Record PV Assumptions Before Comparing Energy Estimates

Summary

Compare PV energy estimates only after recording the module, array, weather, inverter, bifacial, albedo, soiling, shading, snow, and availability assumptions used in the model.

Key Takeaways

• PVWatts estimates depend on inputs and model assumptions; a higher kWh number is not automatically a better design.
• Record model version and weather source because PVWatts V8 uses updated models and NSRDB TMY data compared with V6.
• Module type should be documented as standard, premium, thin film, or bifacial with stated bifaciality where applicable.
• Array type changes how tilt, azimuth, and GCR are interpreted, especially for tracking and backtracking systems.
• Separate loss categories to avoid double counting soiling, shading, snow, degradation, and availability losses.
• SOLARTODO-specific specifications, pricing, certifications, stock, savings, and performance remain unknown unless supported by project documents.

Why the assumptions log matters

Before comparing two solar PV energy estimates, record the modeling inputs first. PVWatts is designed to estimate PV production from basic system information while making assumptions about modules, inverter behavior, and other system parts, so two estimates can differ because the inputs changed rather than because one supplier has a better product. [SAM PVWatts System Design, n.d.] PVWatts V8 also uses updated weather data and upgraded module, inverter, and thermal-effect models compared with Version 6, so comparisons should identify the model version as well as the system assumptions. [PVWatts V8 API, n.d.]

For B2B procurement, the safest practice is to treat any early energy estimate as a model result, not as a guaranteed output. The supplied sources do not provide SOLARTODO module specifications, deployment records, certifications, stock, price, savings, or performance guarantees. Those items should be marked unknown until supported by a project quotation, datasheet, test report, contract, or site-specific simulation.

Module assumptions to record

Record DC nameplate capacity in kWdc, because PVWatts defines system nameplate capacity as the DC power rating of the array at standard test conditions. [SAM PVWatts System Design, n.d.] Also record module type. PVWatts offers standard, premium, and thin-film module options in the API, and the SAM help notes that standard, premium, and thin-film module types use approximate nominal efficiencies of 19%, 21%, and 18%, respectively. [PVWatts V8 API, n.d.; SAM PVWatts System Design, n.d.]

Do not assume premium modules always produce higher annual or monthly output in a same-capacity comparison. The SAM help states that under some irradiance and weather conditions, the premium module type may operate at lower efficiency than standard or thin-film types, and an identical system except for module type may show slightly higher output for standard modules. [SAM PVWatts System Design, n.d.] What can be said from the source is narrower: for the same kW nameplate capacity, premium modules require less area than standard or thin-film modules. [SAM PVWatts System Design, n.d.]

If bifacial modules are modeled, record bifaciality and albedo assumptions. PVWatts V8 added a bifacial module option and inputs for albedo. [PVWatts V8 API, n.d.] The API defines bifaciality as the ratio of rear-side efficiency to front-side efficiency and says typical values are between 0.65 and 0.9 when provided on the bifacial module datasheet. [PVWatts V8 API, n.d.] SAM notes that for PVWatts it assumes a bifaciality factor of 0.7, transmission factor of 0.013, and ground clearance height of 1 meter. [SAM PVWatts System Design, n.d.] If the project’s actual module datasheet is not available, the correct status is unknown, not inferred.

Array geometry and tracking assumptions

Record array type, tilt, azimuth, and ground coverage ratio. PVWatts V8 supports fixed open rack, fixed roof mounted, 1-axis, 1-axis backtracking, and 2-axis array types. [PVWatts V8 API, n.d.] Tilt must be between 0 and 90 degrees in the API, while azimuth must be at least 0 and less than 360 degrees. [PVWatts V8 API, n.d.]

Array type changes how inputs are interpreted. For fixed arrays, tilt and azimuth define the module orientation. For two-axis tracking, SAM ignores tilt and azimuth because the tracker sets orientation to follow the sun. [SAM PVWatts System Design, n.d.] For fixed open rack and 1-axis systems, PVWatts uses ground coverage ratio to estimate self-shading losses; for 1-axis backtracking, it uses GCR to determine when to backtrack. [SAM PVWatts System Design, n.d.] GCR does not apply to fixed roof mount or 2-axis tracking array types in the SAM help. [SAM PVWatts System Design, n.d.]

Do not copy a tilt or azimuth rule from one hemisphere into another without checking the source condition. The SAM help states that north of the equator, a south-facing fixed array uses azimuth 180 degrees, while south of the equator, a north-facing fixed array uses azimuth 0 degrees. [SAM PVWatts System Design, n.d.]

Loss assumptions to separate

Losses should be recorded by category where possible, because PVWatts/SAM can combine multiple loss categories into a total system loss. The SAM help lists soiling, shading, snow, mismatch, wiring, connections, light-induced degradation, nameplate rating, age, and availability as loss categories. [SAM PVWatts System Design, n.d.] PVWatts V8 also has an API input for total system losses and a separate monthly soiling input that reduces incident irradiance. [PVWatts V8 API, n.d.]

Avoid double counting. If monthly soiling losses are specified, SAM notes that the soiling system DC loss category may need to be set to zero. [SAM PVWatts System Design, n.d.] If beam and diffuse shading losses are specified, the shading loss under Losses should be set to zero. [SAM PVWatts System Design, n.d.] If the snow model is enabled, the snow loss category may need to be set to zero. [SAM PVWatts System Design, n.d.] If degradation is represented elsewhere, the Age loss should be set to zero; if operating losses are represented through system availability losses, the Availability loss should be set to zero. [SAM PVWatts System Design, n.d.]

Comparing estimates fairly

A fair comparison table should include model version, dataset or weather file, location input, system capacity, module type, DC-to-AC ratio, inverter efficiency, array type, tilt, azimuth, GCR, bifaciality, albedo, soiling, shading, snow, and availability assumptions. PVWatts V8 response fields can report inputs, warnings, errors, station information, annual AC output, monthly AC output, solar radiation, and capacity factor, which makes it possible to audit what was simulated. [PVWatts V8 API, n.d.]

Weather source must be part of the comparison. PVWatts V8 updates weather data to 2020 TMY data from the NSRDB for locations covered by that database, while Version 6 used NSRDB weather data from around 2015. [PVWatts V8 API, n.d.] The API can use datasets including NSRDB, TMY2, TMY3, and international station data, and the station_info response identifies the weather data source used in the simulation. [PVWatts V8 API, n.d.]

Recommendation: for an inquiry-stage B2B PV project, require each supplier estimate to attach its assumption log before comparing kWh, capacity factor, payback, or storage sizing. If a claim is not traceable to the same model inputs, same location, same weather source, and same dates or dataset version, mark it as not comparable rather than selecting the higher number.

FAQ

What should be recorded before comparing PV energy estimates?

Record model version, weather source, location, system capacity, module type, array type, tilt, azimuth, GCR, inverter assumptions, albedo, bifaciality, and loss assumptions.

Why does PVWatts version matter?

PVWatts V8 uses updated photovoltaic module, inverter, thermal-effect models, and newer NSRDB TMY weather data for covered locations compared with V6.

Can premium modules be assumed to produce more energy?

No. The SAM help says premium modules may produce slightly lower output than standard or thin-film modules under some irradiance and weather conditions.

When should bifaciality be recorded?

Record bifaciality whenever a bifacial module option is modeled, because PVWatts defines it as rear-side efficiency divided by front-side efficiency.

Does GCR apply to every array type?

No. The SAM help says GCR is used for fixed open rack, 1-axis, and 1-axis backtracking, but does not apply to fixed roof mount or 2-axis tracking.

How can loss assumptions be double counted?

Double counting can occur if the same effect is entered both as a specific model input and as a general loss category, such as soiling, shading, snow, age, or availability.

Can one azimuth rule be used globally?

No. The SAM help distinguishes north-of-equator and south-of-equator fixed-array orientation examples, so hemisphere and project location must be checked.

What is unknown from the supplied sources?

SOLARTODO product specifications, project deployments, certifications, stock, price, savings, and performance guarantees are not supported by the supplied PVWatts and SAM documents.

References

• PVWatts V8 API documentation: Used for PVWatts V8 request parameters, datasets, bifaciality, albedo, soiling, response fields, rate limits, and V8 model/weather updates. — https://developer.nlr.gov/docs/solar/pvwatts/v8/
• SAM PVWatts System Design help: Used for module type assumptions, DC-to-AC ratio, inverter efficiency, array orientation, GCR, losses, bifacial, albedo, shading, snow, and availability guidance. — https://samrepo.nlr.gov/help/pvwatts_system_design.html