Files
wesandClaude Opus 4.8 a56f550636 awning: current-sensing auto-retract + true CLOSED state; drop useless OPEN jog
Motion is hold-to-run (one authed opcode runs the motor ~1s). Reproduce the OEM
movement session -- auth once, then stream the opcode @110ms + a page-44
keepalive @510ms -- to sustain continuous retract, and watch node-75 page-3
motor current at 20Hz to stop at the fully-closed stall (~4200 vs ~<1550
running), then mark the cover CLOSED. Backstops: 70s timeout, motion-lost
detector, hold-to-run stop-on-silence. Proven live at the camper 2026-07-01.

- esphome: awning_auto_retract script + 100ms streamer interval + case-0x75
  stall gate + "Awning Motor Current" sensor. Cover CLOSE=auto-retract,
  STOP=abort. Removed open_action (1s jog is useless, no safe timed auto-open).
- bridge: optimistic:true keeps the home HA cover assumed-state so retract is
  always pressable (was greyed when closed); payload_open:null drops OPEN on the
  home (primary) dashboard. Also synced the 6h->15min discovery-cadence drift.
- captures + README: full-retract stall profile and the live auto-retract test.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-01 22:18:15 -04:00

594 lines
28 KiB
YAML

# OneControl IDS-CAN node
#
# ESP32 (native TWAI/CAN) + external SN65HVD230 transceiver, connected to the
# Lippert UNITY X180T CAN bus at the monitor panel's spare CAN *data* port.
# Listens to the modules' broadcasts and republishes them as native HA entities,
# and issues commands using the panel's challenge/response authentication.
#
# IDS-CAN = 250 kbit/s. Read broadcasts are 11-bit standard frames,
# id = (page << 8) | node. Commands and their authentication exchange use 29-bit
# extended frames. See ../README.md for the connection procedure, the node map,
# and the message format. Flash over USB first
# (`esphome run onecontrol-canbus.yaml`), OTA thereafter.
substitutions:
name: onecontrol-canbus
friendly_name: OneControl CAN
# ESP32 GPIOs to the transceiver. Any free non-strapping pins work; these match
# a common SN65HVD230 wiring. tx_pin -> transceiver D/CTX, rx_pin <- R/CRX.
tx_pin: GPIO5
rx_pin: GPIO4
esphome:
name: ${name}
friendly_name: ${friendly_name}
# Command authentication (REMOTE_CONTROL session, key 0xB16B00B5) provides
# ids_can_auth::remote_control_response_bytes(); command_guard.h provides
# onecontrol::command_blocked() — the slides/jacks-are-panel-only safety rule.
includes:
- ids_can_auth.h
- command_guard.h
esp32:
board: esp32dev # classic WROOM-32
framework:
type: esp-idf
wifi:
ssid: !secret wifi_ssid
password: !secret wifi_password
ap:
ssid: "OneControl-CAN Fallback"
password: !secret fallback_ap_password
captive_portal:
logger:
level: DEBUG # global DEBUG so entity "Sending state" publishes and
# the command exchange (idscan) are visible during
# bring-up. Drop the whole thing to INFO once happy.
logs:
canbus: INFO # the esp32_can component logs EVERY received frame at
# DEBUG (~50/s) — that floods the 115200 serial link and
# starves wifi/api/sensor logs. Silence it; our decode
# is what we care about, not the raw component dump.
api:
encryption:
key: !secret api_key
ota:
- platform: esphome
# ---------------------------------------------------------------------------
# CAN bus: ESP32 native TWAI controller + SN65HVD230 transceiver
#
# NOTE: read broadcasts are 11-bit *standard* frames; the command authentication
# challenge is a 29-bit *extended* frame. The node must receive both. Confirm the
# esp32_can trigger accepts both frame types (a single catch-all with
# can_id_mask 0); if your ESPHome build filters by frame type, add a second
# on_frame for standard IDs.
# ---------------------------------------------------------------------------
canbus:
- platform: esp32_can
id: can_bus
tx_pin: ${tx_pin}
rx_pin: ${rx_pin}
bit_rate: 250kbps # IDS-CAN is 250k
can_id: 0 # our own TX id (only matters when we send)
use_extended_id: true # commands use 29-bit IDs
# This ESPHome build filters on_frame by frame type, so a single trigger only
# ever fires for one kind. We need BOTH: read broadcasts (tanks/lights/heater/
# awning) are 11-bit *standard* frames; the command challenge + battery
# telemetry are 29-bit *extended* frames. Two triggers, one shared lambda
# (YAML anchor) — the lambda branches on the id itself, so it's correct for
# either frame type.
on_frame:
- can_id: 0
can_id_mask: 0 # accept every frame, dispatch in the lambda
use_extended_id: true
then: &decode_frame
- lambda: |-
// `can_id` and `x` (data bytes) are provided by the trigger.
uint32_t id = can_id;
// ---- command authentication: page-42 challenge reply ----
// After send_load_command sends the page-42 request, the target
// module returns a fresh 4-byte challenge on the 29-bit ext ID
// (node<<18)|0x10000|0x0142, payload 00 04 CC CC CC CC. Compute the
// REMOTE_CONTROL response, send it on page 43, then send the opcode
// x3. This is the only place we transmit; it acts solely on
// id(g_cmd_node), so no other node is ever touched.
if (id(g_cmd_pending) && x.size() >= 6) {
uint32_t chal_id = ((uint32_t) id(g_cmd_node) << 18) | 0x10000u | 0x0142u;
if (id == chal_id) {
// HARD SAFETY GATE (authoritative). This is the only place a
// command is ever transmitted, so the slides/jacks-are-panel-
// only rule is enforced here, right before TX, regardless of how
// the command was queued. If blocked: send nothing — no
// response, no opcode — and clear the pending command.
uint8_t n = id(g_cmd_node);
if (onecontrol::command_blocked(n, id(g_node_type)[n])) {
ESP_LOGE("idscan", "SAFETY: refusing command to motor node %02X "
"(slides/jacks are control-panel only)", n);
id(g_cmd_pending) = false;
return;
}
uint8_t resp[4];
ids_can_auth::remote_control_response_bytes(&x[2], resp);
// page-43 response: ext ID 0x0004<node>43, payload 00 04 RR RR RR RR
uint32_t resp_id = 0x00040043u | ((uint32_t) id(g_cmd_node) << 8);
std::vector<uint8_t> rframe = {0x00, 0x04, resp[0], resp[1], resp[2], resp[3]};
id(can_bus).send_data(resp_id, true, rframe);
// opcode x3: ext ID 0x0006<node><op>, DLC 0 (op 01=on, 00=off)
uint32_t op_id = 0x00060000u | ((uint32_t) id(g_cmd_node) << 8) | (uint32_t) id(g_cmd_op);
std::vector<uint8_t> opframe; // empty -> DLC 0
for (int i = 0; i < 3; i++) id(can_bus).send_data(op_id, true, opframe);
id(g_cmd_pending) = false;
ESP_LOGI("idscan", "node %02X: challenge %02X%02X%02X%02X -> response %02X%02X%02X%02X, opcode %02X x3",
id(g_cmd_node), x[2], x[3], x[4], x[5],
resp[0], resp[1], resp[2], resp[3], id(g_cmd_op));
return;
}
}
// ---- read broadcasts: 11-bit standard frames, id = (page<<8)|node ----
uint8_t page = (id >> 8) & 0xFF;
uint8_t node = id & 0xFF;
// Track each node's device class from its page-2 identity broadcast
// (x[3] = type byte). The safety gate uses this to refuse any
// motor-class (0x21) node that isn't the awning.
if (page == 2 && x.size() >= 4) {
id(g_node_type)[node] = x[3];
}
// Frame dump — confirmed both frame types decode (2026-06-12), now
// silenced: at DEBUG it logs every frame (~50/s across both
// triggers) and saturates the 115200 serial link. Uncomment to
// re-map the bus.
// ESP_LOGD("idscan", "page=%u node=%02X len=%u %02X %02X %02X %02X %02X %02X",
// page, node, x.size(),
// x.size()>0?x[0]:0, x.size()>1?x[1]:0, x.size()>2?x[2]:0,
// x.size()>3?x[3]:0, x.size()>4?x[4]:0, x.size()>5?x[5]:0);
// page 3 = live value. Layout depends on device class (README):
// tanks (type 0x0A): x[0] = level in percent (0x42 = 66%).
// switched loads (type 0x1E): x[0] bit0 = on/off.
if (page == 3 && x.size() >= 1) {
switch (node) {
// tanks (node addresses from the README node map for this rig)
case 0xE2: id(fresh_tank).publish_state(x[0]); break;
case 0xFE: id(black_tank).publish_state(x[0]); break;
case 0x27: id(grey_tank_1).publish_state(x[0]); break;
case 0x7D: id(grey_tank_2).publish_state(x[0]); break;
// switched loads
case 0xF8: id(interior_lights).publish_state(x[0] & 0x01); break;
case 0x2A: id(exterior_lights).publish_state(x[0] & 0x01); break;
// water heater (node 95): b0 bit0 = on, bit5 (0x20) = DSI/gas
// ignition fault/lockout (healthy: 0x80 off / 0x81 running;
// fault: 0xA0). Confirmed by a forced lockout 2026-06-12.
// x[3] bit7 = burner ACTUALLY FIRING (vs merely enabled): idle
// reads ...00 01, a live burn reads ...00 9X with the low nibble
// climbing as it heats (captured 2026-06-11). The switch state
// above is "enabled"; wh_heating is "actually making heat".
case 0x95: id(water_heater).publish_state(x[0] & 0x01);
id(dsi_fault).publish_state(x[0] & 0x20);
if (x.size() >= 4) id(wh_heating).publish_state(x[3] & 0x80);
break;
// furnace (node 89): propane forced-air, THERMOSTAT-controlled
// (not a Lippert load) — on the bus only to report DSI state, in
// the same type-0x1E encoding as the water heater. b0 bit0 =
// running, bit5 (0x20) = DSI ignition lockout. A lockout also
// trips the bus-wide page-0 fault flag (system_fault). Only idle
// 0x80 was captured (summer bench); 0x81 running / 0xA0 fault are
// inferred by parallel to node 95 — confirm on the first real
// burn cycle (or a forced lockout, propane off).
case 0x89: id(furnace_running).publish_state(x[0] & 0x01);
id(furnace_dsi_fault).publish_state(x[0] & 0x20); break;
// awning H-bridge (node 75): b0 C0 idle / C2 extending (opening)
// / C3 retracting (closing); b2-3 (BE) = motor current (raw
// counts, ~<1550 running incl. inrush, ~4200 at the fully-closed
// stall — captured 2026-07-01). Reflect motion onto the cover and
// publish current; while an auto-retract session is active, watch
// for the stall spike and cut the stream (see interval + script).
case 0x75: {
uint16_t cur = (x.size() >= 4) ? (uint16_t)(((uint16_t)x[2] << 8) | x[3]) : 0;
if (x.size() >= 4) id(awning_current).publish_state(cur);
if (x[0] == 0xC2) id(awning).current_operation = esphome::cover::COVER_OPERATION_OPENING;
else if (x[0] == 0xC3) id(awning).current_operation = esphome::cover::COVER_OPERATION_CLOSING;
else id(awning).current_operation = esphome::cover::COVER_OPERATION_IDLE;
// Stall gate: only while auto-retracting (C3), after the inrush
// blanking window. cur>2500 for 3 consecutive frames (~150ms @
// 20Hz) = fully closed → clear the session flag (interval stops
// streaming → motor halts) and mark the cover CLOSED (the one
// direction we get a real end-stop, so it's no longer assumed).
if (x[0] == 0xC3) id(g_awn_last_c3_ms) = millis();
if (id(g_awn_active) && x[0] == 0xC3 && x.size() >= 4) {
if (millis() - id(g_awn_start_ms) > 1200) {
if (cur > 2500) {
if (++id(g_awn_stall_count) >= 3) {
id(g_awn_active) = false;
id(awning).position = esphome::cover::COVER_CLOSED;
ESP_LOGI("awning", "auto-retract: stall %u -> stop (CLOSED)", cur);
}
} else {
id(g_awn_stall_count) = 0;
}
}
}
id(awning).publish_state();
break;
}
// (water pump 61 is command-blocked — read-only, not exposed.)
}
}
// page 0 b0 bit0 = a bus-wide "system fault present" flag (every
// node mirrors it; healthy 0x02, fault 0x03). Read it from one node
// (the heater) so we publish a single source, not 14 duplicates.
if (page == 0 && node == 0x95 && x.size() >= 1) {
id(system_fault).publish_state(x[0] & 0x01);
}
// Battery voltage rides 29-bit telemetry frames (src 7D/AE, page
// 0x11), payload 00 2B 0D 4x <rolling>; b2..b3 (BE) / 256 = volts.
// The low byte of the 29-bit id is the page (0x11); the "00 2B"
// prefix gates out any other page-0x11 traffic. Match on signature
// rather than the exact source id so either telemetry module
// (7D->FC or AE->01) feeds the same sensor. id > 0x7FF restricts the
// match to 29-bit frames: this lambda also runs for 11-bit broadcasts
// (second trigger), where the low byte is a NODE address — a node
// 0x11 would otherwise spoof the battery reading.
if (id > 0x7FFu && (id & 0xFFu) == 0x11u && x.size() >= 4 && x[0] == 0x00 && x[1] == 0x2B) {
float volts = ((uint16_t) x[2] << 8 | x[3]) / 256.0f;
id(battery_voltage).publish_state(volts);
}
# Second trigger: 11-bit standard frames (the read broadcasts). Same lambda.
- can_id: 0
can_id_mask: 0
use_extended_id: false
then: *decode_frame
# ---------------------------------------------------------------------------
# Command path: authenticated "set switched load"
#
# State shared between send_load_command (sends the request + retries) and the
# on_frame challenge handler (computes the response + sends the opcode). A
# command is queued by setting g_cmd_* and g_cmd_pending; the handler clears
# g_cmd_pending once the exchange completes, which also stops the retry loop.
# ---------------------------------------------------------------------------
globals:
- id: g_cmd_node
type: uint8_t
initial_value: '0'
- id: g_cmd_op
type: uint8_t
initial_value: '0'
- id: g_cmd_pending
type: bool
initial_value: 'false'
# Device class per node, learned from page-2 identity broadcasts (0 = not yet
# seen). Feeds the motor-output safety gate in command_guard.h.
- id: g_node_type
type: 'std::array<uint8_t, 256>'
# Name the type in the initializer: a bare '{}' is ambiguous between the
# GlobalsComponent<T>(T) and (std::array<...>) constructors under the current
# toolchain. zero-initialized -> every node starts as type 0 (not yet seen).
initial_value: 'std::array<uint8_t, 256>{}'
# --- Awning auto-retract session state -----------------------------------
# Set true by awning_auto_retract; the 100ms interval streams the retract
# opcode + page-44 keepalive while true, and the page-3 stall gate (or the
# timeout) clears it. Cleared → streaming stops → motor halts within ~1s
# (hold-to-run), so this bool is the master kill-switch for awning motion.
- id: g_awn_active
type: bool
initial_value: 'false'
- id: g_awn_start_ms
type: uint32_t
initial_value: '0'
- id: g_awn_stall_count
type: uint8_t
initial_value: '0'
- id: g_awn_ka_tick
type: uint8_t
initial_value: '0'
# millis() of the last observed retracting (C3) frame — lets the interval end
# the session ~1s after motion actually stops (whether the stall gate cut it,
# the OEM controller cut at its own limit, or streaming failed to sustain).
- id: g_awn_last_c3_ms
type: uint32_t
initial_value: '0'
script:
- id: send_load_command
# mode: restart -> a new press supersedes an in-flight exchange.
mode: restart
parameters:
node: int
op: int
then:
- lambda: |-
// SAFETY, layer 1 — motor-output gate. Slides/jacks are control-panel
// only; refuse any motor node that isn't the awning. Same rule the
// transmit point enforces (command_guard.h), checked here too so a
// blocked command never even starts the exchange.
if (onecontrol::command_blocked((uint8_t) node, id(g_node_type)[(uint8_t) node])) {
ESP_LOGE("idscan", "SAFETY: refusing command to motor node %02X "
"(slides/jacks are control-panel only)", (uint8_t) node);
id(g_cmd_pending) = false;
return;
}
// SAFETY, layer 2 — exposed-entity allowlist. Nodes wired to entities:
// 2A ext lights, F8 int lights, 95 water heater, 75 awning (cover). Do
// NOT add slide/jack/pump nodes — the gate above refuses them regardless.
if (node != 0x2A && node != 0xF8 && node != 0x95 && node != 0x75) {
ESP_LOGW("idscan", "refusing command to non-allowlisted node %02X", node);
id(g_cmd_pending) = false;
return;
}
id(g_cmd_node) = (uint8_t) node;
id(g_cmd_op) = (uint8_t) op;
id(g_cmd_pending) = true;
# Send the page-42 request, wait ~150 ms for the challenge; retry up to 8x
# (~1.2 s window) to ride out an occasional dropped arm on the busy bus.
# NOTE (confirmed live 2026-06-12): the module imposes a ~2 s cooldown after
# a SUCCESSFUL session — it won't issue a new challenge during it, so a
# second command to the same load <~2 s later is dropped regardless of opcode
# (this is module-side, not a bug; normal HA toggles are spaced far enough).
# The on_frame handler clears g_cmd_pending the moment it answers, so a
# successful exchange short-circuits the remaining iterations.
- repeat:
count: 8
then:
- if:
condition:
lambda: 'return id(g_cmd_pending);'
then:
- lambda: |-
// page-42 request: ext ID 0x0004<node>42, payload 00 04
uint32_t req_id = 0x00040042u | ((uint32_t) id(g_cmd_node) << 8);
std::vector<uint8_t> req = {0x00, 0x04};
id(can_bus).send_data(req_id, true, req);
- delay: 150ms
- if:
condition:
lambda: 'return id(g_cmd_pending);'
then:
- lambda: |-
ESP_LOGW("idscan", "no page-42 challenge from node %02X after 8 tries; command dropped",
id(g_cmd_node));
id(g_cmd_pending) = false;
# -------------------------------------------------------------------------
# Awning auto-retract — retract to the fully-closed hard stop, then stop by
# sensing the motor-current stall spike (no position feedback on the bus).
#
# Movement is HOLD-TO-RUN: a single authenticated opcode runs the motor only
# ~1s. The OEM app sustains motion by streaming the opcode (~110ms) plus a
# page-44 keepalive (~510ms) after ONE auth (captured 2026-06-11). We do the
# same: authenticate once via send_load_command, then the 100ms `interval`
# below streams opcode+keepalive while g_awn_active. The page-3 stall gate
# (case 0x75) clears g_awn_active at the closed stop; the interval's 70s
# timeout is the backstop. Stop streaming = motor stops within ~1s, so this
# fails safe on crash/Wi-Fi loss/stop-press.
- id: awning_auto_retract
mode: single # ignore re-press while a retract is already running
then:
- lambda: |-
id(g_awn_stall_count) = 0;
id(g_awn_ka_tick) = 0;
id(g_awn_start_ms) = millis();
id(g_awn_last_c3_ms) = millis();
id(g_awn_active) = true;
id(awning).current_operation = esphome::cover::COVER_OPERATION_CLOSING;
id(awning).publish_state();
# Authenticate + start motion via the proven single-shot path; the
# interval then keeps it moving until the stall gate or timeout fires.
- script.execute: { id: send_load_command, node: 0x75, op: 2 }
- id: awning_stop
then:
- lambda: 'id(g_awn_active) = false;' # halt the stream (motor stops ~1s)
- script.execute: { id: send_load_command, node: 0x75, op: 0 }
# ---------------------------------------------------------------------------
# Awning motion streamer — while g_awn_active, refresh the hold-to-run opcode
# every 100ms and the page-44 keepalive every ~500ms; enforce the safety
# timeout. Idle (g_awn_active false) → returns immediately, transmits nothing.
# ---------------------------------------------------------------------------
interval:
- interval: 100ms
then:
- lambda: |-
if (!id(g_awn_active)) return;
if (millis() - id(g_awn_start_ms) > 70000) {
id(g_awn_active) = false;
ESP_LOGW("awning", "auto-retract: 70s timeout -> stop");
return;
}
// motion-lost: once past the auth/startup window, if no retracting (C3)
// frame has arrived for >1s the motor isn't moving (stall gate/OEM
// cutoff/failed-to-sustain) — end the session instead of streaming on.
if (millis() - id(g_awn_start_ms) > 2500 &&
millis() - id(g_awn_last_c3_ms) > 1000) {
id(g_awn_active) = false;
ESP_LOGI("awning", "auto-retract: motion ended -> stop");
return;
}
// stream the retract opcode (0006 75 02, DLC 0) — refreshes hold-to-run
uint32_t op_id = 0x00060000u | (0x75u << 8) | 0x02u;
std::vector<uint8_t> op_empty; // DLC 0
id(can_bus).send_data(op_id, true, op_empty);
// page-44 keepalive (0004 75 44, payload 00 04) every ~500ms
if (++id(g_awn_ka_tick) >= 5) {
id(g_awn_ka_tick) = 0;
uint32_t ka_id = 0x00040044u | (0x75u << 8);
std::vector<uint8_t> ka = {0x00, 0x04};
id(can_bus).send_data(ka_id, true, ka);
}
# ---------------------------------------------------------------------------
# Read-back sensors (published by the dispatcher above)
# ---------------------------------------------------------------------------
sensor:
- platform: template
name: "Battery Voltage"
id: battery_voltage
unit_of_measurement: "V"
device_class: voltage
state_class: measurement
accuracy_decimals: 2
# The raw value jitters at 1/256 V every telemetry frame; unfiltered, that's
# constant state churn into both HA recorders + retained MQTT over the WG
# tunnel. Pass real moves (>0.05 V) immediately, else at most one per minute.
filters:
- or:
- delta: 0.05
- throttle: 60s
- platform: template
name: "Fresh Water Tank"
id: fresh_tank
unit_of_measurement: "%"
accuracy_decimals: 0
- platform: template
name: "Black Tank"
id: black_tank
unit_of_measurement: "%"
accuracy_decimals: 0
- platform: template
name: "Grey Tank 1"
id: grey_tank_1
unit_of_measurement: "%"
accuracy_decimals: 0
- platform: template
name: "Grey Tank 2"
id: grey_tank_2
unit_of_measurement: "%"
accuracy_decimals: 0
- platform: template
name: "Awning Motor Current"
id: awning_current
# Raw node-75 page-3 b2-3 counts (not amps): ~<1550 running incl. inrush,
# ramps to a ~4200 plateau at the fully-closed hard stop. Only updates while
# the motor moves (20Hz); holds last value at rest. Feeds the stall gate.
unit_of_measurement: "raw"
accuracy_decimals: 0
state_class: measurement
# ---------------------------------------------------------------------------
# Fault indicators (published by the dispatcher above)
# ---------------------------------------------------------------------------
binary_sensor:
- platform: template
name: "Water Heater DSI Fault"
id: dsi_fault
device_class: problem # node 95 page-3 b0 bit5 — gas ignition lockout
- platform: template
name: "Water Heater Heating"
id: wh_heating
device_class: running # node 95 page-3 x[3] bit7 — burner actively firing
- platform: template
name: "Furnace Running"
id: furnace_running
device_class: running # node 89 page-3 b0 bit0 — thermostat-driven burn
- platform: template
name: "Furnace DSI Fault"
id: furnace_dsi_fault
device_class: problem # node 89 page-3 b0 bit5 — gas ignition lockout
- platform: template
name: "OneControl System Fault"
id: system_fault
device_class: problem # page-0 b0 bit0 — bus-wide "a fault exists"
# ---------------------------------------------------------------------------
# Switches — authenticated command path (see send_load_command above).
# Each turn_on/off queues send_load_command for the load's node; the on_frame
# handler completes the page-42/43 challenge/response and sends the opcode.
#
# Exposed switched loads: exterior lights (2A), interior lights (F8), water
# heater (95). Each must be in the layer-2 allowlist in send_load_command.
#
# SLIDES, JACKS, and the WATER PUMP ARE PANEL/APP ONLY — never over CAN. Hard
# policy in command_guard.h, enforced both here and at the transmit point, so even
# adding a switch for one below cannot actuate it. The awning is permitted by the
# gate but needs a proper cover entity + an attended first test before it's wired.
# optimistic:false: the page-3 (b0 bit0) read-back above publishes true module
# state (verified live 2026-06-12), so HA shows ground truth — a dropped command
# self-corrects within ~1 s instead of an optimistic echo falsely reporting success.
# ---------------------------------------------------------------------------
# restore_mode: DISABLED on all three is load-bearing: ESPHome's default
# restore "applies" the boot state BY EXECUTING THE SWITCH ACTION, which sent
# a real authenticated OFF to the water heater on the 2026-06-12 OTA reboot
# (send_load_command is mode:restart, so the last switch in setup order — the
# heater — won the race). Boot must send nothing: the broadcasts repopulate
# every state within ~1 s and are the only source of truth.
switch:
- platform: template
name: "Exterior Lights"
id: exterior_lights
optimistic: false
restore_mode: DISABLED
turn_on_action:
- script.execute: { id: send_load_command, node: 0x2A, op: 1 }
turn_off_action:
- script.execute: { id: send_load_command, node: 0x2A, op: 0 }
- platform: template
name: "Interior Lights"
id: interior_lights
optimistic: false
restore_mode: DISABLED
turn_on_action:
- script.execute: { id: send_load_command, node: 0xF8, op: 1 }
turn_off_action:
- script.execute: { id: send_load_command, node: 0xF8, op: 0 }
- platform: template
name: "Water Heater"
id: water_heater
optimistic: false
restore_mode: DISABLED
turn_on_action:
- script.execute: { id: send_load_command, node: 0x95, op: 1 }
turn_off_action:
- script.execute: { id: send_load_command, node: 0x95, op: 0 }
# Slides/jacks/pump are command-blocked and must never be wired here.
# ---------------------------------------------------------------------------
# Awning (node 75) — open / close / stop cover.
#
# H-bridge motor: open = extend (op 01), close = retract (op 02), stop (op 00),
# per the captures. assumed_state -> HA shows discrete open/close/stop controls
# (no position slider); current_operation is published from the page-3 motion
# byte in the dispatcher above.
#
# Motion is HOLD-TO-RUN (confirmed 2026-06-11 capture + 2026-07-01 live): one
# opcode runs the motor ~1s. So:
# CLOSE = auto-retract — stream to the fully-closed stall, stop on current
# spike (awning_auto_retract script + interval + stall gate). This is
# the one direction with a real end-stop, so it also marks CLOSED.
# STOP = abort any active retract (clears g_awn_active) + send stop opcode.
# NO open_action ON PURPOSE: a single opcode only jogs the motor ~1s (useless),
# and there's no position/end-stop feedback to safely auto-extend on a timer.
# Extend the awning at the OEM wall switch. Omitting open_action drops SUPPORT_OPEN
# so HA shows only close/stop. assumed_state -> both stay always-pressable (no slider).
# ---------------------------------------------------------------------------
cover:
- platform: template
name: "Awning"
id: awning
device_class: awning
assumed_state: true
close_action:
- script.execute: awning_auto_retract
stop_action:
- script.execute: awning_stop