"How long will my UPS last?" is one of the most common questions people ask before buying a backup power system, and it's also one of the easiest to get wrong, because runtime isn't printed anywhere on most UPS boxes the way VA rating is. Runtime depends on your specific load and battery combination, not on the UPS model alone — the same UPS can have very different runtimes depending on what's plugged in and what battery is attached. This guide walks through what actually determines runtime, gives realistic ballpark figures for common household and office loads, and compares how lead-acid and lithium batteries behave differently under the same runtime math.
The Runtime Formula, in Plain Terms
Runtime comes down to a simple energy balance: how much energy the battery holds, versus how fast the load draws it down. In formula form:
Worked example: A 48V, 100 Ah battery bank at 90% inverter efficiency, supplying a 600W load, runs for (100 × 48 × 0.9) ÷ 600 = 7.2 hours. Cut the load in half to 300W and the same battery stretches to about 14.4 hours — runtime is inversely proportional to load, so trimming what's plugged in during an outage is often the fastest way to extend backup time without buying more battery. Use the UPS Calculator with your own battery and load figures for a more precise estimate.
Why Real-World Runtime Falls Short of the Calculated Figure
The simple formula above gives a theoretical maximum. Three factors consistently pull real-world runtime below that number, and understanding each one helps explain why it's worth building margin into any runtime target rather than trusting the calculated figure to the last minute.
The Peukert effect. Battery Ah ratings are usually measured at a specific, slow discharge rate — commonly the 20-hour rate (C/20) for lead-acid batteries. Pulling more current than that rate implies delivers less total capacity than the nameplate suggests, a phenomenon known as the Peukert effect. Lead-acid batteries commonly show a Peukert exponent above 1.0, with the actual value depending on battery construction and manufacturer test data — meaning a battery discharged much faster than its rated test rate can deliver noticeably less than its full nameplate Ah. Lithium-ion (and especially LiFePO4) generally shows a much smaller rate-dependent capacity loss, with a Peukert exponent much closer to 1.0, so its usable capacity stays closer to nameplate across a wider range of discharge rates — one more reason lithium tends to outperform its Ah rating suggests when powering higher-wattage loads.
Temperature. Battery capacity is rated at a standard temperature, commonly 25°C (77°F), and drops in cold conditions — a battery in an unheated garage or plant room can deliver meaningfully less than its rated Ah in winter, while sustained high temperatures can shorten battery life even though they may temporarily raise capacity. Lithium batteries generally handle a wider operating range but often have a built-in low-temperature charge cutoff, so it's worth checking a battery's temperature rating alongside its Ah rating for any site outside a comfortable, temperature-controlled room.
Aging. Every battery loses capacity over time regardless of chemistry. Lead-acid batteries are generally considered end-of-life once they drop to around 80% of original rated capacity, commonly somewhere in a 3–7 year window depending on usage, depth of discharge, and maintenance; lithium batteries typically retain usable capacity for longer and are often rated in charge cycles rather than years. This is exactly what the aging factor in the required-Ah formula on this page accounts for — sizing only for day-one, brand-new capacity leaves no margin as the battery bank ages.
Sizing a Battery Bank: Series vs Parallel Wiring
Most battery banks aren't a single battery — they're multiple batteries wired together to reach the voltage and Ah capacity a UPS or inverter needs. How you wire them changes the outcome.
Series wiring connects the positive terminal of one battery to the negative terminal of the next, which adds voltage while keeping Ah capacity the same as a single battery. Two 12V, 100Ah batteries wired in series produce a 24V, 100Ah bank — useful for inverters and UPS systems that need a higher DC bus voltage (24V and 48V systems are common for larger backup setups, since a higher voltage means lower current for the same power, which keeps cables thinner and reduces resistive losses).
Parallel wiring connects all the positive terminals together and all the negative terminals together, which keeps voltage the same as a single battery but adds Ah capacity. Two 12V, 100Ah batteries wired in parallel produce a 12V, 200Ah bank.
Larger battery banks often combine both approaches — multiple series-wired strings connected in parallel to reach both a target voltage and a target Ah capacity. A common pitfall with parallel strings is mixing batteries of different age or health: a weaker battery in a parallel string can be repeatedly overworked by stronger neighbours as the bank tries to equalize voltage, shortening its life further and dragging down the whole bank's usable capacity. Most manufacturers recommend replacing an entire string together rather than swapping a single new battery in alongside older ones. When sizing a bank, use the required-Ah figure from the calculator on this page as your total target, then work out the series/parallel combination of your chosen battery's individual voltage and Ah rating that reaches it.
Typical Runtime for Common Loads
To make the formula concrete, here's roughly what a mid-sized home inverter battery setup (a single 150 Ah, 12V battery, a common home inverter configuration) delivers at different load levels, assuming 85% inverter efficiency:
- A few LED lights + a fan (~100W): roughly 15 hours — enough to comfortably cover an evening or overnight outage.
- LED lights + fan + TV + Wi-Fi router (~250W): roughly 6 hours — a typical "keep the essentials running" scenario.
- Adding a refrigerator's average draw (~400W combined): roughly 3.8 hours — refrigerators cycle on and off, so actual runtime is usually somewhat longer than this steady-load estimate suggests.
- A single window AC unit (~1200W): roughly 1.3 hours — air conditioning is by far the fastest way to drain a home battery bank, which is why most home backup setups are sized to skip AC entirely and cover only lighting, fans, and electronics.
These are illustrative estimates for one common battery size — use the UPS Calculator with your own battery Ah, voltage, and load for a figure specific to your setup.
Lead-Acid vs Lithium: How Runtime Behaves Differently
The runtime formula itself is the same for both chemistries, but two practical differences change what you can actually get out of a given Ah rating in real use. First, usable capacity: a common planning assumption is around 50% depth of discharge (DoD) for lead-acid for reasonable service life, and around 80–90% for many lithium systems — but the manufacturer's specified usable DoD should always be used for actual sizing, since it varies by construction and model within both chemistries. Second, voltage sag under load: lead-acid voltage drops more as the battery discharges and as load increases, which can trip a low-voltage cutoff earlier than the simple formula suggests, while lithium holds a flatter voltage curve through most of its discharge, delivering runtime closer to the calculated figure. In practice, this means a lithium battery rated at a lower Ah than a lead-acid battery can still deliver comparable or longer real-world runtime, once usable DoD is accounted for — worth factoring in when comparing battery options by Ah rating alone.
UPS and Inverter Types: Does the Topology Affect Runtime?
The runtime formula on this page applies to any UPS or inverter, but the internal design of the unit affects how efficiently it converts battery DC power to usable AC, and how it behaves the moment grid power fails.
Offline (standby) UPS units pass grid power through to the load directly during normal operation and switch to the battery/inverter only when the grid fails, typically within a few milliseconds. Because the inverter isn't running continuously, standby units tend to be simple and reasonably efficient, though the brief transfer time can occasionally affect very sensitive equipment.
Line-interactive UPS units add automatic voltage regulation (AVR), correcting minor sags and surges by adjusting a transformer tap rather than switching to battery — this means less frequent battery cycling for small grid fluctuations, which can meaningfully extend battery life in areas with unstable but not fully interrupted grid power.
Online (double-conversion) UPS units continuously convert incoming AC to DC and back to AC, so the load always runs from the inverter with zero transfer time during an outage — the trade-off is a small continuous conversion loss even during normal grid operation. This ongoing conversion efficiency is exactly what the "Inverter Efficiency" input in the calculator on this page represents. Higher-quality online units commonly reach 90–95% efficiency, while older or budget units can run meaningfully lower — checking a UPS's actual rated efficiency, rather than assuming a round number, gives a more accurate runtime estimate.
Regardless of topology, the runtime formula itself doesn't change — only the efficiency figure you plug in should reflect your specific unit's actual conversion losses.
How to Measure Your Actual Load Before Sizing
The load figure in the runtime formula should be the real, steady-state Watts your equipment actually draws — not the maximum Watts printed on a nameplate, which is usually a worst-case or peak rating rather than typical draw.
The most reliable way to measure actual load is a plug-in power meter for individual devices, or a clamp meter on the supply cable for a whole panel or circuit. If you're reading current (Amps) directly off a clamp meter rather than a device that reports Watts, remember that Watts = Volts × Amps × Power Factor for AC loads — for a purely resistive load like an incandescent bulb or heater, power factor is close to 1.0, but motor-driven equipment (refrigerator compressors, air conditioners, pumps, fans) typically runs at a power factor of roughly 0.7–0.9, so multiplying Amps by Volts alone without accounting for power factor overstates the real Watts.
Motor-driven and compressor-based equipment also draws a brief but significant inrush or starting current — often several times the running current — for a fraction of a second when it switches on. This doesn't meaningfully affect runtime calculations, which are based on average running load, but it does matter for UPS/inverter sizing, since the inverter itself must be rated to survive that inrush without tripping — a separate consideration from battery runtime, covered on the UPS Size Calculator.
When in doubt, measuring a device directly for a few minutes gives a far more accurate load figure than adding up nameplate Watts for every connected item, which tends to significantly overstate real consumption since equipment rarely runs at its absolute maximum rating continuously.
Battery Maintenance and Testing for Reliable Runtime
A battery bank's actual runtime, months or years into service, depends heavily on how well it's been maintained — a neglected battery can deliver well under its rated Ah even before it's old enough to be considered end-of-life.
For flooded (wet-cell) lead-acid batteries, periodically checking and topping up electrolyte level with distilled water is essential — plates exposed to air lose capacity and can be permanently damaged. Sealed/VRLA (AGM and gel) batteries and lithium batteries don't need this, part of why they're increasingly preferred for low-maintenance installations.
Terminal corrosion is a common, easily overlooked cause of reduced effective runtime — corroded or loose connections add resistance to the circuit, showing up as voltage drop and wasted energy under load, which effectively reduces the usable capacity reaching the inverter even though the battery itself is healthy. A visual check and cleaning of terminals every few months is cheap insurance.
Periodic load testing — deliberately discharging the battery bank under a known load and timing how long it actually lasts — is the only way to verify real-world capacity rather than relying on the battery's age or nameplate rating alone. Many UPS systems include a brief self-test function that does a scaled-down version of this automatically; for critical backup systems, an occasional full manual runtime test under real load is worth the downtime it costs. Finally, ambient temperature control matters over the long term as much as it does for a single discharge — batteries stored or operated in consistently hot environments age faster than the same battery in a temperature-controlled space, shortening the years before capacity drops enough to require replacement.
Practical Tips for Getting the Runtime You Expect
- Size for your actual essential load, not everything in the house. Identify what genuinely needs to stay on during an outage (lights, fans, router, a few devices) and size around that, rather than trying to cover every appliance.
- Account for battery aging. A 3-year-old lead-acid battery may deliver noticeably less than its original rated Ah — don't rely on day-one nameplate figures for long-term planning.
- Avoid deep, repeated discharge on lead-acid. Routinely draining below the recommended DoD shortens battery life significantly faster than shallow, regular cycling.
- Check inverter efficiency, not just battery Ah. A less efficient inverter wastes more energy as heat during conversion, directly reducing usable runtime for the same battery.
- Wire multi-battery banks as complete matched sets. Mixing old and new batteries, or different manufacturers/batches, in the same series or parallel string can shorten the life of the whole bank and produce unpredictable runtime.
- Match your UPS's inverter rating to your inrush current, not just running Watts. A battery sized correctly for runtime is no help if the inverter trips on a motor's starting surge — check the surge rating separately using the UPS Size Calculator.