Undersize an off-grid solar system and the lights go out in February. Oversize it and you have buried $20,000 of battery capacity that will never see a discharge. Both mistakes are common, and both are the result of skipping the only sizing math that matters: worst-month load divided by worst-month sun.
Off-grid solar in 2026 is a different business than it was even three years ago. LiFePO4 cell prices dropped to $60–$80/kWh at the cell level (BloombergNEF, 2026), and installed pack costs for residential storage now sit between $300/kWh and $600/kWh. Modules are 20–24% efficient at the panel level (Fraunhofer ISE, 2026). The components are no longer the limit. The limit is whether your design accounts for December irradiance, surge loads, temperature derating, and a few hundred small details that turn a kit list into a working system.
This guide walks through the calculator method used by professional installers — five steps with explicit formulas, three worked examples covering cabin, mid-size home, and whole-home builds, and an honest cost table for 2026. It then shows where a spreadsheet stops being enough and a design tool takes over.
TL;DR — Off-Grid Solar Sizing
An off-grid solar system is sized in five steps: daily kWh load, worst-month peak sun hours, array kW with a 0.75 derate, battery kWh at 2–3 days autonomy and 80% DoD, and an inverter sized for surge load. A typical US off-grid family home needs 10–15 kW of array and 50–80 kWh of LiFePO4 storage (NREL PVWatts; BloombergNEF, 2026).
In this guide:
- The 5-step off-grid sizing calculator with formulas
- Daily load worksheet and surge factors
- Array sizing for the worst month
- Battery bank sizing: LiFePO4 vs lead-acid in 2026
- Inverter and charge-controller sizing
- Three worked examples: cabin, mid-size home, whole-home
- 2026 cost ranges and component picks
- When to move off the spreadsheet
What Is an Off-Grid Solar System (And When It Makes Sense)
An off-grid solar system is a fully independent power plant: a photovoltaic array, a battery bank, a charge controller, and an inverter, with no electrical connection to the utility grid. It generates, stores, and delivers all the energy a site uses, every day of the year. The 2026 World Bank estimate is that 666 million people still live without grid electricity, which is why off-grid systems remain the dominant rural electrification tool in much of Africa and South Asia (World Bank, Tracking SDG 7, 2025).
The architecture is simple on paper. The PV array generates DC during daylight. A charge controller regulates that DC into the battery bank. The inverter converts battery DC into AC for household loads. When the sun is up, the array runs the loads directly and tops up the batteries. When the sun is down or a cloud sits over the site, the batteries carry the load until the next sunny day.
Off-grid is the right answer in four situations. The first is remote sites where extending utility lines would cost more than $20,000 per kilometre. The second is land that has no grid presence at all — hunting cabins, off-trail homesteads, mountain research stations. The third is mobile applications: RVs, boats, food trucks, expedition vehicles. The fourth is resilience-first homeowners who want to survive a multi-day grid outage without compromise.
It is the wrong answer when grid power is already at the property line. A grid-tied system with battery backup, sometimes called a hybrid system, costs 30–40% less than an equivalent off-grid build because the grid acts as a free, infinite-capacity battery during sunny periods. The choice between off-grid and hybrid usually comes down to grid-extension cost versus storage cost over a 20-year horizon.
The hardest part of off-grid is not the technology. It is the discipline. You can no longer pull from the grid when your washing machine, electric kettle, and well pump all start at the same time on a cloudy December morning. The sizing math has to anticipate that moment.
The 5-Step Off-Grid Solar System Sizing Calculator
The off-grid sizing calculator is a sequence of five formulas. Each one feeds the next. Skip a step or use the wrong input and the whole system is misaligned. The five steps produce, in order: daily energy demand, array size, battery bank size, inverter size, and charge controller size.
The five formulas, in compact form:
- Daily load (Wh) = Σ (running watts × hours per day × days used per week ÷ 7)
- Array (W) = Daily load (Wh) ÷ (Worst-month PSH × derate factor of 0.75)
- Battery (kWh) = (Daily kWh × Days of autonomy) ÷ (Depth of discharge × round-trip efficiency)
- Inverter (W) = max(continuous load, surge load ÷ surge factor)
- Charge controller (A) = (Array Wp × 1.25) ÷ Battery bank voltage
The 0.75 system derate factor in step 2 is the product of typical losses: temperature derate (~10%), soiling (3%), wiring (2%), inverter efficiency (3%), battery round-trip efficiency (10%), and MPPT charge controller efficiency (3%). Multiply 0.90 × 0.97 × 0.98 × 0.97 × 0.90 × 0.97 and you get 0.745. NREL PVWatts uses a similar 14% total system loss assumption by default, which equates to a 0.86 derate before battery losses (NREL, PVWatts v8 documentation).
The depth of discharge in step 3 depends on chemistry. LiFePO4 supports 80–90% DoD without lifetime penalty. Flooded lead-acid is limited to 50% if you want the bank to last more than four years. Round-trip efficiency is 95% for LFP and roughly 80% for FLA (Sandia National Laboratories, energy storage performance reports).
The surge factor in step 4 is the inverter’s ratio of peak surge wattage to continuous wattage. A typical 6 kW pure sine-wave inverter handles a 12 kW surge for 5 seconds, giving a surge factor of 2. Well pumps, air conditioners, and refrigerators can pull 4–8 times their running watts at startup. If a well pump runs at 1,500 W continuous and surges to 6,000 W, the inverter must handle that surge or the pump will fail to start.
Step 5 sizes the charge controller using the array’s short-circuit current with a 1.25 NEC safety factor (NEC 690.8). For a 5,000 W array on a 48 V battery bank, the controller current is 5,000 ÷ 48 × 1.25 = 130 A, which means a 150 A MPPT controller. IEC 62124 covers the same ground for international installs (IEC, Stand-Alone PV Systems standard).
Step 1 — Build the Daily Load Worksheet
The daily load worksheet is a list of every electrical device on the site, with running wattage, hours of use per day, and days of use per week. Multiply, sum, and divide. The output is one number in watt-hours per day. That number drives every other sizing decision in the system, which is why this step is worth an extra hour of care.
Walk the property with a clamp meter or a Kill A Watt monitor. Manufacturer plate ratings are a starting point, but real loads often differ by 20–30%. A US-spec refrigerator labeled at 150 W might draw 90 W steady with 600 W compressor surges. A modern LED bulb labeled at 9 W draws 8.4 W. A cheap modified-sine inverter wastes 15% as heat. Measure where you can.
Record three numbers per device: running watts, hours per day, and days per week. The simplest way to track this is a spreadsheet with one row per device and a total at the bottom. The example table below covers a typical mid-size off-grid home.
| Device | Running W | Hours/day | Days/week | Wh/day |
|---|---|---|---|---|
| LED lighting (whole house) | 60 | 6 | 7 | 360 |
| Refrigerator (compressor avg) | 80 | 24 | 7 | 1,920 |
| Chest freezer | 60 | 24 | 7 | 1,440 |
| Laptop + monitor | 90 | 8 | 7 | 720 |
| Wi-Fi router + ONT | 15 | 24 | 7 | 360 |
| Induction hob (1 burner) | 1,800 | 0.5 | 7 | 900 |
| Microwave | 1,000 | 0.2 | 7 | 200 |
| Washing machine (cold cycle) | 500 | 1 | 3 | 214 |
| Well pump (½ HP) | 750 | 0.6 | 7 | 450 |
| Heat pump (mild climate) | 1,200 | 4 | 7 | 4,800 |
| Phantom loads (TV standby etc.) | 30 | 24 | 7 | 720 |
| Total daily load | ~12,100 Wh |
Phantom loads are easy to ignore and add up to 5–10% of total consumption in most homes. A 30 W standby draw across all devices runs 24 hours a day, which is 720 Wh. Over a year that is 263 kWh — enough to charge an EV 1,000 miles. Use a switched power strip to kill them at night.
Surge loads need a separate column. Well pumps, air conditioners, large refrigerators, and induction cooktops can pull 3–8 times their running watts during startup. The inverter has to handle that peak even if it only lasts 100 milliseconds. Note the surge wattage of the worst three devices on the worksheet — they set the inverter size in step 4.
Seasonal variation is the silent killer of off-grid sizing. In a cabin used only in summer, you can size to summer loads and ignore winter. In a year-round home, December loads typically exceed July loads by 30–50% because of heating, lighting hours, and reduced solar gain through windows. Build the worksheet for December, not for the annual average. That single discipline avoids about 80% of the “my system runs out in winter” complaints on solar forums.
Step 2 — Find Your Worst-Month Peak Sun Hours
Peak sun hours, or PSH, is the daily solar resource at a site expressed as the equivalent number of hours of full 1,000 W/m² irradiance. A site with 4.5 PSH receives the same total daily energy as if the sun shone at full strength for 4.5 hours. For US sites, NREL’s PVWatts v8 calculator is the authoritative source. For European sites, the European Commission’s PVGIS tool serves the same role.
Worst-month PSH is what matters for off-grid. The annual average is a marketing number. December PSH in northern cities can be a third of June PSH. A system sized to the annual average will run out of energy by mid-November and stay out until early February. Always size to the worst-month value.
The table below shows December PSH for ten reference cities, drawn from NREL PVWatts and PVGIS at 30° tilt facing equator-side. The annual averages are included for contrast.
| Location | December PSH | June PSH | Annual avg PSH |
|---|---|---|---|
| Phoenix, AZ | 5.0 | 7.6 | 6.5 |
| Los Angeles, CA | 4.0 | 6.7 | 5.8 |
| Atlanta, GA | 3.2 | 5.7 | 4.8 |
| Denver, CO | 3.8 | 7.0 | 5.5 |
| New York, NY | 2.5 | 5.5 | 4.5 |
| Seattle, WA | 1.5 | 5.8 | 3.8 |
| London, UK | 1.0 | 5.0 | 3.2 |
| Berlin, Germany | 0.8 | 5.5 | 3.5 |
| Mumbai, India | 5.5 | 5.0 | 5.4 |
| Sydney, Australia | 6.0 | 3.4 | 5.0 |
The implications are sharp. A house in Berlin needs roughly six times more array per kWh of December load than the same house in Phoenix. A site in tropical Mumbai actually has its worst month in the monsoon, not December, so always check the local low rather than assuming Dec/Jan.
Tilt angle matters more for off-grid than for grid-tied. A grid-tied roof at 20° tilt loses 5% annual yield versus optimal tilt, which is fine because you net-meter the surplus. An off-grid array at the same suboptimal tilt loses 20–30% of December production, which can collapse the whole sizing budget. For a fixed array, set tilt equal to latitude plus 15° to bias for winter sun. For a winter-priority cabin, latitude plus 25° is even better.
Microclimate effects can shave 10–20% off generic PSH numbers. Coastal fog, mountain shadows, lake-effect cloud, and wildfire smoke are all real losses that PVWatts cannot model. Use the solar shadow analysis software inside SurgePV to layer 3D horizon obstructions onto the irradiance model and confirm your worst-month PSH assumption is realistic at the actual installation site, not just at the regional weather station 30 miles away.
Step 3 — Size the Solar Array
Array sizing converts daily energy demand into installed PV wattage. The formula is one line: daily load in watt-hours divided by worst-month PSH multiplied by a system derate factor, where the derate is typically 0.70–0.80. For a 12,000 Wh/day load and 3.5 PSH, an array at 0.75 derate equals 12,000 ÷ (3.5 × 0.75) = 4,571 W, or roughly 4.6 kW.
The derate factor is not a single number. It is the cumulative product of every loss between the panel face and the AC output side of the inverter. Sandia National Laboratories breaks losses into a stack: temperature (~8–12% in summer for a roof-mounted array), soiling (3–5% on a flat tilt), wiring (1–2%), MPPT efficiency (97%), inverter efficiency (95–97%), and battery round-trip efficiency (95% for LFP, 80% for FLA). The product is your honest derate (Sandia, PV Performance Modeling Collaborative).
Temperature derate is the largest single factor. Crystalline silicon modules lose roughly 0.35% of output per °C above 25°C cell temperature (Fraunhofer ISE module data). On a hot roof in Texas, cell temperatures hit 60–65°C, which means a 15% performance loss versus the STC rating on the data sheet. In December the loss is closer to 5%. Use 10% as an annual average.
Once you have the kW target, the panel count math is simple. Modern modules are 400–550 W. For a 4.6 kW array at 450 W per panel, you need 11 panels. Round up to 12 to keep string voltages clean. SurgePV’s solar design tool handles the layout and string sizing in 2D and 3D — pull a satellite image of the property, drop the panel count, and the tool checks string voltages against the inverter’s MPPT window.
A few sizing rules earned the hard way. First, never size the array at exactly the minimum required by the math. Add 15–20% headroom for module degradation (LFP modules degrade about 0.5% per year), unexpected load growth, and bad-weather weeks. Second, round up to a panel count that gives you clean string voltages — a 24-panel array in two strings of 12 is easier to wire than a 23-panel oddball. Third, if the budget is tight, spend it on more panels rather than more battery: an extra panel is $90–$120 and produces energy for 25 years, an extra kWh of battery is $400–$700 and lasts 15.
The final sanity check is the array-to-battery ratio. For a healthy off-grid system, the array should produce at least 1.2–1.5× the daily load on a worst-month day. If the math gives you a ratio below 1.2, you have an undersized array and the battery will deplete on the first cloudy day. If the ratio is above 2.0, you are wasting capex unless you have specific reasons for the buffer (heavy winter loads, year-round cloudy site, or planned EV charging).
Step 4 — Size the Battery Bank (LiFePO4 vs Lead-Acid in 2026)
Battery bank sizing is the most cost-sensitive step in the whole calculator. The formula is daily kWh load multiplied by days of autonomy, divided by usable depth of discharge multiplied by round-trip efficiency. For a 12 kWh/day load, 2 days of autonomy, 80% DoD, and 95% RTE on LFP, the bank needs (12 × 2) ÷ (0.80 × 0.95) = 31.6 kWh of nameplate capacity.
Days of autonomy is the number of days the bank can carry the loads with zero solar input. A sunny climate like Phoenix or Madrid runs fine on 1.5 days. A temperate climate like Seattle or Berlin needs 2.5–3 days because of multi-day overcast stretches. A cabin used only on weekends can use 1 day because you simply do not show up during a cloudy week. The tradeoff is direct: each extra day of autonomy adds 30–50% to battery cost.
The chemistry choice in 2026 is essentially settled. LiFePO4, also called LFP, dominates new off-grid installations. The reasons are quantitative.
| Spec | LiFePO4 (LFP) | Flooded lead-acid (FLA) |
|---|---|---|
| Usable depth of discharge | 80–90% | 50% |
| Round-trip efficiency | 95% | 80% |
| Cycle life at rated DoD | 6,000+ cycles | 800–1,500 cycles |
| Calendar life | 15–20 years | 5–8 years |
| Maintenance | None | Monthly water top-up |
| Temperature performance | -20°C to +60°C | derates below 0°C |
| Installed cost ($/kWh, 2026) | $300–$600 | $200–$350 |
| 20-year cost per usable kWh-cycle | $0.04–$0.08 | $0.10–$0.18 |
Lead-acid wins on upfront sticker price by 30–50%. LFP wins on every other axis. Over a 20-year off-grid horizon, you replace lead-acid two or three times while a single LFP bank is still under warranty. The 20-year cost-per-kWh-cycle math is the honest comparison, and on that basis LFP is roughly half the cost of lead-acid (BloombergNEF Battery Price Survey, 2026).
Voltage matters at the bank level. A small system under 2 kW can use 12 V or 24 V. A medium system from 2–6 kW should be 24 V or 48 V. Anything over 6 kW belongs at 48 V because the wiring current at lower voltages becomes uneconomic. A 6 kW load at 12 V is 500 A, which requires 4/0 AWG copper at $40 per foot. The same load at 48 V is 125 A, manageable on 1/0 AWG.
The NEC 706 framework governs how off-grid battery banks are installed in the United States, including SCCR labelling, rapid-shutdown integration with PV (NEC 690.12), and clearance from combustibles. NFPA 855 caps indoor residential energy storage at roughly 20 kWh per room with stricter limits for unrated batteries. International installs follow IEC 62124 for stand-alone PV system design. If your sized bank exceeds 20 kWh and lives indoors, plan for a separate battery room or an outdoor enclosure with the appropriate UL 9540 listing.
Step 5 — Size the Inverter and Charge Controller
The inverter and the charge controller are the two power-electronics components that connect the array, the batteries, and the household loads. Both have their own sizing math, and getting either one wrong wastes money or shortens system life.
Inverter sizing has two parts: continuous load and surge load. The continuous rating must equal or exceed the largest sum of running watts that can plausibly run at the same time. Walk through a worst-case morning: well pump on, kettle on, microwave on, fridge cycling. Add the running watts. That is your continuous floor. For a typical mid-size off-grid home, the floor is usually 5–6 kW. For a whole-home system with electric range and central HVAC, plan for 8–12 kW. Pure sine-wave output is mandatory; modified sine wave damages variable-speed motors, switching power supplies, and any equipment with an electronic ballast.
Surge sizing handles the half-second spikes when motors start. Refrigerators, well pumps, air conditioners, and large power tools can pull 3–8 times their running watts at the moment of compressor or motor inrush. A 6 kW continuous inverter that handles a 12 kW surge for five seconds has a surge factor of 2. If your worst surge load is a 1,500 W well pump that spikes to 6,000 W, the inverter must handle a 6 kW surge while still serving the rest of the house. Add the surge to the rest of the active load and check the spec sheet.
Hybrid inverter-chargers (often called all-in-ones) combine the inverter, charge controller, and AC charger into a single unit. The 2026 market leaders include Sol-Ark 12K, EG4 18kPV, Schneider XW Pro 6848, and Victron MultiPlus II 5000. They support split-phase 120/240 V output, AC coupling for grid-connected solar add-ons, and generator integration with auto-start. For most off-grid builds these are the right choice. UL 1741 SB certifies their grid-support and off-grid-capable behaviour.
Charge controller sizing uses the array short-circuit current, the battery voltage, and the NEC 690.8 1.25 safety factor. The formula is (Array Wp ÷ Battery V) × 1.25. A 5,000 W array on a 48 V battery bank needs (5000 ÷ 48) × 1.25 = 130 A, so a 150 A MPPT controller. MPPT controllers are universal in 2026 — they harvest 15–25% more energy than older PWM controllers because they decouple panel voltage from battery voltage.
The choice between AC-coupled and DC-coupled architectures matters at scale. DC-coupled is simpler, more efficient (no double conversion), and the right answer for systems under 10 kW. AC-coupled is the right answer when retrofitting an existing grid-tied system to off-grid, or when scaling above 15 kW with multiple inverters. Sol-Ark and Schneider both support both topologies in a single unit. Stack-ability matters too; the EG4 18kPV stacks up to six units in parallel for 108 kW of capacity, which covers any residential application and most light-commercial off-grid builds.
Stop sizing off-grid systems in spreadsheets
SurgePV runs the array, battery, inverter, and shading math in one workflow — and outputs a client-ready proposal in minutes.
Book a DemoNo commitment required · 20 minutes · Live project walkthrough
Worked Example 1 — 1.2 kW Off-Grid Cabin / RV
A weekend cabin in the Colorado mountains, used Friday-Sunday from May to October and a few winter weekends. The owner wants lights, a fridge, water pump, laptop charging, and a small TV. No heating, no AC, propane for cooking. The owner is on site 60 days per year and absent the rest of the time.
Daily load worksheet for an active weekend day:
| Device | W | Hours/day | Wh/day |
|---|---|---|---|
| LED lighting | 40 | 5 | 200 |
| 12 V DC fridge (compressor avg) | 45 | 24 | 1,080 |
| Water pump (intermittent) | 100 | 0.5 | 50 |
| Laptop charging | 65 | 4 | 260 |
| 32” TV + streaming stick | 60 | 3 | 180 |
| Phone/device charging | 30 | 2 | 60 |
| Phantom loads | 5 | 24 | 120 |
| Total | ~1,950 Wh |
Sizing math, using the five formulas:
- Daily load: 1,950 Wh
- Array: 1,950 ÷ (4.2 PSH × 0.75) = 619 W. Round up to 800 W (two 400 W panels).
- Battery: (1.95 × 2 days) ÷ (0.80 × 0.95) = 5.1 kWh. Use a 5 kWh LiFePO4 server-rack bank.
- Inverter: continuous load peak ≈ 250 W; surge from water pump ≈ 400 W. A 1.5 kW pure sine-wave inverter is overkill in capacity but the right minimum for surge headroom. A 2 kW inverter is the comfortable choice.
- Charge controller: 800 ÷ 24 V × 1.25 = 42 A. Use a 50 A MPPT controller.
Voltage: 24 V battery bank. The system is small enough that 12 V is tempting, but 24 V keeps charge controller current down and supports a 2 kW inverter cleanly.
Bill of materials, 2026 prices:
| Component | Price |
|---|---|
| 2 × 400 W bifacial panels | $360 |
| 5 kWh LiFePO4 24 V bank | $1,800 |
| 50 A MPPT charge controller | $200 |
| 2 kW pure sine-wave inverter | $450 |
| Mounts, cabling, breakers, fuses | $600 |
| Subtotal materials | $3,410 |
| Self-install labour (DIY) | $0 |
| DIY total | $3,410–$4,500 |
| Professional install total | $8,000–$12,000 |
The DIY range covers people who do their own roof and electrical work and skip the permit cost. The professional install range covers a licensed solar contractor handling design, permit, install, and inspection. NEC 690 still applies to off-grid cabins in the US even when there is no utility connection, and most counties enforce permitting for habitable structures.
Worked Example 2 — 6 kWh/day Mid-Size Off-Grid Home
A two-bedroom house in the Texas Hill Country occupied year-round by a couple with home offices. Propane water heating, propane stove, mini-split heat pump for shoulder-season comfort. The site is 14 km from the nearest grid line — utility extension was quoted at $42,000 plus monthly service charges, which made off-grid the cheaper option.
Daily load worksheet:
| Device | W | Hours/day | Wh/day |
|---|---|---|---|
| LED lighting | 80 | 5 | 400 |
| Energy Star fridge | 80 | 24 | 1,920 |
| Chest freezer | 50 | 24 | 1,200 |
| Mini-split heat pump (mild use) | 700 | 2 | 1,400 |
| Two laptops + monitors | 180 | 8 | 1,440 |
| Wi-Fi router + cell booster | 25 | 24 | 600 |
| Microwave + small appliances | 800 | 0.4 | 320 |
| Washing machine | 400 | 0.5 | 200 |
| Well pump | 750 | 0.4 | 300 |
| Phantom loads | 30 | 24 | 720 |
| Total | ~8,500 Wh |
The owners actually use 8.5 kWh/day, which we round to 9 kWh for design margin. December PSH at the site is 4.5 hours.
Sizing math:
- Daily load: 9,000 Wh
- Array: 9,000 ÷ (4.5 × 0.75) = 2,667 W. Round to 3.6 kW (eight 450 W panels) for headroom and a clean two-string layout.
- Battery: (9 × 2.5 days) ÷ (0.80 × 0.95) = 29.6 kWh. Use a 30 kWh LiFePO4 bank.
- Inverter: continuous floor 1,500 W, surge 6,000 W from the well pump. A 6 kW hybrid inverter at 12 kW surge handles both with margin.
- Charge controller: integrated into the 6 kW hybrid (most have an internal 100–150 A MPPT).
Voltage: 48 V battery bank.
Bill of materials, 2026 prices:
| Component | Price |
|---|---|
| 8 × 450 W modules | $1,800 |
| 30 kWh LFP bank (e.g., 6 × EG4 LL-S 5.12 kWh) | $11,400 |
| 6 kW hybrid inverter (Sol-Ark 8K or EG4 6500EX) | $3,200 |
| Roof or ground-mount racking | $1,500 |
| Disconnects, breakers, AC/DC wiring, monitoring | $1,800 |
| Permitting + interconnection prep | $1,200 |
| Install labour (3 days × 2 electricians) | $4,800 |
| Materials subtotal | $19,700 |
| Total installed (2026) | $24,500–$32,000 |
The cost lands at the low end of the $35,000–$65,000 range cited in industry reports for mid-size off-grid homes (EnergySage, Wood Mackenzie, Q1 2026), helped by the homeowner doing some site prep and electrical rough-in. A turnkey contractor build at the same spec lands at $40,000–$48,000 in most US markets.
Worked Example 3 — 25 kWh/day Whole-Home Off-Grid
A four-person family home in coastal Northern California, four bedrooms, one home office, electric heat pump for HVAC, electric range, electric water heater, EV charging two days per week. The site is in a CPUC-area with chronic PSPS shutoffs, which pushed the owners to full off-grid as the long-term resilience play.
Daily load summary (full worksheet omitted for length): roughly 25 kWh on a typical day. In December, with electric heat running and EV charging, the load can spike to 35 kWh. The design target is 25 kWh average with surge capacity for the worst weeks.
Sizing math:
- Daily load: 25,000 Wh design / 35,000 Wh worst case
- Array: 25,000 ÷ (3.5 × 0.75) = 9,524 W on the design load. Size up to 14 kW to cover worst-week loads and EV charging without dipping into reserve.
- Battery: (25 × 2 days) ÷ (0.80 × 0.95) = 65.8 kWh. Use a 60 kWh LFP bank with a generator backup for the rare 3-day overcast event.
- Inverter: 12 kW split-phase to handle electric range plus heat pump plus EV charger plus baseline house. Surge to 24 kW.
- Charge controller: integrated into the inverter; 150 A MPPT × 2 strings.
Voltage: 48 V battery bank with paralleled 12 kW inverters.
Bill of materials, 2026 prices:
| Component | Price |
|---|---|
| 30 × 470 W modules (14.1 kW) | $7,500 |
| 60 kWh LFP bank | $24,000 |
| 2 × 12 kW Sol-Ark or EG4 split-phase inverter | $11,000 |
| Ground-mount racking + foundations | $4,500 |
| Disconnects, breakers, generator transfer switch | $3,500 |
| Backup generator (LP, 14 kW) + auto-start kit | $7,000 |
| Permits, interconnection, engineering stamp | $4,000 |
| Install labour (8 days × 3 crew) | $14,000 |
| Materials subtotal | $75,500 |
| Total installed (2026) | $78,000–$95,000 |
The backup generator runs about 50–80 hours per year (under 1% annual runtime) on long overcast stretches when the bank drops below 20% state of charge. Generator runtime is the off-grid version of grid import — used sparingly, sized for emergencies, and integrated through the inverter’s auto-start logic. Without the generator, you would need to add another 20–30 kWh of battery at $8,000–$12,000, which is more expensive than the $7,000 generator and a year of propane.
The system passes NEC 690 and 706 inspection, qualifies for a state-level resilience rebate in California, and pays back versus the alternative (utility extension plus PSPS-driven backup generator runtime) in 8–11 years depending on PG&E rate trajectory. Use the generation and financial tool to model the IRR and payback under the homeowner’s specific tariff.
Off-Grid Solar System Costs in 2026
Off-grid system costs in 2026 are driven by storage, not by panels. Module prices are at $0.20–$0.25/W wholesale and $0.40–$0.50/W installed. Storage installed costs are $300–$600/kWh. For a typical 12 kWh/day off-grid home, the battery line item is 50–60% of the total bill of materials. The biggest sizing decision in your budget is days of autonomy, not panel count.
The table below summarises 2026 installed-cost ranges across the three archetypes used in this guide. Numbers are pulled from EnergySage’s 2026 marketplace data, Wood Mackenzie Solar + Storage Q1 2026, and BloombergNEF’s Battery Price Survey 2026.
| Archetype | Array | Battery | Inverter | BOS + labour | Total installed |
|---|---|---|---|---|---|
| Cabin / RV (1–2 kW + 5–10 kWh) | $400–$1,200 | $1,800–$4,500 | $300–$700 | $1,500–$3,500 | $8,000–$18,000 |
| Mid-size home (4–8 kW + 25–40 kWh) | $1,800–$4,000 | $9,000–$22,000 | $2,500–$5,000 | $7,000–$15,000 | $30,000–$55,000 |
| Whole-home (10–15 kW + 50–80 kWh) | $5,000–$8,500 | $20,000–$45,000 | $7,000–$15,000 | $15,000–$35,000 | $60,000–$100,000+ |
Two notes on incentives. First, the US federal §25D residential ITC for solar and storage expired on 31 December 2025 and is not available for new residential installs in 2026. Some state programs continue to offer rebates and tax credits — California’s SGIP for storage, New York’s NY-Sun, Colorado’s resilience grants, and a handful of utility-specific programs. Second, the §48E commercial ITC remains available for off-grid systems serving commercial loads (ranches, farms, off-grid offices) at the prevailing tax rate — talk to a CPA before claiming.
Two notes on cost trajectories. First, BloombergNEF projects battery pack prices will fall to roughly $80/kWh at the manufacturing level by end of 2026, which should pull installed prices toward the lower end of the ranges above by Q4. Second, modules show no further significant price decline expected — the bottom of the curve is essentially here, and any future cost reduction comes from labour, soft costs, and BOS rather than the panels themselves (Wood Mackenzie, 2026).
The single biggest lever a homeowner controls is daily load. Cutting load by 30% (LED lighting, Energy Star appliances, propane water heating) reduces the entire system cost by 25–30%. The second lever is days of autonomy: dropping from 3 days to 2 with a backup generator saves $8,000–$15,000 on a typical install. The third lever is DIY versus professional — a careful DIY build saves 30–40% of total cost but requires permit, code, and inspection competence the homeowner has to demonstrate.
Common Off-Grid Sizing Mistakes (and How to Avoid Them)
Most off-grid sizing failures come from one of seven mistakes. Each one is preventable, and each one costs real money — either upfront in oversized capex, or downstream in winter blackouts and warranty arguments.
The first is sizing to the annual average instead of the worst month. PVWatts’ annual average for Berlin is 3.5 PSH. December is 0.8. A system sized to 3.5 PSH will run out of battery by mid-November and stay out until late February. Always size to December (or to the local low if you are in a monsoon climate).
The second is ignoring surge loads. A well pump rated 1,500 W running can spike to 6,000 W at startup. An induction range can spike to 3,600 W on a single burner. A 4 kW continuous inverter that surges to 8 kW will not start that pump and the homeowner will find out in the worst possible way. Always check the surge spec, not just the continuous spec.
The third is overdischarging the battery bank. Lead-acid below 50% SOC degrades dramatically. LFP below 10% SOC stresses the BMS. Set the inverter low-battery cutoff to 20% SOC for LFP and 50% SOC for FLA, no exceptions. Add a second cutoff at 30% SOC that disables non-essential loads (water heater, EV charger, dryer) before the main cutoff fires.
The fourth is skipping the temperature derate. A roof-mounted array in Phoenix at 60°C cell temperature delivers 12% less than the STC rating on the panel data sheet. A bank cabinet at 35°C ambient pushes LFP cycle life down 20–30%. Plan for ventilation in the battery enclosure and use bifacial modules with backsheet ventilation if you can get them.
The fifth is the wrong inverter topology. Modified sine wave inverters cost less but kill variable-speed motors, electronic ballasts, induction cooktops, and audio gear. Pure sine wave is the only choice for a residential off-grid system. Spend the extra $200–$500 — it is cheaper than replacing a heat pump.
The sixth is no autonomy buffer. A bank sized to exactly one day of usage will go to zero on the first cloudy day. Two days is the floor for a year-round occupied home. Three days is the comfortable choice in temperate climates with multi-day overcast stretches. Anything less is a generator-dependent system, which is fine if you accept the runtime cost.
The seventh is no future-load growth. EVs, heat pumps, induction ranges, and heat-pump water heaters are all becoming standard. A bank sized to today’s loads is undersized in three years. Add 20–30% headroom on both array and battery, or leave room in the disconnect panel and inverter to add capacity later. SurgePV’s solar design tool lets you model a future-state system on day one and sequence the install in two phases.
When to Move From a Spreadsheet to Design Software
A spreadsheet is a fine tool for the first pass. You can run the five-step calculator on the back of an envelope, validate the order of magnitude, and produce a rough budget for a homeowner. The trouble starts when the project becomes real and you need a buildable design with site-specific shading, NEC-compliant electrical drawings, a production forecast for client approval, and a proposal that wins the job.
Design software collapses the gap between a sized system and a buildable one. The four steps that matter for installer-grade off-grid work are array layout, shading analysis, production forecast, and proposal generation. Doing them in four different tools means errors compound — the array you laid out in tool A may not fit the panel count tool B used for shading, and the production number tool C generated does not match the system tool D documented in the proposal.
SurgePV runs all four steps in one workflow. Use the solar design software to pull a satellite image and place the sized array. Run shadow analysis on the actual site to confirm the December PSH assumption from PVWatts. Push the design through the generation and financial tool to produce a worst-month production forecast and a 25-year cash-flow model. Generate the solar proposal software output as a branded PDF for the client. Every number flows through automatically — change the panel count and the production, the financials, and the proposal all update.
The hard case for software shows up in three project types. The first is full off-grid family homes where the budget is large enough that a sizing error costs $20,000+ and a shading error costs another $10,000 in production. The second is C&I off-grid (ranches, dairy farms, telecom sites) where the proposal needs an engineering stamp and the documentation has to satisfy an AHJ inspection. The third is multi-system installer businesses doing 30+ off-grid jobs per year — at that volume, the spreadsheet workflow caps out at maybe 10 jobs per crew per year, and revenue scales linearly with how fast you can move from site visit to signed contract.
The decision rule is simple. For one-off DIY cabins under $15,000, a spreadsheet is enough. For installer work over $30,000, software pays for itself on the first job by avoiding a single re-engineering cycle.
Conclusion
- Run the five-step calculator before you order a single component. Daily load, worst-month PSH, 0.75 derate, 2–3 days autonomy at 80% DoD, and an inverter sized for surge. Get those five numbers right and the rest is a parts list.
- Build for December, not the annual average. Worst-month sizing is the difference between a system that runs every day of the year and one that fails in the second week of November.
- For any installer build over $30,000, move off the spreadsheet and onto design software that handles array layout, shading, production, and proposals in one workflow. SurgePV runs the full off-grid workflow in a single platform — book a demo and walk through a live project sizing.
Frequently Asked Questions
How many solar panels do I need to run a house off grid?
A typical 25 kWh/day off-grid home needs roughly 14 kW of array. At 450 W per module, that is 31 panels. The exact count depends on your worst-month peak sun hours, system derate (~0.75), and how many days of battery autonomy you build in. Always size to December irradiance, not the annual average. For a 6 kWh/day mid-size home in a sunny climate, the count drops to 10–12 panels.
What size battery bank do I need for an off grid solar system?
Multiply your daily kWh load by your days of autonomy, then divide by usable depth of discharge multiplied by round-trip efficiency. For 25 kWh/day, 2 days of autonomy, 80% DoD on LiFePO4, and 95% RTE, you need 65.8 kWh of installed nameplate capacity. Add 10–15% extra for cold-weather derate and future load growth. Most full off-grid homes land between 30 kWh and 60 kWh; cabins land at 5–15 kWh.
How much does an off grid solar system cost in 2026?
A small 1.5 kW cabin system runs $8,000–$18,000 installed. A 5–8 kW mid-size off-grid home with 25–40 kWh of LiFePO4 storage runs $30,000–$55,000. A full off-grid family home with 10–15 kW of array and 50–80 kWh of storage runs $60,000–$100,000 or more. Battery capacity drives the cost, not the panels — storage is 50–60% of the typical bill of materials in 2026 (EnergySage; Wood Mackenzie, Q1 2026).
Can an off grid solar system power a whole house?
Yes. The math is the same as any off-grid build, just at a larger scale. A typical four-person US home consumes 25–30 kWh/day. That requires 10–15 kW of array, 50–80 kWh of LiFePO4 storage, and an 8–12 kW pure sine-wave inverter. Plan for a backup propane or diesel generator at under 5% annual runtime to cover two-week cloudy stretches without burying $20,000 in extra battery capacity.
How long do off grid solar batteries last?
LiFePO4 batteries typically last 15–20 years in off-grid duty, rated for 6,000+ cycles at 80% depth of discharge (BloombergNEF, 2026). Flooded lead-acid lasts 5–8 years at 50% DoD. The replacement cost over a 20-year period makes LiFePO4 cheaper on a $/kWh-cycle basis even though the upfront sticker is higher — typically $0.04–$0.08 per usable kWh-cycle for LFP versus $0.10–$0.18 for FLA. Modern LFP batteries from major manufacturers carry 10-year warranties with 70–80% capacity retention guarantees.



