OUTAGE ATLAS / FIELD NOTES

The Math Behind Outage Atlas

The planner is a transparent scenario calculator. Results follow your inputs and should not be read as measurements or a guarantee.

1. Estimate average appliance demand

Watts mode: average W = running W × use percentage ÷ 100
Annual mode: average W = annual kWh × 1,000 ÷ 8,760

Use percentage applies across the selected outage window. Annual energy already includes cycling; we do not apply use percentage to it again. The annual model uses a 365-day year.

2. Account for losses and reserve

Average battery draw = sum of appliance average W ÷ efficiency + inverter idle W
Required nominal Wh = average battery draw × outage hours ÷ (1 − reserve fraction)
Estimated runtime = entered capacity Wh × (1 − reserve fraction) ÷ average battery draw

The default 90% efficiency, 10 W idle draw and 10% reserve are editable scenario assumptions. They are not measured values for any listed model. A new, fully charged battery is assumed, with reserve covering some unavailable energy; there is no explicit model of aging, cold, cutoff behavior or nonlinear efficiency. For no active loads, capacity need is zero and runtime is not calculated.

3. Test a higher-demand scenario

We increase appliance energy by 25% and reduce efficiency by 10 percentage points, with a 50% floor. The same idle draw and reserve remain. This deliberately chosen sensitivity scenario is not a probability distribution or a statistical confidence interval.

4. Check output separately

Running output sums the entered running watts of active appliances, even if they cycle. Startup sums all supplied peak watts, conservatively assuming simultaneous starts. An unknown peak is flagged and prevents a confirmed startup comparison. Zero-use appliances are excluded. Surge duration and exact equipment compatibility still require manufacturer confirmation.

5. Compare reference models

We publish a small set of manufacturer specifications with source links and review dates. Models appear in ascending nominal capacity, not commission order. No prices, discounts, stock claims or laboratory-tested runtime claims are maintained. After 90 days without a source review, calculated model runtime is hidden and specification review is flagged. Ordinary manufacturer source links remain available.

6. Know the boundary

No recharging, solar generation, building wiring, whole-home installation, medical equipment, food safety, UPS certification, port compatibility or return-on-investment claims are included. The runtime result extends average demand uniformly over time; appliance timing can make a specific short outage differ sharply.

Energy-unit reference: Virginia Cooperative Extension. Model sources are linked beside each comparison. Initial review: October 26, 2026.