Home battery — when to charge from grid vs solar

· 2 min read

The default battery logic is simple: charge on midday solar surplus, discharge in the evening. Most days, that’s all you need. But add a dynamic tariff and a second game appears — price arbitrage: storing cheap grid power and using it in expensive hours. The skill is knowing which game to play on which day. The answer lives in two forecasts: prices (known the day before) and production (forecast).

Ground rule: sunshine goes first

Your own roof power is “free” — every stored solar kWh saves the full retail price in the evening. So if the forecast says tomorrow’s surplus will fill the battery: don’t grid-charge overnight. Otherwise the battery wakes up full of cheap power and the midday surplus exports for pennies. That’s the most common beginner arbitrage mistake.

When grid charging makes sense

Scenario 1: overcast tomorrow + expensive evening. Low solar forecast (the battery won’t fill itself) and high evening prices — charge overnight in the price trough exactly as much as the evening ramp needs. Profit = price spread minus round-trip losses (~10–15%).

Scenario 2: negative or zero midday prices. On sunny weekends the wholesale price hits zero or below while your own production can’t max out the charger — topping up from the grid then costs pennies or even earns them.

Scenario 3: an announced price spike. A cold, windless evening with extreme day-ahead prices — enter the evening with a full battery, whatever the charging source.

The five-step evening algorithm

  1. Check tomorrow’s hourly prices (published each afternoon).
  2. Check tomorrow’s production forecast.
  3. Forecast surplus ≥ battery capacity → do nothing; the sun handles it.
  4. Otherwise: compute the deficit and schedule top-up in the cheapest overnight hours.
  5. Always cover the evening peak from the battery, never from the grid.

This takes two minutes a day by hand — and should eventually run itself: exactly this loop of prices, forecast and inverter control is the core job of a home EMS.

What arbitrage players shouldn’t forget

Round-trip losses eat 10–15% of every “turned” kWh — arbitrage pays only from a clear price spread (rule of thumb: 25–30% after losses). Cycles cost battery life, though LiFePO4’s several thousand cycles make that cheaper than commonly assumed. And check your charge power: the overnight trough lasts only a few hours, and the planned amount has to fit inside it.

FAQ

Is grid-charging a home battery allowed? Technically most hybrid systems can; whether it’s enabled depends on inverter settings and sometimes your export scheme — check with your installer or supplier.

What price spread makes arbitrage worthwhile? After losses: from roughly 25–30% between the charge and discharge hour. A typical dynamic-tariff day offers considerably more.

Do I have to do all this manually? Today, largely yes (or semi-automatically in Home Assistant). Automating this exact loop is the core of the next product we’re building, Volcast EMS.