This tool provides educational sizing estimates only. Improper
electrical work can cause fire, injury, or death. Always consult a
qualified electrician before installing any electrical system. You are
responsible for your own safety, the safety of others, and the
protection of property and the environment.
Off-Grid Electric Power Calculator
Size your solar, wind, and battery system with real-world data
Educational tool only, not a substitute for professional
electrical design
About This Tool
A browser-based calculator for sizing off-grid solar, wind, or hybrid
systems paired with a battery bank. Enter or look up a location, then
add your panel array, electrical loads, and battery bank to get
month-by-month adequacy results. It's built for RV owners, cabin
builders, off-grid homesteaders, sailboat crews, and van-lifers. It
pulls real historical weather data from JRC's PVGIS, NREL PVWatts,
NASA POWER, and Open-Meteo to calculate energy generation, depth of
discharge, recharge fraction, and battery duration for every month of
the year; you can also simulate your own forecast for an upcoming trip.
Easy mode includes most features in a streamlined view.
Tap + Add detail on any card to reveal more options without
leaving Easy mode. Switch to Hard to add sun-tracking
mount options (1-axis / 2-axis) on top of everything in Easy. Switch to
Advanced for all features, including system analysis with
historical hourly weather data.
New here? Easy mode is on, enter an address,
panel size, and battery capacity to get started.
for more options.
Try a worked example
Each card loads a complete, working design. Change any field after you load it; the originals aren't saved unless you tap Save Current.
Caribbean sailboat
Cruising catamaran at anchor: fridge, electronics, LED lighting. Naturally clear tropical Kt; modest panel area on a coachroof.
Colorado cabin
Weekend cabin in the foothills above Golden, CO: Mountain West sun, cold and snowy winters. Lithium bank sized for cloudy-week reserve.
Phoenix van life
Reliably clear desert sky most of the year, but a small van-roof area caps the panel array. Vent fan and a brief induction-burner pulse on a 100 Ah LiFePO4 bank.
Vermont homestead
Year-round off-grid house in northern New England; wind picks up where winter solar drops off. Hybrid solar+wind, 48 V LiFePO4 bank.
Please enter system data for calculations. Fields with an asterisk
(*) are required.
Battery Duration
—days, best deficit month (clear skies)
—days, worst month (clear skies)
Charging:
Power generation meets or exceeds the daily load. Batteries
recharge fully; no net discharge.
Deficit:
Daily load exceeds power generation. Batteries net-discharge
over the day; a sustained deficit eventually empties the bank
(shown above as “days to empty”).
OVER DoD:
Daily discharge depth exceeds your configured Depth-of-Discharge
cutoff (“⚠ Over limit” appears in the table).
Repeatedly going below DoD shortens battery life; reduce loads,
add production, or increase battery capacity.
Note: the “⚠ Over limit” flag also appears
next to a daily load row if any single month’s daily DoD
exceeds the cutoff, even when the overall monthly status
is Charging (a “small load” the calculator still
flags).
Month
Irradiance kWh/m²/day
Solar Access
Typical Daily Charge Status
Daily Solar Wh/day
Wind Wh/day
Net Balance Wh/day
Recharge fraction/day
Days to Full from empty
Duration days, clear
Duration days, overcast
Load/Capacity %
Net DoD %
Solar vs Load %
Solar Noon 15th, std time
Scenario Planner — trip / event check
Stress-test a specific trip against a chosen month and sky&wind
conditions. Uses the irradiance you’ve already fetched. Add
an extra ad-hoc load (e.g. a portable air conditioner) and slide
the cloud cover from clear to overcast or select a calm to strong
wind condition to see when your battery hits the DoD cutoff.
Applied to discharge only; charge correction is complex and
rarely material. Load Current is approximated as the 24-hour
average: total daily Wh ÷ System Voltage ÷ 24
hours. This is accurate for steady loads. Intermittent high-draw
loads (e.g., a pump running 30 minutes) experience a higher
instantaneous current and therefore a greater Peukert penalty
than this average reflects; treat those results as
optimistic.
Example: Rated capacity = 100 Ah, k =
1.05 (LiFePO₄), Rated hour H = 20, daily load =
1,200 Wh, Voltage = 12 V
I = 1,200 Wh/day ÷ 12 V ÷
24 h = 4.17 A
→ 100 Ah × (20 h ÷
(4.17 A × 20 h ÷ 100 Ah))0.05
= 100 Ah × 24.00.05 =
117 Ah
(slow draw raises effective capacity above rated)
Source: Peukert, W. (1897). Über die Abhängigkeit der
Kapazität von der Entladestromstärke.
Overcast Sky Factor: 3%
The monthly table's worst-case overcast-day column models a
fully overcast day at 3% of the typical-day harvest for that
month. The typical-day value already reflects the long-term
cloud mix at your site (PVGIS climatology), so this is the
overcast tail of that distribution, not 3% of a clear-sky
ceiling. For a true clear-sky ceiling, use the Scenario Planner
with the sky-condition slider at SKC.
Example: typical-day harvest = 200 W
× 5.2 kWh/m²/day = 1,040 Wh/day
→ 1,040 Wh/day × 0.03 =
31 Wh/day on a fully overcast day
Source: field data from several hundred off-grid systems.
Temperature Derating (γPmax)
PV panel output decreases linearly with cell temperature above
the Standard Test Condition reference of 25 °C. The
power temperature coefficient γPmax
(typically −0.30 to −0.45 %/°C for
crystalline silicon) is printed on every module datasheet.
P(T) = PSTC × [1 + γ × (Tcell
− 25 °C)]
where γ is the signed coefficient (negative for standard
panels) and Tcell is the cell temperature in °C
reported by PVWatts’ thermal model (accounts for
irradiance, ambient temperature, and wind cooling).
PVWatts default replacement (Choice B):
PVWatts already bakes in a default γ =
−0.47 %/°C. Rather than stacking the user’s
value on top (which would double-count), the calculator undoes
the PVWatts default first and applies the user’s
coefficient:
Switch to Advanced mode and set a load Operating Window to see
the day/night split here.
Solar Noon (15th of month, standard time)
Solar noon
is the moment the sun crosses the local meridian,
its highest point in the sky that day, and the midpoint
of the solar window. It is not the same as clock noon: it drifts
by site longitude (4 minutes per degree east/west of your
time-zone meridian) and by the Equation of Time (up to
±16 minutes across the year, from Earth’s axial
tilt and orbital eccentricity).
Solar Noon (local min past midnight) = 720 − EoT(N)
− 4 × (Longitude − 15 × TZ Offset)
where N is the day-of-year for the 15th of the month
(epoch used for the monthly display). TZ Offset is the
site’s standard-time UTC offset;
DST is intentionally ignored so the displayed time is consistent
year-round. Site time zone is resolved from your entered
latitude/longitude.
Example: New York City: Lon =
−74.0°, TZ offset = −5 (EST), Feb 15
(N = 46), EoT(46) = −14 min
→ 720 min − (−14 min) −
4 min/° × (−74.0° −
15 °/h × (−5 h)) = 734 min
− 4 min/° × 1.0° =
730 min = 12:10 PM
Source: Equation of Time, NOAA Solar Position Calculator
methodology.
Battery Duration (days)
Usable Capacity ÷ |Net Daily Deficit|
where Usable Capacity = Effective Capacity × DoD Target.
Shown as ∞ when the system is in daily surplus (charging).
Example: Battery = 300 Ah ×
12 V = 3,600 Wh, DoD target = 80%, net daily
deficit = 1,440 Wh/day
Usable = 3,600 Wh × 0.80 = 2,880 Wh
→ 2,880 Wh ÷ 1,440 Wh/day =
2.0 days
Detailed Inverter Model (partial-load curve)
Real inverters waste a fraction of every watt that passes
through them, and the fraction depends on the load. Below ~10%
of rated power they're inefficient (the conversion circuitry's
idle losses dominate). Around 40–80% load they hit peak
efficiency, then dip slightly at full rated power.
where f = instantaneous load ÷ inverter rated power.
Below 10% load the curve extrapolates from the origin but is
floored at η10 ÷ 2 so a
very small draw (e.g. 5 W on a 2 kW inverter)
never produces an unrealistically low efficiency that would
double-count the no-load loss already captured by the idle-draw
field. Above 100% load (within the surge headroom) the value is
held at η100.
When the largest steady AC load exceeds the inverter's rated
power a red undersized warning fires; a real inverter
would trip overload protection. When the largest load is below
10% of rated, an amber advisory points out that the inverter is
heavily oversized for the system; the math will work but
low-load efficiency dominates and a smaller inverter is usually
the better fit.
No-load draw (the inverter's idle current when
no load is present) is added as a continuous 24/7 baseline,
additive to the existing Standby Draw field. Both feed every
calculation path: the monthly table, the 8760-hour TMY
simulation, and the Scenario Planner.
Example. 2 kW inverter with curve 75% /
92% / 88%, no-load 12 W. A 500 W AC load is at f =
0.25, so η = 75% + (0.25-0.1)/(0.5-0.1) × (92% − 75%)
= 81.4%. Battery sees 500/0.814 = 614 W
draw. Idle adds 12 W × 24 h =
288 Wh/day on top of all loads. A second
35 W load at f = 0.0175 hits the floor: linear
extrapolation gives 13%, but η10/2 = 37.5% wins,
so battery sees 35/0.375 = 93 W for that load.
Surge / peak power (1–2 s instantaneous rating)
triggers an inrush-vs-surge warning when a load's startup inrush
exceeds it. Note: this is the short motor-inrush rating, distinct
from the longer "X-Boost" / sustained-boost spec on portable power
stations; sustained-boost modeling is a separate future
feature.
Wind Power & Capacity Factor
P = ½ ρ A v³ Cp
where ρ = air density (kg/m³), A = rotor swept area
(m²), v = wind speed (m/s), Cp = power
coefficient (max 0.593, the Betz limit; real small
turbines 0.20–0.40 for HAWTs, 0.15–0.22 for VAWTs). Output
follows a three-zone power curve: zero below
cut-in speed, cubic growth from cut-in to rated speed, flat at
nameplate power from rated to cut-out, then zero above
cut-out.
Swept area depends on axis type:
Horizontal-axis (HAWT, propeller):
A = π (d/2)² (circular swept disc).
Vertical-axis (VAWT: Darrieus, H-rotor,
Savonius):
A = diameter × rotor height (the rectangular swept
area traced by the blades around the central column).
For VAWTs the shear-law correction also uses an
effective hub height of mount + rotor height / 2
(the swept-area centroid), since the rotor extends vertically
rather than concentrating at a single hub.
Wind shear (power law):
vhub =
vref (hhub/href)α,
where α ranges from 0.10 (open water) to 0.25 (forested
terrain). Wind speed, and therefore power, rises quickly with
height.
Rayleigh PDF & capacity factor: Real wind
speed is variable. This calculator integrates P(v) over a
Rayleigh probability density (1,000 bins up to
max(2 × vcut-out,
12 v̄)) to find the mean power. The
capacity factor CF = P̄/Prated
shows what fraction of nameplate power is delivered on average.
Yosemite Valley example (15 W
micro-turbine, v̄ = 1.5 m/s, cut-in
2.5 m/s, rated 8 m/s,
Cp = 0.35,
d = 1.0 m, ISA air):
CF ≈ 3.6% →
P̄ ≈ 0.54 W →
~13 Wh/day, a low-wind site where solar dominates.
Frequently Asked Questions
How do solar panels work?
A solar panel converts sunlight into electricity through the
photovoltaic effect, the semiconductor cousin of the
photoelectric effect Einstein won the 1921
Nobel Prize for explaining. In both, a photon (a particle of
light) striking a material can knock an electron loose, but
only if the photon carries more energy than the material's
binding threshold, called the bandgap, in
semiconductors. The two effects differ in what happens next:
In the pure photoelectric effect, the
electron is ejected entirely from the material into vacuum
or another medium. This is the principle behind
night-vision tubes,
photomultiplier detectors in PET scanners and
airport gamma-ray scanners, and old vacuum phototubes,
the everyday device closest to Einstein's original
experiment. (Heads-up: most things people loosely call
"photoelectric" today, like camera sensors, photodiodes
in TV remotes, and dusk-to-dawn yard lights, actually
use the photovoltaic variant below.)
In the photovoltaic effect used by solar
panels, the electron is freed inside a thin
silicon wafer that's been doped to create a P-N junction.
The junction's built-in electric field sweeps the freed
electron toward the N-side and the matching "hole" toward
the P-side, separating the charges before they can
recombine. Connect a wire between the two sides and you
have current, DC electricity, at roughly
0.5–0.6 V per cell.
A typical "solar panel" wires 60–72 cells in series to
reach the usable 18–40 V range. From there: a
charge controller (MPPT or PWM) regulates
voltage and current to match the battery's needs, the
battery stores excess energy for night-time
or cloudy days, and an inverter converts the
battery's DC back to AC for household loads (or DC loads run
straight off the battery). This calculator models every step
of that chain (irradiance reaching the panel, the
panel's STC-rated conversion, controller efficiency,
round-trip battery losses, inverter losses, and final load
delivery) so you can see exactly where energy is lost
between sunshine and the outlet.
How much battery do I need for my off-grid system?
It depends on your daily load in watt-hours, your target depth of
discharge (DoD), and how many days of autonomy you want. A common
starting point is: divide your daily load by your DoD to get the
minimum usable capacity, then add a safety margin. For example,
1,000 Wh/day at 80% DoD requires at least 1,250 Wh
of nominal battery capacity. This calculator sizes both battery
and solar together against real monthly irradiance data so you can
see which months fall short.
What is depth of discharge (DoD) and why does it matter?
Depth of discharge is the percentage of a battery's capacity that
has been used relative to its total capacity.
Discharging a battery too deeply too often shortens its lifespan;
lithium chemistries typically allow 80–100% DoD, while
lead-acid batteries are usually limited to 50% for longevity.
Setting a conservative DoD target in the calculator gives you a
buffer above the hard cutoff, so the system flags months where
your battery would dip into that buffer zone.
How does cloud cover affect solar panel output?
Cloud cover reduces the amount of solar irradiance reaching your
panels.
In the Scenario Planner's geometric mode, the sky-condition
slider sets a clearness index (Kt) at the front of the physics
chain: SKC ≈ 0.75 (clear),
SCT ≈ 0.475 (scattered, ~63% of the clear-sky
ceiling), OVC ≈ 0.15 (overcast, ~20% of the
ceiling). These are midpoints of the METAR Kt bins
(Iqbal/Reindl), so you can stress-test your system against any
sky state without re-fetching irradiance. The three weather modes
(PVGIS-monthly, TMY-hourly, Forecast) use real observed
irradiance and ignore the slider.
What's the difference between PVGIS and PVWatts data?
Both are free, government-backed datasets of historical solar
irradiance, but they draw on different satellite and
ground-station sources. PVGIS (from the EU's Joint Research
Centre) has strong coverage for Europe, Africa, and Asia. PVWatts
(from NREL) is optimized for the Americas and also supports
tracking array types (1-axis, 2-axis). This calculator lets you
choose which source to use per panel plane, and supports hourly
TMY data from PVWatts for 8760-hour simulations.
Can this calculator size an RV or cabin solar system?
Yes, it's designed for exactly that. Enter your location (or any
location you'll be traveling to), your panel array wattage and
tilt, and your electrical loads with their daily hours. The
calculator returns month-by-month results showing whether your
system is adequate, approaching your DoD limit, or undersized for
that month. The Scenario Planner can also simulate a specific trip
duration under chosen weather conditions.
What is TMY data?
TMY stands for Typical Meteorological Year, a synthetic dataset assembled from many years of historical
weather records to represent a statistically typical year at a
given location. It provides hourly values for solar irradiance,
temperature, and wind. Using TMY data gives a more realistic
picture of annual system performance than a single year of
measurements, but it is not a forecast; actual production in
any given year will vary. This calculator uses TMY data for both
its monthly irradiance lookups and its optional 8760-hour hourly
simulation mode.
Why does a 100 Ah battery deliver less than 100 Ah at high current?
The Ah rating on a battery is measured at a slow, standardized
discharge rate, typically C/20 (5 A for a 100 Ah battery, drained
over 20 hours). At higher currents, internal resistance,
electrolyte concentration gradients, and the chemistry's
ion-transport speed all limit how much of the stored charge you
can actually pull out before the voltage collapses. This is the
Peukert effect: a 100 Ah lead-acid battery discharged at
50 A might only deliver ~75–80 Ah before
hitting cutoff. The Peukert exponent k captures how
steeply capacity falls off: k ≈ 1.05
for lithium (small effect),
k ≈ 1.10–1.30 for AGM/flooded
lead-acid. This calculator applies a Peukert correction whenever
your average load current exceeds the rated-hour discharge rate.
Should I get an MPPT or PWM charge controller?
MPPT (Maximum Power Point Tracking) is almost always the better
choice for off-grid systems.
A PWM controller forces the panel to operate at the battery's
voltage, typically well below the panel's optimal voltage,
so you lose the difference. On a 100 W nominal
panel feeding a 12 V battery, a PWM controller often
delivers ~75 W while MPPT delivers ~95–97 W:
a 25–30% harvest gap. PWM is cheaper and may make sense
for very small systems (under ~100 W) where the
controller cost difference matters more than the harvest
difference. Never direct-connect a panel to a battery without a
controller: you'll either overcharge lead-acid (gassing, water
loss, damage) or trip a lithium BMS. This calculator's
Charge Controller dropdown applies the
appropriate efficiency factor.
Does panel tilt matter more than panel size?
Both matter, but tilt affects when you generate power,
while wattage affects how much.
A panel matched to local latitude maximizes annual harvest, but
if your worst-month deficit is in winter, tilting steeper
(latitude + 15°) can recover 20–40% more
December production at the cost of summer output,
sometimes a better trade than buying a bigger panel. A flat
(0°) panel collects more in summer but loses heavily in
winter when the sun is low. Azimuth matters too: a 30°
east-of-south offset typically costs only 5–10% of annual
harvest, but 90° (due east or west) can cost 20–30%.
Use this calculator's Tilt Optimization helper
to see annual-best, summer-best, and winter-best tilt angles for
your specific latitude.
Why does my fridge show a 70 W draw when it's rated at 35 W?
Fridge nameplate ratings are usually an average over a full
day,
accounting for the fact that the compressor only runs about half
the time (~50% duty cycle). When you measure with a Kill-A-Watt
or clamp meter and the compressor is on, you'll see the actual
instantaneous draw, roughly 2× the nameplate. So a
fridge rated at 35 W average is really a 70 W
compressor cycling on/off. Add another 1–3× brief
surge at startup (motor inrush) and the peak can momentarily hit
150–200 W. This distinction matters: your battery
sizing depends on the daily average (35 W ×
24 h = 840 Wh/day), but your inverter has to
handle the 70 W running peak plus the inrush
surge. Sizing the inverter to just the nameplate will trip
overload protection every time the compressor kicks on.