A bifacial module does not carry one universal "bifacial wattage." Its front-side nameplate power, rear-to-front response and project energy gain answer different questions. A datasheet table showing 5%, 10% or 20% backside gain is not a promise that the project will gain that amount. Actual rear irradiance depends on array geometry, height, pitch, ground reflectance, shading, season, soiling, snow, terrain and mounting hardware. Electrical current and inverter loading can also constrain how much rear-side contribution becomes useful AC energy.
A defensible RFQ separates module measurement evidence from project yield assumptions. First, compare the exact model's front-side STC values, power tolerance, rear-side I-V measurements and bifaciality coefficients. Then compare the site model: front and rear irradiance, albedo, geometry, shading, mismatch, temperature, clipping and losses. Finally, check whether cables, connectors, fuses, combiner inputs and inverter MPPTs were designed for the applicable current cases.
IEC TS 60904-1-2:2024+AMD1:2026 describes I-V measurement procedures for single-junction bifacial PV devices. The IEC page says it applies to encapsulated cells, subassemblies and complete modules. It also says these are additional requirements beyond the monofacial procedures in IEC 60904-1. The 2026 consolidated version changed several details, including the equivalent-irradiance calculation basis. IEC TS 60904-1-2:2024+AMD1:2026.
This guide complements the STC versus NMOT datasheet guide. It also complements the flash-test and EL traceability guide. It does not estimate energy for a real project or select protection, cable or inverter equipment. No statement here claims an unverified SINAWATTS module, bifaciality, power bin, certificate, yield, test capability, warranty, stock, price, MOQ or lead time.
Direct answer: what should a bifacial-module RFQ require?
Require two linked evidence packages.
The module package should include:
- exact manufacturer, model code, suffix, bill-of-material revision and manufacturing site;
- front-side STC Pmax, Vmp, Imp, Voc and Isc with tolerance and measurement uncertainty;
- rated bifaciality coefficients for Pmax, Isc and Voc, each with tolerance and test method;
- front-only and rear-only I-V reports or the controlled measurement route used by the applicable standard;
- any dual-side or equivalent-irradiance table with all front and rear irradiance conditions;
- calibration chain, simulator class, spectral and nonuniformity treatment, device temperature and report date;
- nameplate and serial-number format tied to flash data;
- qualification and certification documents covering the exact construction;
- mechanical drawing, frame, glass, cell, encapsulant, back surface, junction box, leads and connectors;
- installation manual revision and approved mounting or rear-side obstruction limits;
- power warranty basis and a statement of whether it covers front-side nameplate power, total output or another metric; and
- change-control rules for cells, glass, encapsulant, back surface, frame, junction box, leads, connectors and factory route.
The project package should include:
- weather and irradiance dataset identity, period and spatial source;
- ground-reflectance or albedo data, including seasonal and uncertainty treatment;
- fixed-tilt or tracker geometry, pitch, ground coverage ratio, hub or lower-edge height and row length;
- module transparency or gap assumptions;
- torque tube, rail, clamp, junction-box, cable and other rear-shading geometry;
- front and rear plane-of-array irradiance method;
- rear irradiance nonuniformity and electrical mismatch treatment;
- module temperature model and front or rear sensor assumptions;
- DC design current and voltage cases, inverter current limits, MPPT allocation and clipping;
- soiling, snow, vegetation, flooding and degradation assumptions;
- software name, version, transposition and bifacial-model settings;
- loss diagram, uncertainty budget and sensitivity cases; and
- model validation or independent review route.
A bidder should not place front-side nameplate watts, a modeled rear gain and a dual-side laboratory output in the same comparison column without labels. Keep every number attached to its measurement or simulation condition.
Four values that buyers must not confuse
On small screens, swipe or scroll sideways to read every column. Keyboard users can focus the table and use the arrow keys.
| Term | What it describes | What it does not prove |
|---|---|---|
| Front-side STC Pmax | Module power under the stated front-side standard test condition | Annual energy or rear gain |
| Bifaciality coefficient | Rear response divided by front response for a defined characteristic and test method | Rear irradiance available at the project |
| Dual-side or BNPI output | Electrical output under a stated front and rear irradiance combination | Guaranteed field output at another geometry |
| Bifacial energy gain | Modeled or measured energy difference against a defined reference | A universal property of the module alone |
Front-side STC power is useful for nameplate capacity and model identity. It should be compared with its test condition, bin and tolerance. The STC versus NMOT guide explains why STC and operating-condition values are not interchangeable.
Bifaciality is a ratio, not an irradiance. Pmax bifaciality, Isc bifaciality and Voc bifaciality are different characteristics. Require each symbol and equation to be stated rather than accepting a single unlabeled percentage.
A dual-side table can help equipment design and modeling only when front irradiance, rear irradiance, spectrum, temperature and calculation method are known. A field energy-gain number adds site geometry, weather and loss assumptions. It cannot be transferred unchanged from another project.
Use the current bifacial I-V measurement reference
The IEC webstore page checked on 2026-09-30 identifies the consolidated edition as IEC TS 60904-1-2:2024+AMD1:2026. It states that the document covers current-voltage characterization in natural or simulated sunlight. Its scope includes complete modules as well as encapsulated cells and subassemblies.
The IEC page lists significant changes from the 2019 first edition. It says irradiance-nonuniformity requirements now refer to IEC 60904-9 classifications. It notes a revised non-irradiated-background requirement. It says spectral mismatch correction is not mandatory unless another standard requires it, but spectral mismatch belongs in measurement uncertainty. It also says equivalent irradiance is based on Isc bifaciality rather than the minimum of Isc and Pmax bifaciality.
These points matter in procurement. A report citing only "IEC 60904-1-2" should identify edition and amendment status. The RFQ should ask whether the report is a direct test, a calculation from one-sided tests or another permitted method. It should state which side was illuminated, how the other side was controlled and how nonuniformity and uncertainty were handled.
Do not reproduce a paywalled standard from a datasheet summary. Require an authorized laboratory or responsible technical party to apply the purchased document. Ask the bidder to disclose deviations and the decision rule used for pass or rating.
Read the manufacturer datasheet without turning examples into guarantees
Manufacturer datasheets often present a front-side STC table plus rows for assumed backside gains. These rows are useful only with their notes.
For example, an official Canadian Solar BiHiKu6 datasheet checked on 2026-09-30 presents front-side STC electrical data. It also shows additional rows for 5%, 10% and 20% bifacial gain. Its footnote says the backside contribution depends on mounting structure, height, tilt and ground albedo. Those values apply to the named CS6W-MB-AG family and its document revision. They are not generic SINAWATTS values and do not predict another project. Canadian Solar BiHiKu6 datasheet.
An official Trina Solar datasheet provides another model-specific format. It labels STC, NOCT and a bifacial nameplate irradiance condition and lists model-specific bifaciality coefficients. The presence of different tables across manufacturers is exactly why the RFQ needs common comparison fields. Do not compare a BNPI output from one bid with front-side STC Pmax from another. Trina Solar bifacial module datasheet.
For each offered model, extract:
- document number and revision;
- model range and exact offered power bin;
- front-side test conditions;
- dual-side or rear-side conditions;
- tolerances for Pmax, Voc, Isc and bifaciality;
- whether the displayed gain is measured, calculated or illustrative;
- mechanical drawing and rear obstructions;
- installation-manual reference; and
- statement that specifications may change, followed by contractual document control.
A screenshot without the footer, revision and notes is not sufficient evidence.
Treat rear irradiance as a site variable
Rear irradiance is produced by a scene, not by the module alone. The scene includes ground and sky contributions, reflections, adjacent rows and objects. The IEA PVPS Task 13 report describes rear-side irradiance modeling using sky, ground and array contributions. Its detailed method belongs to that report and model; the general procurement lesson is to document the full scene. IEA PVPS bifacial modules and systems report.
The official NREL bifacial irradiance toolkit page lists model inputs and effects such as albedo, tilt, pitch, clearance height and module spacing. It also includes shading obstructions, electrical mismatch, spectral effects and finite-array edge effects. This demonstrates why "rear gain = bifaciality x albedo" is not a complete yield model. NLR bifacial irradiance and performance modeling toolkit.
Request a site-input schedule:
On small screens, swipe or scroll sideways to read every column. Keyboard users can focus the table and use the arrow keys.
| Input | Required evidence | Sensitivity question |
|---|---|---|
| Albedo | Site measurement, dataset or justified source | What happens at low seasonal values? |
| Height | Dimension from reflective surface to module | Does settlement or vegetation reduce clearance? |
| Pitch or GCR | Issued layout | How does tighter spacing shade ground and rear side? |
| Tilt or tracking | Issued geometry and control strategy | How do morning, evening and backtracking change exposure? |
| Row length | Actual table and boundary layout | Are edge rows modeled differently? |
| Rear obstructions | Torque tube, rail, clamp, box and cable geometry | Is nonuniform shading represented? |
| Ground condition | Soil, gravel, membrane, crop, snow or water | How stable is reflectance over life? |
| Transparency | Cell gaps and module construction | Is the value tied to the exact BOM? |
| Soiling or snow | Front and rear treatment | Are asymmetric losses included? |
| Terrain | Slope and row-to-row variation | Is a simple 2D model suitable? |
Use measured albedo when practical and preserve the instrument, location, interval and cleaning method. If a generic value is used, label it as an assumption and test a conservative range. A high-reflectance ground treatment also needs durability, drainage, glare, maintenance and cost review. Do not count an albedo improvement that is not in the project scope.
Model geometry and rear shading from issued drawings
The rear side can be shaded by torque tubes, rails, brackets, junction boxes, cables and neighboring rows. A percentage labeled "structure shading" is not reviewable unless it maps to actual hardware.
Require:
- module orientation and cell-string layout;
- exact frame and rear-glass drawing;
- junction-box count, size and position;
- mounting-hole or clamp zones used;
- tracker torque-tube size and position;
- rail or purlin dimensions and offsets;
- cable route and clips;
- table gap and module-to-module gap;
- lower-edge and axis height;
- row pitch and terrain;
- representation of edge rows; and
- rear-irradiance sampling grid or model resolution.
PVsyst's official bifacial documentation explains that its view-factor model evaluates ground-reflected and sky contributions. It says rear irradiance is weighted by module bifaciality in its electrical conversion. The documentation also identifies structural shading and rear nonuniformity or mismatch as inputs or losses. These are software-specific descriptions, not a mandatory project method. PVsyst bifacial systems documentation.
Check the chosen model's limitations. PVsyst's procedure page states that its regular-system approach uses a 2D cross-section with simplifications. The NREL toolkit offers ray-tracing and finite-array options. A regular flat project and an irregular carport, rooftop or agrivoltaic array may need different representations. Record why the selected method fits the geometry.
The mechanical-load and mounting guide should be reviewed at the same time. Moving a rail to reduce rear shading can change load paths and approved mounting conditions. Yield optimization cannot override structural or warranty requirements.
Separate bifaciality from annual bifacial gain
A module with higher bifaciality may not produce proportionally higher annual energy if rear irradiance is low or nonuniform. Conversely, a moderate bifaciality module in an open, elevated, reflective layout may receive a larger rear resource. The comparison must hold the project scene constant.
Use this sequence:
- compare front-side Pmax and efficiency for the exact size and bin;
- compare Isc, Pmax and Voc bifaciality with tolerance and test method;
- run each exact module in the same weather, geometry and loss model;
- use each model's current, voltage, temperature coefficients and dimensions;
- apply rear irradiance, mismatch and structure-shading methods consistently;
- check inverter and DC equipment constraints;
- compare annual energy and uncertainty, not only a headline gain; and
- rerun sensitivities for albedo, height, mismatch, soiling and clipping.
Do not compare a bidder's optimistic model for one module against a conservative independent model for another. Ask for the input file or an export detailed enough to reproduce the result.
Use a bounded calculation without calling it a yield forecast
Consider a hypothetical module with front-side STC Pmax of 600 W and Isc bifaciality of 80%. Assume a simplified instant with front irradiance normalized to 1000 W/m^2 and rear irradiance of 120 W/m^2. An equivalent-irradiance concept based on Isc bifaciality would add a rear contribution of 0.80 x 120 = 96 W/m^2. The simplified effective level would be 1096 W/m^2 before other corrections.
This is not a statement that power becomes 657.6 W or annual energy rises 9.6%. Actual I-V response, temperature, spectral conditions, rear nonuniformity, mismatch, angle of incidence, ohmic losses and inverter clipping intervene. The example only shows why the bifaciality definition and rear irradiance must use compatible units and conditions. The current IEC method and responsible project model should govern the actual calculation.
A useful bid comparison asks every supplier to provide front values and bifaciality, while the buyer or independent engineer runs one common project model. This removes one major source of inconsistent assumptions.
Design DC current from the applicable bifacial case
Rear illumination primarily increases current. It can affect string current, parallel-circuit aggregation, connector loading, cable ampacity, voltage drop, fuse selection, combiner inputs and inverter MPPT limits. The design cannot use only the front-side Isc if the applicable code, manual or engineering method requires a bifacial adjustment.
An official Canadian Solar bifacial installation manual checked on 2026-09-30 distinguishes listed current at 0% bifacial gain and correction-factor treatment. It directs appropriate factors to qualified project designers under relevant codes and simulation results. That instruction is manufacturer and model-family specific. It is not a universal multiplier for other products. Canadian Solar bifacial installation manual.
The RFQ should require an electrical design schedule showing:
- front-side Isc and Imp from the offered datasheet;
- Isc bifaciality and tolerance;
- rear-current or bifacial correction method;
- code-required continuous-current and other factors;
- maximum source-circuit and output-circuit current;
- connector and cable current ratings under installed temperature;
- string-fuse need, rating and module maximum-series-fuse constraint;
- combiner and isolator input limits;
- inverter maximum input current and short-circuit-current limits per MPPT;
- DC-to-AC ratio and clipping under bifacial cases; and
- accountable designer and document revision.
The PV string-fuse sizing guide covers Isc, reverse current and module maximum-series-fuse evidence. The PV connector current and derating guide covers mated-pair current and temperature evidence. Use exact product limits and the governing electrical code.
Do not add a rear-gain percentage directly to Voc. The datasheet or accepted design method should state how voltage characteristics are treated. Cold-weather maximum string voltage remains a separate check under the cold-Voc guide.
Account for rear nonuniformity and mismatch
Rear irradiance is often less uniform than front irradiance because of hardware shadows and ground view. Cell and substring currents interact with bypass-diode topology and series connections. An average rear irradiance can therefore overstate useful power if a strongly shaded region limits current.
Require the model to state:
- spatial resolution on the module rear;
- treatment of cell and substring layout;
- position of junction boxes or opaque areas;
- torque-tube and rail shadow;
- module-to-module and row-edge effects;
- mismatch model or loss factor;
- validation source for that factor; and
- sensitivity when nonuniformity increases.
PVsyst documentation says a rear-side mismatch loss is applied because rear irradiance is nonuniform. Its default or chosen value is a software assumption, not a measured property of every project. The NREL toolkit page identifies electrical mismatch as a modeled effect. Compare model assumptions with the actual module and rack.
For an unusual structure, request a higher-resolution study or field mock-up. Document the sensor or scan layout. Do not hide a large modeled mismatch inside a generic "other loss" line.
Include temperature, spectrum, soiling, snow and degradation boundaries
Bifacial output does not occur at a fixed 25 C cell temperature in the field. Rear contribution can increase current while module temperature changes power. Glass-glass construction, wind exposure and mounting geometry may affect temperature. Use the exact module temperature coefficients and a documented model.
Front and rear spectra can differ. The current IEC page places spectral mismatch in measurement uncertainty when correction is not otherwise required. A project model should state whether it treats spectral effects and whether the available data support that treatment.
Soiling can be asymmetric. Dust may settle mainly on the front, while rear surfaces can collect splash, condensation or site-specific contamination. Snow can raise ground reflectance and also cover the module or shade the lower rear. Use site evidence and operational rules rather than selecting only the beneficial effect.
Degradation claims need an explicit metric. A front-side power warranty does not automatically guarantee a constant rear gain. Ask how bifaciality, rear transparency, encapsulant, glass and cell technology are covered after aging. Use the PID, LID and LeTID guide to separate mechanisms, test conditions and warranty language.
Tie every module to flash, BOM and certificate evidence
The offered power bin should map to a serial-numbered flash record. For a bifacial module, the buyer should know whether production flash testing is front-only, dual-side, equivalent-irradiance or another controlled route. Require the test condition, simulator, calibration, uncertainty and calculation fields.
The flash-test and EL guide explains how serial identity connects reports to delivered modules. Bifacial procurement adds rear-response and bifaciality evidence. Do not accept a type-test report from one construction as a production record for another.
Map qualification and certification evidence to:
- exact model and power range;
- cell type and supplier-control rule;
- front and rear glass or transparent back surface;
- encapsulant;
- frame;
- junction box and bypass diodes;
- output leads and connectors;
- factory or manufacturing route;
- maximum system voltage and fire or safety classification where applicable; and
- approved mounting configurations.
The glass and backsheet BOM change-control guide covers model mapping and retest logic. A change in rear glass, encapsulant, cell metallization or opaque area can affect rear response even when the front power bin is unchanged.
Compare bids using one evidence hierarchy
Use the following order:
- exact model, revision and front-side nameplate data;
- current-edition bifacial measurement report and uncertainty;
- production control for bifaciality and serial flash records;
- qualification or certificate mapping to the exact BOM;
- issued-layout project model with transparent inputs;
- electrical design using applicable current cases;
- sensitivity and uncertainty analysis;
- field acceptance and monitoring plan; and
- contractual warranty and change-control terms.
A higher headline gain should not outrank traceable lower-bound evidence. A module with a wide bifaciality tolerance may need a conservative model input. A layout using an unverified high albedo may need sensitivity or a maintenance obligation.
Use a bounded bid comparison
Assume three bids offer modules with similar front-side power. The project layout and weather file are fixed for comparison.
Bid A advertises "up to 20% extra power." It gives no rear irradiance condition, bifaciality tolerance, model file or current-design case. Its claim is not comparable.
Bid B supplies a datasheet table at a stated backside gain and a Pmax bifaciality value. It models a generic albedo and omits torque-tube shading, rear mismatch and inverter current limits. Its module evidence is partly useful, but the project result is incomplete.
Bid C provides exact model and BOM identity, front and rear I-V evidence under the current IEC technical specification and coefficients with tolerances. It supplies a reproducible model using issued geometry, measured or bounded albedo, rear obstruction and mismatch treatment. It also returns DC current checks, sensitivity cases, serial flash mapping and change control.
Bid C offers the strongest decision package. The buyer still needs qualified technical review and commercial comparison. This hypothetical example does not approve a vendor or predict a project yield.
RFQ evidence matrix
On small screens, swipe or scroll sideways to read every column. Keyboard users can focus the table and use the arrow keys.
| Decision field | Buyer requirement | Supplier return | Release evidence |
|---|---|---|---|
| Model identity | Exact code, suffix, bin, BOM and site | Controlled offer | Datasheet and certificate mapping |
| Front rating | Pmax, Vmp, Imp, Voc, Isc and tolerance | Exact values | Serial flash method |
| Bifaciality | Pmax, Isc and Voc coefficients and tolerance | Exact definitions | Current-edition test report |
| Dual-side values | Front and rear irradiance, temperature and method | Labeled table | Laboratory report |
| Geometry | Pitch, height, tilt, gaps and row length | Issued layout | Model input export |
| Rear obstruction | Tube, rail, box, clamp and cable shadow | Detailed representation | Drawings and loss result |
| Ground | Albedo source, season and uncertainty | Bounded values | Measurement or cited dataset |
| Mismatch | Spatial method and loss | Explained treatment | Sensitivity or validation |
| Electrical | Current cases and equipment limits | Calculation schedule | Design review |
| Energy | Software, version, losses and clipping | Reproducible result | Model files and report |
| Production | Serial, flash and bifaciality sampling | Control plan | Sample delivery records |
| Change control | Inputs that reopen review | Written process | Approved change matrix |
Commissioning and field-monitoring evidence
IEC 61724-1:2021 outlines terminology, equipment and methods for PV-system performance monitoring. The IEC page says its second edition introduced monitoring of bifacial systems and updated irradiance-sensor requirements. This makes monitoring design part of the procurement package, not an afterthought. IEC 61724-1:2021.
A 2024 NREL-hosted research paper evaluated rear-irradiance approaches for bifacial system performance and capacity tests. It describes challenges caused by nonuniform rear irradiance and discusses sensor type and placement. Its findings apply to the studied systems and methods, not every project. The practical RFQ lesson is to state how rear irradiance will be measured or modeled during acceptance. NLR research record for the 2024 paper on irradiance monitoring for bifacial PV systems.
A commissioning plan should identify:
- front and rear irradiance sensors or reference modules;
- calibration and uncertainty;
- placement by row and height;
- shading exclusions and representativeness;
- cleaning and inspection;
- module and ambient temperature sensors;
- inverter and string data;
- time synchronization and data completeness;
- expected-energy model and weather correction;
- treatment of snow, clipping, curtailment and outages; and
- acceptance metric, period and decision rule.
Do not use a front-only performance ratio without explaining how rear energy is treated. Do not place a rear sensor in an unusually bright or shaded position and call it representative. Preserve photographs and coordinates of the sensor installation.
Contract language and warranty questions
Define which value controls payment, acceptance and warranty. Front-side nameplate Pmax, bifaciality, modeled energy and measured system output are separate obligations.
Ask:
- Is module pricing based on front-side nameplate watts?
- Is a minimum bifaciality coefficient guaranteed, and at what confidence or tolerance?
- What report proves that coefficient for the offered model?
- Is production bifaciality sampled, calculated or tested?
- Does warranty degradation apply to front Pmax, bifaciality or both?
- Which project-yield assumptions are supplier commitments?
- Who owns error in weather, albedo, layout or model inputs?
- What happens if delivered BOM or factory differs?
- Which records accompany each shipment?
- What remedy applies to a failed module or system acceptance test?
Avoid a contract that guarantees "bifacial gain" without defining the monofacial reference, period, weather correction, availability and test method. A bankable metric must be reproducible.
Change control and revalidation
Reopen the comparison when any of these change:
- module model, suffix, power bin or factory;
- cell technology, layout, metallization or supplier;
- front or rear glass, transparent back surface or encapsulant;
- frame, junction box, leads, connectors or opaque rear area;
- bifacial test method, edition, simulator or calculation;
- tracker or fixed-rack geometry, pitch, height or row length;
- torque tube, rail, clamp or cable routing;
- ground treatment, vegetation or drainage design;
- weather or albedo dataset;
- modeling software, version or loss assumption;
- inverter, MPPT allocation, DC-to-AC ratio or current limit;
- applicable code or manufacturer manual; or
- warranty, qualification or certificate scope.
A supplier statement that the change has no front-side power impact is not enough when rear response or shading can change. Map the changed attribute to measurement, energy, electrical and qualification consequences.
Source boundaries checked on 2026-09-30
The current IEC TS 60904-1-2 consolidated page and IEC 61724-1 page were checked on 2026-09-30. The NREL bifacial toolkit and 2024 monitoring paper, IEA PVPS Task 13 report and PVsyst official documentation were also checked. The Canadian Solar and Trina Solar materials were used only as model-specific examples of datasheet and manual conventions. No manufacturer value was transferred to SINAWATTS or another product. No site energy gain, current multiplier, certification, warranty or test capability was inferred.
Send the exact module datasheets, site geometry, albedo evidence and inverter limits for a structured bifacial PV RFQ. Include project location, rack or tracker drawings, electrical one-line, applicable code and acceptance method. Bidders can then return comparable module and project evidence. The actual supplier and qualified designer must confirm capability, energy model, protection, price, quantity and lead time.
Buyer FAQ
Is the highest bifaciality module always the best choice?
No. Compare front power, bifaciality tolerance, geometry, rear resource, mismatch, electrical limits, qualification, warranty, cost and modeled annual energy.
Does 10% bifacial gain on a datasheet guarantee 10% more annual energy?
No. It is usually a stated condition or illustrative table. Field gain depends on site geometry, albedo, shading, temperature, losses and the comparison baseline.
Which bifaciality coefficient should the RFQ request?
Request Pmax, Isc and Voc bifaciality with definitions, tolerances and test method. For equivalent-irradiance calculations, use the method required by the current applicable standard.
Can I compare BNPI output with another module's STC power?
Not directly. BNPI or another dual-side condition includes rear irradiance. Front-side STC and dual-side output must remain in separate labeled columns.
Why is Isc important for bifacial system design?
Rear illumination can increase current. That can affect cables, connectors, fuses, combiner inputs and inverter current limits. Use the applicable code, exact manual and qualified design method.
Does bifacial rear gain increase cold-weather Voc by the same percentage?
Do not apply a gain percentage to Voc. Use exact datasheet characteristics and the accepted voltage-design method. Keep cold-temperature string-voltage calculation separate.
How should albedo be specified?
State source, location, surface, instrument or dataset, time interval, seasonality, uncertainty and maintenance. Run conservative sensitivity cases when evidence is weak.
What rear-side obstructions matter?
Torque tubes, rails, purlins, clamps, junction boxes, cables and neighboring rows can create nonuniform rear shading. Model issued hardware, not a generic percentage alone.
Is a flash report enough?
No. It supports serial power traceability. You also need bifaciality or rear-response evidence, exact BOM and certificate mapping, project modeling and electrical checks.
What should be measured during commissioning?
At minimum, use the project monitoring and acceptance plan. It may include front and rear irradiance, module temperature, string or inverter output, weather, availability and synchronized data quality.
What should trigger a new energy model?
Changes to the module BOM, layout, racking, ground, weather data, software, loss assumptions, inverter or current limits require documented review. Re-run affected cases before acceptance.