Sizing Your Solar System
Sizing Your Solar System
Proper solar sizing means the system producesreliably enoughrecharges energy to run indefinitely throughwithin your worst-case season whileand maintainingholds enough battery reserve to survive multipleyour design number of consecutive cloudyno-sun days. No solar system runs unattended forever: cells fade with cycling, panels degrade, and a long overcast stretch can outlast any fixed reserve. Plan for battery replacement every few years and use remote monitoring (see Monitoring Battery State) so you catch a failing node before it dies.
Step-by-Step Sizing Process
Step 1: Determine Daily Energy Consumption
FromDaily energy depends heavily on platform, firmware, TX duty cycle, and whether the Currentscreen/Bluetooth/Wi-Fi Draware page:on, so always use your own measured value where possible. As representative planning figures (measure your own): an nRF52840 repeater (RAK4631, T-Echo) averages ~10–15 mA, while an always-on ESP32 board (Heltec V3) averages ~40–80 mA (higher with Wi-Fi/MQTT). Compute daily energy as: Daily energy (Wh/day) = average current (mA) × 24 h × system voltage (V) ÷ 1000. The worked example below uses a typicalHeltec LoRaV3 (ESP32) repeater consumesdrawing approximatelya 2.22conservative Wh/day (~25 mA average at 3.7V7 overV: 25 × 24 hours)× 3.7 ÷ 1000 = 2.22 Wh/day. UseA higher-duty ESP32 node can easily draw 2–3× this, so re-run the calculation with your actual measured value if available.current.
Step 2: Find Worst-Case Peak Sun Hours
Peak sun hours (PSH) vary by location and season. UseSize the panel against your worst month (December value for year-round reliability in the northern hemisphere.hemisphere). Do not assume a flat "4 PSH year-round" — northern-US winter PSH is often only ~1–2.5 h/day. Look up your locationexact site on the NREL PVWatts calculator or(which useuses thesethe referenceNREL values:NSRDB dataset) rather than relying on a single state-wide number, because PSH varies sharply by city and dataset. The values below are rough orientation only — replace them with a PVWatts result for your coordinates:
| Location (approx.) | December PSH (approx.) | Annual Average PSH (approx.) |
|---|---|---|
| North Dakota / Fargo ( | ~2.5 h/day | ~4.5 h/day |
| Minnesota (~45°N) | ~2.8 h/day | ~4.6 h/day |
| Texas (~30°N) — varies widely by region; central/eastern TX (e.g. Austin) can be ~2.7 h in December | ~2.7–4. | ~5. |
| Pacific Northwest / Seattle, Portland (~47°N) | ~1.5 h/day | ~4.2 h/day |
| Florida (~28°N) | ~4.8 h/day | ~5.5 h/day |
Note: winter PSH can run anywhere from ~7% to ~42% below the 12-month average depending on latitude and climate, and a single state (especially Texas) is too large to treat as one number. Always confirm your site in PVWatts. (Figures as of 2026-06-08.)
Step 3: Calculate Required Panel Size
Panel size (W) = Daily energy (Wh) ÷ (PSH × efficiencysystem derate factor)
Use efficiencya system derate factor of 0.7075 to— accountthis forsingle planning factor covers charge-controller, wiring, temperature, and soiling losses. (We use 0.75 consistently across all Mesh America sizing pages; size your panel temperature,against wiring losses,this and chargeconfirm controllerproduction losses.in PVWatts.)
Example (North Dakota repeater):
Panel = 2.22 Wh ÷ (2.5h5 h × 0.70)75) = 2.22 ÷ 1.75875 = ~1.27W18 W minimum
A 6W6 W panel providesis 4.7×roughly 5× the nameplate minimum required— -but treat this as a comfortablenameplate marginratio, thatnot accountsreal forsurplus. In deep winter at high latitude (low sun angle, cold, dirt, snow, and a cheap controller running off its maximum-power point), a panel rarely delivers anything close to its nameplate wattage, and a snow-covered panel produces essentially zero regardless of its rating. The large nominal oversize is therefore needed, not luxury: it buys back winter harvest losses and helps recover after a cloudy or snowy stretch. The real protection against multi-day storms and snow shading,cover is battery reserve (Step 4), not panel degradation, and future firmware changes that may increase power use.margin.
Step 4: Calculate Battery Reserve
Battery capacity (Wh) = Daily energy × Reserve days ÷ Usable fraction
- Reserve days:
33–5 days for most locations;5 - 75–7+ days for high-latitude winterdeploymentsor emergency-comms nodes (panels do not help during a multi-day overcast, so reserve is your only protection). - Usable fraction: 0.80 for
Li-LiFePO4 (plan to 80% depth of discharge for longevity; this is the single value used across all Mesh America sizing pages so you get the same battery size on every page). Lithium-ion (dischargeLiPo/NMC)tocells20%aretoalsopreservecommonlycellplannedlife);at ~0.9080.forDoLiFePO4not apply a second derate on top of this unless you explicitly state and justify it.
Example (3-day reserve, Li-ion)LiFePO4 at 0.80):
Battery = 2.22 × 3 ÷ 0.80 = 8.33 Wh minimum (nameplate)
A single 3500mAh3500 mAh 18650 cell (e.g. Samsung INR18650-35E, 3500 mAh nominal, ~3.6–3.7 V nominal) = 3.5Ah5 Ah × 3.7V7 V = 12.95 Wh.Wh Thisnameplate, providesor ~10.4 Wh usable at the required0.80 fraction. That comfortably satisfies the 3-day reserverequirement (8.33 Wh) with margin. It does not satisfy the 5-day example below.
For a 5-day reserve in(LiFePO4 coldat climate with 50% cold-weather derate:0.80):
Battery = 2.22 × 5 ÷ 0.80 ÷ 0.50 = 27.7513.9 Wh minimum (nameplate)
→ Two 3500mAh3500 mAh 18650 cells in parallel = 25.9 Wh nameplate (marginally adequate); three cells = 38.85~20.7 Wh usable at 0.80) — comfortably covers the 13.9 Wh requirement with real positive margin. A single 12.95 Wh cell (comfortable)~10.4 Wh usable) is not enough for 5 days.
Cold-climate charging warning: For any winter/cold-climate build, never charge lithium (including LiFePO4) below 0 °C (32 °F) — sub-freezing charging causes lithium plating, permanent capacity loss, and a hidden internal-short fire risk (discharging in the cold is fine). The CN3791 used in the example build has no low-temperature charge cutoff, so a cold-climate node built around it must add a BMS with low-temp protection or a charge controller with a battery temperature sensor. See the Cold-Weather Operation page.
Complete Sizing Example: North Dakota Year-Round Repeater
| Parameter | Value |
|---|---|
| Node | Heltec |
| Average current draw (representative — measure your own) | ~25 mA (conservative for an ESP32 board; higher-duty configs draw more) |
| Daily energy | 2.22 Wh/day |
| Location | Fargo, ND ( |
| December PSH (approx.) | ~2.5 h/day |