← Knowledge

Measuring Solar Surplus Correctly

Why “what flows into the grid” is the wrong number, and how to calculate the surplus you can actually distribute.

The obvious number, and why it's wrong

Anyone building PV surplus charging instinctively reaches for the grid meter: if power is flowing into the grid, there's apparently something left over, so charge the car with that. This definition is baked into a surprising number of systems. And it has a fundamental design flaw:

Grid export is what's left over after everyone else has already helped themselves. The home battery charging at 5 kW? The heating element pulling 3 kW? Then the grid meter shows almost nothing, and "surplus charging" for the car never kicks in, even though the roof is delivering 9 kW. In practice, you end up with a car that doesn't charge on a bright summer day, while the battery has been full since 11 a.m.

The correct calculation

The distributable surplus is:

Surplus = PV production − non-controllable house load

The non-controllable house load is the fridge, the stove, the TV: everything you can't shift in time. The controllable consumers (car, water heater, battery, pool pump) are not subtracted: their allocation is precisely the point. In practice, this is calculated from grid power, battery power, and the measured power of the controllable devices, which requires a grid measuring point plus one measurement per controllable device.

The second decision: who gets it?

With the right number, the real question begins: in what order are the car, battery, and hot water served? That's not a technical question but a preference question:

• Battery first means: evening reserve, but the car charges more slowly.
• Car first means: cheap kilometers, the battery fills up later, or on some days, not completely.
• The compromise many people want: "battery first, but only up to X%": secure a reserve, then everything goes to the car.

A clean system models this as a waterfall: each consumer in line gets its share, and the rest flows onward. An interesting detail with the home battery: you often don't even need to actively control it. If the car gets priority, less is automatically left for the battery, and its own energy management handles the rest.

Three practical traps

1. The 6-amp floor: under the charging standard, wallboxes can't charge below 6 A; three-phase, that's a minimum of roughly 4.1 kW. Smaller surpluses can only be used by the car itself, provided it's current-controllable. Systems that ignore this either don't charge at all at a 2 kW surplus, or charge with 50% grid power.
2. Cloud flicker: a system that stops the instant a cloud passes and starts again the instant the sun returns switches the wallbox contactor every minute or two: wear with no benefit. It needs hysteresis and minimum run times.
3. Flying blind on meter failure: if the grid meter goes offline (classically: the router assigned the meter a new IP), naive systems keep calculating with the last known number, and "solar charge" for hours out of the battery and the grid. The honest response is: pause, and say so.

Checklist for a DIY build (or comparing systems)

• Is the surplus calculated from production − non-controllable load (not from grid export)?
• Is the allocation order configurable, including "battery up to X%, then car"?
• Is there hysteresis against cloud flicker, and restart protection for the contactor?
• What happens on meter failure: flying blind, or an honest pause?
• Can sub-6A surplus be used via the car?