Fill Factor Numerical Evaluation: Measurement Principles, Standards, and Applications

Fill factor (FF) is a critical performance metric that quantifies the "squareness" of a photovoltaic device's current–voltage characteristic and directly determines the achievable power conversion efficiency. Professional numerical evaluation of fill factor combines calibrated I–V curve measurement, standardized correction procedures, and rigorous uncertainty analysis performed by accredited third-party laboratories. This article explains the technical fundamentals, applicable standards, sample types, testing items, measurement methods, and practical applications of fill factor numerical evaluation.

Technical Overview

Fill factor numerical evaluation refers to the quantitative determination and interpretation of the fill factor (FF), a dimensionless parameter that characterizes the quality of the current–voltage (I–V) characteristic of photovoltaic (PV) devices. The fill factor is defined as the ratio of the maximum output power to the theoretical power limited by the open-circuit voltage and short-circuit current:

FF = Pmax / (Voc × Isc) = (Vmp × Imp) / (Voc × Isc), where Pmax is the maximum power point, Voc is the open-circuit voltage, Isc is the short-circuit current, and Vmp and Imp are the voltage and current at the maximum power point. Together with Voc and Isc, the fill factor determines the overall power conversion efficiency, so any uncertainty in FF propagates directly into the reported efficiency of a cell or module.

Typical fill factor values range from approximately 0.74 to 0.80 for conventional crystalline silicon cells, 0.80 to 0.86 for advanced architectures such as PERC, TOPCon, and heterojunction designs, 0.60 to 0.75 for most thin-film technologies, and 0.70 to 0.82 for state-of-the-art perovskite and tandem devices. The theoretical fill factor limit for silicon, governed by Auger recombination and intrinsic carrier physics, approaches 0.89, while the detailed-balance (Shockley–Queisser) limit yields an ideal fill factor of approximately 0.87 for a single junction under standard illumination.

Numerical evaluation of the fill factor therefore serves two purposes: first, to determine FF accurately from measured or corrected I–V data, and second, to diagnose the loss mechanisms that reduce FF relative to the theoretical maximum. The dominant loss mechanisms include:

• Series resistance (Rs), which arises from contact resistance, emitter resistance, metallization grid resistance, and interconnect resistance, causing a voltage drop that rounds the I–V curve near the maximum power point.
• Shunt resistance (Rsh), which originates from edge leakage, crystal defects, pinholes, or process contamination, and reduces FF by allowing current to bypass the junction.
• Recombination losses in the depletion region or at surfaces, reflected in an elevated diode ideality factor that lowers the curve slope at Voc.
• Non-ohmic contact behavior and parasitic junction effects, which distort the curve shape beyond the predictions of the ideal single-diode model.
• Temperature and irradiance deviations, which shift Voc and the operating point and must be corrected before comparison.

Because fill factor is highly sensitive to these mechanisms, it is widely used as a sensitive process-quality indicator in research, production, and field-degradation studies. The single-diode (five-parameter) lumped model and the two-diode model are commonly fitted to measured data to extract Rs, Rsh, ideality factor, and diode current, enabling root-cause analysis when FF falls below specification.

Testing Samples

Crystalline Silicon Solar Cells

Monocrystalline and multicrystalline silicon cells, including Al-BSF, PERC, TOPCon, IBC, and heterojunction (HJT) architectures, are the most frequently evaluated sample type. Full-area cells, half-cut cells, and small-area research cells with designated illumination areas are tested, typically under standard test conditions (STC: 1000 W/m², 25 °C cell temperature, AM1.5G spectrum) in accordance with IEC 60904-1 and ASTM E948.

Thin-Film Solar Cells and Modules

Cadmium telluride (CdTe), copper indium gallium selenide (CIGS), and amorphous/microcrystalline silicon devices exhibit metastable behavior and often require light soaking, dark annealing, or controlled preconditioning prior to I–V measurement to obtain a reproducible fill factor value, following the stabilization procedures defined in IEC 61215 series standards.

Emerging Photovoltaic Devices

Perovskite solar cells, organic photovoltaic devices, quantum-dot cells, and monolithic or mechanically stacked tandem devices are evaluated for fill factor using small-area test layouts with defined aperture masks. These samples frequently show hysteresis in the I–V sweep, so scan-rate, scan-direction, and preconditioning protocols are documented and reported together with the FF value.

PV Modules and Mini-Modules

Full-size laminated modules, glass–glass and glass–backsheet constructions, bifacial modules, and mini-modules used for process qualification are measured under continuous or pulsed solar simulators in accordance with IEC 60904-1, ASTM E1036, and the power rating procedures of IEC 61853-1.

Testing Items

Typical items covered in a fill factor numerical evaluation program include:

• Numerical determination of fill factor (FF) at STC, including maximum power point location (Vmp, Imp) via mathematical fitting of the P–V curve.
• Companion electrical parameters: Voc, Isc, Pmax, and conversion efficiency with their measurement uncertainties.
• Series resistance and shunt resistance extraction using the single-diode model, the two-diode model, or the comparison method of IEC 60891.
• Diode ideality factor and reverse saturation current from dark I–V characterization.
• Pseudo fill factor (pFF) from Suns–Voc measurement to separate series-resistance losses from recombination losses.
• Fill factor temperature coefficient and irradiance dependence across the operating envelope.
• Statistical evaluation of FF distribution across production batches, including mean, standard deviation, and process capability indices.
• Uncertainty budget prepared in accordance with ISO/IEC Guide 98-3 (GUM) covering calibration, spectral mismatch, temperature, and simulator non-uniformity contributions.

Testing Methods

I–V Curve Measurement Under Simulated Illumination

The primary method measures the illuminated I–V characteristic using a solar simulator meeting the classification requirements of IEC 60904-9, with spectral match, spatial non-uniformity, temporal instability, and (for pulsed systems) flash duration verified. The device under test is held at 25 °C on a temperature-controlled chuck or stage, and a calibrated reference cell traceable to a recognized metrology institute is used to set the irradiance, following IEC 60904-2. Four-wire (Kelvin) connections eliminate lead resistance, and the FF is computed numerically after the maximum power point is identified by polynomial or spline fitting of the measured P–V curve.

Dark I–V Measurement and Model Fitting

Dark I–V data are acquired over several decades of current and fitted to the single-diode equation with lumped Rs and Rsh using the Lambert W function solution or iterative numerical minimization. The extracted diode parameters allow reconstruction of the ideal (loss-free) I–V curve and prediction of the fill factor that the device would achieve if resistive losses were eliminated, providing a diagnostic reference value.

Suns–Voc Measurement for Pseudo Fill Factor

In the Suns–Voc technique, the device is illuminated at varying intensity while open-circuit voltage and a calibrated reference signal are recorded simultaneously. Because the measurement is performed at effectively zero current, series resistance does not influence the result, yielding a pseudo I–V curve and a pseudo fill factor (pFF). The difference FF − pFF quantifies the series-resistance loss, while the gap between pFF and the theoretical FF quantifies recombination losses.

Corrections and Normalization

Measured I–V curves are corrected to STC using the translation procedures of IEC 60891 ( Procedures for temperature and irradiance corrections to measured I-V characteristics), applying Procedure 1 (linear parameter correction) or Procedure 2 (internal diode model interpolation) as appropriate. Spectral mismatch between the simulator spectrum, the reference cell response, and the device under test is corrected according to IEC 60904-7, and the equivalent cell temperature is determined per IEC 60904-5 where direct contact measurement is impractical. For bifacial devices, the rear-side irradiance is controlled and reported in accordance with IEC TS 60904-1-2.

Uncertainty Analysis and Validation

Each reported fill factor value carries a combined standard uncertainty assembled from the reference cell calibration, spectral mismatch factor, temperature deviation, simulator non-uniformity and instability, voltmeter and ammeter accuracy, and maximum power point fitting error. Repeatability and reproducibility checks, including round-robin intercomparison where required, validate the laboratory's measurement capability. National and international frameworks such as IEC 60904 series, ASTM E948, ASTM E1036, EN 50513, GB/T 6495 series, and JIS C 8904 series define the accepted procedures, and ISO/IEC 17025 accreditation confirms the competence of accredited third-party laboratories performing the evaluation.

Applications

Fill factor numerical evaluation supports a wide range of engineering and commercial activities:

• Research and development of new cell architectures, where FF is tracked to quantify gains from passivation, metallization, and contact engineering.
• Manufacturing quality control, where inline FF screening identifies process drift in firing, plating, and texturing steps before modules leave the line.
• Incoming inspection and supplier qualification, comparing measured FF against datasheet values and agreed acceptance limits.
• Certification and type approval testing under the IEC 61215 and IEC 61730 series, where initial and post-stress FF values demonstrate design qualification and safety performance.
• Energy rating and yield modeling under IEC 61853, where FF dependence on irradiance and temperature feeds performance simulation software.
• Failure and degradation analysis, where FF decline over time, combined with pFF and Rs trends, distinguishes contact degradation, corrosion, and recombination-driven power loss in fielded modules.
• Insurance claims, engineering due diligence, and bankability assessments requiring independent, standards-based FF verification from accredited third-party laboratories.

FAQ

What is considered a good fill factor value?

For commercial crystalline silicon cells, a fill factor above 0.78 is generally acceptable, with premium TOPCon and HJT products reaching 0.82 to 0.86. Thin-film and perovskite devices typically achieve 0.65 to 0.80. The appropriate target depends on the technology, cell area, and design, so results should be compared against the theoretical FF and manufacturer specifications rather than a universal threshold.

How is the fill factor calculated from an I–V curve?

The measured I–V curve is converted to a power–voltage (P–V) curve, and the maximum power point (Pmax = Vmp × Imp) is located by numerical curve fitting rather than simply selecting the largest sampled point. FF is then Pmax divided by the product of Voc and Isc obtained from the same corrected curve. Using a fitted maximum avoids underestimating FF when the sampling interval near the knee of the curve is coarse.

What are the most common causes of a low fill factor?

High series resistance from poor contacts, busbars, or interconnects; low shunt resistance from edge leakage, cracks, or shunted regions; and elevated recombination shown by a high ideality factor. Comparing measured FF with pseudo fill factor from a Suns–Voc measurement is the fastest way to determine whether resistive or recombination losses dominate.

Which standards govern fill factor measurement?

The IEC 60904 series defines measurement principles, reference devices, spectral mismatch correction, and simulator classification; IEC 60891 defines correction procedures; IEC 61215 covers module qualification including initial FF determination; and IEC 61853 addresses performance testing and energy rating. Equivalent procedures are provided by ASTM E948 and ASTM E1036, the GB/T 6495 series, and JIS C 8904 series. Laboratories performing these tests are typically accredited to ISO/IEC 17025.

How accurate can a fill factor measurement be?

With a class AAA solar simulator, calibrated reference cell, four-wire connections, and controlled device temperature, the combined expanded uncertainty (k = 2) on Pmax and FF is typically within 1 to 2 percent for cells and 1.5 to 3 percent for full modules. The uncertainty budget must be documented because FF uncertainty contributes directly to reported efficiency and power ratings.

Why do perovskite cells show different fill factors depending on the scan direction?

Perovskite devices exhibit ionic migration and interfacial capacitance effects that produce I–V hysteresis, so forward and reverse voltage sweeps yield different apparent FF values. Reporting protocols therefore require defined scan rates, pre-bias and light-soaking conditions, and stabilized maximum power point tracking to obtain a representative fill factor.