Inverter efficiency explained: weighted efficiency and clipping

Inverter datasheets list several efficiency numbers, and a surprising number of installers and buyers only ever look at one — the peak figure at the top of the spec sheet. Peak efficiency is real but almost never representative of how the inverter performs across a normal day, because a PV system spends most of its operating hours well below its rated maximum power, at partial load, under whatever conditions the sun and clouds provide. Two weighting schemes exist precisely to correct for this: European weighted efficiency and CEC (California Energy Commission) weighted efficiency.

This article explains what those two figures actually represent, how MPPT efficiency is a separate and distinct number from conversion efficiency, and how clipping and DC/AC ratio decisions interact with all of the above to determine real annual yield. It builds on the tracking behaviour covered in MPPT explained.

The practical payoff is being able to read an inverter datasheet critically, and to size the DC array against the inverter's AC rating with a clear understanding of the trade-off being made.

Why a single efficiency number is misleading

An inverter's DC-to-AC conversion efficiency is not constant — it varies with load. Efficiency typically starts low at very light loads (a few percent of rated power), climbs through a broad plateau somewhere in the 20–100% load range, and often peaks around 50–75% of rated power before tapering slightly at full load, largely because internal losses have both a fixed component (switching losses, control electronics power draw) and a load-dependent component (resistive losses scaling with the square of current).

A PV array's output over a day, and over a year, spends the large majority of its hours at partial load — mornings, evenings, cloudy periods, and off-peak seasons — with only a modest fraction of hours near full rated output. So the efficiency that matters for annual energy yield is a weighted average across the realistic load distribution, not the single best point on the curve.

European weighted efficiency

European weighted efficiency was developed to reflect the load profile typical of Central European solar irradiance conditions, weighting the inverter's measured efficiency at several load points with fixed coefficients that emphasise the mid-range loads where the inverter spends most of its time:

η_EU = 0.03·η(5%) + 0.06·η(10%) + 0.13·η(20%) + 0.10·η(30%) + 0.48·η(50%) + 0.20·η(100%)

The dominant weight on the 50% load point (0.48) reflects the assumption that Central European irradiance conditions keep the inverter operating around half its rated capacity for a large share of productive hours. This weighting was defined in early standardisation work and has become the conventional shorthand figure quoted on European datasheets, typically 0.5–1.5 percentage points below the peak efficiency figure for a well-designed modern string inverter.

CEC weighted efficiency

CEC weighted efficiency, used primarily in the US market, applies a different weighting scheme reflecting Californian irradiance conditions, which run higher and flatter across the day than much of Central Europe:

η_CEC = 0.04·η(10%) + 0.05·η(20%) + 0.12·η(30%) + 0.21·η(50%) + 0.53·η(75%) + 0.05·η(100%)

Note the heavier weight at 75% load rather than 50%, reflecting the assumption of stronger, more consistent irradiance driving the inverter closer to full output more often. Because of this different weighting, the same physical inverter can report a European weighted efficiency and a CEC weighted efficiency that differ by several tenths of a percentage point, without either number being wrong — they describe different assumed load distributions.

MPPT efficiency is a separate number entirely

Both weighted efficiency figures describe DC-to-AC conversion loss only, measured with the DC input already held precisely at the true maximum power point by external test equipment. They say nothing about how well the inverter's own MPPT algorithm finds and holds that point in real operation — that is captured by a separate MPPT efficiency or static/dynamic MPPT efficiency figure, occasionally listed on more detailed datasheets, typically above 99.5% for modern inverters under standard test irradiance profiles. A full accounting of real-world yield loss therefore has at least three components: MPPT tracking efficiency, DC-to-AC conversion efficiency (weighted), and clipping losses from oversizing the DC array relative to the inverter's AC limit, discussed next.

DC/AC ratio and clipping

The DC/AC ratio (also called the inverter loading ratio, ILR) is the ratio of the array's total STC DC rating to the inverter's rated AC output — for example, a 6.6 kWp array on a 5 kW inverter has a DC/AC ratio of 1.32.

Oversizing the DC array relative to the inverter (a ratio above 1.0) is deliberate and common practice, because the array rarely produces its full STC rating in the field (real irradiance is usually below 1000 W/m², and modules run hotter than 25°C, both of which reduce output relative to nameplate). A modestly oversized DC array lets the inverter run closer to its efficient mid-load range for more of the day and captures more energy during low-light morning and evening hours, when the array's own output is naturally below the inverter's ceiling anyway.

The cost of oversizing is clipping: on the relatively rare occasions when the array's actual output would exceed the inverter's AC rating (clear, cool, high-irradiance conditions, typically around solar noon in the shoulder seasons), the inverter caps its output at its rated maximum, and the excess DC power is not converted — the MPPT tracker is deliberately pushed off the true maximum power point to limit AC output, so the lost energy is a genuine, unrecoverable loss rather than being stored or shifted.

Typical DC/AC ratios in temperate European conditions run around 1.1–1.3 for residential systems without inverter-level clipping becoming a material loss (commonly under 1–2% of annual yield in well-chosen ratios), but the appropriate ratio depends on latitude, typical irradiance, module orientation spread, and local temperature.

Worked example: clipping loss at DC/AC ratio 1.3

Take a 5 kW-rated inverter paired with a 6.5 kWp DC array (DC/AC ratio = 1.3). Assume a clear summer day where the array's actual output profile, before any inverter limit, would peak at 5.6 kW for a two-hour window around solar noon, tapering linearly to the inverter's 5 kW ceiling over an additional hour on each side.

Energy above the 5 kW cap during the flat 2-hour peak: (5.6 − 5.0) kW × 2 h = 1.2 kWh clipped

Energy above the cap during the two 1-hour tapering shoulders (triangular, averaging half the peak excess of 0.6 kW): 2 × (0.5 × 0.6 kW × 1 h) = 0.6 kWh clipped

Total clipped energy for that day ≈ 1.2 + 0.6 = 1.8 kWh

If the array's total unclipped output for that day would have been approximately 42 kWh, the clipping loss represents:

1.8 / 42 ≈ 4.3% loss, for that specific clear summer day

Across a full year, only a limited number of days reach conditions steep enough to clip at all, so the annual clipping loss for a 1.3 ratio in a temperate climate is typically much lower than this single-day figure — commonly in the 1–3% range — while the gain from running the inverter more often in its efficient mid-load band, and from capturing more shoulder-season and early/late-day energy, frequently outweighs the clipping cost. Sizing this trade-off correctly is one of the calculations behind the inverter sizing calculator.

Clipping loss by DC/AC ratio: rule-of-thumb ranges

These figures vary substantially with latitude, local cloud climatology, and module orientation (a system with east/west split arrays clips far less than an equivalent south-facing array of the same DC/AC ratio, since the two halves peak at different times of day).

Standards and references

• IEC 62109-1/-2: safety requirements for PV power converters, within which efficiency test methods are referenced.
• EN 50530: European standard defining the test method for overall efficiency of grid-connected PV inverters, including the MPPT efficiency and European weighted efficiency calculation procedure.
• IEC 61683: general procedure for measuring power conversion efficiency of power conditioning equipment used in PV systems.
• IEEE 1547: interconnection requirements for distributed energy resources, relevant to inverter behaviour and limits during grid support and curtailment.

Frequently asked questions

Is a higher peak efficiency number always better?

Not necessarily. Peak efficiency reflects a single load point that the inverter may rarely operate at in practice. Two inverters with identical peak efficiency can have meaningfully different European weighted efficiency if one holds its efficiency plateau more consistently across the 20–75% load range where most annual energy is actually produced.

Does clipping damage the inverter or shorten its life?

No. Clipping is a controlled limiting of output at the inverter's rated capacity, a normal operating mode the equipment is designed for, not a fault or stress condition. The inverter simply holds its AC output at the rated ceiling rather than allowing it to rise further.

Should DC/AC ratio be the same for every roof orientation?

No. A single south-facing array benefits from a more conservative ratio to avoid concentrated midday clipping, whereas east/west split arrays, or arrays with several different orientations feeding separate MPPT inputs, can often tolerate a higher DC/AC ratio because the combined output profile is flatter and less likely to breach the inverter's AC ceiling at any one time.

Why do some datasheets list both CEC and European weighted efficiency for the same inverter?

Manufacturers selling into both US and European markets test and certify to both weighting schemes since each region's certification and incentive programmes often require or expect the locally standard figure. The two numbers are not interchangeable and should not be compared directly against each other without noting which scheme produced them.

How does inverter efficiency interact with the payback calculation for a system?

Weighted efficiency and clipping losses both feed directly into the annual energy yield estimate used in a payback calculation — see the honest payback calculation for how yield assumptions, not just up-front cost, drive the economics.

Sources

• PVEducation: Inverters and efficiency — background on conversion efficiency and load curves.
• Fraunhofer ISE Photovoltaics Report — inverter efficiency benchmarking and technology trends.
• NREL: PV Performance Modeling Collaborative — modeling resources including inverter efficiency curves and clipping analysis.
• IEA PVPS — international performance and reliability reporting for grid-connected PV systems.