May 18, 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

5

Plan writes

2

No-change acknowledgements

3

Pending

0

EventDeliveredTypeStatusResultResolvedNote
Morning planning cycle05:43SUNRISEplan_writteniris-20260518-054305:46Wrote public plan iris-20260518-0543. Resolved 05:46 MDT.
Solar peak planning checkpoint12:58SOLAR_MAXacked-12:58Acknowledged: SOLAR_MAX reviewed live climate, forecast, active plan, and setpoints. Live solar (~198 W/m2) is above the immediate overcast forecast but still modest and below the SUNRISE plan's expected weak-solar ceiling; house remains IDLE, temp 73.5F within 72-78F band, VPD 0.98 kPa below vpd_high 1.30 with no VPD-high stress, RH 65%, and dew margin 12.5F. Existing cold/wet suppression plan covers the afternoon and no tunable nudge is warranted. Resolved 12:58 MDT.
Peak Stress14:58TRANSITIONacked-14:58Acknowledged: Peak-stress transition checked: current house is in band overall (75.9F, VPD 1.14 vs vpd_high 1.23) with safe dew margin 13.1F, IDLE mode, no fans/vent/misters/fog active, and only heat1 on. South VPD is locally high at 1.71 but matches the south target context and average VPD is below the high band; no VPD-high or heat stress is climbing beyond the active cold/wet suppression plan... Resolved 14:58 MDT.
Decline19:13TRANSITIONacked-19:14Acknowledged: Decline transition checked: climate is tracking the cold/wet suppression plan. House is IDLE at 64.3F, VPD 0.47 kPa, RH 78%, dew margin 7.1F; outdoor remains cold/saturated, fog/misters/vent/fans are off, heat1 is providing mild support, and the existing plan already has a 22:00 conservative overnight waypoint. No concrete signal warrants a tunable change. Resolved 19:14 MDT.
Evening planning cycle20:12SUNSETplan_writteniris-20260518-201220:15Wrote public plan iris-20260518-2012. Resolved 20:15 MDT.

🌅 Morning Cycle (5:46 AM) — iris-20260518-0543

Status

validated

Outcome score

7/10

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-20260517-2014

Previous hypothesis: SUNSET overnight plan for Sunday 2026-05-17. Today’s score improved sharply versus the 7-day trend (80.1 vs 50.4) with both-axis compliance 76.6%, low cost USD 0.85 vs USD 4.79, and no dew-point risk. VPD-high was still the largest stress (3.23h), but the current and forecast risk has shifted: house VPD is now 0.50 kPa with RH 78% and a safe 7.1F dew margin, while the next 12h are cold, saturated, cloudy/rainy, and fall to 41-45F with outdoor VPD near 0.05-0.10 kPa. I evaluated the governing SUNRISE plan at 6/10 versus anchor 3; my grade is higher because the operator-level VPD-low, dew-risk, water, and VPD-high targets mostly succeeded, while the anchor penalized 171 guardrail events and the plan missed heat/compliance targets after actual solar exceeded forecast… Result: Plan-window scorecard for the 2026-05-18 SUNRISE cold/wet suppression plan: score 73.7 with 75.0% both-axis compliance, temp compliance 75.3%, VPD compliance 84.3%, heat stress 0.22h, cold stress 0.90h, VPD-high 0.00h, VPD-low 0.05h, cost about USD 0.60 in the plan window. Full-day partial score by sunset improved to 87.0 with 88.6% both-axis compliance, temp 88.9%, VPD 99.5%, 0 dp-risk hours, and only 8 gal water. Hypothesis was mostly validated: conservative moisture posture kept VPD-low and condensation essentially at zero, while bias_cool +4 and gas staging bounded cold stress. Remaining score loss was mostly temp/cold-band alignment and gas cost from a cold saturated day, not VPD control failure. Score: 7/10

New finding: Cold wet overcast days with outdoor VPD near 0.1-0.3 kPa validate high moisture thresholds, conservative fog escalation, wide VPD hysteresis, short sealed-mist residence, and bias_cool +4 with gas staging; success should be judged by VPD-low and dp-risk avoidance plus bounded cold stress rather than water use alone. → Added to Lessons Learned

Hypothesis

Testing: All-day cold/wet suppression with daytime bias_heat trim: test whether high moisture thresholds plus bias_cool +4 and gas staging keep VPD-low/dp-risk at zero while reducing heat overshoot versus the overnight posture. Expected outcome: By tonight: planner score >80, both-axis compliance >80%, VPD compliance >95%, VPD-low 0h, dp-risk 0h, cold stress <1.5h, heat stress <1.0h, mister/fog water near zero, and total daily cost below the 7-day average.

Conditions

Outdoor peak

44.0°F

RH minimum

85.0%

Solar peak

294.0 W/m²

Cloud average

96.0%

Cold wet overcast day: outdoor 38-44F, RH 85-100%, raw outdoor VPD 0.00-0.13 kPa, rain probability high through afternoon, and current house is 64.5F, VPD 0.48 kPa, RH 77%, dew margin 7.4F, IDLE with heat1 on. Primary risks are VPD-low/condensation pressure and heater-to-vent/heat overshoot, not dry-air VPD-high.

Expected stress windows

vpd_lowmedium · 2026-05-18T05:45:00-06:00 to 2026-05-18T23:00:00-06:00

Keep mister_engage_kpa 2.35-2.45, mister_all_kpa 2.5, 60s gaps, 900s mist delays, fog_escalation 1.0, vpd_hysteresis 0.50, and mist_max_closed_vent_s 120s.

coldmedium · 2026-05-18T05:45:00-06:00 to 2026-05-18T23:00:00-06:00

Use d_heat_stage_2 3.0, heat_hysteresis 1.5, dwell gate ON, and bias_cool +4.0; keep bias_heat modest during day and restore 1.75 in the evening as outdoor falls toward the high 30s.

heatlow · 2026-05-18T10:30:00-06:00 to 2026-05-18T15:30:00-06:00

Avoid over-driving heaters through the small solar/rainy midday window by trimming bias_heat to 1.25 while keeping wide cooling bias to prevent vent dumps.

Parameter rationale

bias_heat1.75 → 1.25; Cold but not drying day with only weak solar pulses; partial morning already shows some heat stress while crop-band morning cold scoring is structurally hard to eliminate.

Reduce heater overshoot/heat stress while keeping cold stress under about 1.5h using gas staging and slab retention.

bias_cool4.0 → 4.0; Forecast stays below 45F with heat demand likely; validated lessons use +3 to +4 to prevent heater-to-vent oscillation.

Prevent vent dumps after heater overshoot and keep cold stress bounded.

mister_engage_kpa2.4 → 2.4; Outdoor VPD remains near zero and indoor RH is already high; lesson 98 says high mist thresholds on cool high-RH days protect against VPD-low and condensation.

Keep VPD-low near 0h and misting water near zero unless an unexpected live VPD rise occurs.

fog_escalation_kpa1.0 → 1.0; Fog window exists during the day but forecast RH is saturated and fog is unsafe/unnecessary unless VPD unexpectedly rises far above band.

Avoid fog-driven condensation and keep dp_risk_hours at 0.

mist_max_closed_vent_s120.0 → 120.0; Dew margin is safe now but forecast outdoor RH is 85-100%; sealed-mist should not trap humid air for long if entered.

Maintain minimum dew margin above 5F.

sw_dwell_gate_enabled1.0 → 1.0; Recent equipment runtime showed many heat and vent/fan transitions; dwell gate has shipped and thermal relief preempts it.

Reduce ordinary mode churn without blocking safety or thermal-relief exits.

Setpoints

Monday May 18

Primary crop-band changes:

TimeValuesNote
05:50hyst 0.5Immediate sunrise cold/wet posture: suppress mist/fog, keep
08:30hyst 0.5Morning rain block: outdoor VPD stays near zero; keep moistu
11:30hyst 0.5Wet midday with possible small solar pulse: trim heat bias t
14:00hyst 0.5Peak rain/cold period and forecast solar blip: remain humidi
18:30hyst 0.5Evening cold returns with little solar: restore stronger hea
22:00hyst 0.5Cold saturated overnight handoff: continue gas-friendly heat

Tactical tunable changes:

TimeValuesNote
05:50d_cool 3; engage 2.4; all 2.5; pulse 60; gap 60; wt 2Immediate sunrise cold/wet posture: suppress mist/fog, keep
08:30d_cool 3; engage 2.4; all 2.5; pulse 60; gap 60; wt 2Morning rain block: outdoor VPD stays near zero; keep moistu
11:30d_cool 3; engage 2.4; all 2.5; pulse 60; gap 60; wt 2Wet midday with possible small solar pulse: trim heat bias t
14:00d_cool 3; engage 2.45; all 2.5; pulse 60; gap 60; wt 2Peak rain/cold period and forecast solar blip: remain humidi
18:30d_cool 3; engage 2.4; all 2.5; pulse 60; gap 60; wt 2Evening cold returns with little solar: restore stronger hea
22:00d_cool 3; engage 2.4; all 2.5; pulse 60; gap 60; wt 2Cold saturated overnight handoff: continue gas-friendly heat

Changed secondary parameters:

TimeParameterChange
05:50bias_coolinitial 4
05:50bias_heatinitial 1.5
05:50d_heat_stage_2initial 3
05:50dwell_gate_msinitial 300000
05:50enthalpy_closeinitial 1
05:50enthalpy_openinitial -2
05:50fog_escalation_kpainitial 1
05:50heat_hysteresisinitial 1.5
05:50min_fog_off_sinitial 180
05:50min_fog_on_sinitial 45
05:50min_heat_off_sinitial 180
05:50min_heat_on_sinitial 120
05:50min_vent_off_sinitial 60
05:50min_vent_on_sinitial 60
05:50mist_backoff_sinitial 2400
05:50mist_max_closed_vent_sinitial 120
05:50mist_thermal_relief_sinitial 90
05:50mister_all_delay_sinitial 900
05:50mister_engage_delay_sinitial 900
05:50mister_water_budget_galinitial 600
05:50outdoor_staleness_max_sinitial 600
05:50sw_dwell_gate_enabledinitial 1
05:50sw_fog_closes_ventinitial 1
05:50sw_fsm_controller_enabledinitial 1
05:50sw_mister_closes_ventinitial 1
05:50sw_summer_vent_enabledinitial 1
05:50temp_hysteresisinitial 2
05:50vent_prefer_dp_delta_finitial 5
05:50vent_prefer_temp_delta_finitial 5
05:50vpd_watch_dwell_sinitial 120
11:30bias_heat1.5 → 1.25
18:30bias_heat1.25 → 1.75

🌆 Evening Cycle (8:15 PM) — iris-20260518-2012

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-20260518-0543

Previous hypothesis: SUNRISE plan for Monday 2026-05-18. Yesterday improved sharply: daily score 81.7, both-axis compliance 78.8%, temp compliance 85.2%, VPD compliance 86.1%, cost USD 1.04 vs 7-day average USD 4.79, and 0 dp-risk hours. VPD-high and heat were yesterday’s dominant stresses, but the completed overnight plan validated the cold/wet suppression posture: plan-window score 85.4, VPD compliance 100%, VPD-low 0h, dp-risk 0h, cold 0.20h. My prior self-score 8 deviated from anchor 5 because the anchor penalized guardrail/dispatched events while operator safety goals were fully met; I am carrying the operational lesson forward but being conservative about heat overshoot today. Today’s forecast is cold, saturated, rainy, and overcast with raw outdoor VPD 0.00-0.13 kPa, peak solar only about 294 W/m2, and temperatures 38-44F… Result: Overnight plan-window result: score 85.5 with 83.4% both-axis compliance, temp compliance 83.4%, VPD compliance 100.0%, heat stress 0.04h, cold stress 0.32h, VPD-high 0.00h, VPD-low 0.00h, and plan-window cost about USD 0.394. The full previous day scorecard was 86.6 with 88.8% both-axis compliance, temp 89.1%, VPD 99.6%, 2.12h cold stress, 0.48h heat stress, 0.10h VPD-low, 0 VPD-high, cost USD 3.36 mostly gas-driven, 8 gal water, and 0 dp-risk hours. Hypothesis was validated: high mist thresholds, conservative fog, short sealed-mist cap, wide VPD hysteresis, bias_cool +4, and modest gas staging kept VPD-low and condensation at zero overnight while bounding cold stress. Remaining score loss was the structural spring-morning temp/cold-band alignment rather than humidity control failure. Score: 8/10

New finding: Cold saturated nights with outdoor VPD near 0.05-0.15 kPa validate conservative moisture suppression plus bias_cool +4 and modest gas staging: keep mister_engage_kpa 2.35-2.45, mister_all_kpa 2.5, fog_escalation_kpa 1.0, 60s gaps, 300/600s mist delays, 120s sealed cap, and dwell gate on to maintain 0 VPD-high/VPD-low and 0 dp-risk while keeping cold stress under about 0.5h. → Added to Lessons Learned

Hypothesis

Testing: Cold saturated overnight restoration: test whether high mist thresholds plus bias_cool +4 and modest bias_heat +2 at the coldest hours keep VPD-low and dp-risk near zero while reducing cold stress versus today. Expected outcome: By Tuesday sunrise: VPD-high 0h, VPD-low under 0.25h, dp-risk 0h, cold stress near or below 1h, paired compliance above 80% for the overnight window, and cost dominated by gas but remaining proportional to sub-40F weather.

Conditions

Outdoor peak

40.0°F

RH minimum

84.0%

Solar peak

24.0 W/m²

Cloud average

100.0%

Cold saturated overnight: outdoor 40F at sunset falling to 37-38F pre-dawn, RH 84-89%, raw outdoor VPD 0.09-0.12 kPa, 100% cloud, no meaningful solar. Current house is 63.3F, VPD 0.46 kPa, RH 78%, dew margin 7.2F, IDLE with heat1 on. Main challenge is stable gas/electric heating and avoiding VPD-low/condensation, not VPD-high.

Expected stress windows

coldmedium · 2026-05-18T22:00:00-06:00 to 2026-05-19T07:30:00-06:00

bias_cool 4.0, bias_heat 1.75 rising to 2.0 at coldest hours, d_heat_stage_2 3.0, heat_hysteresis 1.5, dwell gate ON, and normal nonzero heat timers to prevent heater-to-vent oscillation while allowing sustained gas support.

vpd_lowlow · 2026-05-18T20:20:00-06:00 to 2026-05-19T08:30:00-06:00

mister_engage_kpa 2.35-2.45, mister_all_kpa 2.5, 60s pulse gaps, 300/600s mist delays, fog_escalation 1.0, vpd_hysteresis 0.50, and 120s sealed-mist cap. This also protects dew-point margin because structured hypothesis only allows heat/cold/vpd_high/vpd_low stress kinds.

Parameter rationale

mister_engage_kpa1.1 → 2.4; Outdoor VPD 0.09-0.12 kPa with RH 84-89% and no solar; lessons 98/99 favor high thresholds on cool saturated nights.

Prevent physical misting overnight and keep VPD-low under 0.25h with dp-risk at 0h.

mister_all_kpa1.3 → 2.5; No dry-air window overnight; all-zone mist escalation would only add condensation pressure.

Keep all-zone mist unavailable unless an unexpected severe VPD spike occurs.

mister_engage_delay_s30.0 → 300.0; Cool high-RH night; recent clamp history shows 300s is the safe upper engage-delay invariant.

Avoid short humidity pulses during transient VPD noise.

mister_all_delay_s60.0 → 600.0; Cool high-RH night; recent clamp history shows 600s is the safe upper all-delay invariant.

Avoid all-zone mist rotation overnight.

mister_pulse_gap_s30.0 → 60.0; Humidity holds better when sealed overnight and VPD-high risk is absent.

Reduce VPD-low and water use without increasing VPD-high stress.

bias_heat1.75 → 2.0; Forecast falls to 37-38F from 02:00-07:00; heating lesson favors modest bias plus gas staging for sustained cold.

Keep cold stress near or below 1h overnight while maintaining dew margin above 5F.

bias_cool4.0 → 4.0; Heaters are expected to run; high-confidence lesson uses +3 to +4 to prevent heater-to-vent oscillation.

Avoid vent dumps after heater overshoot and keep heat/cold whipsaw low.

mist_max_closed_vent_s120.0 → 120.0; Dew margin is safe at 7.2F but outdoor RH is saturated; sealed residence should remain short if entered.

Maintain dp_risk_hours at 0.

vpd_hysteresis0.5 → 0.5; Localized VPD spread is 0.50 kPa and humid conditions can oscillate around the low band.

Reduce dehum/mist chatter while preserving safety.

Setpoints

Monday May 18

Primary crop-band changes:

TimeValuesNote
20:20hyst 0.5evening_settle: restore conservative saturated-night moistur
22:30hyst 0.5mid_evening_cold_deepening: outdoor near 39F, saturated and

Tactical tunable changes:

TimeValuesNote
20:20d_cool 3; engage 2.4; all 2.5; pulse 60; gap 60; wt 2evening_settle: restore conservative saturated-night moistur
22:30d_cool 3; engage 2.4; all 2.5; pulse 60; gap 60; wt 2mid_evening_cold_deepening: outdoor near 39F, saturated and

Tuesday May 19

Primary crop-band changes:

TimeValuesNote
02:00hyst 0.5coldest_saturated_block: forecast 38F, RH near 89%, outdoor
05:30hyst 0.5pre_dawn_hold: outdoor still 38F with no solar; hold heating
08:15hyst 0.5morning_handoff_fallback: if the sunrise cycle is delayed, r

Tactical tunable changes:

TimeValuesNote
02:00d_cool 3; engage 2.45; all 2.5; pulse 60; gap 60; wt 2coldest_saturated_block: forecast 38F, RH near 89%, outdoor
05:30d_cool 3; engage 2.45; all 2.5; pulse 60; gap 60; wt 2pre_dawn_hold: outdoor still 38F with no solar; hold heating
08:15d_cool 3; engage 2.35; all 2.5; pulse 60; gap 60; wt 2morning_handoff_fallback: if the sunrise cycle is delayed, r

Changed secondary parameters:

TimeParameterChange
20:20bias_coolinitial 4
20:20bias_heatinitial 1.75
20:20d_heat_stage_2initial 3
20:20dwell_gate_msinitial 300000
20:20enthalpy_closeinitial 1
20:20enthalpy_openinitial -2
20:20fog_escalation_kpainitial 1
20:20heat_hysteresisinitial 1.5
20:20min_fog_off_sinitial 180
20:20min_fog_on_sinitial 45
20:20min_heat_off_sinitial 180
20:20min_heat_on_sinitial 120
20:20min_vent_off_sinitial 60
20:20min_vent_on_sinitial 60
20:20mist_backoff_sinitial 2400
20:20mist_max_closed_vent_sinitial 120
20:20mist_thermal_relief_sinitial 90
20:20mister_all_delay_sinitial 600
20:20mister_engage_delay_sinitial 300
20:20mister_water_budget_galinitial 600
20:20outdoor_staleness_max_sinitial 600
20:20sw_dwell_gate_enabledinitial 1
20:20sw_fog_closes_ventinitial 1
20:20sw_fsm_controller_enabledinitial 1
20:20sw_mister_closes_ventinitial 1
20:20sw_summer_vent_enabledinitial 1
20:20temp_hysteresisinitial 2
20:20vent_prefer_dp_delta_finitial 5
20:20vent_prefer_temp_delta_finitial 5
20:20vpd_watch_dwell_sinitial 120
22:30bias_heat1.75 → 1.85
02:00bias_heat1.85 → 2
08:15bias_cool4 → 3.5
08:15bias_heat2 → 1.5

End-of-Day Summary

Climate

Temperature

61.7–76.5°F; avg 67.8°F

VPD

0.33–1.23 kPa; avg 0.67 kPa

Relative humidity

60.9–83.4%

Stress Hours

  • Heat stress (>85°F): 0.5h
  • VPD stress (>2.0 kPa): 0.0h
  • Cold stress (<55°F): 2.1h

Economics

Electric

USD 5.87

Gas

USD 3.27

Water

USD 0.040

Total

USD 9.18

Equipment Runtimes

EquipmentRuntimeReading
Fan 10 minPrimary exhaust runtime.
Fan 20 minSecondary exhaust runtime.
Vent0 minIntake vent runtime.
Fog0 minFogger runtime.
Heat 1 electric1312 minElectric heater runtime.
Heat 2 gas315 minGas heater runtime.
Grow lights1757 minSupplemental lighting runtime.
Mister south0.00hSouth mister runtime.
Mister west0.00hWest mister runtime.
Mister center0.01hCenter mister runtime.

Water

  • Total: 8 gal
  • Mister: 0 gal

Crop Health (Gemini Vision)

CropZoneHealthObservationsNote
Canna Liliessouth80%4Observation notes are collapsed below to avoid publishing partial vision snippets.
Vanda Orchidscenter80%4Observation notes are collapsed below to avoid publishing partial vision snippets.
lettuceeast80%4Observation notes are collapsed below to avoid publishing partial vision snippets.
peppereast70%4Observation notes are collapsed below to avoid publishing partial vision snippets.
strawberryeast80%4Observation notes are collapsed below to avoid publishing partial vision snippets.
Vision observation notes
Canna LiliesGemini Vision notes

Canna Lilies appear healthy with robust foliage visible in the IR image.

Vanda OrchidsGemini Vision notes

Hanging orchids appear healthy with green foliage.

lettuceGemini Vision notes

Seedlings appear healthy and are establishing well.

pepperGemini Vision notes

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

strawberryGemini Vision notes

Seedlings are growing steadily.

Hourly Pattern

HourTemperatureVPDRH
12:0075.7°F1.03 kPa66.1%
13:0073.3°F0.85 kPa70.0%
14:0077.1°F1.23 kPa62.3%
15:0079.1°F1.46 kPa57.7%
16:0076.0°F1.01 kPa67.3%
17:0076.2°F0.93 kPa70.1%
18:0076.3°F1.05 kPa66.3%
19:0070.5°F0.97 kPa62.1%
20:0066.2°F0.53 kPa76.4%
21:0064.0°F0.51 kPa75.6%
22:0062.9°F0.42 kPa78.6%
23:0063.7°F0.39 kPa80.5%
00:0063.4°F0.38 kPa81.1%
01:0063.3°F0.38 kPa81.0%
02:0063.3°F0.40 kPa80.0%
03:0063.4°F0.41 kPa79.4%
04:0063.3°F0.41 kPa79.3%
05:0064.0°F0.44 kPa78.5%
06:0065.9°F0.54 kPa75.8%
07:0068.1°F0.63 kPa73.5%
08:0070.8°F0.79 kPa69.9%
09:0072.5°F0.94 kPa66.4%
10:0073.5°F1.03 kPa64.3%
11:0073.8°F1.04 kPa64.0%

7-Day Stress Context

DateHeatVPD highCold
2026-05-128.0h7.9h0.7h
2026-05-1312.0h8.0h0.6h
2026-05-1411.4h10.9h0.0h
2026-05-155.8h6.3h0.2h
2026-05-169.1h8.6h0.1h
2026-05-173.1h3.2h0.4h
2026-05-180.5h0.0h2.1h