Inverter efficiency curves matter far more than the single peak number on the front of a datasheet. That headline figure is the best result the unit achieves at one specific load and one specific input voltage. A real array spends most of its hours well below rated power, at input voltages that drift with temperature and irradiance, and the efficiency it actually achieves over a year comes from the whole curve, not from its highest point.
What the curve actually shows
Plot efficiency against output power as a fraction of rating and you get a characteristic shape: poor at very low load, rising steeply, flattening across a broad middle plateau, then easing down slightly near full output. The low-load penalty exists because the inverter has fixed overheads that do not scale with throughput. Control electronics, gate drive, cooling fans, monitoring and the magnetizing losses in transformers and inductors all consume power whether the array is delivering a hundred watts or five thousand. At dawn, at dusk, and on overcast days, those fixed losses are a large fraction of what is coming in. Near full output, conduction and switching losses in the semiconductors dominate instead, which is why the curve droops slightly at the top. A good datasheet publishes the whole family of curves, usually at several input voltages.
Why input voltage changes the answer
Most inverters have a direct-current input voltage at which their conversion topology is happiest, and efficiency falls off on either side of it. That is why the datasheet shows separate curves for low, nominal and high input voltage. String design interacts with this directly: a string sized so the array operates near the inverter’s optimal voltage through the middle of the day will convert more efficiently than one that spends the day at the bottom of the tracking window. Because module voltage falls as cells heat, a string that sits comfortably in the sweet spot on a mild morning may drop toward the low end on a hot afternoon. Design software accounts for this, which is another reason to trust modeled annual output over back-of-envelope reasoning.
How this interacts with array-to-inverter sizing
Designers commonly install more module capacity than the inverter’s alternating-current rating, because the array rarely reaches its nameplate and because a slightly loaded inverter is an inefficient one. Pushing the ratio up keeps the inverter operating on the flat part of its curve for more hours, which raises the annual conversion efficiency and improves the economics of the inverter itself. Push it too far and the inverter spends midday hours limiting output, which throws away energy. The optimum depends on the climate, the array orientation, the inverter’s own curve and the value of the energy. There is no universal correct ratio; the inverter manual states the maximum the manufacturer permits, and modeling with local weather data finds the sensible point within that limit.
Heat, derating and the curve you get in August
Published efficiency curves are measured under controlled conditions. An inverter mounted in direct sun on a south wall, or in an unventilated closet, or behind a fence that blocks airflow, will run hotter and may reduce its own output to protect itself. Thermal derating shows up as a flat-topped production curve on the hottest, sunniest days, which owners often misread as clipping. Placement is therefore part of efficiency: shaded, ventilated, at the installed height and clearance the manual specifies, away from heat sources, with the cooling fins and filters accessible for cleaning. This is one of the few efficiency levers an owner can still pull after commissioning.
Reading a datasheet like a designer
Look for the full curve family rather than a single number, note the load fraction where the plateau begins, and check the input voltage at which peak efficiency occurs against where your string will actually operate. Then look at the weighted efficiency figures, which compress the curve into a single number using an assumed distribution of operating conditions. Those weighted figures are far more representative of a real year than the peak. Finally, check the night-time or standby consumption, because an inverter that draws power all night gives back some of what it earned during the day.
What to check before you act
- Ask for the full efficiency curve family, not the peak number.
- Check where your string voltage sits relative to the inverter’s best input.
- Mount the inverter shaded and ventilated per the manual’s clearances.
- Compare weighted efficiency figures rather than headline peaks.
Common questions
Does a bigger inverter always waste less?
No. An oversized inverter spends more hours at low load, where the curve is worst. Matching the inverter to the array and the local climate matters more than buying headroom.
Is a one percent efficiency difference worth paying for?
Rarely, compared with siting, shading, string design and installation quality. Those factors move annual production far more than a small difference in the conversion curve.
The practical takeaway
Judge an inverter by the shape of its efficiency curve across load and input voltage, not by its best single measurement. Size the array-to-inverter ratio with modeling rather than rules of thumb, and mount the unit where it can stay cool. Installation and placement are qualified-installer work.
References and further reading
- NREL: photovoltaics research including inverter and system performance
- NREL PVWatts: models annual production including system loss assumptions
- NREL: a technical report from the laboratory’s photovoltaics publication library
Numerical scenarios are illustrative unless identified otherwise. Follow the exact product instructions; component ratings and local installation requirements can differ.
