The direct answer: The inverter in a UPS (Uninterruptible Power Supply) is the power electronics stage that converts stored DC battery energy back into clean AC mains voltage (120V or 230V) to keep connected loads running during a grid outage. While the battery stores the energy, the inverter dictates the quality, surge capacity, and efficiency of the power delivered to your equipment.

Whether you are specing a rackmount unit for a home lab or building a custom 48V DC-coupled backup system, understanding the inverter's role—and the math required to support it—is the difference between a seamless failover and a catastrophic voltage sag. Below, we break down the system topology, the exact sizing math including Peukert's law, and the hard limits of battery configurations.

The Core Function: What Is the Inverter in a UPS?

To understand the inverter, you have to look at the entire system block description from source to load. In a standard Online Double-Conversion UPS, the power path flows like this:

  1. Source (Grid AC): Dirty or fluctuating mains power enters the unit.
  2. Rectifier/Charger: Converts AC to DC, simultaneously charging the battery bank and feeding the DC bus.
  3. DC Bus: The central high-voltage or low-voltage DC link (often 24V, 48V, or up to 384V DC in large enterprise units).
  4. Inverter Stage: Uses an H-bridge topology of MOSFETs or IGBTs switching at high frequencies (typically 16kHz to 20kHz) with Sinusoidal Pulse Width Modulation (SPWM) to synthesize a flawless 60Hz/50Hz AC sine wave.
  5. Load: Your servers, networking gear, or medical equipment receives clean, uninterrupted power.

In Line-Interactive units (like the popular CyberPower CP1500 or APC Back-UPS Pro), the inverter is bidirectional. It acts as a charger when grid power is present, and instantly switches to DC-to-AC inversion via an internal transfer relay when the grid drops. The transfer time is typically 4 to 8 milliseconds—fast enough that most ATX power supplies don't register the blip.

UPS Topology and Inverter Behavior Spec-Sheet
TopologyInverter State During Normal OperationTransfer TimeWaveform Output
Offline (Standby)Off (Only charging)8 - 12 msStepped Approximation / Square
Line-InteractiveOff or Low-Power Charging4 - 8 msStepped or Pure Sine Wave
Online (Double-Conversion)Always On (Inverting DC Bus)0 ms (Seamless)Pure Sine Wave (<3% THD)

Sizing the Inverter and Battery Bank for Your Load

Sizing a UPS inverter isn't just about matching the wattage on the back of your PC. You must account for inverter efficiency, power factor, and the non-linear discharge curve of batteries. Let's run the sizing math for a realistic scenario: a home networking rack and a workstation drawing a continuous 500W real power load.

Step 1: Inverter and Charger Sizing

First, calculate the DC draw. Inverters are not 100% efficient; a good high-frequency inverter stage operates at about 85% to 90% efficiency under typical loads.

  • DC Power Required: 500W / 0.85 (efficiency) = 588W
  • Surge Margin: Switch-mode power supplies (SMPS) draw high inrush currents. Add a 20% surge margin: 588W * 1.2 = 705W

You need an inverter rated for at least 705W. In UPS terms, assuming a standard 0.8 Power Factor (PF), you need a unit rated for roughly 1000VA / 800W minimum. A 1500VA unit (like the APC Smart-UPS 1500) is the pragmatic choice to keep the inverter operating in its peak efficiency curve (around 50-60% load).

Step 2: Battery Sizing and Peukert's Law

Assume a 24V DC bus system (two 12V batteries in series) to keep DC current manageable.

  • DC Current Draw: 588W / 24V = 24.5 Amps
  • Target Runtime: 30 minutes (0.5 hours)
  • Ideal Capacity: 24.5A * 0.5h = 12.25 Ah

If battery chemistry were perfect, a 15Ah battery would suffice. But lead-acid (AGM/Gel) batteries suffer from Peukert's Law, which states that as the discharge current increases, the available capacity decreases. A battery rated at 18Ah is rated at the 20-hour discharge rate (C20, or 0.9A). When you pull 24.5A (a rate faster than 1C), the effective capacity plummets to roughly 55-60% of its nameplate rating.

To get 12.25Ah of usable capacity at a high 25A draw, you need a battery with a C20 nameplate of at least 22Ah to 24Ah. Therefore, the correct spec is two 12V 24Ah AGM batteries wired in series to create a 24V 24Ah bank.

Wire Sizing Note: Pulling 25A continuously on a 24V bus requires minimum 10 AWG THHN copper wire, but 8 AWG is strongly recommended to minimize voltage drop across the bus bars and prevent the inverter's low-voltage cutoff (LVC) from triggering prematurely.

Battery Configuration: Series vs. Parallel and Discharge Limits

How you wire your cells fundamentally changes the system's voltage and amp-hour (Ah) capacity, which directly impacts the inverter's DC-to-AC conversion efficiency.

Series vs. Parallel Consequences

  • Series Wiring: Voltages add; Ah capacity remains the same. Wiring four 12V 100Ah batteries in series yields 48V at 100Ah (4,800Wh total). This is the preferred method for high-power inverters because higher DC voltage means lower DC current, reducing I²R heat losses in the cabling and MOSFETs.
  • Parallel Wiring: Ah capacity adds; Voltage remains the same. Wiring four 12V 100Ah batteries in parallel yields 12V at 400Ah. This is generally discouraged for loads over 1000W, as pulling 2000W from a 12V bus requires over 166 Amps of DC current, necessitating massive 2/0 AWG cables and posing severe thermal risks.
Lithium Fire-Safety & Mismatch Warning: Never parallel mismatched lithium cells, and never mix old and new battery packs in parallel. Differences in internal resistance and state-of-charge (SoC) will cause massive cross-currents to flow between the packs, bypassing the Battery Management System (BMS) and leading to thermal runaway. If you must parallel lithium packs, they must be identical models, purchased in the same batch, and each must have its own independent BMS. Always comply with NFPA 855 standards for stationary energy storage installations.

Charge/Discharge Limits: C-Rate and Depth of Discharge (DoD)

The inverter can only pull what the chemistry safely allows. Exceeding the C-rate (the rate of discharge relative to battery capacity) causes voltage sag and permanent degradation.

Battery Chemistry Limits for UPS Inverter Applications
ChemistryMax Continuous Discharge (C-Rate)Recommended DoD (Standby)Cycle Life at Rated DoD
Lead-Acid (AGM/VRLA)0.2C to 0.5C (up to 3C for seconds)50% (Cyclic) / 80% (Emergency)300 - 500 cycles
Lead-Acid (Tubular Gel)0.2C60%800 - 1200 cycles
LiFePO4 (LFP)1.0C (Some up to 3C)80% - 90%3000 - 6000 cycles

If your inverter demands 50A from a 50Ah AGM battery (a 1C discharge rate), the battery voltage will sag below the inverter's 21V low-voltage cutoff almost immediately, and you will severely damage the lead plates. Always size the battery bank so the continuous draw remains at or below 0.3C for lead-acid.

Frequently Asked Questions About UPS Inverters

What is the difference between a UPS inverter and a standalone solar inverter?

A UPS inverter is optimized for fast transfer times (milliseconds) and handling the high inrush currents of IT equipment power supplies. It typically operates in a float-charge standby mode. A standalone solar inverter (like a Victron MultiPlus or Fronius) is optimized for maximum power point tracking (MPPT), continuous daily cycling, and grid-tie synchronization. While hybrid inverter/chargers blur this line, a dedicated UPS inverter prioritizes zero-transfer-time and clean total harmonic distortion (THD) over solar harvesting efficiency.

Can I replace the internal UPS inverter batteries with a larger external bank?

Technically yes, but it introduces significant risks. The internal charging circuit of a standard 1500VA UPS is usually limited to 10W–20W of charge current, designed to recharge small internal 9Ah or 18Ah SLA batteries. If you connect a 200Ah external LiFePO4 bank, the internal charger will run at 100% duty cycle for days, overheat, and likely fail. If you expand the bank, you must add an external AC-to-DC smart charger (like a Victron Blue Smart) wired directly to the batteries to handle the bulk charging, leaving the UPS internal charger only for float maintenance.

Why does my UPS inverter output a stepped approximation instead of a pure sine wave?

Cost and thermal management. Generating a pure sine wave requires high-frequency SPWM switching, complex LC output filters, and faster, more expensive IGBTs or MOSFETs. Stepped-approximation (modified sine wave) inverters simply switch the DC bus polarity at varying intervals to create a blocky waveform that mimics the RMS voltage of a sine wave. While this is fine for resistive loads (heaters, incandescent bulbs) and basic switch-mode power supplies, it will cause severe humming, overheating, and potential failure in AC motors, laser printers, and audio equipment due to the high harmonic distortion.