Sezibwa Solar + Battery Operating Plan

8 × 600 W panels (4.8 kWp), one string on inverter PV1 · prepared 26 Oct 2026 
 
 ≈19.5 kWh solar per average day (15–18 in rainy weeks) 
 ≈28 kWh site load per day (19–25 Oct, measured) 
 ≈9.6 kWh midday surplus to store in the battery 
 ≈65% of the load from solar + battery on a good day 
 
 A day at Sezibwa: solar vs load 
 0 kW 1 kW 2 kW 3 kW 4 kW 00:00 03:00 06:00 09:00 12:00 15:00 18:00 21:00 24:00 night grid charge (today) evening peak (battery covers) solar surplus ≈ 9 kWh → battery expected solar, avg day (≈19.5 kWh) site load, 19–25 Oct average (≈28 kWh) 
 
 Time of day Load Share Who supplies it (plan) 
 Night 00–07 5.8 kWh 21% Grid (cheap night rate) 
 Day 07–18 10.6 kWh 38% Solar directly; surplus charges the battery 
 Evening 18–24 11.8 kWh 42% Battery (solar-charged), then grid 
 
 The plan 
 
 Prove the output. Remote PV1 read 27 Oct 09:30 EAT, then midday checks: expect ~300–360 V and 2–4 kW in sun. 
 Make room for the sun (key decision). Today the grid charges the battery 00:00–06:00, so it's near full at sunrise and midday surplus is wasted (no export). Stop or cap the night charge (~50%) so solar fills the battery for free. 
 Keep the evening discharge. The solar-charged battery covers the 18:00–24:00 peak, the dearest hours. 
 Keep a backup floor. Hold 15% (higher in the rainy season) so outages stay covered. 
 Make solar visible. OpenEMS reads PV as production (logger #95), then add solar to the daily brief. 
 Later: cloudy-day logic and OpenEMS solar-aware charging controllers. 
 
 Still open on site: earthing (no earth point yet) for the surge protector and panel rails. Watch: trees beside the array can shade the string; one shaded panel limits all eight. 
 Decision with Aaron: where to set the 00:00–06:00 grid charge once PV is proven: stop it, or cap it at ~50% for cloudy days. 
 Commissioning log (26 Oct 2026) 8 × 600 Wp in one row on two continuous rail lines (vent removed, end clamps used at one joint for the missing inter-clamp pair). 6 mm² PV cable (25 m −, 15 m +); cable sizing checked: 0.4% drop. FMPV32 4-pole DC isolator wired per the manual's 4B layout (cables on top 1/3/5/7, links 2–4 and 6–8, go by stamped numbers); 1000 V DC SPD across +/−. Open-circuit string 379 V at 16:45 EAT; inverter PV1 read 151.8 V / 0 A at 18:39 EAT (sunset). PV1 registers 0x0107 V×0.1, 0x0108 A×0.1, 0x0109 W. Open: earthing (no earth point on site), full-output check 27 Oct, tree shading at one end of the array.