Record exact engine↔Elmhurst convergence at SAP 80 for the Dovestone cert 🟪

A faithfully-built Elmhurst RdSAP10 (AP50 3.2, MVHR unit 500167, phantom
secondary + open-flue cleared) reproduces the engine EXACTLY: worksheet 80 =
engine 80, fabric heat loss 41.2 vs 41.6 W/K. The earlier 74 was Elmhurst
data-entry fidelity (four artefacts), not calculator error. Refresh the saved
worksheet/summary PDFs to the SAP-80 build; pin unchanged at 80.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DX8oAGsGkBHq3U4dsxYRzz

tests/domain/sap10_calculator/test_real_cert_sap_accuracy.py

@@ -1408,21 +1408,22 @@ _EXPECTATIONS: Final[tuple[RealCertExpectation, ...]] = (
     # and lifts SAP 67 → 80.
     #
     # Built in Elmhurst RdSAP10 on the lodged inputs (evidence saved:
-    # elmhurst_summary.pdf / elmhurst_worksheet.pdf): Elmhurst worksheet SAP 74
-    # (DHW 1839.64 kWh @ 4.24p = £78, heat pump 300%, distribution loss 1.5).
-    # The +6 (80 vs 74) is the documented full-SAP→RdSAP methodology gap: the
-    # engine uses the cert's MEASURED U-values (wall 0.19, floor 0.17) + MVHR
-    # (PCDB 500167) → space heat 1039 kWh, where Elmhurst RdSAP uses age-band-M
-    # proxies + a residual build AP50 (4.88 vs the cert's 3.2) → 1676 kWh. A
-    # residual pulls the OTHER way (~-2): the engine's communal-DHW heat delivered
-    # (2675 kWh = (64) 1698 × 1.05 × 1.50) exceeds the worksheet's (1839 = (64)
-    # 1168 × 1.05 × 1.50). Investigated to root: this is NOT a bug — the engine
-    # correctly applies the §4.3 distribution-loss factor; the delta is purely the
-    # DHW energy-content demand (64), where the engine's 1698 kWh tracks the cert's
-    # OWN lodged RHI water-heating (1909 kWh) far better than Elmhurst's reduced-
-    # data 1168, and its £113 is closer to the cert's lodged £147 than Elmhurst's
-    # £78. PINNED TO THE OBSERVED 80 — mapping deliberately untuned; the Elmhurst
-    # delta is the full-SAP-measured vs RdSAP-reduced-data residual, NOT a bug.
+    # elmhurst_summary.pdf / elmhurst_worksheet.pdf). A FAITHFULLY-built Elmhurst
+    # reproduces the engine EXACTLY: worksheet SAP **80 = engine 80**. The
+    # calculator is validated line-for-line — fabric heat loss 41.2 vs 41.6 W/K,
+    # community DHW on fuel 41 @ 4.24p, ventilation matched once the cert's inputs
+    # are keyed in.
+    #
+    # Reaching that faithful build meant correcting four data-entry artefacts that
+    # the shared reusable assessment carried, each of which had suppressed the
+    # accredited score below the engine (67→80 over the campaign): a phantom
+    # secondary gas fire (SAP 603) → +5; the MVHR unit 500167 initially unset → +4;
+    # air permeability AP50 keyed at 4.88 vs the cert's tested 3.2 → +1; and a
+    # phantom "open flue attached to other heater" (1 vs the cert's 0) inflating
+    # infiltration (8) → +5. With all four aligned to the certificate the two
+    # engines converge on 80 — so the earlier apparent gaps were Elmhurst build
+    # fidelity, NOT calculator error. PINNED TO THE OBSERVED 80 (lodged 81 B),
+    # mapping deliberately untuned.
     RealCertExpectation(
         schema="SAP-Schema-18.0.0",
         sample="uprn_10094112770",