How to Size a Solar System for Your Business: A Step-by-Step Load Calculation Guide

Nobody sizes a solar system by asking how many panels fit on the roof, and yet that is exactly how most quotations in Lagos, Cairo and Khartoum get written. The number that matters is not on your roof. It is in your building, spread across every appliance you switch on and the hours you leave it running.

If you have ever been handed a quote for a "5kVA solar system" with no explanation of where 5kVA came from, your instinct that something was missing was correct. This guide shows you how to size a solar system properly: how to build a load audit, split day loads from night loads, allow for surge, then calculate the array, battery bank and inverter in that order. We carry one small business through every step with real arithmetic, so you can copy the method against your own numbers.

We size systems this way because we install and maintain them afterwards. Getting the solar load calculation wrong does not cost us a sale; it costs us a return trip.

How to Size a Solar System: The Six Steps That Actually Matter

Solar sizing runs in one direction, and the direction is not negotiable. Loads first, then energy, then generation, then storage, then conversion. Every step depends on the one before it. Choose an inverter first and you are guessing; choose a panel count first and you are decorating a roof.

The six steps are: build the load audit, separate day from night, account for surge, size the array against peak sun hours, size the battery bank, then size the inverter. A sanity check at the end confirms the three components can actually work together across a full 24 hours.

When Chinedu opened his print and design shop in Ikeja in March 2025, a vendor sold him a "5kVA solar system with 4 batteries" for a flat price. Nobody visited the shop. Nobody asked how long his two large-format printers ran, or noticed that his security lights and CCTV stayed on all night.

The system worked beautifully until 9pm every evening, when the batteries hit their cut-off and the shutters, alarm and cameras died. He had bought a competently built system sized against nothing at all.

Not sure where your own numbers land? MIMAH runs professional load audits and feasibility studies as a standalone engagement. See how our engineering services work before you commit capital to a specification.

Step 1: Build a Load Audit Table (Your Solar Load Calculation)

A load audit is a list of every electrical item in the building, what it draws, and how long it actually runs. That is the whole idea. The rigour comes from being honest about the hours.

The core arithmetic never changes:

Watts x hours per day = watt-hours per day (Wh/day)

Multiply by quantity where you have several of the same item, then add every row together. That total is your daily energy consumption, and it is the single most important number in the entire project.

Three rules separate a load audit that holds up from one that does not. Use running wattage, not the nameplate rating on the sticker, because nameplate is usually the maximum a device can ever draw. Use observed hours, not assumed hours. And treat cycling loads such as fridges and freezers by their average draw across 24 hours, since a unit rated at 200W may only run its compressor 35 to 40 percent of the time.

The cheapest instrument in this process is a plug-in energy meter costing a few dollars. Leave it on your three biggest suspects for a week.

The Load Audit Template You Can Copy

Set up seven columns in a spreadsheet and fill one row per appliance:

Appliance: the specific item, not a category

Quantity: how many units

Running watts: measured or manufacturer running load, not nameplate peak

Hours per day: observed, not assumed

Day or night: which window the load sits in

Wh/day: quantity x watts x hours

Surge multiple: starting current factor for motors and compressors, 1 for everything else

Here is that template filled in for our worked example, a small business we will carry through every remaining step. Call it a design and print shop in Ikeja, Lagos, open 07:00 to 19:00, with security and refrigeration running overnight.

Daytime loads, 07:00 to 19:00:

LED lighting, 12 units at 18W = 216W for 10 hours = 2,160 Wh

Inverter split AC, 1.5 hp, 1,100W for 6 hours = 6,600 Wh

Desktops and monitors, 4 units at 200W = 800W for 9 hours = 7,200 Wh

Laser printer, 350W for 1 hour = 350 Wh

Water dispenser and kettle, 900W for 1 hour = 900 Wh

Refrigerator, 70W average draw for 12 hours = 840 Wh

CCTV, NVR and router, 60W for 12 hours = 720 Wh

Pressure pump, 0.75 kW (1 hp), 750W for 1.5 hours = 1,125 Wh

Daytime subtotal: 19,895 Wh

Night and standby loads, 19:00 to 07:00:

How the worked example's 22,535 Wh daily load splits between daytime and night — the ratio that keeps the battery bank small. Source: MIMAH worked example.

Security lighting, 6 units at 15W = 90W for 12 hours = 1,080 Wh

Refrigerator, 70W average draw for 12 hours = 840 Wh

CCTV, NVR and router, 60W for 12 hours = 720 Wh

Night subtotal: 2,640 Wh

Total daily energy: 22,535 Wh, or roughly 22.5 kWh per day.

Notice how ordinary the biggest number is. Four computers running nine hours consume more energy per day than the air conditioner. That is typical, and it is the kind of finding that only ever surfaces from a real solar load calculation.

Step 2: Separate Day Loads From Night Loads

This split is where amateur sizing and engineering sizing part company, because day loads and night loads are paid for by completely different parts of the system.

Daytime loads are served directly by the array while the sun is up. Every watt-hour consumed between roughly 08:00 and 17:00 never has to enter a battery, never suffers charge and discharge losses, and never consumes battery cycle life. Nighttime loads are the opposite: generated during the day, stored, then released later, paying an energy toll of 8 to 15 percent on the round trip and consuming cycles from an expensive asset.

In our worked example, 19,895 Wh sits in daylight and only 2,640 Wh sits at night. That ratio is what makes the design affordable. Reverse those two figures and the same 22.5 kWh business needs a battery bank several times larger, roughly doubling project cost.

Which is why load shifting is a design tool, not a lifestyle inconvenience. Filling a storage tank at 11:00 instead of 20:00 costs nothing and removes 1,125 Wh from the battery bank's job. On commercial sites we regularly find 20 to 30 percent of the night load can simply move into daylight, and every watt-hour moved is battery capacity you never have to buy.

Step 3: Account for Surge Loads on Motors, Pumps and Air Conditioners

Anything with a motor or compressor draws far more current at the instant it starts than it does while running. Energy sizing ignores this because the spike lasts a second or two. Inverter sizing cannot ignore it, because that is exactly when systems trip.

Typical starting multiples, applied to running wattage:

Refrigerators and freezers: 3 to 5 times running watts

Non-inverter (fixed-speed) air conditioners: 3 to 5 times

Inverter air conditioners with soft start: 1.2 to 1.5 times

Water pumps and borehole pumps: 3 to 6 times

Workshop machinery, compressors, lathes: 3 to 7 times

Lighting, computers, TVs, routers: 1 times, no meaningful surge

The failure mode is coincidence. One motor starting is survivable; two starting within the same second while the base load is already running is what pushes an inverter past its surge rating and into shutdown.

Hala runs a pharmacy in Omdurman with a 1 hp borehole pump feeding a rooftop tank. Her installer sized a 3.5 kW inverter against a measured peak demand of 2.9 kW, which was arithmetically correct.

But the pump's 750W running load surged to roughly 2,600W on start, and it usually started while the vaccine fridge compressor was already cycling. The inverter tripped several times a week for four months before anyone connected the pattern to the pump. A soft starter costing a fraction of the inverter's price fixed it permanently.

Record the surge multiple in your audit now. You will need it in Step 6, and you will not remember it later.

Step 4: How Many Solar Panels Do I Need? Sizing the Array

Now the calculation turns outward, to the resource. Two inputs govern array size: peak sun hours and system losses.

Peak sun hours (PSH) is the number of hours per day of equivalent full-strength sunlight at 1,000 W/m². It is not daylight hours. A location with 12 hours of daylight might deliver only 5 peak sun hours, because early morning and late afternoon sun arrives at a shallow angle through more atmosphere.

Approximate annual daily averages across our regions, which you should verify for your exact coordinates using the Global Solar Atlas:

Lagos, Nigeria: roughly 4.3 to 4.8 PSH

Abuja, Nigeria: roughly 5.0 to 5.4 PSH

Cairo, Egypt: roughly 5.6 to 6.0 PSH

Khartoum, Sudan: roughly 6.0 to 6.5 PSH

Design against the worst month, not the annual average. Harmattan haze in December and January can strip 20 to 30 percent off West African output, and a system sized on the annual mean will underperform every year during precisely the months you notice it.

System losses are the second input, and they are larger than most buyers expect. A realistic derating stack for hot, dusty climates runs: heat 8 to 12 percent, soiling and dust 3 to 8 percent, wiring and mismatch 2 to 4 percent, MPPT and conversion 3 to 5 percent, inverter efficiency 3 to 5 percent. Combined, that is a total system efficiency of roughly 70 to 75 percent. We design at 0.70 for Nigerian and Sudanese conditions because dust and heat are relentless there.

The formula:

Daytime energy by appliance in the worked example — four desktops consume more per day than the air conditioner. Source: MIMAH worked example.

Array size (kWp) = Daily energy (kWh) ÷ (Peak sun hours x System efficiency)

Applied to our Ikeja print shop, using 4.5 PSH for a conservative Lagos design:

22.535 kWh ÷ (4.5 x 0.70) = 22.535 ÷ 3.15 = 7.15 kWp minimum

With 550W modules that is 13 panels. We would specify 14 x 550W = 7.7 kWp, arranged as two strings of seven, because symmetrical strings simplify MPPT configuration and the extra module buys headroom for future load growth. Check your string voltage before finalising: seven modules at roughly 49.5V open-circuit gives 346V, and applying a cold-morning correction factor of 1.1 gives 381V, comfortably inside a 500V MPPT input window.

If you want a second opinion on annual yield before signing anything, model your array in NREL's PVWatts Calculator, which uses satellite-derived weather data and lets you adjust tilt, azimuth and loss factors independently.

Step 5: Size the Battery Bank

The battery bank's job is defined narrowly: carry the night load, plus whatever daytime deficit you want covered during poor weather or grid outages. Size it against anything vaguer than that and you will overspend badly.

Battery capacity (kWh) = (Night load x Autonomy days) ÷ (Depth of discharge x Round-trip efficiency)

Two variables drive the result. Autonomy days is how long the bank must run with no meaningful generation, typically 0.5 to 1 day where a grid or generator backs the system up, and 2 to 3 days for a fully off-grid site with no fallback. Depth of discharge (DoD) is how much of the nominal capacity you may actually use: around 0.5 for tubular lead-acid, 0.8 to 0.9 for LiFePO4.

For our print shop, with a night load of 2.64 kWh, 1.5 days autonomy, LiFePO4 at 0.90 DoD and 0.92 combined round-trip and inverter efficiency:

(2.64 x 1.5) ÷ (0.90 x 0.92) = 3.96 ÷ 0.828 = 4.78 kWh minimum

That covers the night and nothing more. Because Lagos grid outages routinely hit during working hours, we would specify 10.24 kWh as two 51.2V, 100Ah LiFePO4 modules. That delivers roughly 8.3 kWh of usable energy: 2.64 kWh for the night, leaving around 5.6 kWh to ride through a cloudy afternoon without shedding the air conditioner.

The chemistry decision changes the physical bank dramatically. Delivering that same 8.3 kWh of usable energy from tubular lead-acid at 0.5 DoD and 0.85 round-trip efficiency requires roughly 19.5 kWh nominal, which is eight 12V 220Ah batteries in two parallel strings, occupying several times the floor space and needing replacement two to three times sooner. We break that trade-off down properly in our comparison of lithium versus tubular batteries for solar.

Specifying components rather than a turnkey build? Our equipment shop lists the panels, batteries and inverters we specify on our own projects, sized against the same calculations shown here.

Step 6: Size the Inverter for Continuous and Surge Load

The inverter is sized against instantaneous power in watts, not daily energy in watt-hours. This is where the load audit's wattage and surge columns finally get used.

Continuous rating must cover the largest realistic simultaneous demand, which is not the sum of every appliance in the building. Nobody runs everything at once. Add the loads that genuinely overlap, then apply a growth factor of 20 to 25 percent.

For our worked example, the realistic worst case is the AC (1,100W), four computers (800W), lighting (216W), pump (750W), printer (350W), fridge (200W), CCTV and router (60W), and occasionally the kettle (900W). That totals 4,376W. Adding 25 percent headroom gives 5.47 kW, so we specify a 6 kW hybrid inverter.

Surge rating must absorb the worst starting event. The pump's 2,600W inrush landing on top of roughly 2,000W of base load produces a momentary 4.6 kW demand. A 6 kW inverter typically surges to double its rating for a few seconds, so this passes with margin.

Check the DC to AC ratio last. A 7.7 kWp array on a 6 kW inverter gives a ratio of 1.28, which sits inside the normal 1.1 to 1.3 band. Ratios above roughly 1.35 start clipping output at midday; ratios below 1.1 leave inverter capacity idle. Our guide to choosing a solar inverter in Africa covers hybrid versus off-grid topology, MPPT count and thermal derating in more detail.

Sanity-Check the Whole Design Before You Buy

Three components sized in isolation can still fail as a system. Run these checks before anyone raises a purchase order.

Check one: does the array actually generate enough? 7.7 kWp x 4.5 PSH x 0.70 = 24.3 kWh per day against a requirement of 22.5 kWh. That is an 8 percent margin. Acceptable.

Approximate annual-average daily peak sun hours across the four cities discussed in the guide. Source: MIMAH analysis based on Global Solar Atlas (World Bank Group / Solargis).

Check two: does the array cover day loads and still charge the battery? Day generation of 24.3 kWh minus day loads of 19.9 kWh leaves 4.4 kWh surplus. Storing the 2.64 kWh night load at 0.92 efficiency requires 2.87 kWh of that surplus. It fits, with room to spare.

Check three: can the inverter carry the peak? 6 kW continuous against a 4.4 kW realistic peak and a 4.6 kW momentary surge, on an inverter rated to 12 kW surge. It holds.

If any check fails, go back a step rather than oversizing the next component. An undersized array cannot be rescued by a bigger battery, and an oversized battery on a weak array simply never reaches full charge, which shortens its life.

Common Mistakes That Wreck a Solar System Sizing Calculation

Sizing from the electricity bill alone. In Nigeria, Egypt and Sudan, bills reflect what the grid delivered, not what the business needed. Where outages are routine, as the IEA's Nigeria country profile documents, the bill systematically understates demand.

Using nameplate wattage instead of running wattage. This inflates every row in the audit and can oversize a system by 30 percent or more.

Ignoring the day and night split. It is the single fastest route to an oversized, overpriced battery bank.

Forgetting surge on motor loads. An inverter that is arithmetically correct on continuous load will still trip weekly if the surge maths was never done.

Using annual average peak sun hours. Design on the worst month or accept seasonal underperformance.

Applying optimistic loss factors. European derating assumptions of 15 percent do not survive a Sudanese June or a Harmattan January.

Sizing before fixing efficiency. Tarek runs a textile workshop in 6th of October City, Cairo. His first load audit came to 41 kWh per day, until it showed that an ageing chest freezer and two fixed-speed air conditioners accounted for 38 percent of consumption. Replacing all three cost less than a fifth of the solar capacity they were forcing him to buy, and cut the array requirement from 13 kWp to 9 kWp. Efficiency first, then generation. Always.

Frequently Asked Questions About How to Size a Solar System

How many solar panels do I need for a business? It depends entirely on daily consumption and location, but the arithmetic is fixed: divide daily kWh by peak sun hours multiplied by system efficiency to get array kWp, then divide by your module rating. A business using 22.5 kWh per day in Lagos needs around 14 panels of 550W. The same business in Khartoum, with better irradiance, needs around 10.

Can I size a solar system from my electricity bill? Only as a rough starting point, and only where supply is reliable. A bill records delivered energy, so if your grid fails for six hours a day, the bill understates your real demand by roughly that proportion. Always validate against a load audit.

Should I oversize the system for future growth? Oversize the array and inverter modestly, by 20 to 25 percent, because retrofitting roof space and re-running DC cable is disruptive. Do not oversize the battery bank, since batteries degrade with age whether or not you use the capacity.

What is the difference between kVA and kW? kW is real power; kVA is apparent power. Inverters are often marketed in kVA, and a 5 kVA unit at a 0.8 power factor delivers only 4 kW of usable continuous output. Always size against the kW figure.

Do I need batteries at all? If your loads are almost entirely daytime and you have a stable grid connection to fall back on, a grid-tied system without storage will be substantially cheaper. Batteries earn their cost where outages are frequent or where night loads are critical. Our breakdown of commercial solar installation costs in Nigeria shows how much of a typical budget storage consumes.

From Spreadsheet to Specification

Sizing a solar system is not guesswork, and it is not a product you pick from a catalogue. It is a sequence: audit your loads honestly, split them by day and night, allow for surge on every motor, size the array against local peak sun hours and realistic losses, size the battery bank against the night load and your true autonomy requirement, then size the inverter against continuous and surge power.

Do those six steps in order and sanity-check the result, and you will know exactly what you need and exactly why. You will also be able to read any quotation and tell within minutes whether the vendor did the same work.

Where professional engineering earns its fee is in the details this article can only summarise: metered load data instead of estimates, site-specific irradiance and shading analysis, cable and protection design, and a financial model that survives contact with reality.

Ready to turn your numbers into a specification you can build? Talk to MIMAH's engineers about a load audit and feasibility study. We are engineers, not resellers, and we would rather tell you the honest number before you spend the money than explain it afterwards.