Files
znetsixe f18f3cc673 feat(mgc-dashboard): -1 OFF sentinel on per-pump % control chart
fn_chart_pump_a/b/c now emit -1 on the ctrl output when the cached pump
state is off/idle/maintenance, instead of the residual ctrl% (which would
sit at 0 and be indistinguishable from a pump genuinely running at 0%).
ui_chart_pumps_ctrl ymin set to -5 so the OFF rail is visible below the
0-100 band.

Adds test/integration/per-pump-ctrl-fanout.integration.test.js covering
both chart outputs of all three pumps in populated (running), OFF
(off/idle/maintenance), and degraded (missing state/ctrl/flow, pre-tick,
NaN, ctrl-only delta) states per .claude/rules/output-coverage.md. Updates
test/_output-manifest.md to document the previously-undocumented per-pump
fan-out functions.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-27 18:16:50 +02:00
..

machineGroupControl - Example Flows

Import-ready Node-RED examples for machineGroupControl (MGC). MGC is not a standalone node — it needs at least one rotatingMachine child to dispatch demand to. Both flows below ship three child pumps.

Files

File Tier What it shows
01-Basic.json 1 One MGC + three rotatingMachine pumps driven by inject buttons. Setup once-fires virtualControl + cmd.startup on all three pumps; mode / scaling / demand are then driven by buttons.
02-Dashboard.json 2 Same command surface driven by a FlowFuse Dashboard 2.0 page — mode + scaling buttons, demand slider, live status rows, three trend charts, and a raw-output table.

Prerequisites

  • Node-RED with the EVOLV package installed (machineGroupControl and rotatingMachine registered).
  • For 02-Dashboard.json: @flowfuse/node-red-dashboard (Dashboard 2.0).

Load a flow

curl -X POST -H 'Content-Type: application/json' \
  --data @nodes/machineGroupControl/examples/01-Basic.json \
  http://localhost:1880/flows

Or in the editor: Menu → Import → drag the file → Import.

Canonical command surface

Topic Aliases Payload What it does
set.mode setMode "optimalControl", "priorityControl", "prioritypercentagecontrol", "maintenance" Switch dispatch strategy
set.scaling setScaling "normalized", "absolute" Interpret demand as 0100 % vs m³/h
set.demand Qd number Operator demand setpoint
child.register registerChild child node id (string) Manually register a child (Port 2 wiring does this automatically)

01-Basic — what to try

  1. Deploy. After ~1.5 s the Setup group auto-fires, putting all three pumps in virtualControl mode + sending cmd.startup to each.
  2. Click set.demand = 50 % — MGC's optimalControl picks the best pump combination by BEP-gravitation and dispatches flowmovement to the selected pumps.
  3. Click set.demand = 100 % — MGC switches to a higher combination, possibly engaging an extra pump.
  4. Switch mode to priorityControl and try the same demands — pumps now run equal-flow by priority order.
  5. Switch scaling to absolute — set.demand is now interpreted as m³/h (capped at the group min / max).
  6. set.demand = 0 — MGC calls turnOffAllMachines, all pumps shut down.

02-Dashboard — what to try

  1. Deploy → open http://localhost:1880/dashboard/mgc-basic.
  2. The dashboard auto-initialises the pumps; the Initialize pumps button on the page re-runs the setup manually.
  3. Drag the Demand slider — MGC dispatches and the Flow / Power / BEP charts react.
  4. Switch modes and scalings via the buttons; the Mode / Scaling rows in the Status panel reflect the change.
  5. Inspect the Raw output table for the full Port 0 surface (every field MGC emits, including flowCapacityMax, machineCountActive, absDistFromPeak, relDistFromPeak).