A quotation promising excellent cloudy-weather performance leaves a purchasing team with several unanswered questions. Does the claim describe measured power at one low irradiance, relative efficiency against the same module's reference power, or modeled annual energy at a particular location? Those are different deliverables. An attractive percentage cannot be compared until its denominator, temperature and measurement conditions are visible.
Direct answer: ask for an identified module's irradiance–temperature power matrix, the normalization used for any low-light percentage, measurement uncertainty and a traceable test report. If annual energy is claimed, request a separate calculation package with its weather, mounting, electrical-system and loss assumptions. A low-irradiance power advantage at one test point does not establish the same percentage advantage in annual delivered electricity.
This guide concerns the evidence behind a low-light purchasing claim. The STC and NMOT datasheet guide explains reference-condition tables. The power tolerance and current-binning guide addresses shipment allocation. Neither substitutes for a performance surface covering different irradiance and temperature conditions.
The examples below are hypothetical procurement calculations, not measurements, yield forecasts or SINAWATTS product specifications. Actual model availability, testing services, certification, commercial terms and performance commitments require written confirmation for the offered configuration.
Define the claim before comparing the numbers
Give every claim a short definition in the bid comparison. Irradiance means the incident power per unit area under the specified measurement basis. Irradiation is accumulated incident energy over time. Module power is an instantaneous electrical quantity; module energy is power accumulated over time. Mixing these quantities is one reason a convincing chart can support the wrong buying decision.
Ask the supplier to complete this sentence: “For model ___, under conditions ___, quantity ___ is compared with ___, using method ___.” A response that cannot fill these fields belongs in the clarification column, not in the scored performance column.
On small screens, swipe or scroll sideways to read every column. Keyboard users can focus the table and use the arrow keys.
| Quotation wording | Evidence to request | Purchasing interpretation |
|---|---|---|
| High power at low light | Measured watts, irradiance and temperature | Absolute output at identified conditions |
| High relative efficiency | Formula, reference point and measured inputs | Retention relative to the chosen baseline |
| Better than another module | Matched conditions, samples and uncertainty | A bounded comparison, not universal superiority |
| Higher annual yield | Energy model, weather and complete assumptions | A project-dependent prediction |
| IEC 61853 tested | Part, edition, report and model mapping | Evidence whose scope still needs inspection |
Do not translate “low light” into a single purchasing threshold without agreement. A buyer interested in overcast operation may need different comparisons from a buyer evaluating early-morning production. Describe the intended decision and request relevant data points, while allowing the laboratory to identify the method and valid test range.
Understand what the IEC evidence actually addresses
The public scope of IEC 61853-1:2011 describes module power evaluation across irradiance and temperature conditions, including maximum-power characterization. It supports requesting a power matrix, rather than treating one reference wattage as a complete performance description. The public listing alone does not verify that a particular supplier followed every requirement or that an offered model is covered by a report.
For complementary characterization, IEC 61853-2:2026 covers incidence-angle effects, operating temperature under specified ambient and mounting conditions, and spectral responsivity. The official listing identifies edition 2.0, published on August 5, 2026. Its scope includes bifacial technologies while noting limitations for time-dependent behavior and significant modeling nonlinearities. Record the actual edition on the supplier's report; an older report needs a documented relevance review, not an automatic rejection or an assumed update.
The public scope of IEC 61853-3:2018 describes annual module energy and climatic-specific rating calculations using reference climatic profiles from Part 4. That comparison framework is different from a project-specific forecast of electricity delivered through the installed system. Request the named profile and calculation basis when a quotation presents an energy rating.
These are scope summaries of publicly accessible official listings checked on October 10, 2026. They are not a reproduction of paid standards, a complete test procedure or a conformity determination. The responsible laboratory or technical reviewer must resolve normative requirements using the applicable documents.
Request the matrix as usable data
A screenshot can show the shape of a curve without preserving enough information to reproduce the comparison. Request a readable report together with a table or machine-readable export. Preserve units, axis labels, temperature definition, blank cells and notes. A missing measurement should remain missing until its treatment is explained.
For each condition, request the measured maximum power and whatever electrical parameters the agreed method reports. Ask whether entries represent individual measurements, averages, corrected results or modeled values. Separate those categories explicitly. A calculated value may be useful, but it should not be presented as an independently measured point.
The data package should identify the module model, serial number, rated power class, relevant construction revision, report number and test date. Record sample conditioning and stabilization information supplied by the laboratory. If the sample represents a model family, ask who establishes that relationship and which differences are included. A test report for another wattage or cell construction needs an explanation connecting it to the offered module.
Ask for the full reported range before selecting the rows used in a commercial comparison. Seeing only the best low-light point can conceal a crossover at another temperature or irradiance. The buyer does not need to invent a new laboratory grid; the useful request is the original result set and the report's explanation of its coverage.
Compare temperatures as carefully as irradiances
A low-irradiance number at one temperature is not directly comparable with another supplier's number at a different temperature. Also identify whether a temperature refers to the cell, module back surface or ambient air. A spreadsheet heading labeled only “temperature” is insufficient for an auditable comparison.
Create a common comparison subset using conditions actually supported by both reports. If interpolation is needed, request its method, range and validation. Preserve the original measurements alongside interpolated results. If extrapolation is proposed outside the characterized range, flag it separately and ask the responsible modeler to justify its suitability.
Do not assume that a single reference temperature coefficient reproduces an entire low-irradiance matrix. Ask the supplier or modeler to demonstrate the intended calculation against the available data. The procurement objective is a defensible comparison, not a more detailed spreadsheet whose additional precision has no evidence behind it.
Also separate a controlled laboratory comparison from outdoor interpretation. Two products can be compared at a common test temperature while their installed operating temperatures depend on the actual mounting and environment. A technical team may need both comparisons for different decisions. Keep the controlled comparison in the product-evidence section and the operating-temperature assumptions in the project-model section.
Normalize low-light power without hiding the denominator
A useful comparison-sheet calculation is relative efficiency at a specified temperature and irradiance. For a consistent illuminated-area basis, define a reporting ratio as measured power divided by the product of reference power and the irradiance ratio. Written explicitly, the denominator is reference power × measured irradiance / reference irradiance. This is an explanatory arithmetic convention, not a declaration that it reproduces every laboratory's reporting method.
For example, assume a module has measured reference power of 500 W at 1,000 W/m² and produces 96 W at 200 W/m² at the same stated temperature. The linearly scaled reference is 100 W. The reporting ratio is therefore 96 / 100 = 0.96, or 96%. It does not mean the module converts 96% of sunlight into electricity. Nor does it mean it produces 96% of its reference watts under low light.
The following hypothetical comparison deliberately separates size from retention:
On small screens, swipe or scroll sideways to read every column. Keyboard users can focus the table and use the arrow keys.
| Assumed result at one common temperature | Module A | Module B |
|---|---|---|
| Measured reference power at 1,000 W/m² | 500 W | 520 W |
| Measured power at 200 W/m² | 96 W | 98.8 W |
| Linearly scaled reference denominator | 100 W | 104 W |
| Relative efficiency by the stated convention | 96% | 95% |
Module B has greater absolute low-light power in this example; Module A has better normalized retention. Neither statement selects the better project purchase without considering module area, installed capacity, electrical constraints, cost and project energy modeling. Check whether the denominator uses nameplate or measured reference power. As another explicit assumption, keep the low-irradiance result at 96 W but suppose the measured reference is 510 W while the nameplate is 500 W. Normalizing to measured power gives 96 / (510 × 200 / 1,000), approximately 94.1%, instead of 96% against nameplate. Nothing about the assumed low-light measurement changed; only its baseline did. Keep both inputs visible and compare suppliers on the same agreed basis.
Treat uncertainty as part of the comparison
Ask whether the stated uncertainty is associated with absolute power, a normalized ratio, corrected measurements or the difference between products. These uncertainties are not interchangeable. Where common references or corrections create correlated errors, the uncertainty of a difference needs appropriate treatment by the laboratory or qualified analyst.
An apparent one-percentage-point advantage is not automatically a defensible ranking. Obtain the coverage convention, repeatability information, sample basis and the laboratory's explanation of what comparison the result supports. Avoid creating a universal uncertainty allowance from a different report. The actual method, instrument chain and reporting conditions matter.
Agree a decision rule before commissioning comparison testing. For example, the purchasing team can require a technical finding of “distinguishable under the agreed method,” “not distinguishable,” or “insufficient evidence.” Those are proposed decision categories, not standard-prescribed acceptance classes. Commercial staff can then distinguish a documented advantage from a difference that remains unresolved.
If a supplier will not release detailed results publicly, discuss an appropriate confidential review path and the minimum report fields available to the buyer. A protected report can still support a decision if authorized reviewers can verify it. A marketing percentage with no reviewable basis cannot receive the same evidence status.
Send your low-light comparison requirements for an RFQ with the model list, intended comparison conditions and required evidence. Identify which documents you already hold and which items remain subject to supplier confirmation.
Keep module characterization separate from annual energy
Sandia's PV Performance Modeling Collaborative modeling guide organizes performance modeling from weather and system inputs through plane-of-array irradiance, module temperature, electrical behavior and system output. Its structure illustrates why a module power characteristic is one input in a longer calculation chain. It does not establish a yield advantage for the products in a quotation.
For a project comparison, request the location, weather source and period, array geometry, thermal model, electrical configuration, inverter assumptions and loss treatment. Ask whether the reported result is module DC energy, array DC energy or delivered AC energy. Name the capacity denominator when results use energy per installed kilowatt. The characterization describes available module power at the reported maximum-power operating point. Ask the project modeler how the proposed electrical system uses that power during weak irradiance, including relevant inverter start/stop behavior, tracking range and conversion-efficiency assumptions. Require those inputs to match the proposed equipment and configuration. A module comparison by itself cannot demonstrate that the installed system operates at the characterized point during every interval. This is a request to review the model's scope, not an assertion that a particular inverter will or will not operate.
Hold common inputs constant wherever the comparison intends to isolate module behavior. Where a different module requires a different layout or string arrangement, document those changes instead of calling the calculation a pure technology comparison. Both comparisons can be useful, but they answer different questions: intrinsic behavior under common assumptions versus the outcome of two feasible project designs.
A modeled benefit must not be counted twice. If the matrix already describes low-irradiance behavior, an additional generic “low-light bonus” needs a justified modeling role. Ask the modeler to show where every performance adjustment enters the calculation and which reported test result supports it.
Use a bounded energy example to challenge broad claims
Consider two hypothetical modules of equal reference rating. Assume the first produces 100 arbitrary energy units during a selected comparison period. For illustration only, 20 units arise from a defined low-irradiance subset and 80 units from all other conditions. Assume the second module produces 2% more energy in that subset and exactly the same elsewhere, with every other effect excluded.
The second total becomes 20 × 1.02 + 80 = 100.4 units. The overall difference is 0.4%, not 2%. If the first module's low-irradiance subset instead contributes only 5 units, the same artificial 2% subset advantage changes the total to 100.1 units. These calculations show weighting; they are not predictions for cloudy climates or specific module technologies.
The weights must represent energy contribution under an explicit baseline, not merely a percentage of operating hours. A long period of weak irradiance can contribute less energy than a short period of stronger irradiance. A real calculation must also account for conditions that do not fit this simplified two-bucket picture.
Use this example during bid clarification: ask the supplier to identify the modeled energy contribution associated with its claimed advantage, explain how the advantage varies across conditions, and provide the full result. If that evidence is unavailable, retain the verified point comparison and remove the unsupported annual percentage from the purchasing summary.
Distinguish shading, spectrum and rear-side contributions
Uniformly reduced irradiance, nonuniform shading and a change in spectral distribution are different comparison conditions. A low-light power curve should not become a promise about a tree shadow, partial obstruction or every cloudy-sky spectrum. Ask which conditions were measured and which require separate evidence.
For bifacial products, identify front and rear measurement conditions and the normalization basis. A result influenced by rear irradiance should not be compared against a front-only result as though the sole variable were low-light efficiency. The bifacial module datasheet guide covers those separate rating and rear-input questions.
Record spectrum and incidence-angle treatment in the report review without assigning unverified gains to either. If a supplier attributes an advantage to diffuse light, ask how that claim was separated from temperature, spectrum, incidence angle and total irradiance. Where the evidence does not isolate a mechanism, describe the observed result under the tested conditions instead of inventing the cause.
This distinction also improves replacement decisions. A module that appears favorable in one comparison may have different mechanical dimensions, electrical characteristics or connector requirements. Low-light evidence is one evaluation category; it does not authorize substitution in an approved system design.
Audit the link between report, model file and offered product
Sandia's public datasets page makes module characterization reports, model files and related field datasets available for research. It demonstrates that these are distinguishable evidence objects. Public reference data can help a reviewer understand a package, but a dataset for another product cannot substantiate the offered model's low-light claim.
Request a short lineage table connecting each offered model to its test report, measured data export, fitted model file and energy calculation. Include revision dates and the owner of each transformation. If the report is unchanged but the model file changes, request the reason and the resulting effect on the comparison.
Inspect whether the model file was fitted to the submitted measurements or selected from a generic database entry. A generic representation may be adequate for an early feasibility study, but that limitation must be visible before it supports a product-specific performance premium. Ask the project analyst to compare modeled values against the submitted matrix at representative conditions.
Control changes after award. A new cell layout, glass treatment, construction revision or model substitution should trigger an assessment of whether the prior characterization remains applicable. Do not prescribe a universal retest list; require the evidence owner to document the scope decision and the buyer's authorized reviewer to accept it.
Build an RFQ and award record that can be checked
Use a staged request so the supplier understands which evidence affects shortlisting and which deliverables are conditions of award. This avoids paying for a detailed comparison before the basic product identity is established.
- Identify the proposed models, exact revisions and comparison purpose.
- Request the original characterization report and usable matrix with units and conditions.
- Record normalization, uncertainty, conditioning and sample coverage.
- Select a common supported subset for technical comparison.
- Request separate energy-rating or project-model evidence for any energy claim.
- Have the technical reviewer document limitations and unresolved differences.
- Attach accepted documents, revision control and change notification to the order.
The commercial comparison should separately show module price, evidence cost where quoted, modeling scope, sample requirements, review responsibilities and unresolved assumptions. Do not assume that a product quotation includes independent testing or that a test schedule matches the delivery schedule. Ask for the proposed sequence and obtain actual terms in writing.
Keep a brief award rationale: which benefit was established, under which conditions, which remaining uncertainty was accepted and who approved the decision. That record is useful when a later quotation offers a substitute with a similar headline wattage but a different evidence package.
Buyer FAQ
1. What is the first document to request for a low-light claim?
Request the exact model's test report with an irradiance–temperature power matrix and the formula behind the advertised percentage. A curve without units, reference power or temperature is insufficient for a comparison. Identify the report's sample and construction coverage before using its numbers in an award decision.
2. Does higher relative efficiency mean more watts?
Not necessarily. Relative efficiency uses a stated baseline; a larger module can deliver more watts while retaining a smaller fraction of its scaled reference power. Compare both absolute power and normalized behavior, then consider the project constraints that make either quantity commercially relevant.
3. Is one low-irradiance test point enough?
It can substantiate a narrowly worded point claim if conditions and uncertainty are adequate. It does not describe the full performance surface. Request additional supported conditions when the purchasing decision depends on changing irradiance and temperature, and keep unsupported regions visible.
4. Does an IEC 61853 report guarantee annual production?
No. Identify the specific part and report scope. Characterization and reference-climate energy ratings support defined comparisons; actual project energy depends on additional site and system inputs. A contractual energy commitment requires its own agreed scope, assumptions, measurement basis and responsible parties.
5. Can an older Part 2 report still be useful?
It can provide relevant historical evidence. The technical reviewer must identify its edition, method, sample coverage and compatibility with the intended analysis. The 2026 edition should not be silently assigned to an older report, nor should an edition difference alone be treated as proof that the measurements are unusable.
6. Should we compare nominal or measured reference power?
Use the reference basis required by the comparison and disclose it. If one supplier normalizes to nominal power and another to measured power, the ratios are not directly equivalent. Request enough data to reconcile the difference without altering the original report.
7. Is low-light performance the same as shade tolerance?
No. A uniform irradiance reduction does not reproduce every nonuniform shadow pattern. Treat shading behavior as a separate technical question with an appropriate test or model. Do not extend a low-light marketing statement into an unsupported claim about partial shading or system output.
8. What if the claimed difference is very small?
Request the uncertainty and decision basis relevant to that difference. The result may support a clear ranking, an indistinguishable comparison or a request for further evidence. Avoid paying a performance premium solely because two rounded figures appear different in a brochure.
9. Can a model file replace the laboratory report?
It can support modeling, but it does not independently establish where its parameters came from. Ask for lineage, revision and validation against measurements. If confidentiality limits disclosure, agree an authorized review process and record any evidence gaps that remain.
10. Should every delivered module receive matrix testing?
That is a project-specific quality and commercial decision. Distinguish product characterization from production or shipment checks. Ask the supplier and laboratory to propose appropriate sampling and traceability, then agree scope, cost and schedule before ordering. This guide does not prescribe a universal sample count.
11. How should low-light performance affect price comparison?
First establish the bounded technical benefit. Then ask the project team whether a justified energy or layout comparison changes the economic decision. Keep evidence costs and unresolved assumptions visible. Do not convert a point-efficiency percentage directly into the same percentage price premium or lifetime return.
12. What information should accompany the RFQ?
Include the model list, project comparison purpose, available reports, desired output metrics, proposed acceptance process and required document revisions. If annual energy matters, provide the agreed site and system assumptions or identify who will develop them. Ask separately for confirmed price, quantity and delivery terms.
Send your module matrix and energy-evidence checklist for an RFQ so the quotation can distinguish verified characterization, proposed modeling work and information still required for the buying decision.