionally chronological and may include in-progress cycles before validation.

Planner Execution Ledger

Planner checkpoints can acknowledge that the active plan is still suitable without writing a new public plan ID. Those no-change decisions are part of the audit trail.

Delivery events

Plan writes

No-change acknowledgements

Pending

Event	Delivered	Type	Status	Result	Resolved	Note
Morning planning cycle	05:44	SUNRISE	plan_written	iris-20260517-0544	05:46	Wrote public plan iris-20260517-0544. Resolved 05:46 MDT.
Forecast deviation	12:33	FORECAST_DEVIATION	acked	-	12:33	Acknowledged: FORECAST_DEVIATION assessed: solar is far below forecast, but current climate is stable and in-band (75.2F, VPD 0.99 kPa, RH 67%, dew margin 11.9F) with ESP32 in IDLE. Equipment state shows only heat1 on, vent/fans/fog/misters off, so this is not an equipment-induced cooling or moisture failure. Direction matches the known solar forecast overshoot bias, though magnitude indicates an earlier cloud/weather shift; no tunable change warranted while VPD is below high band and condensation margin is healthy... Resolved 12:33 MDT.
Solar peak planning checkpoint	12:58	SOLAR_MAX	acked	-	12:58	Acknowledged: SOLAR_MAX checked live climate, forecast, active plan, and setpoints. Solar is tracking/under forecast rather than overshooting (live 38 W/m2 vs 13:00 forecast 39 W/m2 before calibration), house is IDLE at 74.3F with VPD 0.88 kPa below active vpd_high 1.30, dew margin is safe at 10.6F, and the active SUNRISE plan already stages a moderate 14:45-16:30 afternoon moisture shoulder for the forecast solar/VPD pulse. No tunable change warranted. Resolved 12:58 MDT.
Forecast deviation	13:21	FORECAST_DEVIATION	acked	-	13:22	Acknowledged: FORECAST_DEVIATION assessed: observed weather is much cooler, more humid, and cloudier than forecast, matching a weather/forecast solar-cloud miss rather than equipment failure. Current house remains stable in IDLE at 76.2F, 67% RH, VPD 1.03 kPa with 11.7F dew margin; all climate equipment is off. No tunable change made because average VPD is in band, dew margin is safe, and the south zone only touches the high edge; retrieved cool/cloudy lessons advise not carrying aggressive dry-day mist/fog into cl... Resolved 13:22 MDT.
Forecast deviation	13:26	FORECAST_DEVIATION	acked	-	13:27	Acknowledged: FORECAST_DEVIATION assessed with live climate/equipment. Actual weather is much cooler, wetter, and darker than forecast, but the house is stable in IDLE: 75.4F, VPD 1.01 kPa, RH 66%, dew-point margin 12.0F, no mist/fog/vent/fans active and only heat1 on. Cause appears to be forecast/weather nowcasting error in cloud/precip regime rather than equipment fault. Current active tunables are already conservative for cool humid conditions, so no immediate set_tunable change is warranted. Resolved 13:27 MDT.
Forecast deviation	13:55	FORECAST_DEVIATION	acked	-	13:55	Acknowledged: No tuning change warranted. Live climate is stable and in safe posture despite forecast deviation: indoor 72.1F, VPD 0.76 kPa near/inside band, RH 72%, dew margin 9.6F, controller IDLE. Equipment state is coherent for cool humid conditions: heat1 on, vent/fans/fog/misters off, no relay anomaly. The deviation is a weather/forecast miss toward cooler, wetter, lower-solar conditions, matching today’s overcast/rain signal rather than an equipment fault; active plan already has conservative humid-window mo... Resolved 13:55 MDT.
Forecast deviation	14:12	FORECAST_DEVIATION	acked	-	14:13	Acknowledged: FORECAST_DEVIATION reviewed: observed solar 267 W/m2 versus forecast ~942 W/m2, with current house 73.4F, VPD 0.81 kPa, RH 71%, dew margin 9.9F, and controller IDLE. Equipment state is consistent with cool/cloudy conditions: heat1 on, vent/fans/fog/misters off. This is a weather/forecast-solar overcall in the same direction as the known positive solar forecast bias, not equipment failure... Resolved 14:13 MDT.
Peak Stress	14:57	TRANSITION	plan_written	iris-oneshot-20260517-1458	14:58	Wrote non-archive plan iris-oneshot-20260517-1458. Resolved 14:58 MDT.
Forecast deviation	16:43	FORECAST_DEVIATION	acked	-	16:44	Acknowledged: FORECAST_DEVIATION assessed: solar is far below forecast (32-36 W/m2 observed vs 765 W/m2 forecast) because cloud cover/rain regime arrived, not equipment failure. Current house is stable and not dry-hot: 75.7F, VPD 0.94 kPa, RH 69%, dew margin 10.8F, mode IDLE. Equipment confirms no cooling/misting/fog demand; only heat1 is on. Active posture is already conservative for the cool/cloudy evening and overnight, so no tunable change is warranted. Resolved 16:44 MDT.
Decline	19:13	TRANSITION	acked	-	19:14	Acknowledged: Decline transition checked: greenhouse is IDLE at 70.5F, VPD 0.85 kPa, RH 67%, dew margin 11.4F, all climate equipment off. Existing posture is already unwound for evening/overnight with conservative mist/fog thresholds and the 21:30 cold-front waypoint pending; no concrete stress signal warrants a tunable change. Resolved 19:14 MDT.
Evening planning cycle	20:14	SUNSET	plan_written	iris-20260517-2014	20:16	Wrote public plan iris-20260517-2014. Resolved 20:16 MDT.

🌅 Morning Cycle (5:46 AM) — `iris-20260517-0544`

Status

validated

Outcome score

6/10

Changed parameters

bias_cool, bias_heat, d_cool_stage_2, d_heat_stage_2, dwell_gate_ms, enthalpy_close, enthalpy_open, fog_escalation_kpa ...

Reflection

Validating previous cycle: iris-20260516-2011

Previous hypothesis: SUNSET overnight plan for 2026-05-16. Today improved materially versus the 7-day trend but the dry-day experiment was mixed: score 69.1 and cost 2.35 beat the 7-day averages, water was controlled, and VPD-high stayed near 2.5h in the plan window, but both-axis compliance missed target because heat and VPD-high remained paired in VENTILATE and the anchor score penalized 223 moisture-guardrail events. My self-score of 6 vs anchor 2 was intentionally higher because the anchor heavily penalized guardrail activity while the operator-level goals of water control and VPD-high under 3h were partially achieved; nevertheless the mismatch means tonight should stop asking for conservative values that the guardrail must rewrite. Current VPD is slightly above band with a healthy 12… Result: Window scorecard for the 2026-05-17 SUNRISE plan: plan-window score 67.7 with 64.7% both-axis compliance, temp compliance 71.2%, VPD compliance 70.5%, heat stress 1.46h, cold stress 0.14h, VPD-high 1.65h, VPD-low 0.04h, cost about USD 0.43. Structured forecast was directionally right about a cool/cloudy wet day but under-forecast the solar peak (773 predicted vs 1234 actual), so the brief moderate moisture shoulders kept VPD-high below 2h and avoided VPD-low/dew risk, but lowered cooling bias/solar handling did not prevent heat stress. The cold-front overnight handoff was only partly observable by sunset, but current conditions show safe dew margin and low VPD-high risk. Self-score 6 because safety, water, VPD-low, and VPD-high targets largely succeeded, while the plan missed the >75 score, >70% compliance, and heat-stress goals. Score: 6/10

New finding: On cool/cloudy wet days with high indoor RH, live-confirmed moderate moisture shoulders can keep VPD-high under about 2h without causing VPD-low, but if actual solar materially exceeds forecast, do not lower bias_cool too far during the afternoon shoulder; heat stress can become the limiting axis even when the day is not hot-dry. → Added to Lessons Learned

Hypothesis

Testing: Test whether a cool/cloudy-day posture with only two moderate, live-confirmed moisture shoulders can reduce VPD-high breakouts without repeating overnight guardrail-held conservative values or causing VPD-low/dew risk. Expected outcome: For the 05:50-21:30 daytime window, target score above 75, both-axis compliance above 70%, VPD-high under 2.0h, VPD-low under 0.5h, heat stress under 1.0h, dp_risk_hours 0, water under 200 gal, and total cost below the 7-day average. Overnight handoff should keep cold stress under 1h and prevent heater-to-vent oscillation.

Conditions

Outdoor peak

65.0°F

RH minimum

38.0%

Solar peak

773.0 W/m²

Cloud average

67.0%

Cool humid dawn, showery/cloudy day with raw outdoor VPD briefly 1.31 kPa late morning and 0.88 kPa at the afternoon solar pulse; forecast calibration says not to pre-stage aggressive mist until live VPD confirms. Current dew margin is only 5.3F, so condensation safety and VPD-low avoidance matter. Tonight turns cold/wet, falling toward the low 40s.

Expected stress windows

vpd_lowmedium · 2026-05-17T05:45:00-06:00 to 2026-05-17T08:30:00-06:00

High mist thresholds, 60s pulse gap, fog_escalation 0.95, long mist delays, and no aggressive dawn misting while indoor VPD is 0.34 kPa and dew margin is near 5F.

vpd_highmedium · 2026-05-17T09:00:00-06:00 to 2026-05-17T11:45:00-06:00

Moderate, guardrail-aware shoulder only if live VPD rises: engage 0.95, all 1.10, 30s gap, 45/90s delays, fog_escalation 0.30; sealed windows remain 150s to avoid heat traps.

vpd_lowmedium · 2026-05-17T12:00:00-06:00 to 2026-05-17T14:30:00-06:00

Return to conservative moisture during rain/cloud recovery: engage 1.80, all 2.10, gap 60, fog_escalation 0.95, watch dwell 90s.

vpd_highlow · 2026-05-17T14:45:00-06:00 to 2026-05-17T16:30:00-06:00

Brief afternoon solar fallback, still moderate because temperatures remain cool: engage 1.05, all 1.20, gap 35, fog_escalation 0.35.

coldmedium · 2026-05-17T21:30:00-06:00 to 2026-05-18T07:00:00-06:00

Cold-front posture: bias_heat 1.5, bias_cool 4.0, d_heat_stage_2 3.0, heat_hysteresis 1.5, conservative mist/fog, and wide VPD hysteresis to prevent heater-to-vent and humidity oscillation.

Parameter rationale

mister_engage_kpa1.8 → 0.95; Only during 09:00-11:45 raw VPD pulse to 1.31 kPa; live dawn VPD is 0.34 so this is delayed until the confirmed dry shoulder.

Reduce VPD-high during the short dry pulse without causing dawn VPD-low.

mister_all_kpa2.1 → 1.1; Active vpd_high is about 0.83 and guardrail evidence says all-zone support should stay near vpd_high+0.25 during real VPD-high/VENTILATE stress.

Let all-zone assist engage if VPD rises during the pulse while avoiding all-zone water use in humid periods.

mister_pulse_gap_s60.0 → 30.0; Moderate short dry pulse, not a hot-dry peak; use 30s rather than 15s because temperature stays near 65F and dew margin is near 5F.

Provide recovery capacity without over-humidifying the cool house.

fog_escalation_kpa0.95 → 0.3; PR-A makes fog useful in VENTILATE VPD-high, but today is cool/cloudy and fog should not be aggressive outside confirmed pulses.

Avoid stale conservative guardrail holds during actual VPD-high while preventing unnecessary fog during humid windows.

bias_cool3.5 → 0.5; Brief solar pulses may warm the low crop band greenhouse; lower cooling bias only during dry/solar windows, then restore +4 for night.

Reduce heat/VENTILATE VPD pairing during solar pulses without creating heater-to-vent chatter overnight.

bias_heat0.75 → 1.5; Cold front tonight with 41-43F and rain; retrieved cold saturated night plans use modest warmth plus wide cooling bias rather than raising bands.

Hold stable warmth and dew margin with gas-efficient staging while keeping cold stress under 1h overnight.

d_heat_stage_25.0 → 3.0; Forecast low drops below 45F tonight; high-confidence heating lesson favors gas staging for sustained cold.

Bring gas heat in earlier during sustained cold instead of relying on electric heat1 alone.

vpd_hysteresis0.5 → 0.5; Cool humid day with risk of VPD-low and oscillation; retrieved lessons favor wide hysteresis on overcast humid days.

Avoid short-cycle mist/dehum transitions while keeping band compliance stable.

Setpoints

Sunday May 17

Primary crop-band changes:

Time	Values	Note
05:50	hyst 0.5	Humid cool dawn: protect dew margin and avoid carrying dry-d
08:30	hyst 0.45	Morning VPD ramp watch: loosen from dawn but do not use peak
10:15	hyst 0.4	Late-morning dry pulse: guardrail-aware moderate support, no
12:15	hyst 0.5	Rain/cloud recovery: return to conservative moisture posture
14:45	hyst 0.45	Afternoon solar fallback: brief moderate moisture support if
17:00	hyst 0.5	Fog window closing and humidity returning: unwind moisture s
21:30	hyst 0.5	Cold wet overnight handoff: suppress mist/fog, protect dew m

Tactical tunable changes:

Time	Values	Note
05:50	d_cool 3; engage 1.85; all 2.1; pulse 60; gap 60; wt 2	Humid cool dawn: protect dew margin and avoid carrying dry-d
08:30	d_cool 2.5; engage 1.3; all 1.6; pulse 60; gap 45; wt 2.2	Morning VPD ramp watch: loosen from dawn but do not use peak
10:15	d_cool 2.5; engage 0.95; all 1.1; pulse 60; gap 30; wt 2.4	Late-morning dry pulse: guardrail-aware moderate support, no
12:15	d_cool 3; engage 1.8; all 2.1; pulse 60; gap 60; wt 2	Rain/cloud recovery: return to conservative moisture posture
14:45	d_cool 2.5; engage 1.05; all 1.2; pulse 60; gap 35; wt 2.3	Afternoon solar fallback: brief moderate moisture support if
17:00	d_cool 3; engage 1.85; all 2.1; pulse 60; gap 60; wt 2	Fog window closing and humidity returning: unwind moisture s
21:30	d_cool 4; engage 2; all 2.2; pulse 60; gap 60; wt 1.8	Cold wet overnight handoff: suppress mist/fog, protect dew m

Changed secondary parameters:

Time	Parameter	Change
05:50	bias_cool	initial 3.5
05:50	bias_heat	initial 0.75
05:50	d_heat_stage_2	initial 5
05:50	dwell_gate_ms	initial 300000
05:50	enthalpy_close	initial 1
05:50	enthalpy_open	initial -2
05:50	fog_escalation_kpa	initial 0.95
05:50	heat_hysteresis	initial 1.3
05:50	min_fog_off_s	initial 120
05:50	min_fog_on_s	initial 45
05:50	min_heat_off_s	initial 180
05:50	min_heat_on_s	initial 120
05:50	min_vent_off_s	initial 60
05:50	min_vent_on_s	initial 60
05:50	mist_backoff_s	initial 900
05:50	mist_max_closed_vent_s	initial 150
05:50	mist_thermal_relief_s	initial 90
05:50	mister_all_delay_s	initial 300
05:50	mister_engage_delay_s	initial 180
05:50	mister_water_budget_gal	initial 600
05:50	outdoor_staleness_max_s	initial 600
05:50	sw_dwell_gate_enabled	initial 1
05:50	sw_fog_closes_vent	initial 1
05:50	sw_fsm_controller_enabled	initial 1
05:50	sw_mister_closes_vent	initial 1
05:50	sw_summer_vent_enabled	initial 1
05:50	temp_hysteresis	initial 1.8
05:50	vent_prefer_dp_delta_f	initial 5
05:50	vent_prefer_temp_delta_f	initial 5
05:50	vpd_watch_dwell_s	initial 90
08:30	bias_cool	3.5 → 1.5
08:30	bias_heat	0.75 → 0.5
08:30	fog_escalation_kpa	0.95 → 0.55
08:30	heat_hysteresis	1.3 → 1.2
08:30	min_fog_off_s	120 → 90
08:30	mist_backoff_s	900 → 700
08:30	mister_all_delay_s	300 → 180
08:30	mister_engage_delay_s	180 → 90
08:30	temp_hysteresis	1.8 → 1.6
08:30	vpd_watch_dwell_s	90 → 60
10:15	bias_cool	1.5 → 0.5
10:15	bias_heat	0.5 → 0
10:15	fog_escalation_kpa	0.55 → 0.3
10:15	heat_hysteresis	1.2 → 1.1
10:15	min_fog_off_s	90 → 60
10:15	mist_backoff_s	700 → 600
10:15	mister_all_delay_s	180 → 90
10:15	mister_engage_delay_s	90 → 45
10:15	temp_hysteresis	1.6 → 1.5
10:15	vpd_watch_dwell_s	60 → 45
12:15	bias_cool	0.5 → 2.5
12:15	bias_heat	0 → 0.5
12:15	fog_escalation_kpa	0.3 → 0.95
12:15	heat_hysteresis	1.1 → 1.3
12:15	min_fog_off_s	60 → 120
12:15	mist_backoff_s	600 → 900
12:15	mister_all_delay_s	90 → 300
12:15	mister_engage_delay_s	45 → 180
12:15	temp_hysteresis	1.5 → 1.8
12:15	vpd_watch_dwell_s	45 → 90
14:45	bias_cool	2.5 → 0.5
14:45	bias_heat	0.5 → 0
14:45	fog_escalation_kpa	0.95 → 0.35
14:45	heat_hysteresis	1.3 → 1.1
14:45	min_fog_off_s	120 → 75
14:45	mist_backoff_s	900 → 650
14:45	mister_all_delay_s	300 → 120
14:45	mister_engage_delay_s	180 → 60
14:45	temp_hysteresis	1.8 → 1.5
14:45	vpd_watch_dwell_s	90 → 45
17:00	bias_cool	0.5 → 3.5
17:00	bias_heat	0 → 0.75
17:00	fog_escalation_kpa	0.35 → 0.95
17:00	heat_hysteresis	1.1 → 1.3
17:00	min_fog_off_s	75 → 120
17:00	mist_backoff_s	650 → 900
17:00	mister_all_delay_s	120 → 300
17:00	mister_engage_delay_s	60 → 180
17:00	temp_hysteresis	1.5 → 1.8
17:00	vpd_watch_dwell_s	45 → 90
21:30	bias_cool	3.5 → 4
21:30	bias_heat	0.75 → 1.5
21:30	d_heat_stage_2	5 → 3
21:30	fog_escalation_kpa	0.95 → 1
21:30	heat_hysteresis	1.3 → 1.5
21:30	min_fog_off_s	120 → 150
21:30	min_heat_on_s	120 → 150
21:30	min_vent_off_s	60 → 90
21:30	mist_backoff_s	900 → 1200
21:30	mist_max_closed_vent_s	150 → 120
21:30	mister_all_delay_s	300 → 360
21:30	mister_engage_delay_s	180 → 240
21:30	temp_hysteresis	1.8 → 2

🌆 Evening Cycle (8:16 PM) — `iris-20260517-2014`

Status

validated

Outcome score

8/10

Changed parameters

bias_cool, bias_heat, d_cool_stage_2, d_heat_stage_2, dwell_gate_ms, enthalpy_close, enthalpy_open, fog_escalation_kpa ...

Reflection

Validating previous cycle: iris-20260517-0544

Previous hypothesis: SUNRISE plan for Sunday 2026-05-17. Yesterday’s score was 60.8 with paired heat and VPD-high stress; the daytime plan self-score of 6 deviated from anchor 2 because the anchor heavily penalized 223 moisture-guardrail events, while operator goals of lower water/cost and VPD-high under prior extremes were partially met. The overnight plan self-score of 8 deviated from anchor 5 for the same reason: 328 guardrail events remained, but the key safety goals were met with 0 VPD-low, 0 dp-risk, negligible cold stress, and very low cost. Today is not another hot-dry day: current house is cool/humid at 62.6F, VPD 0.34 kPa, RH 83%, dew margin 5.3F, IDLE with heat1 on. Forecast is cool, cloudy/showery, with brief VPD/solar pulses around 09:00-11:00 and 15:00 but a cold wet night ahead… Result: Overnight plan-window score was 85.4 with 82.9% both-axis compliance, temp compliance 82.9%, VPD compliance 100.0%, heat stress 0.17h, cold stress 0.20h, VPD-high 0h, VPD-low 0h, cost about USD 0.26 in the plan window, and daily partial dew margin minimum 5.2F with 0 dp-risk hours. Hypothesis was strongly validated: conservative mist/fog plus short sealed residence prevented VPD-low and condensation; bias_cool +4 with gas staging kept cold stress under 1h. One resolved heat-staging-inversion alert appeared, so heater interlock should remain a watch item, but it did not materially hurt climate outcomes. Self-score 8 because all safety, VPD, cold, and cost targets were met, though both-axis compliance stayed below 90% due mainly to crop-band morning heat alignment rather than actual plant danger. Score: 8/10

New finding: Cold wet high-RH nights with outdoor VPD near 0.05-0.15 kPa validate a conservative humidity posture: mister_engage_kpa 2.2-2.5, mister_all_kpa 2.5, fog_escalation_kpa 0.95-1.0, 60s gaps, long mist delays, short 120s sealed cap, plus bias_cool +4 and modest gas staging can hold VPD-low and dp-risk at 0 while keeping cold stress under 1h. → Added to Lessons Learned

Hypothesis

Testing: Cold/wet overnight suppression: test whether conservative mist/fog plus bias_cool +4 and gas staging keeps dp risk at 0, VPD-low under 1h, cold stress under 1h, and avoids heater-to-vent chatter. Expected outcome: By Monday morning: dp_risk_hours 0, min dew margin >5F, VPD-low stress under 1h, cold stress under 1h, no VPD-high breakout, mist/fog water near zero, and cost dominated by modest gas heat while remaining below the 7-day average.

Conditions

Outdoor peak

54.0°F

RH minimum

76.0%

Solar peak

0.0 W/m²

Cloud average

100.0%

Cold wet overnight: 54F at 21:00 falling to 41-42F by dawn, RH 76-95%, raw outdoor VPD 0.34 falling to 0.05 kPa, 100% cloud/rain. Current house is 66.9F, VPD 0.50 kPa, RH 78%, dew point 59.8F, dew margin 7.1F, IDLE. Main risks are VPD-low/condensation pressure and heater-to-vent oscillation, not dry-air VPD-high.

Expected stress windows

vpd_lowmedium · 2026-05-17T20:20:00-06:00 to 2026-05-18T08:00:00-06:00

Suppress mist/fog with engage 2.2-2.4, all 2.5, 60s gaps, 900s mist delays, fog_escalation 0.95-1.0, vpd_hysteresis 0.50, and short 120s sealed cap.

coldmedium · 2026-05-18T01:00:00-06:00 to 2026-05-18T08:00:00-06:00

Use bias_heat 1.5-1.75, bias_cool 4.0, d_heat_stage_2 3.0, heat_hysteresis 1.5, and dwell gate on to use gas staging for sustained cold without heater-to-vent chatter.

Parameter rationale

bias_cool3.5 → 4.0; Forecast falls below 45F after 03:00 with heating likely; high-confidence lessons use +3 to +4 to prevent heater-to-vent oscillation.

Keep cold stress under 1h overnight and avoid vent dumps after heater overshoot.

bias_heat0.75 → 1.5; Cold wet night reaches 41-42F with rain/high RH; use modest warmth rather than raising crop bands.

Hold stable 63-66F greenhouse and keep dew margin above 5F with gas/slab support.

d_heat_stage_25.0 → 3.0; Sustained <45F period from about 03:00-08:00 validates the heating lesson favoring gas for real cold.

Shift sustained cold support to gas staging and avoid excessive electric-only runtime.

mister_engage_kpa1.3 → 2.2; Current VPD is 0.50 and forecast outdoor VPD collapses to 0.05-0.10; retrieved lessons say avoid mist until live VPD actually rises.

Prevent new VPD-low and keep mist water near zero overnight.

mister_all_kpa1.5 → 2.5; No dry-air window overnight; all-zone mist should be effectively unavailable unless an unexpected VPD-high event occurs.

Avoid over-humidifying the cool house.

fog_escalation_kpa1.0 → 1.0; Fog is blocked overnight and conditions are saturated; conservative fog posture should persist until a live daytime dry ramp is observed.

No unnecessary fog/condensation pressure overnight.

mist_max_closed_vent_s120.0 → 120.0; Dew margin is safe but forecast RH is 90-95%; keep sealed-mist residence short if it is ever entered.

Keep dp_risk_hours at 0.

vpd_hysteresis0.5 → 0.5; Cool wet night with localized VPD spread and likely low-band edge behavior; preserve wide deadband.

Reduce humidity/dehum oscillation and false mist starts.

heat_hysteresis0.0 → 1.5; Current readback showed a manual zero row; safety lesson forbids zero timers/rails and prior plan used nonzero heat hysteresis.

Restore stable heater clearing behavior and prevent rapid heat cycling.

Setpoints

Sunday May 17

Primary crop-band changes:

Time	Values	Note
20:20	hyst 0.5	evening_settle: current dew margin is safe but humidity is h
23:00	hyst 0.5	midnight_posture: rain/high RH arrives; prioritize VPD-low a

Tactical tunable changes:

Time	Values	Note
20:20	d_cool 3; engage 2.2; all 2.4; pulse 60; gap 60; wt 2	evening_settle: current dew margin is safe but humidity is h
23:00	d_cool 3; engage 2.3; all 2.5; pulse 60; gap 60; wt 2	midnight_posture: rain/high RH arrives; prioritize VPD-low a

Monday May 18

Primary crop-band changes:

Time	Values	Note
03:00	hyst 0.5	pre_dawn_cold: forecast bottom near 41-44F with saturated ai
07:00	hyst 0.5	morning_hold: Monday remains cold, overcast, and wet with ra

Tactical tunable changes:

Time	Values	Note
03:00	d_cool 3; engage 2.4; all 2.5; pulse 60; gap 60; wt 2	pre_dawn_cold: forecast bottom near 41-44F with saturated ai
07:00	d_cool 3; engage 2.4; all 2.5; pulse 60; gap 60; wt 2	morning_hold: Monday remains cold, overcast, and wet with ra

Changed secondary parameters:

Time	Parameter	Change
20:20	bias_cool	initial 4
20:20	bias_heat	initial 1.25
20:20	d_heat_stage_2	initial 4
20:20	dwell_gate_ms	initial 300000
20:20	enthalpy_close	initial 1
20:20	enthalpy_open	initial -2
20:20	fog_escalation_kpa	initial 0.95
20:20	heat_hysteresis	initial 1.4
20:20	min_fog_off_s	initial 180
20:20	min_fog_on_s	initial 45
20:20	min_heat_off_s	initial 180
20:20	min_heat_on_s	initial 120
20:20	min_vent_off_s	initial 60
20:20	min_vent_on_s	initial 60
20:20	mist_backoff_s	initial 1800
20:20	mist_max_closed_vent_s	initial 120
20:20	mist_thermal_relief_s	initial 90
20:20	mister_all_delay_s	initial 900
20:20	mister_engage_delay_s	initial 900
20:20	mister_water_budget_gal	initial 600
20:20	outdoor_staleness_max_s	initial 600
20:20	sw_dwell_gate_enabled	initial 1
20:20	sw_fog_closes_vent	initial 1
20:20	sw_fsm_controller_enabled	initial 1
20:20	sw_mister_closes_vent	initial 1
20:20	sw_summer_vent_enabled	initial 1
20:20	temp_hysteresis	initial 2
20:20	vent_prefer_dp_delta_f	initial 5
20:20	vent_prefer_temp_delta_f	initial 5
20:20	vpd_watch_dwell_s	initial 90
23:00	bias_heat	1.25 → 1.5
23:00	d_heat_stage_2	4 → 3.5
23:00	fog_escalation_kpa	0.95 → 1
23:00	heat_hysteresis	1.4 → 1.5
23:00	mist_backoff_s	1800 → 2400
23:00	vpd_watch_dwell_s	90 → 120
03:00	bias_heat	1.5 → 1.75
03:00	d_heat_stage_2	3.5 → 3
07:00	bias_heat	1.75 → 1.5

End-of-Day Summary

Climate

Temperature

61.9–81.1°F; avg 69.7°F

VPD

0.32–1.92 kPa; avg 0.77 kPa

Relative humidity

47.7–84.0%

Stress Hours

Heat stress (>85°F): 3.1h
VPD stress (>2.0 kPa): 3.2h
Cold stress (<55°F): 0.4h

Economics

Electric

USD 3.98

Gas

USD 0.73

Water

USD 0.220

Total

USD 4.93

Equipment Runtimes

Equipment	Runtime	Reading
Fan 1	131 min	Primary exhaust runtime.
Fan 2	125 min	Secondary exhaust runtime.
Vent	130 min	Intake vent runtime.
Fog	4 min	Fogger runtime.
Heat 1 electric	925 min	Electric heater runtime.
Heat 2 gas	70 min	Gas heater runtime.
Grow lights	1026 min	Supplemental lighting runtime.
Mister south	0.07h	South mister runtime.
Mister west	0.13h	West mister runtime.
Mister center	0.78h	Center mister runtime.

Water

Total: 45 gal
Mister: 14 gal

Crop Health (Gemini Vision)

Crop	Zone	Health	Observations	Note
Canna Lilies	south	65%	4	Observation notes are collapsed below to avoid publishing partial vision snippets.
Vanda Orchids	center	78%	5	Observation notes are collapsed below to avoid publishing partial vision snippets.
lettuce	east	80%	4	Observation notes are collapsed below to avoid publishing partial vision snippets.
pepper	east	70%	4	Observation notes are collapsed below to avoid publishing partial vision snippets.
strawberry	east	83%	4	Observation notes are collapsed below to avoid publishing partial vision snippets.

Vision observation notes

Canna LiliesGemini Vision notes

Canna Lilies on the south floor look generally healthy, though soil moisture is reported at 0.0%.

Vanda OrchidsGemini Vision notes

Hanging orchids appear healthy with green leaves.

lettuceGemini Vision notes

Seedlings appear healthy and are establishing well on the shelf.

pepperGemini Vision notes

Seedlings are growing, but some variation in size is visible.

strawberryGemini Vision notes

Seedlings are growing steadily.

Hourly Pattern

Hour	Temperature	VPD	RH
12:00	76.8°F	1.24 kPa	61.0%
13:00	75.1°F	1.07 kPa	64.1%
14:00	78.9°F	1.40 kPa	58.9%
15:00	79.0°F	1.42 kPa	58.8%
16:00	77.4°F	1.20 kPa	62.8%
17:00	75.9°F	1.23 kPa	60.6%
18:00	71.0°F	1.04 kPa	60.0%
19:00	69.5°F	0.91 kPa	62.9%
20:00	68.0°F	1.02 kPa	56.1%
21:00	66.7°F	0.97 kPa	56.8%
22:00	67.0°F	1.06 kPa	52.9%
23:00	65.9°F	0.93 kPa	57.1%
00:00	64.4°F	0.88 kPa	57.2%
01:00	63.2°F	0.53 kPa	73.5%
02:00	63.2°F	0.44 kPa	77.9%
03:00	62.4°F	0.39 kPa	80.0%
04:00	63.3°F	0.58 kPa	70.8%
05:00	63.0°F	0.35 kPa	82.2%
06:00	66.6°F	0.51 kPa	77.4%
07:00	67.6°F	0.54 kPa	77.1%
08:00	71.3°F	0.74 kPa	71.9%
09:00	73.2°F	0.87 kPa	68.8%
10:00	76.0°F	1.13 kPa	63.2%
11:00	76.5°F	1.02 kPa	67.5%

7-Day Stress Context

Date	Heat	VPD high	Cold
2026-05-11	10.9h	11.5h	1.1h
2026-05-12	8.0h	7.9h	0.7h
2026-05-13	12.0h	8.0h	0.6h
2026-05-14	11.4h	10.9h	0.0h
2026-05-15	5.8h	6.3h	0.2h
2026-05-16	9.1h	8.6h	0.1h
2026-05-17	3.1h	3.2h	0.4h

Lab

May 17, 2026

Planner Execution Ledger

🌅 Morning Cycle (5:46 AM) — iris-20260517-0544

Reflection

Hypothesis

Setpoints

Sunday May 17

🌆 Evening Cycle (8:16 PM) — iris-20260517-2014

Reflection

Hypothesis

Setpoints

Sunday May 17

Monday May 18

End-of-Day Summary

Climate

Stress Hours

Economics

Equipment Runtimes

Water

Crop Health (Gemini Vision)

Hourly Pattern

7-Day Stress Context

🌅 Morning Cycle (5:46 AM) — `iris-20260517-0544`

🌆 Evening Cycle (8:16 PM) — `iris-20260517-2014`