5 Ways to Read PVSyst's System Losses｜Basics of Loss Rates

When advancing photovoltaic (PV) system design and generation forecasting in practice, it is easy to focus on annual generation and PR, but it is extremely important to understand the downstream losses that reduce the final result. What you should check in that case is PVSyst's System Loss. In PVSyst's documentation, the loss framework organizes, in addition to array-side losses, AC wiring losses, external transformer losses, auxiliary consumption, and downtime losses, and the main variables in the simulation results separately list inverter losses, AC ohmic loss, transformer loss, Aux_Lss, and UnavLss.

However, System Loss is not simple enough to be understood the moment you view it as a single value. There are losses from the inverter’s efficiency curve, as well as losses caused by low-output thresholds, overloads, and voltage and current limits. Furthermore, on the downstream side of the inverter, AC wiring, transformers, auxiliary equipment, and downtime have an impact, and the final E_Grid is evaluated at the point of injection. In other words, System Loss needs to be read as "an aggregation of losses in the downstream portion of the system."

Especially for practitioners searching for information on "how to read PVSyst", judging System Loss solely by whether the loss rate is large or small can lead to misprioritization of improvements. This is because, if it is confused with DC-side array losses, it becomes unclear which problems are due to irradiance conditions and which are due to equipment configuration or issues up to the injection point. The Loss diagram also displays each loss as a proportion of the energy at the preceding stage and explicitly states that they cannot be simply added.

This article organizes and explains how to read PVSyst's System Loss in practical work from five perspectives. First, it clarifies the position of System Loss, then summarizes how to view its breakdown, how to interpret differences between configured values and annual results, how to assess it on the Loss diagram, and how to trace back to monthly figures and on-site conditions. The basics of loss rates are arranged with an emphasis on readings that can be used directly in design, comparison, and explanation.

Table of Contents

• What does System Loss refer to?

• Interpretation 1｜Consider Array Loss and System Loss separately

• Interpretation 2｜Read the breakdown as "Inverter", "AC wiring", "Transformer", "Auxiliary equipment", "Shutdown"

• Interpretation 3｜Do not confuse the loss rate settings with the annual results

• Interpretation 4｜Read the Loss Diagram in terms of the sequence of losses

• Interpretation 5｜Refer back to monthly results and design conditions when evaluating

• Common misinterpretations of System Loss in practice

• The accuracy of on-site conditions affects the perceived validity of the loss rates

• Summary

What does System Loss refer to?

In the official PVSyst documentation, losses are organized under the broad heading "Array and system losses", and, in addition to IAM, soiling, irradiance, temperature, LID, module quality, mismatch, and DC wiring losses, AC wiring losses, external transformer losses, auxiliary consumption, and unavailability losses are listed. Furthermore, in the grid system result variables, EArrMPP denotes the MPP energy after array losses, EOutInv the inverter output, EacOhmL the AC wiring losses, ETrfLss the external transformer losses, Aux_Lss the auxiliary losses, and UnavLss the unavailability losses. In practice, it is easier to organize by reading this latter grouping as the System Loss.

What’s important here is to understand "System Loss" as a group of losses occurring downstream of the equipment. On the array side, factors closer to power generation—irradiance conditions, temperature, module characteristics, and DC wiring—have an effect. By contrast, System Loss refers to the losses that occur at the stage where that power is converted to AC by the inverter, transported through cabling, passed through transformers, used to operate required auxiliary equipment, and reduced by downtime, before ultimately becoming the amount injected into the grid. Since E_Grid is evaluated at the point of injection, System Loss should be treated as losses that directly lower the amount of electricity sold or injected into the grid.

Also, because PVSyst’s loss diagram defines each loss rate with reference to the immediately preceding energy amount, it is risky to casually interpret System Loss as a single aggregate rate. Just because one loss is large, you cannot understand the overall picture by simply adding it to other loss rates. From a practical standpoint, it is better to view System Loss as a concept for seeing what is affecting the system at each stage, rather than as a mere total percentage.

Approach 1｜Treat Array Loss and System Loss Separately

The first step in reading System Loss is to clearly distinguish it from Array Loss. PVSyst’s Array Loss deals with how much the ideal module output under STC conditions is reduced by shading, IAM, irradiance, temperature, module quality, mismatch, optimization equipment, DC wiring, and so on. Even in the official documentation, array losses are defined as the factors that reduce the PV module nominal power specified for STC conditions below.

In contrast, System Loss refers to losses that occur on the downstream side of the energy produced by the array. In practice, whether you can separate these affects the order in which improvements are made. For example, when power generation is lower than expected, unless you first determine whether the cause is array losses or System Loss, you won't know whether to revise the layout and module conditions or to review the inverter and AC-side conditions.

In PVSyst's "array losses, general considerations", it is explained that curtailment losses, energy not utilized because the inverter input voltage is outside its allowable range, and energy unused due to overload are, for MPP purposes, usually treated as inverter losses, i.e. on the system losses side. In other words, even the same "reduced" energy is split within PVSyst into losses assigned to the array side and losses assigned to the system side. Reading along with this classification makes it easier to see where the boundary lies between issues on the generation side and those on the power processing/transmission side.

A practical viewpoint in the field is to first check EArrMPP and the array loss group to confirm "how cleanly power has been generated up to this point," and then look at System Loss to confirm "how much was lost from that." With this order, the meaning of System Loss becomes immediately much more concrete. Treating Array Loss and System Loss separately is the fundamental basis of loss-rate analysis.

How to Read 2 | Read the breakdown as "Inverter", "AC wiring", "Transformer", "Auxiliary equipment", "Shutdown"

When reading System Loss in practice, it's easier to organize if you divide the breakdown into five parts. The first is inverter loss. In PVSyst variables, InvLoss is the global inverter loss, and as breakdowns IL_Oper, IL_Pmin, IL_Pmax, IL_Vmin, IL_Vmax, IL_Night, etc. are provided. In other words, the structure allows tracking not only the normal losses from the inverter efficiency curve but also output thresholds, overload, the MPPT voltage window, and night-time consumption separately.

The second is AC wiring losses. In PVSyst, the AC wiring losses from the inverter to the point of injection are determined from each conductor’s length, cross-sectional area, and metal type, with the basic formula evaluated as R×I². The documentation further explains that even if you set a nominal loss fraction, the actual annual energy loss depends on the irradiance distribution and is typically around 60% or so of the nominal loss fraction you set. In other words, you should read it with the assumption that the percentage you set and the percentage of the annual result will not match.

The third is transformer losses. In PVSyst, external transformer losses are treated as iron losses and copper losses. Iron loss is the no-load loss that occurs even when unloaded, and copper loss is the resistive loss proportional to the square of the current. The documentation also states that transformer losses can be specified as a percentage based on the reference nominal power, and in practice it is recommended to define them using the nominal power, iron loss, and global loss from the datasheet. In other words, when looking at System Loss, it is important not to lump transformers together, but to understand that no-load losses and load-dependent losses are mixed.

The fourth is auxiliary equipment loss, and the fifth is downtime loss. Auxiliary equipment loss is the energy used for system management such as fans, air conditioning, electronic equipment, and lighting, and is explained as being deducted from PV produced energy. A constant daytime load, an output-proportional load, and a fixed nighttime consumption can be set separately, and inverter-specific night loss is managed separately as IL_Night. Downtime loss can be defined as a time fraction or in days for maintenance or failure stoppages, but the actual energy loss depends on season, time of day, and weather, so it is said to have only statistical meaning. When reading System Loss, it is important not to compress these into a single loss rate, and to distinguish what is constant, what is condition-dependent, and what is statistical.

Reading 3 | Do not confuse the configured loss rate with the annual results

The most common misunderstanding about System Loss is treating the configured loss rate and the annual result as the same thing. In PVSyst, nominal or reference values are set during design for items such as wiring losses, transformer losses, and auxiliary losses. However, the annual losses accumulated by the simulation vary with the load and output at each time step, so the configured values do not directly equal the annual loss rate.

In the official documentation's ohmic losses section, it explains that because the loss fraction is proportional to power, the nominal loss factor alone is not reliable, and the annual result is often around 60% of the configured value. This is because wiring losses do not always occur at rated output; they vary over time according to R×I². In other words, reading “if AC wiring losses are set to 1.5% then annual losses will also be 1.5%” is incorrect.

The same applies to transformer losses. Iron (core) losses may appear constant even without load, but their effect on annual energy depends on operating hours and output conditions. Copper losses change with the square of the current, so the annual average is governed by the equipment’s operating distribution. Auxiliary losses also change the annual outcome depending on whether they are constant, only operate when a threshold is exceeded, or are proportional to output. In other words, the System Loss setting is the "input to the calculation," not "the result itself."

What's important in practice is that after reviewing the set values you always check the annual results using the loss diagram, monthly results, and simulation variables. Even if the set values are reasonable, if the actual results are larger than expected the cause may be the design conditions or the operating distribution. Conversely, settings that look large at first glance may not have that much effect on the annual results. Once you can understand this difference, interpreting System Loss becomes far more practical.

How to Read 4｜Read before-and-after relationships in a Loss Diagram

When checking System Loss, it is very important to read the loss diagram in terms of the before-and-after sequence. PVSyst's loss diagram visualizes the major losses used to assess the quality of a system's design, and can be displayed not only annually but also monthly. Most importantly, each loss percentage is defined as a proportion of the preceding energy amount. In other words, loss percentages cannot be simply added together; if you don't view them in the context of the sequential flow, you will misinterpret their meaning.

If you apply this way of thinking to System Loss, downstream losses always act on the energy that has already been subject to Array Loss. For example, in projects where output is heavily reduced by shading or temperature, the absolute amount of System Loss will look different even with the same inverter efficiency and the same AC wiring conditions. Conversely, in projects where the array side is very neatly arranged, System Loss tends to stand out relatively. That is precisely why you should not judge System Loss solely by its percentage; you must always check what the immediately preceding stage looks like.

A practical way to read a loss diagram is to split the loss diagram into "up to array losses" and "after system losses." First, check how the energy decreases from received light to EArrMPP, then track inverter losses up to EOutInv, and then follow AC wiring, transformer, auxiliaries, and unavailability up to the point of injection. Doing so makes it easier not only to see where System Loss is large but also to judge whether that magnitude is reasonable given the upstream conditions.

Also, because loss diagrams can be viewed by month, they are suitable for checking seasonal differences in System Loss. Items that are not visible from annual values alone—such as inverter overload, the load dependence of AC wiring, and day–night patterns of auxiliary equipment—become easier to interpret when broken down by month. If you want to understand System Loss, it is important to use the loss diagram not as a single result figure but as a diagram for reading the flow.

How to Read 5｜Return to the monthly results and design conditions to make a judgment

If you evaluate System Loss using only the annual value, you cannot tell which losses are affecting which seasons. PVSyst lets you check the loss diagram on a monthly basis, so when interpreting System Loss it is important to return to the monthly results. Since irradiance distribution and output levels change with the seasons, the effects of inverter efficiency, overload, AC wiring, and auxiliary equipment losses also change.

For example, if AC wiring loss and inverter overload-related losses stand out only during the summer, you should suspect load concentration at the output peak or the DC:AC ratio setting. For inverter-related items, IL_Pmax is defined as overload loss, IL_Pmin as low-output threshold loss, and IL_Vmin and IL_Vmax as losses due to the MPPT voltage window. Therefore, when viewed month by month, whether Pmax-related effects are pronounced only in midsummer or Vmin-related effects are pronounced only in winter will determine which design conditions need to be reviewed.

How "System Loss" is interpreted ultimately needs to be tied back to the design conditions. If wiring losses are a concern, you should review the distance, conductor cross-sectional area, and route design. If transformer losses are a concern, you should check how capacity is selected and how no-load losses are treated. If auxiliary equipment losses are large, there is room to re-examine parameter settings and operating conditions. Because downtime losses are treated statistically, you should confirm whether they are not being applied too heavily in individual cases and whether they are appropriate as a forecast.

What is most effective in practice is, after reviewing the System Loss, to clarify which design conditions you should revert to. Rather than stopping at the annual results, check monthly trends and breakdowns, and establish a workflow that feeds back into inverter selection, the DC:AC ratio, AC routing, transformer conditions, and auxiliary equipment settings so that reading the loss rate directly translates into improvement actions.

Common Misinterpretations of System Loss in Practice

The most common practical misunderstanding of System Loss is to lump it together with Array Loss. If you describe “system losses” as being large while including losses due to shading and temperature, it becomes unclear what should be corrected. In PVSyst’s documentation, array losses are factors that reduce the ideal array output under STC, while system-side losses such as inverter losses and AC losses are listed separately. If these are not separated, you will choose the wrong order of improvements.

Another common mistake is to assume that the nominal set value and the annual result are the same. Especially for wiring losses, it's easy to think “I set it to 2%, so it will be 2% for the year,” but PVSyst accumulates R×I² hour by hour, so the actual annual energy loss depends on the output distribution. The same applies to transformers and auxiliaries: the annual result changes depending on how the set conditions apply across operating hours and output ranges.

Furthermore, it is incorrect to add the loss rates in the loss diagram to describe the total loss. PVSyst shows each loss as a percentage of the energy at the previous stage, so summing the detailed losses does not yield the overall rate. When reading System Loss, you must always be aware of "which stage the percentage is relative to."

Another common mistake is to think of System Loss as merely an equipment-efficiency issue. In reality, stoppage losses are statistical, auxiliary equipment losses depend on operating conditions, transformer losses are a mix of no-load components and load-dependent components, and inverter losses are a mix of the efficiency curve and limiting conditions. Therefore, System Loss should not be dismissed with the single phrase "equipment performance"; it is important to read the characteristics by breakdown.

The accuracy of on-site conditions affects confidence in loss rates

Since System Loss is a downstream loss, it may appear to have little direct relation to on-site conditions. However, in practice the accuracy with which on-site conditions are understood greatly affects the perceived validity of System Loss. The reason is simple: inverter placement, AC wiring distances, transformer locations, the route to the injection point, and the operating conditions of auxiliary equipment all depend on the site’s spatial relationships and equipment planning. Even in PVSyst, AC wiring losses are calculated from conductor length and cross-sectional area, and E_Grid is evaluated at the injection point. In other words, if the accuracy of the on-site layout is low, the assumptions underlying System Loss themselves become ambiguous.

Even an AC route that looks short on the drawings can end up longer in reality because of cable routing, switchgear locations, or transformer placement. That difference directly affects the annual AC ohmic loss. For external transformers, wiring conditions and the way losses appear change depending not only on capacity and connection configuration but also on where they are located. Auxiliary equipment consumption may also require re-evaluating settings if cooling systems or equipment room conditions on site change.

Therefore, if you want System Loss to be a reliable figure in practice, it is important not just to assign a loss rate on paper but to grasp the on-site positional relationships with high precision. If the relationships among inverters, panels, transformers, and injection points are known precisely, it becomes easier to refine the assumptions for AC wiring and equipment placement, and confidence in the System Loss figures increases. Downstream loss is not an "abstract rate independent of the site" but also a reflection of the actual layout plan.

In that sense, in practical work where you want to grasp on-site positional relationships with high precision, you naturally turn to LRTK, the iPhone-mounted GNSS high-precision positioning device. Being able to organize on-site, with high accuracy, the positional relationships of equipment layout, cable routes, and injection points makes it easier to tighten the assumptions for System Loss entered into PVSyst. When you don't want System Loss to remain just a set value but want to bring it closer to numbers used in detailed design, this level of precision in on-site understanding is a great help.

Summary

When reading PVSyst's System Loss, first separate and position it relative to Array Loss; then check the breakdown into "Inverter", "AC wiring", "Transformer", "Auxiliaries", and "Shutdown", avoid confusing configured values with annual results, view the loss diagram in terms of sequence, and finally return to the monthly results and design conditions before making a judgment. By keeping these five perspectives in mind, System Loss becomes not merely a loss rate but practical information for assessing the quality of the downstream design.

The important thing is not to treat System Loss as a single number. Inverter efficiency, overloading, voltage window, night consumption, AC wiring, transformers, auxiliary equipment, and shutdowns each have different characteristics. If you separate and examine them, you can see which are configuration issues, which are operating-condition issues, and which are statistical assumptions. The basis of loss rates is not memorizing the rate itself, but understanding the meaning of the rate in context.

And to make that interpretation even more reliable, it is essential to grasp the on-site positional relationships with high precision. If you want to organize the positional relationships of inverters, panels, transformers, and injection points more accurately, the perspective of utilizing LRTK, an iPhone-mounted GNSS high-precision positioning device, is also useful. By combining the ability to correctly read PVSyst’s System Loss with the ability to accurately capture the site, it becomes easier to arrive at loss-rate settings and power generation forecasts that are more convincing.