Every storage and generation decision downstream depends on one number: how much energy your system needs in a day, in watt-hours. A load profile is how you get that number honestly — not by guessing, and not by reading a single figure off an appliance and multiplying. This guide gives you a repeatable method, a reusable worksheet, and a fully worked example you can check line by line.

The work here is planning and measurement, not wiring. If you use a plug-in energy meter to measure appliances, follow its instructions and your appliance manuals, and stop if any lead, outlet, or meter is damaged, hot, or smells.

What a load profile is (and what it feeds)

A load profile is a list of everything the system powers, with each item turned into energy per day (Wh/day). Summed, it gives your daily energy demand — the figure that later drives:

  • how much usable storage you need for your target days of autonomy;
  • how much generation (solar, alternator, shore, generator) must replace each day;
  • whether your design survives a stretch of poor charging.

Before starting, make sure the distinction between power and energy is solid; the whole method rests on Energy = Power × Time — see energy vs. power.

The method in six steps

  1. List every load. Walk through a realistic day and write down everything that draws power, including the easy-to-forget ones: standby devices, an always-on router, the inverter itself.
  2. Find each load's power. Use the label, the data sheet, or — best — a measurement. Note whether it is an AC or DC load.
  3. Estimate effective run time per day. Not how long it is switched on, but how long it actually draws power. A thermostatically controlled device runs only part of the time it is "on."
  4. Compute energy per load: Wh/day = W × effective hours/day.
  5. Handle cycling, standby, and inverter losses (the three places profiles usually go wrong — covered below).
  6. Total it, then add a margin you can see and justify.

The worksheet

Copy this table and fill one row per load. Keep AC and DC loads distinguishable, because AC loads need an extra adjustment (step 5c).

Load AC/DC Rated power (W) Effective run time (h/day) Energy (Wh/day) How the figure was obtained
e.g. LED lights ×4 DC 32 4 128 label
e.g. Fridge (12 V) DC 45 (compressor) — (cycles) 430 measured over 24 h
e.g. Inverter idle DC 8 12 96 data sheet
…
Daily total sum

Three columns matter more than they look:

  • Effective run time, not switch-on time.
  • How the figure was obtained — a measured value and a guess should not look identical on paper.
  • AC/DC, because it decides whether an inverter-loss adjustment applies.

Step 5a: cycling and fluctuating loads

Loads like fridges, freezers, and pumps switch on and off. Their compressor might draw 45 W while running, but it does not run all day. Two traps follow:

  • The naive trap: taking the running wattage and multiplying by 24 hours. A 45 W compressor treated this way gives 45 W × 24 h = 1 080 Wh — more than double reality, because the compressor is off most of the time.
  • The guessing trap: picking a duty cycle out of the air.

The reliable approach follows U.S. Department of Energy measurement guidance for fluctuating loads: measure energy over a period of time and divide by that period to get average power. DOE's standby-power method notes that a reading can be taken directly only when it is stable — within about 5% of the mean over 5 minutes — and that otherwise you should accumulate energy over a timeframe and compute the average. A cycling fridge is the textbook unstable case.

In practice, run a plug-in or DC energy meter across a representative period — ideally a full day that spans several complete cool-down cycles at a realistic ambient temperature — and read the accumulated watt-hours. Suppose the meter shows 430 Wh over 24 hours. That is your fridge's Wh/day directly. As a cross-check, its average power is 430 Wh ÷ 24 h ≈ 17.9 W — far below the 45 W compressor figure, which is exactly the point.

If you genuinely cannot measure, a labelled duty-cycle estimate is the fallback: 45 W × 24 h × 0.40 (assumed) = 432 Wh. Mark it as an estimate. Note that a fridge's duty cycle rises sharply in hot weather, so an estimate made in spring can understate summer demand.

Step 5b: do not forget standby and idle loads

Small always-on loads are individually trivial and collectively decisive, because they run every hour of every day. A 5 W standby device is 5 W × 24 h = 120 Wh/day — more than four LED lights running for an evening. Two to watch:

  • Device standby: routers, alarms, sensors, and "off but not unplugged" electronics. DOE's guidance exists precisely because these draws are real and measurable.
  • Inverter idle draw: many inverters consume power whenever switched on, even with nothing plugged in. If yours idles at 8 W and stays on 12 hours a day, that is 8 W × 12 h = 96 Wh/day before a single appliance runs. Turning the inverter off when unused is often the single easiest saving — see AC vs. DC.

Step 5c: adjust AC loads for inverter losses

Watt-hours listed for an AC appliance are what the appliance uses. The energy pulled from your DC battery to supply it is larger, because the inverter loses a little in conversion. Convert AC-load energy to DC draw with:

DC energy drawn = AC load energy ÷ inverter efficiency

Efficiency is a property of your specific inverter and its operating point, so take it from the data sheet. As an illustration only, a 200 Wh/day AC load through a 90%-efficient inverter draws 200 Wh ÷ 0.90 ≈ 222 Wh from the battery. Apply this to AC loads; native DC loads need no such adjustment. (This is one reason running a load natively on DC, where possible, saves energy.)

Worked example: a small off-grid day

Putting it together with illustrative figures — your numbers will differ:

Load AC/DC Power (W) Effective h/day Energy (Wh/day)
LED lights ×4 DC 32 4 128
Phone/laptop charging DC 45 3 135
Water pump DC 60 0.3 18
Ventilation fan DC 12 6 72
Inverter idle DC 8 12 96
Router / misc standby DC 5 24 120
Fridge (12 V, cycling) DC 45 running — 430 (measured)
Daily total ≈ 999 Wh ≈ 1.0 kWh

Each row is a simple W × h, except the fridge, which is measured. Summing the constant loads gives 128 + 135 + 18 + 72 + 96 + 120 = 569 Wh; adding the measured fridge gives 569 + 430 = 999 Wh, about 1.0 kWh per day. Every product is independently checkable, and the fridge is a measured value rather than an inflated 45 W × 24 h.

Step 6: add a visible margin

Real days are not average days. It is reasonable to add a margin for higher use, measurement error, and aging — but make it explicit rather than hiding it inside an inflated load. State it plainly, for example: "daily demand ≈ 1.0 kWh; design target 1.2 kWh (a 20% margin)." That way the margin is a decision anyone can see and adjust, not a number smuggled into the loads. How large a margin is appropriate depends on how tolerant your use case is of a shortfall, which the sizing guide takes up directly.

Common mistakes this method prevents

  • Multiplying a cycling load's running wattage by 24 hours.
  • Omitting standby and inverter idle draws.
  • Costing AC loads at appliance watt-hours without the inverter adjustment.
  • Confusing switch-on time with effective run time.
  • Burying a safety margin inside the load figures so no one can review it.

Key takeaways

  • A load profile turns appliances into one reproducible number: daily energy in watt-hours.
  • For cycling or fluctuating loads, measure accumulated energy over a representative period and divide by time — do not read an instantaneous watt figure (DOE FEMP method).
  • Standby and inverter idle loads run 24 hours a day and often dominate; include them.
  • Divide AC-load energy by inverter efficiency to get the real DC draw.
  • Keep any safety margin visible and separate from the loads.

Where to go next