Solar Simulators — Abridged Guide

Quick-reference guide to solar simulator technology — classification, light sources, performance metrics, and spectral mismatch correction. For full derivations and worked examples, see the Comprehensive Guide.

Comprehensive Solar Simulators Guide →

1.Introduction to Solar Simulation

A solar simulator is an artificial light source calibrated to approximate the spectral irradiance of natural sunlight under controlled laboratory conditions, enabling standardized photovoltaic and materials testing independent of weather, location, or time of day.

The three-letter classification (e.g., AAA) rates spectral match, spatial uniformity, and temporal stability — in that fixed order. "Class AAA" means all three metrics are Class A.

Solar simulators were developed in the late 1970s to standardize photovoltaic device measurements. Every published solar cell efficiency is measured under solar simulator illumination at standard test conditions: 25 °C cell temperature, 1000 W/m² irradiance, AM1.5G spectrum. Applications extend beyond PV to materials weathering, photocatalysis, space qualification, and photobiology.

2.Solar Spectrum and Air Mass

Air Mass

\text{AM} = \frac{1}{\cos\theta_z}

Air mass quantifies the atmospheric path length sunlight traverses. AM0 is the extraterrestrial spectrum (1366 W/m²); AM1.5G is the terrestrial standard (1000 W/m², zenith angle 48.2°).

AM1.5G includes both direct and diffuse radiation and is used for flat-plate PV testing. AM1.5D (direct only, ~900 W/m²) is for concentrator PV. AM0 is for space applications.

The sun radiates approximately as a 5800 K blackbody. Atmospheric Rayleigh scattering, molecular absorption (O₃, H₂O, CO₂), and aerosol scattering attenuate and reshape the spectrum as it passes through the atmosphere. The standard AM1.5G reference spectrum was normalized to 1000 W/m² for convenience and is defined in ASTM G173-03.

3.Classification Standards

Spectral Match Ratio

R_{\text{SM},i} = \frac{E_{\text{sim},i}\,/\,E_{\text{sim,total}}}{E_{\text{ref},i}\,/\,E_{\text{ref,total}}}

Each spectral bin must fall within 0.75–1.25 for Class A, or 0.875–1.125 for Class A+ (IEC 2020). The overall class is set by the worst bin.

IEC 60904-9:2020 uses equal-irradiance bins (300–1200 nm) and introduces two new metrics — spectral coverage (SPC) and spectral deviation (SPD) — especially relevant for LED simulators.

Metric	A+	A	B	C
Spectral match	0.875–1.125	0.75–1.25	0.6–1.4	0.4–2.0
Spatial non-uniformity	≤ 1%	≤ 2%	≤ 5%	≤ 10%
Temporal instability (STI)	≤ 0.25%	≤ 0.5%	≤ 2%	≤ 5%

Solar Simulator Evaluator →

4.Light Sources

Xenon arc lamps provide the best native spectral match to sunlight (~5800 K blackbody) but have short lifetimes (~1000 hr) and require AM filters. LED arrays offer 50,000–100,000 hr lifetime and spectral tunability but may have spectral gaps between emission peaks.

For standard silicon PV testing, xenon is the proven workhorse. For multi-junction or perovskite cells requiring spectral tunability, LED arrays are increasingly preferred.

Parameter	Xenon Arc	LED Array
Lifetime	1,000–2,000 hr	50,000–100,000 hr
Spectral match	A+ (filtered)	A+ (multi-channel)
Tunability	No (filter-based)	Yes (per channel)
Warm-up	30–60 min	Seconds
Explosion risk	Yes	No
IR coverage > 1100 nm	Yes	Limited

Metal halide lamps offer lower cost but weaker collimation. QTH lamps supplement other sources for extended IR coverage but cannot match the solar spectrum alone due to their low (~3400 K) color temperature.

5.Optical System Design

The optical system collects lamp emission (ellipsoidal reflector), homogenizes spatial distribution (integrating rod or fly-eye array), conditions the spectrum (AM filter), and delivers a collimated beam to the test plane.

Spatial uniformity improves with better homogenization but typically degrades as illumination area increases. Always verify uniformity over the full rated area, not just at the beam center.

Illumination areas range from 2 × 2 inches (51 × 51 mm) for small-cell testing to 12 × 12 inches (305 × 305 mm) for large benchtop simulators. Larger areas require higher lamp power — a 150 W xenon lamp covers ~2 × 2 in. at 1 sun, while 1600 W is needed for 12 × 12 in. Natural sunlight has a divergence half-angle of ~0.27° (32′ arc); most benchtop simulators have larger divergence unless specifically designed for CPV or space simulation.

6.Performance Metrics

Spatial Non-Uniformity

\text{SNU} = 100 \times \frac{\max(E) - \min(E)}{\max(E) + \min(E)} \quad (\%)

Temporal Instability

\text{TIE} = 100 \times \frac{\max(I_T) - \min(I_T)}{\max(I_T) + \min(I_T)} \quad (\%)

Spatial non-uniformity uses min/max irradiance across the test plane (64+ measurement positions per ASTM). Temporal instability uses min/max of a time-series. IEC 2020 separates short-term (STI) and long-term (LTI) instability.

The 2020 IEC metrics of spectral coverage (SPC) and spectral deviation (SPD) are not yet required for classification but reveal spectral quality that the binned match hides — especially important for LED simulators.

7.Calibration and Mismatch Correction

Spectral Mismatch Factor

M = \frac{\int E_{\text{ref}} S_{\text{ref}}\,d\lambda \;\cdot\; \int E_{\text{sim}} S_{\text{test}}\,d\lambda}{\int E_{\text{sim}} S_{\text{ref}}\,d\lambda \;\cdot\; \int E_{\text{ref}} S_{\text{test}}\,d\lambda}

The mismatch factor M corrects for the difference between the simulator spectrum and the reference spectrum, weighted by the spectral responses of the reference and test cells. Apply M when the reference cell technology differs from the device under test.

For silicon-on-silicon measurements with a Class A xenon simulator, |M − 1| < 0.5% — often negligible. For perovskite or organic cells measured against a silicon reference, M correction is essential.

8.Steady-State vs. Pulsed

Steady-state simulators (xenon or LED) provide continuous illumination — simple but thermally loading. Pulsed (flash) simulators deliver 1–100 ms pulses, reducing thermal load and enabling multi-sun testing up to thousands of suns.

Flash simulators are standard for production-line module testing. Steady-state is preferred for research where spectral stability during the measurement is critical. LED flash offers microsecond control without the spectral transience of xenon flash.

9.Applications

Applications span PV characterization (silicon, thin-film, perovskite, multi-junction), materials weathering, photocatalysis, space qualification (AM0), photobiology, and concentrated solar power research.

Match the simulator spectrum to the application: AM1.5G for terrestrial PV, AM0 for space, UV-focused for weathering and photobiology. Emerging technologies (perovskites, organics, tandems) are more spectrally sensitive than silicon and may require higher spectral fidelity or tunability.

10.Selection Guide

Identify standard

IEC 60904-9 (global) or ASTM E927 (North America).

Classification level

A+AA/AAA for research, ABA for production, B–C for education.

Illumination area

Size to fully cover the largest DUT with margin.

Reference spectrum

AM1.5G (terrestrial), AM0 (space), custom (weathering).

Source technology

Xenon (broadband), LED (tunable), hybrid (widest range).

Reference cell match

Match ref cell technology to DUT or apply mismatch correction.

Select a solar simulator by first identifying the governing standard and required classification, then sizing the illumination area, choosing the source technology (xenon vs. LED vs. hybrid), and establishing a re-verification schedule for spectral, uniformity, and stability drift.

Lamp-based simulators have lower upfront cost but higher consumables cost (lamp replacement every ~1000 hr). LED simulators have higher capital cost but near-zero consumables over their 50,000+ hr life. Factor in total cost of ownership, not just purchase price.

Need	Best Source
Broadband coverage > 1100 nm	Xenon arc or hybrid
Spectral tunability	LED array
Lowest operating cost	LED array
High collimation (< 1°)	Xenon arc
Multi-sun (> 10 suns)	Xenon flash or LED flash
Production-line throughput	Xenon flash

Continue Learning

Comprehensive Guide →Solar Simulator Evaluator →

The Comprehensive Guide includes 6 worked examples, 5 SVG diagrams, 3 data tables, and 10 references.