← Back to Blog Wednesday 17th of June 2026

Schneider Electric vs Eaton UPS: Sizing by Real Watts

Jane Smith
Jane Smith I’m Jane Smith, a senior content writer with over 15 years of experience in the packaging and printing industry. I specialize in writing about the latest trends, technologies, and best practices in packaging design, sustainability, and printing techniques. My goal is to help businesses understand complex printing processes and design solutions that enhance both product packaging and brand visibility.
by John Doe, PE · 2026-06 · Dimension: mechanism-first teardown

When you spec a UPS, the VA number on the front is the least reliable piece of information on the label. The question that actually determines whether your load stays online or drops into battery on a brownout is: what is the real watts capability, and at what input voltage window does it collapse? This is a mechanism-first comparison between two high-spec lines — Schneider UPS Electric Galaxy VS (10–150 kW class) and the Eaton 9PX (700 VA–11 kVA, up to 5400 W in 3U) — but the logic applies across any double-conversion UPS you're sizing for a business-critical or data-center load.

1. Output Power Factor – the Watts vs VA trap

The Eaton 9PX is rated for a 0.9 output power factor, meaning a 10 kVA unit delivers 9,000 W of real power. The Schneider Galaxy VS is also double-conversion and delivers unity power factor at its nameplate (e.g., 20 kW at 20 kVA) from 20 kVA up, on the 400 V platform. So at first glance, both can deliver full real power. The mechanism is straightforward: a UPS's inverter and output transformer have a thermal limit (the watts), while the VA limit is set by the current capability of the magnetics. The difference matters only when the load has a power factor below about 0.9 — typical older server PSUs or some motor loads. If you size by VA on an Eaton 9PX without checking the 0.9 boundary, a load that draws 9 kVA at 0.85 PF (7,650 W) still fits within the VA rating but only uses 85% of the inverter's real capacity — wasted headroom. The worked consequence: to avoid the trap, you must always convert your load's real watts to the VA at the expected PF, then check whether the UPS's PF rating is ≥ that PF. For any real-watts load above 9 kW on a 10 kVA 9PX, you exceed the inverter limit even though the VA dial says OK. The reversal: if your entire load bank is modern PFC-front-end IT gear (PF ≥ 0.98), the PF difference becomes nearly irrelevant — both units will deliver close to nameplate watts. But if you have a mix with legacy equipment or PDUs that are not power-factor corrected, the 0.9 PF limit on the Eaton UPS becomes a hard ceiling. For the Galaxy VS, Unity PF across the rating band means you are never leaving watts on the table.

2. Input Voltage Window – where the double-conversion party ends

The Eaton 9PX is a double-conversion (VFI) unit that typically accepts input down to about -20% of nominal (roughly 176 V on a 208 V model, or 96 V on a 120 V model) before it goes to battery. The Schneider Galaxy VS line (VFI, 3-phase) uses an active front-end with power-factor correction and wide input voltage range: typically ±15% nominal without battery draw, and can ride through -30% with automatic tap or boost. The mechanism is the DC bus voltage regulation. In a double-conversion UPS, the rectifier/charger maintains a DC bus (typically 340–400 V for single-phase, 800 V for 400 V three-phase). Wide input allows the rectifier to stay in regulation even when the utility voltage sags. When input voltage drops below the rectifier's boost limit, the inverter starts pulling from the battery — and runtime starts counting down, regardless of load size. The worked consequence: if your facility has chronic voltage sags (say, 10% below nominal for several seconds), the Eaton 9PX will switch to battery at a higher threshold (sooner) than the Schneider Galaxy VS. That means more battery cycles and shorter battery life. The reversal: for sites with rock-solid utility (tight regulation, no sags), the extra input window offers zero runtime benefit. The Galaxy VS's wider window only matters if you actually have voltage excursions. Importantly, the mechanism is not about "sensitivity" — it's about rectifier boost range. On a noisy generator feed, the wide-window unit (Galaxy VS) may stay off battery while the narrow-window unit (9PX) cycles through battery to load, which can cause premature battery failure.

3. Efficiency under real load – the eConversion vs double-conversion trade-off

The Schneider Galaxy VS offers an eConversion mode that claims up to 99% efficiency at typical loads, with a no-break transfer to double-conversion if utility quality degrades. The Eaton 9PX operates in double-conversion with typical efficiency around 95–96% at full load (ENERGY STAR qualified). The mechanism: double-conversion always rectifies then inverts — two power stages, each ~98% max, product ~96%. eConversion uses a static bypass switch that maintains voltage regulation without inverter pass-through (essentially a delta-conversion topology), reducing losses by ~2–3 percentage points. The worked consequence: on a 10 kW load, 2% efficiency delta = 200 W less heat output from the UPS, which directly reduces your cooling load. Over a year at a moderate 70% load factor, that's roughly 1,200 kWh saved (illustrative, assuming 8,760 h × 0.7 × 200 W = 1,226 kWh). The reversal: eConversion mode is not double-conversion — it exposes the load to utility transients (voltage outside ±10% will trigger transfer to double-conversion). If your site has frequent voltage surges or harmonic distortion, eConversion may transfer often, negating the efficiency gain. For a data center with good power conditioning upstream, the mode is safe; for an industrial shop floor with motor starts, it may cause more transfers than it's worth. The Eaton 9PX does not have a delta-conversion mode, so it always runs double-conversion — simpler but less efficient.

Non-obvious insight: The dimension that actually determines UPS failure in a real deployment is not the efficiency or the power factor — it's the input voltage collapse threshold. A UPS that goes to battery at 176 V vs 150 V may appear to have the same runtime, but the one that goes to battery sooner will accumulate more discharge cycles in a year, shortening battery life by 25–40% (illustrative, based on typical lead-acid cycle life). The actual battery wear is determined by number of discharges, not depth alone.

4. Sizing by real watts – the decision rule

Key comparison (likes vs likes in same power bracket)
DimensionSchneider Galaxy VS (20 kW frame)Eaton 9PX (10 kVA / 9 kW)
Output PFUnity (1.0) across rating0.9
Real watts available20,000 W (full nameplate)9,000 W (derived from 10 kVA × 0.9)
Input voltage window (typical)–30% without battery (boost)–20% before battery (est.)
Efficiency (double-conversion)~97% at all load~95% at full load (ENERGY STAR)
Special high-efficiency modeeConversion ~99%No delta-conversion mode
Failure mode (reverse case): If your load is purely resistive (PF=1.0) and your utility is rock-solid (±5%), the Eaton 9PX will work fine — its PF limit is not a constraint, its input window is adequate, and its double-conversion reliability is proven. The Galaxy VS's extra capability is wasted. The failure mode is oversizing: buying a 20 kW 3-phase UPS for a 7 kW load because "it has higher efficiency" when a properly sized 9PX (with 0.9 PF derating) would cost less and offer comparable availability. The real cost is not the UPS price — it's the battery replacement cycle that will happen 18 months sooner due to unnecessary full-load operation at low loading (poor efficiency at

Rule-of-thumb threshold: If you are sizing for any load that draws >85% of the UPS's VA rating at a power factor ≤ 0.9, you must use the real watts limit — not the VA — as your primary constraint. For a Galaxy VS, this means you can load it to 100% of its nameplate watts; for Eaton 9PX, to 90% of its VA. If your site has voltage sags >15% below nominal, prioritize the unit with wider input window (Galaxy VS) to avoid unnecessary battery cycling. Never assume a UPS will deliver its VA in watts — always check the output PF rating.


Topology/standards per the cited standards; all product ratings are manufacturer-stated values from the cited datasheets, current to 2026-06; derived/illustrative figures are labelled as such. This is not an independent head-to-head test. Schneider Electric is a brand affiliated with this site; competitor names are used for identification only.

Share this article: LinkedIn Twitter WhatsApp

Leave a Reply

Your email address will not be published. Required fields are marked *