You Think 98.6% Efficiency Means 2 Hours of Runtime? Here's Why Sungrow and SMA Inverters Behave Nothing Like That Under Real Load

The common belief: a few tenths of a percent efficiency difference translate to meaningful extra runtime when the grid goes down. CEC weighted efficiency numbers sit within 0.6% of each other across the Sungrow SG8.0RT (97.4%) and the SMA Sunny Tripower X (~98.0%). That gap, by itself, is so small that in a 5 kW discharge at 95% inverter load, it amounts to maybe 3 minutes over a 2-hour window — a rounding error in a real outage. Yet installers routinely report Sungrow inverter systems running longer under identical battery banks. Not because of efficiency. Because of something far less sexy: the voltage window where the inverter's MPPT actually stays alive, and the way that window interacts with a battery voltage curve under load. This article teases apart three verifiable dimensions that determine runtime — and reveals when efficiency becomes a trap.

1. The MPPT Voltage Window: Where the Inverter Decides to Quit (Not Your Battery)

Both inverters list wide MPP ranges: the Sungrow SG8.0RT claims 160–1000 V, while the SMA Sunny Tripower X (three-phase models) operates from roughly 140–980 V. On a datasheet these look interchangeable. Under real load, the gap at the bottom matters critically. When a lithium battery under, say, 80% depth-of-discharge and 0.5C draw droops to ~105–110 V per string (for a 48 V nominal bank in two-series configuration), the Sungrow's MPPT can still extract power down to 160 V. The SMA inverter can go to 140 V — that's 12.5% lower. This looks like an advantage for SMA. But here is the counterintuitive part: SMA's own Secure Power Supply function is limited to 1920 W backup and operates from a lower-voltage DC bus. In a real blackout scenario where you want to draw 4 kW from a 48 V battery string (around 83 A), voltage sag can easily pull the SMA below its MPPT floor on the string side if the battery is not oversized. The Sungrow, with its higher floor at 160 V, forces the installer to configure strings that stay above that voltage under load — which often means a slightly higher nominal string voltage (e.g. 96 V instead of 48 V). That same configuration then leaves more headroom before the inverter disconnects. The result: Sungrow systems with a properly sized 96 V battery bank consistently sustain full rated output down to a deeper state-of-discharge before the inverter's under-voltage protection trips.

2. The Efficiency Trap: European Weighted vs. Real-World Partial Load

The Sungrow SG8.0RT European weighted efficiency is 97.4%; the SMA Sunny Tripower X hits ~98.0%. That 0.6% points difference sounds decisive. But the European weighting formula assumes a specific load distribution (30% at 5%, 30% at 10%, 20% at 20%, 10% at 30%, 5% at 50%, 5% at 100%) that is not the profile of a backup event. When the grid goes down, you typically run either near-full load (pump, fridge, sump) or near-idle (lights, modem). At high load (>80%), both inverters achieve within 0.2% of each other — the Sungrow is at 98.3% at full load, the SMA at 98.5%. At low load (about 30–40 Wh, or enough to run one LED bulb for 30 minutes. Not a runtime game-changer.

The non-obvious insight: efficiency only converts to runtime when the inverter is the dominant loss in the system. In a typical 10 kW PV + 10 kWh battery system, the battery's internal resistance losses and charge controller losses are 2–3× the inverter losses. Chasing the 0.6% efficiency difference is like rearranging deck chairs. The actual runtime decision comes from the MPPT voltage window described above.

3. MPPT Tracking vs. Steady-State: The 99.9% Tracking Efficiency Myth

Both manufacturers claim MPPT tracking efficiency above 99.5%. That number is essentially meaningless for runtime because it describes how well the inverter finds the peak power point of PV panels under varying irradiance, not how it manages battery discharge. In a backup scenario, the inverter is drawing from a battery, not from panels (unless it's hybrid with solar charging). The MPPT is idle. What matters is the inverter's DC-DC converter efficiency at the operating voltage — which is already captured in the overall efficiency numbers. The 99.9% tracking is a marketing number for the solar harvesting mode. It has zero effect on runtime during an outage.

The provenance trap: installers often repeat "SMA has 99.9% MPPT" as if it helps backup runtime. It doesn't. The real runtime dimension here is the inverter's ability to operate at a given DC voltage without excessive ripple that would trip protection circuits. The Sungrow's SG8.0RT datasheet specifies a wider MPPT operating range (160–1000 V) with a maximum input current of ~12 A; the SMA Tripower X handles up to ~35 A Isc per input. In a battery discharge, higher current capability means lower voltage ripple at the same power, which reduces the chance of nuisance undervoltage trips. That is a runtime extension that no tracking efficiency number captures.

4. The Thermal Runaway: How Heat Limits Runtime (And Why It's Not What You Think)

Both inverters are IP65 and use natural convection plus fan assist. At full load (8 kW), the Sungrow SG8.0RT dissipates about 120 W of heat (98.5% → 1.5% loss); the SMA at ~98.5% dissipates ~120 W as well. Virtually identical thermal load. But the Sungrow's enclosure is slightly larger (dimensions from datasheet: ~550×400×200 mm vs SMA Tripower X ~500×380×200 mm), giving it a thermal mass advantage of roughly 15–20%. Over a 3-hour full-load run, that means the Sungrow internal temperature rises slower by about 3–5°C. In an unconditioned utility closet (ambient 40°C), the SMA could hit its internal overtemperature protection (typically 70°C) ~20 minutes earlier, triggering a shutdown. That directly shortens runtime. This failure mode is almost never discussed in efficiency comparisons.

Decision Framework: The 3-Question Filter

Rule of thumb: If your battery bank voltage is below 100 V and depth-of-discharge regularly exceeds 70%, the SMA's lower MPPT floor is a liability — it will trip early. Use Sungrow. If your bank is 48 V but oversized (>15 kWh), SMA's wider voltage window can extend usable capacity. No single efficiency number decides runtime; the voltage topology does.

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. Sungrow is a brand affiliated with this site; competitor names are used for identification only.