Clipping in Photovoltaics · Glossary

In a photovoltaic system, clipping refers to the limiting of the AC output power by the inverter. It occurs when the solar modules on the direct current (DC) side could provide more electrical power than the inverter can convert into alternating current (AC) and deliver. The potential generation above this limit is not used. In the generation profile this appears as a flattened power peak: the curve rises to the maximum inverter output and stays there, even though irradiance would temporarily allow a higher module output. The English term "clipping" (cutting off) describes exactly this picture.

DC side, AC side and what happens technically

The separation between the direct current side and the alternating current side is central to the concept. A solar module generates direct current, which the inverter converts into grid-compatible alternating current. Module output is usually stated in kilowatt peak (kWp), i.e. as output under standardised test conditions. For the inverter, by contrast, a maximum AC output is specified. It can deliver this continuously, but at high device temperatures often only to a limited extent. Clipping occurs when the DC power the modules can currently provide exceeds the usable AC power of the inverter.

Physically, the surplus energy is generally not first generated and then "destroyed" in the device. A grid-connected inverter normally seeks the maximum power point of the modules (maximum power point tracking). During clipping it deliberately leaves this point and shifts the operating point, usually to a higher voltage. The modules then deliver only as much power as the inverter can process. The modules do absorb the rest of the incident solar energy, but it is not converted into electricity and additionally heats the modules. Clipping is therefore usually a controlled operating state and neither a fault nor an overload of the device, as long as the inverter's DC input limits are observed, in particular maximum voltage and maximum current.

Deliberate oversizing: the DC/AC ratio

In practice, clipping is not a rare edge phenomenon. Many systems are deliberately planned with more module capacity than inverter capacity. This ratio is called the DC/AC ratio, internationally also the inverter loading ratio (ILR). An example: 12 kWp of module capacity on an inverter with 10 kW of AC output gives a ratio of 1.2. At first this looks like a loss, but it can increase the yield per kilowatt of inverter capacity and improve economic viability. The reason: modules only rarely deliver exactly their rated output. For large solar farms, the trend towards higher DC/AC ratios is well documented. A planning document by the US organisation Northwest Power and Conservation Council, for example, uses a ratio of 1.4 for reference plants.

The rated power of a module is a test value and not a permanently available operating state on a roof or in a solar farm. The power actually available depends, among other things, on:

Irradiance and clouds
Module temperature
Orientation and tilt
Dirt accumulation (soiling) and shading
Cable and mismatch losses
Module ageing

In many hours the DC power is therefore well below the rated power. A relatively smaller inverter then operates more often close to its rated output, while clipping occurs only in comparatively few sunny hours. What matters economically is therefore not whether clipping can be avoided entirely. The decisive question is whether the avoided peak losses justify the additional costs of a larger inverter, a stronger grid connection or other system components.

How large are the losses?

How much energy is clipped depends strongly on location, orientation, tracking, DC/AC ratio and loss assumptions. There is no single blanket figure. An NREL study of US solar farms with single-axis tracking, at a then-typical DC/AC ratio of 1.34 and an assumed 10 percent of other losses, arrives at an average of roughly 3.5 to 4 percent of energy clipped. For a tracking plant with a ratio of 1.4 at a high-irradiance site in Oregon, the Northwest Power and Conservation Council cites, based on model calculations, around 15 percent of potential annual generation. These values apply to specific large-scale plants and cannot be transferred directly to Central European rooftop systems. They do show, however, how strongly clipping grows with the DC/AC ratio and site conditions.

The calculation method also matters. Yield models using hourly values average out short irradiance peaks, for example when clouds pass by. They therefore tend to underestimate clipping losses and overestimate energy yield, especially at high DC/AC ratios. An NREL study demonstrates this using real measurement data. For planning heavily oversized systems, finely resolved simulations or measurement data are therefore advisable.

Distinction from curtailment, shading and faults

Clipping in the narrow sense must be distinguished from curtailment. With clipping, the technical design of the system limits the AC output, i.e. the interplay of module capacity, inverter capacity and operating strategy. Curtailment usually means a reduction of feed-in for reasons outside the conversion technology. Examples are interventions by the grid operator, statutory or contractual feed-in limits, or market-driven throttling during negative electricity prices. In the power curve, both processes can look similar because the same inverter often limits the output. Their causes, however, are different. In general usage, the capping caused by a fixed feed-in limit is sometimes also called "clipping". In that case it should be stated clearly which limit is meant.

Shading, degradation and technical faults are also something different. Shading reduces the power that can be generated because less light reaches the modules or partial shading affects entire strings. Degradation is the gradual ageing of the modules. A defect in the inverter can likewise reduce feed-in. Clipping, by contrast, precisely presupposes that the modules could deliver more than is output on the AC side. There are, however, interactions: in oversized systems, clipping can initially mask part of the soiling or ageing losses, because the modules still reach the inverter limit despite losses. According to the NREL analysis, this effect diminishes over the course of the system's lifetime.

Feed-in limit in Germany: the 60 percent rule

The topic becomes legally relevant where regulations are tied to power values. In Germany, § 9 paragraph 2 EEG 2023, in the version amended by the so-called Solarspitzengesetz (Solar Peak Act), requires the following for systems receiving a feed-in tariff or the tenant electricity surcharge: until an intelligent metering system with control device has been installed and successfully tested by the grid operator, the maximum active power feed-in at the point of connection to the grid must be limited to 60 percent of the installed capacity. This applies to systems below 25 kW. For systems from 25 kW to below 100 kW, the obligation of remote controllability is added. From 100 kW, obligations apply for remotely controlled reduction and for retrieval of the actual feed-in. Plug-in solar devices up to 2 kW of installed capacity and 800 volt-amperes of inverter output are exempt from the 60 percent limit. Under § 100 paragraph 3b EEG, the limit does not apply to older existing systems. According to the Verbraucherzentrale, it affects systems that went into operation from 25 February 2025.

The reference point is important. The limit applies to the feed-in at the grid connection point, not to the generation of the modules. Electricity consumed in the house or charged into a battery storage system does not fall under it. Dynamic limiting with measurement at the grid connection can therefore keep the loss considerably smaller than a fixed limit at the inverter. Higher self-consumption around midday works in the same direction. A limit of 60 percent of capacity does not mean that only 60 percent of the annual yield remains. Only the hours in which feed-in would otherwise exceed the limit are clipped. Changes in the law are frequent in this area, so the version currently in force should be checked before planning.

Significance for the grid and the power system

For the power system, clipping shows that installed capacity is not to be equated with maximum feed-in power. Debates, grid planning and market analyses often work with gigawatts of installed PV capacity. This figure says only a limited amount about how much power reaches the grid at a given point in time. That also depends on the DC/AC ratio, geographical distribution, orientation, inverter design, storage, self-consumption and feed-in limits. Two systems with the same module capacity can therefore have different maximum grid feed-in and different annual yields.

A system that cuts off some peaks at midday can deliver more energy in the morning, in the afternoon and in diffuse light than a system with fewer modules on an inverter of the same size. For grid operation this is often favourable, because high feed-in peaks are dampened and more energy flows through the same connection. Limited feed-in capacity can prompt operators to install more module capacity behind a limited connection. This can dampen the need for grid expansion in low- and medium-voltage grids. At the same time, part of the technically possible solar energy remains unused. For an honest assessment of renewable electricity generation, it should be disclosed what share is lost through design or requirements.

Two common misconceptions

First: clipping is pure waste. Viewed in isolation, every clipped kilowatt hour looks like a loss. Planning, however, evaluates costs and yields over the lifetime. If additional modules are cheaper than additional inverter and grid capacity, moderate clipping can make economic sense. A larger inverter would absorb the peaks, but in many hours would be only partly utilised.

Second: clipping is generally insignificant. With high DC/AC ratios, high-irradiance sites, tracking systems or additional feed-in limits, the amount of clipped energy can become considerable. East-west systems, by contrast, are usually less affected. Their daily profile is broader and their midday peak lower, which is why they are often deliberately planned with higher DC/AC ratios. Inverter temperature, reactive power provision and grid-technical requirements can also additionally restrict the usable active power. An inverter that has to provide reactive power has less headroom for active power at the same apparent power.

Power and energy

Clipping initially affects power, i.e. the electrical work currently delivered per unit of time. Over time, the loss adds up to an amount of energy, measured in kilowatt hours. An example: if 2 kilowatts are clipped for 15 minutes, 0.5 kilowatt hours are lost (2 kW × 0.25 h). For the annual balance, the sum of such situations counts. For grid operation, it additionally matters when they occur and whether they coincide with already high feed-in in the grid area.

Battery storage and hybrid inverters

With a battery storage system, the assessment changes. A storage system can absorb part of the energy that would otherwise be clipped if it has free capacity and sufficient charging power at that moment. It does not, however, automatically prevent clipping. If it is full, too small or already charged in the morning, the limit continues to apply. How much a storage system can capture also depends on the coupling. A DC-coupled storage system can absorb surplus module power before the inverter. An AC-coupled storage system can only use what has already passed the inverter. With a hybrid inverter, separate internal limits often apply for PV inputs, battery charging power, AC output power and grid feed-in. It must therefore be checked carefully which of these quantities a stated power limit refers to.

Conclusion

Clipping makes visible that photovoltaic systems cannot be understood solely through their installed module capacity. What matters are the power limits along the entire chain:

Module and string
Inverter
Storage
House connection and grid connection
where applicable, a statutory feed-in limit

The term does not describe a fault but a boundary between possible DC generation and usable AC output. Whether this boundary is a sensible compromise or an avoidable yield loss can only be judged on the basis of the specific system concept, the costs of the alternatives and the role of the system in the grid.

Sources

Erneuerbare-Energien-Gesetz (EEG 2023), § 9 Technische Vorgaben — Bundesministerium der Justiz / Bundesamt für Justiz (gesetze-im-internet.de); unbekannt (konsolidierte Fassung, abgerufen am 2026-11-11). § 9 Abs. 2 Satz 1 Nr. 1–3 und Satz 4; Fußnote zu § 100 Abs. 3b
The Effect of Inverter Loading Ratio on Energy Estimate Bias: Preprint (NREL/CP-5K00-82812) — National Renewable Energy Laboratory (NREL); 2022. Abstract
Quantifying the Impact of Inverter Clipping on Photovoltaic Performance and Soiling Losses (NREL, Dokumentnummer 87894) — National Renewable Energy Laboratory (NREL); 2024 (Geschäftsjahr laut Dokumentnummer, genaues Datum unbekannt). Abstract und Ergebnisteil (nur über Suchauszug geprüft)
2021 Power Plan: Utility-Scale Solar PV – Generating Resource Reference Plants — Northwest Power and Conservation Council; 2021 (genaues Datum unbekannt). Abschnitt Solar PV Trends, Considerations, and Resource Potential
EEG 2023/24: Was heute für Photovoltaik-Anlagen gilt — Verbraucherzentrale; unbekannt (Seite laut Suche 2026 aktualisiert). Abschnitt zu intelligenten Messsystemen und Steuerboxen

Related glossary entries

Edited by eCMS on 11.11.2026 17:28