How to Set Losses in PVSyst | Explained in 6 Beginner-Friendly Items

Table of Contents

• Why Loss Settings Are Important in PVSyst

• Basic Approach to Keep in Mind Before Configuring Loss Settings

• Loss Item 1: Losses Due to Incidence Angle

• Loss Item 2: Losses Due to Soiling

• Loss Item 3: Losses Due to Temperature

• Loss Item 4: Losses Due to Mismatch

• Loss Item 5: Losses Due to Wiring

• Loss Item 6: Losses Due to Downtime

• Configuration Mistakes Beginners Often Make

• How to Proceed in Practice

• Summary

Why Loss Settings Are Important in PVSyst

In PVSyst, losses are treated not simply as correction values but as central information for interpreting simulation results. The results screen includes a loss diagram that lets you progressively confirm, starting from the nominal potential energy production, how much energy is reduced by each factor. Because monthly, daily, and hourly values can be reviewed, the structure makes it easier to understand in which seasons or under what conditions losses are effective than by looking only at the annual total. In other words, loss settings are both an input task and an analytical process for exposing weaknesses in the design.

A common situation in practice is that orientation and tilt settings, equipment configuration, and the selection of irradiance data are handled carefully, while losses are left at their default values. However, even with the same layout and the same capacity, results change depending on how easily soiling accumulates, ventilation conditions, wiring length, per-row variability, and the presence or absence of periodic shutdowns. PVSyst itself provides multiple loss categories—incidence angle, soiling, temperature, mismatch, wiring, shutdowns, and so on—to deal with the effects of losses in detail. Therefore, rather than judging a system solely by its energy production figure, checking the breakdown of the losses that led to that figure is the quickest way to improve practical reliability.

Basic Principles to Keep in Mind Before Setting Losses

What beginners should know first is not to try to finalize the loss settings in one go. The official PVSyst documentation also recommends that loss parameters are populated with reasonable initial values in the early stages, and that you should run the first simulation and then refine the loss factors to match project-specific conditions. This is a very practical approach. If you try to nail down all the details from the start, you will end up deciding numbers before the assumptions are settled, which can lead to misguided settings.

Another important point is to structure loss items so their meanings don't overlap. For example, if you have already evaluated shading effects in the detailed shading settings, adding the same effect again as a separate loss item leads to double counting. Electrical losses from partial shading are handled by a dedicated mechanism that includes module layout and string connections, and this is typically evaluated at the final stage once shading and system definitions are finalized. Therefore, for loss settings aimed at beginners, it's easier to stay organized by first covering the six basic items and managing detailed shading evaluations separately.

Furthermore, input values should prioritize justification over appearance. Use values that have some explainable basis—such as field performance, past projects, maintenance plans, drawings, cable lengths, installation methods, and regional characteristics—and be sure to note provisional values so they can be reviewed later. Treat loss settings not as a task of finding a single perfect answer but as updatable assumptions that become more realistic as your understanding of the project grows; this makes them easier to handle in practice.

Loss Item 1 Loss due to Angle of Incidence

The first item is the incidence-angle loss that occurs when light enters a surface obliquely. In PVSyst, incidence-angle loss is treated as a measure of how much the irradiance reaching the cell surface is reduced compared with sunlight entering perpendicularly. The main cause is reflection at the surface: as the incidence angle increases, reflection increases and the transmitted energy decreases. Because this affects not only the direct component on clear days but also the diffuse component and the ground-reflected component, it influences annual energy production more than it might appear.

What beginners should do for this item is first to understand the default handling. PVSyst calculates incidence-angle losses as an angle-dependent function and has a standard model. Therefore, in the early stages of a project it is safer to use the default model to observe the overall results rather than forcibly replace it with custom values, and only revisit it later if there are grounds based on the specifications of the selected components or on measured data. In particular, finely adjusting this item alone while other losses are still uncertain is unlikely to improve overall accuracy.

As a practical approach, it is appropriate to treat incidence angle loss as a fundamental optical loss that reflects surface conditions and installation conditions. For projects with low solar elevation in winter, morning and evening incidence conditions, or many oblique incidences on sloped surfaces, checking how losses occur by month as well as the annual total increases confidence in design decisions. In PVSyst you can track these changes with loss diagrams and monthly results, so it is important not to pick a single number and stop, but to adopt an attitude of examining seasonal differences.

Loss Item 2: Losses Due to Soiling

The second is losses due to soiling. In PVSyst's official documentation, soiling losses are said to be strongly influenced by rainfall, and therefore can be defined as monthly values. Also, in simulations, soiling losses are treated as losses of solar irradiance. This is important. In other words, soiling should be considered not simply as a reduction in downstream power, but as a factor that reduces the energy that can be received in the first place.

A common mistake beginners make is entering a single uniform value for the entire year and stopping there. Of course, a uniform value is fine for a rough estimate early in a project, but if you want to improve accuracy in practice, you should at least set values that account for seasonal differences—considering dry periods, times with a lot of yellow sand or dust, periods when cleaning by rainfall can be expected, and whether regular cleaning occurs. The reason PVSyst supports monthly inputs is precisely because soiling is seasonally dependent. In environments with little rainfall and a lot of dust, exhaust, or deposits from agricultural land, underestimating soiling tends to lead to an overestimation of annual results.

As a practical matter for setting parameters, the starting point is to first review the maintenance plan and organize the cleaning frequency and timing. On that basis, combine the regional conditions with the installation environment. Locations along main roads, near factories, on reclaimed or newly developed land, around quarries, or close to the coast will differ in both the nature of the soiling and the accumulation rate. Conversely, for projects where regular rainfall can be expected and surrounding dust is minimal, overly conservative soiling assumptions can lead to underestimation. What is important is not to reuse a uniform conventional value for soiling losses, but to determine them tied to the operational plan and site conditions.

Loss Item 3 Losses Due to Temperature

The third is losses due to temperature. PVSyst explains that the nominal performance of the modules is specified at 25℃, whereas during actual operation the array temperature becomes considerably higher than that, causing temperature losses. The official documentation indicates that losses of roughly −0.2% to −0.4% per 1℃ increase can occur, which has a non-negligible impact on annual energy yield. Furthermore, PVSyst employs a two-layer thermal model that handles steady-state temperature and actual transient temperature, calculating temperatures using heat transfer coefficients and similar parameters.

The essential point beginners should grasp in this topic is that temperature losses are not determined by ambient temperature alone. Installation method, ventilation quality, rear-side airflow, clearance from the roof, racking conditions, and layout density all affect temperature rise. Therefore, even within the same area, results differ between installations where wind passes through easily and installations where heat tends to accumulate. The reason PVSyst uses heat transfer coefficients to handle temperature is precisely because the installation structure affects the balance between heat generation and heat dissipation. If temperature losses are underestimated in rooftop projects or densely packed layouts, the apparent power generation will tend to be higher.

In practice, when refining temperature-related losses, it is important not to make judgments based solely on the average temperature from meteorological data. First, capture the overall trend using standard settings, and then prioritize revisions where the mounting method departs from typical conditions. For example, a ground-mounted installation with good rear ventilation and a rooftop installation where heat tends to accumulate will not exhibit the same loss characteristics even at the same capacity. If measured array temperatures are available, PVSyst can be operated to use measured temperatures instead of calculated ones. The larger the project and the more strongly temperature assumptions affect profitability, the more worthwhile it is to handle this carefully.

Loss Item 4: Losses Due to Mismatch

The fourth is losses due to mismatch. In PVSyst, mismatch losses are defined as the difference between the sum of the maximum output of the individual, independent submodules and the maximum output obtained from the I-V characteristic of the entire array as actually assembled. In other words, although each unit could produce more on its own, connecting them in series or parallel causes them not to operate in unison, reducing the overall output. Causes include variations in module characteristics, non-uniform soiling, differences due to aging, partial shading, voltage differences between strings, and temperature differences.

For beginners, what matters is not to overestimate mismatch excessively and to clarify what should be included in this item. PVSyst has default values for mismatch, and its official documentation explains that losses due to variability in module characteristics are quite small when variability is below 2% and often stay under 0.5%, while they increase sharply when variability is large. In other words, assigning large values without justification tends to produce results that are harsher than reality. For projects with stable procurement quality and installation under uniform conditions, it is appropriate to start with the default or conservative values and update them if there are operational records or acceptance data.

Also, electrical mismatch caused by partial shading is handled in specialized detailed shading calculations. Therefore, in projects that evaluate shading in detail, substantially adding the same effect to general mismatch losses can result in double counting. At the beginner stage, it is safer to separate basic mismatch — which accounts for module variability and differences between columns — from the detailed electrical losses that originate from shading. In other words, mismatch loss is a convenient lumped coefficient, but precisely because it is convenient, it is important not to apply it at a large magnitude while leaving its contents ambiguous.

Loss Item 5: Losses Due to Wiring

The fifth is losses due to wiring. In PVSyst, wiring losses are treated as Joule losses based on conductor resistance, and the basic formula is determined by the resistance and the square of the current. What is important here is that, even if it appears as a percentage on the input screen, it is treated internally as a resistance value. Furthermore, the official documentation explains that because operational losses are evaluated according to the current at each moment, the specified nominal loss rate and the actual annual energy loss do not match, and in annual results the actual loss is often around 60% of the specified nominal loss rate.

This item is a part that beginners tend to set based on intuition. However, wiring losses are influenced by cable length, cross-sectional area, circuit configuration, temperature, and current conditions, so they should properly be determined based on drawings and wiring plans. In particular, for projects with large facility scale, long cable runs, or a high degree of consolidation, errors in wiring loss are likely to translate directly into errors in revenue forecasts. Rather than setting values simply by the customary allowable voltage-drop practice, reviewing them based on the actual lengths and cross-sections makes it easier to fulfill the simulation’s accountability.

A practical tip for field work is to use rough estimates in the early stages of a project and update them once wiring diagrams and construction plans become concrete. In PVSyst, wiring losses can be tracked as an independent loss within the system rather than treated as a mere final correction. Therefore, re-evaluating at three stages—initial design, detailed design, and pre-construction verification—not only improves the accuracy of the numbers but also makes the impact of design changes easier to see. Especially for beginners, prioritizing the ability to explain why a given wiring loss value was chosen, rather than trying to make the loss appear small, will make the approach more practical in real work.

Loss Item 6 Losses Due to Operational Downtime

The sixth is losses due to downtime. In PVSyst, the system’s unavailability can be defined as a time fraction or as a number of days, and the system is treated as stopped during that period. You can also set specific downtime periods, and for unpredictable outages there is an option to generate periods randomly. This is useful when you want to reflect scheduled maintenance, planned shutdowns, fault-related stoppages, or grid-related outages in power generation forecasts.

Beginners tend to simulate power generation facilities on the assumption that they will run continuously. However, in practice, planned values that assume zero downtime can be difficult to handle both when explaining them and in operation. In particular, for projects where annual inspections are specified in the maintenance contract, projects prone to shutdown work in certain seasons, or projects that have experienced shutdowns due to grid constraints or troubles in the past, it is not realistic to ignore downtime losses entirely. PVSyst treats this item separately in order to reflect operational realities in the planned values of power generation facilities.

One tip for setting this is not to overestimate downtime and be overly conservative. Downtime loss reflects how you view the equipment’s reliability and maintenance regime. If you have past performance data, use it; if not, it’s better to start with a modest provisional assumption aligned with your maintenance plan and inspection frequency. Wanting to guard against generation shortfalls by inflating soiling, mismatch, temperature, and downtime all at once leads to an underestimation with little basis. It’s important to treat downtime as a separate, independent category and to distinguish its role from other losses.

Common configuration mistakes beginners often make

One common mistake beginners make is trusting default values too much. PVSyst's default values are convenient for starting an initial study, but they do not automatically reflect project-specific conditions. The official guidance also advises defining each loss factor to match the project after the first simulation. Therefore, rather than treating the default values as the values to submit, you should use the initial results as a starting point and check which losses will vary by project.

The second mistake is entering losses redundantly. It is especially common to lump soiling-induced variability, partial-shading–induced variability, and general mismatch losses together into one large coefficient. In PVSyst, electrical losses from soiling, mismatch, and shading are handled by separate mechanisms, so failing to distinguish their roles leads to double-counting. Loss settings do not become more precise by adding more items; accuracy only improves when each role is properly separated and entered.

The third mistake is looking only at annual values. PVSyst’s results can be checked by month, by day, and by hour, and loss diagrams are also available. For example, for soiling check whether it is biased toward the dry season; for incidence angle check whether it is biased toward winter or mornings and evenings; for temperature check whether it becomes excessive in summer; and for downtime check whether planned stoppages in specific months are reflected—inspecting these makes it easier to verify the validity of the input values. Even if the annual values look plausible, if the underlying details are unnatural you should review the settings.

How to proceed in practice

A practical recommendation for day-to-day work is to first create a single baseline case. In it, using default values as the foundation, minimally reflect only the items likely to vary significantly between projects—dirt, temperature, wiring, shutdowns, etc. Look at the results, check the loss diagram, and identify which items are dominant. If any loss is larger than expected, trace back and inspect the assumptions for that item. This allows you to improve accuracy while preserving the overall structure, rather than diving straight into the details.

Next, update the loss items each time more project information becomes available. When the wiring plan is decided, update the wiring losses; when the mounting structure is finalized, review the temperature losses; when the maintenance plan is decided, update soiling and downtime; and when acceptance quality and construction conditions are known, reassess mismatch. Treating losses as something you cultivate in line with design progress makes simulation results easier to explain. PVSyst has a structure that makes results easy to compare, so it pairs well with the practice of laying out and checking multiple cases with different assumptions.

Finally, prioritize well-founded reasonableness over fine-grained detail that lacks justification. For example, even complex loss settings listed down to decimal places are not persuasive if their basis is vague. Conversely, settings that can be explained—such as monthly soiling, temperatures appropriate to the installation method, wiring losses based on wiring diagrams, and shutdowns aligned with the maintenance plan—are stronger in practice. Simulations are for decision-making, not for show. For that reason, loss settings should be structured so they can fulfill accountability.

Summary

When setting losses in PVSyst, you don't need to decide every detail from the start. First run a simulation, grasp the overall picture from the loss diagram, and then review the six items—angle of incidence, soiling, temperature, mismatch, wiring, and downtime—in that order to match the project conditions. Simply following this workflow will greatly increase the credibility of the power generation forecast. Setting losses is not a process of lowering numbers; it is the process of transferring the realities of the site into the simulation.

If you really want to improve the accuracy of loss settings, it is important not only to rely on desk-based inputs but also on how accurately you can ascertain on-site conditions. If terrain undulation, verification of installation positions, geotagging of recorded photos, and understanding of surrounding conditions remain vague, assumptions about soiling, temperature, wiring, and shutdowns also tend to become blurred. For practitioners who want to streamline such on-site assessment, the smartphone-mounted high-precision GNSS positioning device LRTK is also a strong option. If you can accurately capture and record the site’s location information, it becomes easier to iterate between organizing simulation assumptions and on-site verification, which in turn helps improve the accuracy of power generation forecasts.