redzilla
All tools
Power

UPS Runtime

Estimate the backup minutes of a UPS from the load and the battery bank, accounting for inverter efficiency, depth of discharge (DoD) and aging. Includes an inverse mode to size the Ah you need for a target runtime.

What do you want to calculate
Load to back up
W
Battery bank
V
Ah
×

Ah and voltage per battery, at 20 °C. The bank energy is Vb × Ah × N.

DC→AC losses. Typical 8595 %.

How much of the battery you use. Lead-acid 5080 %; lithium up to 100 %.

Real capacity vs. rated. Used batteries ~80 %.

Examples
redzilla.cl — ups
 
Estimated runtime
In minutes
Usable energy
after DoD, efficiency and aging
Runtime
Minutes

Calculation breakdown

ItemValueDetail
How it is calculated · Wh, DoD and Peukert

1. Bank nominal energy: Wh_bank = Vb × Ah × N.

2. Energy actually deliverable to the load: Wh_usable = Wh_bank × DoD × efficiency × aging (each factor as a fraction).

3. Runtime: minutes = Wh_usable / load_W × 60, with load_W = VA × PF if you enter VA.

4. Sizing: Ah = load_W × (min/60) / (Vb × DoD × eff × aging × N).

! This is an estimate. Real discharge is not linear: at high currents capacity drops (the Peukert effect), so the actual runtime is usually somewhat lower than calculated. Also consider temperature and the batteries' state of health.

Runs locally in your browser · no sign-up · nothing leaves your browser.

How it works

The calculator estimates the backup minutes of a UPS from the load (in W directly, or in VA multiplied by the power factor) and the battery bank: voltage and Ah per battery, and how many batteries there are. The nominal bank energy is Wh = V × Ah × N, but not all of it reaches the load: a derating is applied for depth of discharge (DoD, 50-80% for lead-acid, up to 100% for lithium), inverter efficiency (typically 85-95%) and battery aging (~80% for used batteries).

The runtime comes from minutes = usable_Wh / load_W × 60, where usable_Wh = bank_Wh × DoD × efficiency × aging. The inverse mode solves for the Ah per battery needed to reach a target runtime. It is a first-order estimate: real discharge is not linear (at high currents the effective capacity drops due to the Peukert effect), so the actual time is usually somewhat shorter, especially with large loads on small batteries.

Example: 300 W on a single 12 V, 7 Ah battery

  1. Nominal energy: 12 V × 7 Ah × 1 = 84 Wh.
  2. Combined derating: DoD 80% × efficiency 90% × aging 80% = 0.576, so the usable energy is 84 × 0.576 ≈ 48.4 Wh.
  3. Runtime: 48.4 / 300 × 60 ≈ 9.7 minutes: enough for a clean shutdown, not for continued operation.

Frequently asked questions

How many minutes does a 1000 VA UPS last at full load?
With the typical internal battery (one or two 12 V, 7-9 Ah units), a 1000 VA UPS at its real maximum load (about 600 W with PF 0.6) usually lasts 4 to 8 minutes. Runtime grows almost linearly as the load drops: at half load you can expect double or more. Hours of backup require external battery banks.
What is the difference between VA and W on a UPS?
VA is apparent power and W is real power: they relate through the power factor, W = VA × PF. A 1000 VA UPS with PF 0.6 can only feed 600 real watts. Runtime depends on the real power in W, which is why the calculator converts VA using the PF you enter.
Why should I not discharge the battery to 100%?
On lead-acid batteries (the usual ones in UPS units), deep discharges drastically shorten service life: working at 50-80% depth of discharge multiplies the available cycles. Lithium batteries tolerate DoD close to 100%. The DoD slider in the calculator reflects exactly that design criterion.
Why is the real runtime shorter than the calculated one?
Because of the Peukert effect: at high discharge currents, the effective Ah capacity is lower than the nominal one (which is specified at a 20-hour discharge rate). Temperature and battery health also matter. Treat the result as optimistic and leave some margin. The whole calculation runs in your browser, with no data sent anywhere.
Was this tool useful?
Disclaimer We take great care to keep every tool accurate and review it thoroughly; even so, we can't guarantee it is free of errors or take responsibility for how the results are used. We recommend double-checking anything critical.
Found an error? Let us know →