It's a common assumption: buy a bigger UPS, get longer backup time. In reality, a 5 kVA UPS with a small internal battery pack can run out of power faster than a 1.5 kVA unit with a larger external battery bank. The reason comes down to a distinction that's easy to overlook — kVA tells you how much load the UPS's inverter can support, while runtime is governed almost entirely by a completely different number: the battery's amp-hour (Ah) capacity. Mixing these two up is one of the most common — and most expensive — UPS sizing mistakes.
kVA vs. Watts: Two Different Limits
Every UPS carries two ratings: VA (or kVA) and Watts. The VA rating defines the maximum apparent power the inverter can physically deliver, while the Watts rating reflects the real, usable power after accounting for the UPS output power factor — typically 0.7 to 0.9 for many consumer and line-interactive units. A 1500VA UPS at a 0.6 output power factor only delivers 900W of real capacity, not 1500W. Connecting a load that draws more than the Watt rating will overload the UPS even though the VA headroom looks fine on paper, so the Watt figure — not the VA figure — is what actually limits how much equipment you can safely connect.
Runtime Is a Battery Question, Not a kVA Question
Once the UPS can support your load's real power draw, the next question — how long will it last on battery — depends on a separate set of variables entirely: battery voltage, amp-hour (Ah) capacity, and inverter efficiency. A UPS with a small internal battery might support a large load comfortably from a kVA standpoint but only bridge a few minutes of backup, while a smaller-kVA unit paired with an external battery bank can sustain a modest load for an hour or more. This is exactly why a "bigger" UPS by kVA rating doesn't automatically mean longer runtime — the two numbers answer different questions.
Worked example: A 48V battery bank rated at 100 Ah, running at 90% inverter efficiency, supplying a 600W load: Runtime = (100 × 48 × 0.9) ÷ 600 = 7.2 hours. Halve the load to 300W and the same battery bank stretches to roughly 14.4 hours — runtime scales with load and battery energy, completely independent of the UPS's kVA rating, as long as the kVA rating is large enough to support the load in the first place. Try your own numbers in the calculator above.
Sizing the Battery for a Target Runtime
When the requirement is a specific backup window — say, 30 minutes for a graceful shutdown, or several hours for a critical process — the formula runs in reverse to find the required Ah capacity, with extra margin built in for battery aging and depth-of-discharge limits. Switch the calculator above to "Find Required Ah" to run this in reverse.
For a 1,000W load needing 30 minutes (0.5 hours) of runtime from a 48V battery bank at 90% efficiency and an aging factor of 0.8: Ah = (1,000 × 0.5) ÷ (48 × 0.9 × 0.8) ≈ 14.5 Ah — and in practice, a battery rated somewhat above this figure would be selected to leave margin for real-world conditions.
Why Real Runtime Often Falls Short of the Formula
The basic runtime formula assumes ideal conditions, but several real-world factors reduce usable capacity below the nameplate number:
- Battery aging — lead-acid batteries typically lose around 20% of their capacity over a 5–7 year service life, so sizing only for day-one capacity leaves no margin near end of life.
- Temperature — both UPS output capacity and battery capacity derate in hot or cold environments; a UPS rated for full output at 25°C may deliver noticeably less at 35°C.
- High discharge rates — batteries deliver less total energy when discharged quickly (the Peukert effect) than the simple Ah × Voltage figure suggests.
- Depth of discharge limits — for long cycle life, many lead-acid applications use conservative DoD limits; the appropriate value depends on battery type, duty cycle, and manufacturer recommendations, which affects how much of the rated Ah is safely usable.
The Sizing Sequence That Avoids This Mistake
A reliable UPS specification follows the load all the way through, rather than jumping straight to a single VA number. First, total the real Watts of every device that must stay powered, and convert to VA using each device's power factor. Second, size the UPS kVA/Watt rating with headroom for inrush current and future growth — typically 20–30% above the calculated steady-state load. Third, and separately, define the required backup runtime and size the battery Ah using the runtime formula with aging and efficiency margins included. Treating these as two distinct sizing steps — UPS capacity for the load, battery capacity for the runtime — is what prevents the classic mismatch of a UPS that handles the load fine but dies in minutes when the power actually goes out.
Conclusion
kVA and battery Ah answer two different engineering questions, and conflating them is how oversized-on-paper UPS systems end up underperforming in the field. Size the UPS's VA/Watt rating to comfortably support the connected load with margin for inrush and growth, then size the battery bank's Ah capacity separately against your actual target runtime, with realistic allowances for aging, temperature, and discharge rate. Get both right independently, and the backup system will behave the way the spec sheet promised when the power actually fails.