Controls — Recipes

Field	Detail
Primary artifact(s)	controls/water/recipes/recipes.json, controls/water/recipes/recipes.csv, controls/water/st/delta-recipe.st
Spec	RO-SPEC-001 §5.5 ; WIRE-001 —

Summary

Ten named HMI presets (indices 0–9 → RCP_0_* … RCP_9_* in tags.csv). Indices 0–3 mirror the Example Factory-Loaded Presets table in system overview §5.5; indices 4–7 are JCWS service profiles; indices 8–9 are operator NV slots (Custom 1 / Custom 2). Dosing rates are illustrative until concentrate bench calibration.

idx	id	display_name	MP-101	MP-102	MP-103	TDS	§5.5
0	sca_gold_cup	SCA Gold Cup	2.4	3.1	1.8	120	yes
1	light_roast_high_ext	Light Roast — High Extraction	3.6	1.5	2.0	100	yes
2	medium_roast_balanced	Medium Roast — Balanced	2.0	2.5	2.2	130	yes
3	rao_perger_water	Rao/Perger Water	3.0	0.0	1.5	85	yes
4	espresso_standard	Espresso — Standard JC	2.4	3.1	1.8	120	—
5	espresso_perla_negra	Espresso — Perla Negra	3.0	3.8	2.6	135	—
6	matcha_water	Matcha Service Water	1.9	2.2	2.8	95	—
7	dw_still_service	Still Drinking Water	1.2	2.8	2.0	110	—
8	custom_1	Custom 1	—	—	—	—	NV
9	custom_2	Custom 2	—	—	—	—	NV

Barista ppm deltas (RCP_DELTA_PPM_ADJ_*) convert to mL/gal offsets in delta-recipe.st via dimensionless gains kMg / kCa / kNa (placeholders 0.02 / 0.018 / 0.015); dosing.st applies the same factors per pump when computing stroke words.

Test

Import JSON or CSV into CM5 NV / PLC retain array; run controls/water/sim-tests/sim-delta-recipe-propagation.md and sim-recipe-change-mid-fill.md.

Offline validation (from repo root):

python3 - <<'PY'
import csv, json
from pathlib import Path

json_path = Path("controls/water/recipes/recipes.json")
csv_path = Path("controls/water/recipes/recipes.csv")

recipes = json.loads(json_path.read_text(encoding="utf-8"))
assert len(recipes) == 10, len(recipes)

fields = (
    "id", "display_name",
    "mp101_ml_per_gal", "mp102_ml_per_gal", "mp103_ml_per_gal",
    "target_tds_ppm", "tds_tolerance_ppm", "notes",
)
with csv_path.open(encoding="utf-8") as f:
    rows = list(csv.DictReader(f))
assert len(rows) == 10
for i, (r, j) in enumerate(zip(rows, recipes)):
    assert int(r["idx"]) == i
    for k in fields:
        got = r[k] if k.startswith("mp") or k.endswith("_ppm") else r[k]
        if k.startswith("mp"):
            assert float(got) == j[k]
        elif k.endswith("_ppm"):
            assert int(float(got)) == j[k]
        else:
            assert got == j[k]
ids = [r["id"] for r in recipes]
assert len(ids) == len(set(ids))
print("OK: 10 presets, JSON/CSV mirror, unique ids")
PY

Changelog

• 2026-05-24 — Reordered presets: §5.5 factory rows at indices 0–3; JC service profiles 4–7; custom_1 / custom_2 at 8–9. Expanded delta-recipe.st per-mineral mL/gal math and lab-cal placeholder comments.

Open items

• Lab calibrate all dosing rates and kMg / kCa / kNa gains against Levitt concentrate prep (metering pumps §6)
• Wire RCP_APPLY_TRIG when HMI commissioning wires Apply button (see hmi/screens.md)

Reviewer sign-off

• Recipe preset table reviewed against RO-SPEC-001 §5.5 — _______________