How it works

The authority split, the ladder, photoperiod, sensor trust, plants and stages, journals, learning, and the grow tools. Brain stopped or unplugged? Go to When the brain stops: the hub and the plugs. The planner, who is in charge and rolling back, and how the brain reaches devices have a page each below. Timed runs are in Using the dashboard.

Who decides, and what happens when the brain stops

• 01Who decides what — hub vs brainThe authority split: hub-owned safety and demand, brain-owned targets, tasks, rules and the Sonoff driver; why demand proposals are logged, not written; what happens when the brain or hub goes away.
• 02When the brain stops: the hub and the plugsWhat the hub keeps doing from its own sensors and clock, its hard limits, and what each smart plug does when it stops hearing from the brain.
• 03The climate ladder, failsafe and reality gatesHow the hub turns targets into appliance demand: the rung order, min-off, hysteresis, ladder waits, reality gates (ineffective, relay mismatch), the failsafe, priority tent and the shared exhaust.

Light schedules and sensor trust

• 04Photoperiod, holds and catch-upThe hub's lights clock, auto photoperiod, sunrise/sunset ramps, the dark-period guard, manual light hold, catch-up and its surprises, the two 4×8 lamps and lamp plugs, and how schedule shifts are confirmed.
• 05Sensor trust — offsets, clamps, held and staleThe pipeline a reading goes through before it is shown or acted on, and the horizons that turn a value into HELD, stale or rejected.

Plants, stages and device roles

• 06Plants, stages, harvest window and DLIHow a strain profile becomes Want bands, how expected stage and the harvest window are computed, what the DLI estimate is made of, and the plant · probe · assignment model.
• 07Zigbee and Tuya — roles, tasks and placementHow a paired device becomes entities: capability class, role, zone and placement, tasks and policy state, climate deltas, and the additive rule for plugs.

Records, learning and reports

• 08Journals, reminders and the grow recordThe four journal layers, frozen snapshots, typed actions and custom actions, photos on disk, reminders anchored to entries, retention, export and the archive written on retire.
• 09Energy, learning and calibration curvesNameplate vs learned airflow, the anemometer method, CFM ownership, Phase A/B climate learning, the space energy model and tariff cost, and what soft cal does and does not do.
• 10Runoff trend and grow alert sensitivityHow runoff minus input is charted per plant, when it raises an alert, and the one setting that sets the grow tools' alert thresholds together.
• 11The run report and its gradeWhat the brain freezes when you log a harvest, how yield is normalised, and what the execution grade does and does not judge.

Drying, curing and nutrient tools

• 12Drying and curing zonesWhat flipping a zone to Dry or Cure starts, the dry strategies, mould risk, jar pucks in a curing chamber, and what the brain never controls.
• 13Procedures — SOP checklistsShort standard procedures that open when something happens, tick themselves from the journal or the sensors, and write their close to the journal.
• 14The reservoir Mix panelChecking a nutrient mix against the plants it feeds and your own earlier mixes — verification, not prediction.

The planner, who is in charge and connections

• 15The planner: what it decides and whyHow the planner works out the cheapest safe set of actions across all your devices, and where it writes down its reasons.
• 16Who is in charge and rolling backThe Control screen: who is in charge of each device class, the evidence needed to hand one over, and the one-tap way back.
• 17Connections: how the brain reaches devicesEach connection the brain uses to reach devices, what it carries and what the screens show when it is healthy or not.

Guide contents  ·  Troubleshooting  ·  Firmware  ·  What runs today  ·  Digital Stealth Care

DSC-Brain user guide · edited on this site · How it works