How this calculation works
Choosing a UPS answers two separate questions: capacity (how much power it delivers continuously) and runtime (how long the battery lasts). The tool computes both with common design formulas.
1. Total load
Each device given in VA is converted to watts with the power factor, and vice versa:
Prefer measured draw or vendor configurator figures (HPE Power Advisor, Dell EIPT) over PSU nameplate ratings — power supplies are usually much larger than the real load.
2. UPS capacity
Standard sizes considered: 1, 1.5, 2, 3, 5, 6, 8, 10, 15, 20, 30, 40, 60, 80, 100 kVA and up. For N+1 two options are shown: two equal units in parallel (1+1) or, above 15 kVA, a modular UPS with 10–50 kVA modules plus one spare module.
3. Battery energy
The 1.25 factor makes sure the battery still delivers the runtime at end of life (≈ 80 % of initial capacity). The DC bus voltage is assumed from typical classes (e.g. 240 V for 6–10 kVA, 384–480 V above) — the exact value depends on the UPS model.
Assumptions & limits
- VRLA capacity drops at high discharge rates (Peukert effect). For runtimes under 30 minutes, final selection must use the battery maker’s constant-power-per-cell tables; the figure above is an energy estimate.
- The runtime curve uses a 1.15 exponent relative to the design point — for comparison, not a guarantee.
- Battery rooms above 25 °C shorten battery life sharply (roughly halved per +10 °C).
- Motor loads (air conditioners, pumps) should not be on the server-room UPS.