diff --git a/backend/documents_parser/tests/test_pashub_sap_accuracy_lrha_wave3.py b/backend/documents_parser/tests/test_pashub_sap_accuracy_lrha_wave3.py
index db5e9056b..d734b17d7 100644
--- a/backend/documents_parser/tests/test_pashub_sap_accuracy_lrha_wave3.py
+++ b/backend/documents_parser/tests/test_pashub_sap_accuracy_lrha_wave3.py
@@ -52,6 +52,15 @@ _MANIFEST_PATH = _FIXTURES_DIR / "manifest.json"
 # dwellings → **51 computed / 53 blocked** (within-0.5 41.2% = 21/51, MAE
 # 2.948). The newly-computable oil configs land outside 0.5 per the
 # coverage-growth convention above — floor re-baselined 0.46 → 0.41.
+# 2026-07-16 (#1622 sub-issues #1630-#1637): cylinder Medium band, heat-pump
+# Group 2 controls + category 4, heat-network charging controls
+# 2306/2314/2313, MVHR, biomass community fuel 43, HHR-control downgrade →
+# **77 computed / 27 blocked** (within-0.5 42.9% = 33/77, MAE 2.960; ratchets
+# hold). Residual blocks: 'Exact cylinder volume' ×16 (extractor+dataclass,
+# tracked separately), the line-wrap-truncated '…room thermostat and'
+# controls label ×6 (extractor fix, tracked separately), heat_network_age_band
+# '2007 - 2011' ×4 (calculator cascade dict, surfaced behind the biomass
+# fuel), room-in-roof gable 'None' ×1.
 # Treat the constants as a regression tripwire, NOT a cohort accuracy figure;
 # re-baseline (coverage growth is not loosening) as further mapper fixes unblock
 # fixtures.