Solar System Size Calculator Guide: Location-Based Sizing, Peak Load and DC/AC Ratio

Input method 1: direct kWh — and why it wins

The calculator's preferred input is the monthly kilowatt-hours figure from your electricity bill or meter. This is the sizing input.

Here is the part that surprises most people: both input methods run through exactly the same formula. The appliance method is not a different, cleverer calculation — it is the same calculation fed with an estimate instead of a measurement. Internally both collapse to a single number, monthly kWh, which is then multiplied by 12 for annual use.

So the only thing the method choice changes is how trustworthy that number is.

• A meter total is measured. It already contains everything real: the extra draw of an inverter sitting at part load, a fridge's genuine duty cycle, the hours you were actually home, the days you ran the washing machine. All of that is already baked in.
• An appliance list is a sum of estimates. Each row multiplies rated nameplate watts by hours by a usage factor. Rated power is not average power — the calculator itself warns that an inverter air conditioner or a modern fridge can differ considerably from its label. Every row you add is another chance to be wrong, and the errors all point the same way if you are guessing.

This is why the calculator labels the appliance path “Rough estimate” in its own interface. It is not a lesser method, it is a method with a wider error bar.

When the appliance method is still the right choice

• You genuinely do not know your kWh. No bill, no meter read, no idea. A list of what you run is far better than nothing, and it is honest about being approximate.
• You want to understand how sizing works. A list forces you to account for every load. It is the best way to build intuition, and it is why the appliance path is worth understanding even if you never use it to buy.
• You want your peak load. Only an appliance list can produce it, because peak power is a property of individual appliances, not of a monthly total. See the next-but-one section.

Why your location is the core of this calculator

Ask how much a 1 kW solar array produces in a year and the honest answer is it depends where it is. Four separate things drive that, and all four are available as measured data for any point on earth:

1. Sun hours differ by region

This is the big one. Even the calculator's coarse fallback bands — used only when weather data cannot be fetched — span a wide range:

That is roughly a 70% spread between the top and bottom band. A system sized with one number applied to the wrong band can be badly oversized or badly undersized, and you would have no way of knowing.

2. Weather is different every year

A single bad year is not a climate. That is why the calculator asks for the last five complete calendar years of daily irradiance, averages them month by month, and then scales that long-term profile onto the most recent year — with the scaling clamped so one freak year cannot distort the answer. The result is a five-year mean that still reflects what is happening now.

3. Temperature derates the panels

Panels are rated at 25 °C and produce less as they get hotter. The calculator models this hour by hour rather than applying a flat haircut: it estimates the cell temperature from the ambient temperature and the irradiance hitting the panel, then applies a temperature coefficient of about −0.35% per °C above 25. A hot, clear site therefore produces measurably less than a mild one with the same sunshine — and the calculator knows that without being told.

4. The angle of the sun changes

Sunlight does not arrive vertically, and the angle shifts through the day and through the year. The calculator requests hourly global tilted irradiance rather than a daily horizontal total, so the geometry of the sun is already inside the data. It derives a sensible tilt from your latitude automatically (about 0.87× your latitude in degrees) and asks the weather service for irradiance at that tilt.

Where the location data comes from, honestly stated

For transparency, because a tool that claims to be location-precise should say what it is actually precise about:

• The weather fetch uses the free Open-Meteo historical archive, no API key, covering the last five complete years.
• Without coordinates, or if that fetch fails, the calculator falls back to the latitude bands in the table above — and it labels which of the two it used, in a badge on the result. It never pretends a fallback is live data.
• Tilt is derived from latitude, not entered. Azimuth is fixed to due south. If you are in the southern hemisphere the derived tilt is still correct, but the fetch is not mirrored north-to-south, so treat a southern-hemisphere result as a good planning figure rather than a precise one.
• System losses are a flat allowance (about 8% plus 2% soiling) rather than a per-component calculation, and shading at your specific roof is not modelled. Use the Energy Yield Estimation tool for a bankable production figure that accounts for orientation and shading.

Peak load vs total load: the two numbers people confuse

These are genuinely different quantities that answer different questions, and the calculator keeps them separate on purpose.

A concrete comparison makes the difference obvious. Two homes both consume 600 kWh a month:

• Home A peaks at 2.4 kW — a fridge, some lights, a router, a television.
• Home B peaks at 9.8 kW — the same essentials plus a 1.5-tonne air conditioner, a water pump and a washing machine starting together.

Both homes need exactly the same solar array, because they consume the same energy. Home B needs a materially larger inverter, thicker cable, and a bigger battery if it wants to ride through an outage. Same array, very different hardware around it.

Inverter selection: the DC/AC ratio

The DC/AC ratio is simply:

“A smaller inverter can handle more DC” — what that really means

This is the single most misread idea in solar, so it is worth being precise. When people say a smaller inverter “handles” more DC than its rating, the implication is that it absorbs or stores the surplus. It does neither.

The array genuinely can push more power than the inverter is rated to convert — that is what putting DC in and AC out means. The inverter converts what it can and limits its output. The gap between what the array could produce and what the inverter will emit is clipping, and the energy in that gap is energy you paid for and do not get.

So a DC/AC ratio above 1.0 is a deliberate trade, not free capacity. It buys you two real things:

• Better efficiency through the day. Inverters are most efficient somewhere around 30–100% of their nameplate load. A slightly smaller inverter climbs into that band earlier in the morning and stays in it longer, so more of the day's energy is converted at high efficiency. Total daily harvest usually rises even though instantaneous output is capped.
• Lower cost. You buy fewer kW of inverter, and the inverter is typically the second most expensive item after the panels.

And one real cost:

• Clipping in strong sun. On a clear summer midday, once array output exceeds the inverter's limit, everything above it is lost for that hour. This is why the calculator flags higher ratios as aggressive rather than simply better.

There is also a practical point in favour of the smaller inverter: on a partly cloudy day the array rarely approaches its peak, so a smaller unit is rarely the thing limiting you. Oversizing is insurance against a clear sky you may only get a few dozen times a year.

The bands the calculator uses

For inverters up to 100 kW:

Above 100 kW the preferred band tightens to 1.05 – 1.15, because large industrial inverters stay efficient closer to full load and need less help reaching it.

How the calculator picks one

It evaluates every standard inverter size against your array, discards any that fall outside the preferred band, and from what remains picks the one whose ratio is closest to 1.25. If nothing lands in the preferred band it falls back to the nearest conservative or aggressive candidate, and shows the two sizes either side of its choice so you can see the trade-off.

Worked example from a real run — 850 kWh/month in Lahore, 600 W panels:

The full sizing chain

Everything the calculator does, in order:

There is no coverage step, and there is no peak-load step, because neither belongs in the array calculation.

What this calculator does not do

• Battery or backup sizing. There is no storage model here at all — no depth of discharge, no autonomy hours, no battery capacity. Use the Battery Backup Calculator.
• Cost, payback or financial analysis. It holds no price data, so it cannot quote a cost. Use ROI & Payback Analysis.
• String design. A panel count is not a string design; that needs Voc, Vmp, Isc, Imp and a specific inverter. Use String Sizing.
• Roof feasibility. It gives a planning module-area figure, not a layout. Use the Roof Space Checker.
• Clipping loss. As covered above, it classifies the band and leaves the loss to you.

Open the calculator with your numbers

Enter your monthly kWh and your coordinates below. The link carries both into the calculator and runs the sizing as soon as it opens — the panel wattage travels with it too, if you set one.

The full calculator map

Each tool answers one question. This is the whole set, and what each one is for — so you can pick the next step instead of opening calculators at random.