Solar simulators for photovoltaic metrology and laboratory calibration require spectral matching, spatial uniformity, and temporal stability to meet IEC 60904-9, ASTM E927-05, and JIS C 8912 standards. The evaluated system achieves spectral match ratios of 0.82–1.17 across the 400–1100 nm band, spatial non-uniformity better than ±2%, and temporal stability down to Class A (±0.5%) on select models, with irradiance adjustable from 0.7 to 1.2 suns. This article examines the optical engineering trade-offs behind these figures, the honest limitations of single-lamp designs, and what buyers should validate before procurement.

Why Spectral Matching Became a Hard Constraint

During a qualification run at a PERC-to-TOPCon transition fab, a process engineer noticed that cells passing the simulator-based I-V test were underperforming in flash-test verification. The root cause was spectral mismatch in the 900–1100 nm band, where the legacy simulator deviated by 18% from AM1.5G. At 24.3% average cell efficiency, that deviation translated to a 0.15% absolute efficiency error—enough to misgrade thousands of cells per shift.

This is not an isolated incident. As cell architectures shift from PERC to TOPCon and HJT, quantum efficiency curves become steeper at the band edges. The 400–500 nm and 900–1100 nm bands, once considered "good enough" if within ±20%, now directly distort open-circuit voltage and short-circuit current measurements. Spectral matching has moved from a secondary specification to a pass/fail criterion.

Optical Design: The Single-Lamp Trade-Off

Why One Lamp Instead of Many

The evaluated system uses a single xenon arc lamp feeding an ellipsoidal reflector, a metal turning mirror, an AM1.5G filter, an optical integrator (field lens + projection lens), and finally a glass reflector with collimating optics. This is a high-integration path: fewer optical nodes mean fewer alignment drift sources over time.

Multi-lamp systems can push total irradiance higher, but each lamp ages at a different rate. After 500 hours, spectral drift between lamps can exceed the Class A tolerance. The engineering team prioritized path efficiency and thermal management over raw power, focusing tolerance budgets on the ellipsoidal reflector surface figure and the microlens array uniformity of the integrator.

Beam Geometry and Power Scaling

Four spot sizes are offered: 50×50 mm, 50×50 mm (

Lamp power scales from 300 W to 1000 W. Nominal output is 100 mW/cm² (1 sun), adjustable from 0.7 to 1.2 suns via a knob-style power controller. The 1000 W lamp paired with the 160×160 mm spot operates near its thermal safety boundary at 1.2 suns—an honest boundary that buyers should note if they plan to run extended high-irradiance campaigns.

AM1.5G Filter Engineering

The AM1.5G filter shapes the xenon continuum into the standard terrestrial solar spectrum. Measured spectral match ratios across six bands are:

Wavelength Band (nm)	Match Ratio	IEC Class A Range
400–500	1.04	0.75–1.25
500–600	1.02	0.75–1.25
600–700	1.01	0.75–1.25
700–800	0.82	0.75–1.25
800–900	1.17	0.75–1.25
900–1100	0.95	0.75–1.25

The 700–800 nm ratio of 0.82 sits at the edge of the Class A window (0.75–1.25). For silicon cells, this band lies on the quantum efficiency decline slope, so fill-factor measurement error is limited. However, in perovskite/silicon tandem structures, this band gains weight. The engineering team has disclosed that the next filter formulation targets >0.90 in this band.

The 50×50 mm base model and 160×160 mm infrared-optimized model achieve Class A (0.7–1.2) across all bands. The 100×100 mm model reaches Class A+ (0.875–1.125) in select bands.

Spatial Uniformity and Temporal Stability

Uniformity: The Repeatability Gatekeeper

Spatial non-uniformity is rated at Class A: better than ±2%. This means the peak-to-valley irradiance difference across the active spot does not exceed 4% of nominal. For standard 156 mm or 182 mm cell formats, this uniformity level caps efficiency measurement error from spot variation at approximately 0.1% absolute.

Temporal Stability: Model-Dependent Performance

Temporal stability splits by model. The 50×50 mm base unit achieves Class B (±2%), while the 100×100 mm and larger models reach Class A (±0.5%). The divergence comes from arc wander in the 300 W lamp under sustained load, versus the tighter power-feedback loops in the 500 W–1000 W configurations. For line-sorting applications, Class B is typically sufficient. For R&D or contract testing requiring tighter uncertainty budgets, the power controller module can be upgraded to push the base model to Class A.

Application Flexibility: Beyond Downward Illumination

Configurable Beam Directions

The system supports downward (standard), upward, and side-exit beam configurations. Upward exit allows mounting beneath a glovebox port for inert-atmosphere cell processing and immediate testing. Side exit integrates with spectrometers or electrochemical workstations without folding mirrors that degrade uniformity.

Direction changes are not mechanical rotations of the entire chassis. They require replacement of the terminal optical assembly and re-sealing of the thermal convection path. The modular design preserves the internal optical train, so conversion does not require factory re-alignment.

Ecosystem: Closing the Test Loop

A solar simulator alone does not complete a measurement. The vendor offers three companion products: a temperature-controlled sample stage, a 2 cm×2 cm monocrystalline silicon reference cell, and a precision source-measure unit.

The reference cell is vacuum-encapsulated on an aluminum base with an integrated Pt100 RTD and four-wire Kelvin connections to eliminate lead resistance. The source-measure unit performs I-V sweeps with simultaneous sourcing and sensing. Together, these form a closed loop from photon delivery to electrical parameter extraction. For buyers, single-vendor integration reduces interface risk and shortens qualification timelines.

Honest Limitations: Where Single-Lamp Designs Stop

Despite the AAA-class headline figures, single-lamp architecture has hard edges.

Irradiance ceiling.A 1000 W xenon lamp at 1.2 suns on a 160×160 mm spot is near its thermal limit. Concentrated photovoltaics (CPV) testing at hundreds of suns requires multi-lamp or LED array architectures. This system is not a fit for CPV.

Consumable lifetime.Xenon lamps degrade. The constant-power feedback loop slows decay but does not eliminate it. Replacement intervals run from several hundred to one thousand hours, depending on operating current. Buyers must model this into total cost of ownership (TCO). LED-based simulators offer 10×–100× longer source life, though they currently trail in spectral continuity and near-infrared coverage.

Spectral lock-in.The AM1.5G match data assumes standard terrestrial conditions. AM0 (space) or AM2.0 (high-latitude) applications require custom filter sets and re-calibrated spectral response functions. The standard filter cannot be re-purposed across atmospheric mass conditions without re-qualification.

Industry Context: Supply Chain Resilience in Metrology

The release of these specifications reflects a broader pattern: emerging

For research institutes, the advantage of emerging suppliers often lies in customization turnaround and spare-part availability. For production-line buyers, procurement cost and TCO typically outweigh marginal parameter differences. The cross-industry applicability—material aging, photocatalysis, photobiology—depends on the multi-directional beam design and the promise of non-photovoltaic filter kits in late 2026.

Frequently Asked Questions

What defines a Class AAA solar simulator?

A Class AAA rating requires simultaneous compliance with Class A in spectral match (0.75–1.25 per IEC 60904-9), spatial non-uniformity (<±2%), and temporal stability (<±0.5%). The evaluated system meets or approaches these thresholds depending on model and configuration.

How does single-lamp architecture affect long-term spectral drift?

Fewer optical nodes reduce alignment drift, but the single arc source means all spectral power scales together as the lamp ages. The constant-power controller maintains irradiance, yet the spectral shape shifts subtly with electrode erosion. Re-calibration against a reference cell every 50–100 hours is standard practice.

Can the system be used for perovskite or tandem cell testing?

Yes, with a caveat. The 700–800 nm match ratio of 0.82 is acceptable for silicon but may introduce error in tandem structures where this band carries higher current-generation weight. Buyers should validate against their specific quantum efficiency curves or wait for the next filter revision targeting >0.90 in this band.

What is the realistic lamp replacement cost and interval?

At typical production-line duty cycles (16–24 hours/day), lamp replacement occurs every 6–12 months. Replacement cost varies by regional logistics but should be budgeted at 8–12% of initial equipment cost annually for consumables and re-calibration.

How can I independently verify spectral match and uniformity claims?

Request a third-party calibration certificate traceable to NIST or a national metrology institute. For spectral match, demand raw spectroradiometer data (not just ratio tables) across the 300–1100 nm range. For uniformity, ask for a 49-point irradiance map measured with a calibrated radiometer at the specified working distance and temperature. Run the same test on your reference cell at receipt and after 30 days of operation to establish baseline drift.

About This Guide

Data Sources:IEC 60904-9:2020, ASTM E927-05, JIS C 8912, product technical documentation, and industry public information aggregated from published specifications.

Author:Optical Metrology Content Operations, Precision Measurement Equipment Communications.

Disclosure:Jingyi Optoelectronics manufactures solar simulators, spectrometers, optical integrating spheres, and LiDAR calibration targets. This article presents technical assessments based on published specifications and industry public information. No compensation was received from third-party brands mentioned.

Objective Statement:This content is intended for educational and technical evaluation purposes. Equipment selection should always include independent proof-of-concept validation under your specific process conditions and spectral requirements.

Last Updated:July 2026

For detailed specifications and application notes on solar simulators, search "Jingyi Optoelectronics solar simulator" or visit the technical library.