Solar Simulator Classification and Calibration

For a light source to be classed as a solar simulator, it must be evaluated according to one of three standards, and comply with the specifications set out within. The three organisations that provide solar simulator standards are:

• ASTM International (ASTM E927-19 Standard Classification for Solar Simulators for Electrical Performance Testing of Photovoltaic Devices)
• Japanese Industrial Standards (JIS C 8904-9 Solar simulator performance requirements)
• International Electrotechnical Commission (IEC 60904-9:2020 Classification of solar simulator characteristics)

While there are subtle differences between the standards, their overall approach and classification system is largely the same.

Solar simulators are assigned a rating in three different performance tests:

1. Spectral match to a solar spectrum
2. Spatial non-uniformity of irradiation
3. Temporal instability of irradiation

In each area, the solar simulator will receive a rating between A and C (or A+ and C in the case of the latest IEC standard) depending on its performance. A is the highest rating that can be achieved, and C is the lowest. The total solar simulator rating is therefore a 3 letter grade for example ABB. For clarity, the order of these gradings must be in the following order: spectral match, spatial non-uniformity and temporal instability. For example, a solar simulator may be given an ABB rating, indicating it has achieved an A for spectral match, a B for spatial non-uniformity and a B for temporal instability. Below we will review the requirements to achieve the classifications in each performance test, and how they are calculated. The Ossila Solar Simulator is measured against the IEC 60904-9:2020 standard, so we use this as a basis, but the principles are common to all three standards.

Spectral Match to Solar Spectrum

---

A solar simulator spectrum should approximate the spectrum of light received on Earth i.e., the relative intensities of ultraviolet, blue, green, red, and infrared light emitted by a solar simulator should ideally be the same as the relative intensities of those received from the sun. This is complicated somewhat by the fact that the light received by the sun varies greatly depending on several factors including location, time of day, weather, time of year etc. To simplify and standardise matters, a series of reference solar spectra are defined — such as the AM1.5 Spectra. Solar simulators are measured against the AM1.5G spectral irradiance standard as defined in ATSM G173-03 and IEC 60904.

The latest IEC 60904-9:2020 standard only considers the wavelength range between 300 nm and 1200 nm. This is largely because the responsivity of most solar cell technologies falls within this range. The only exception are cells made from small band gap materials such as germanium or indium nitride which can absorb light up to a wavelength of approximately 1800 nm. This part of the spectrum is separated into six ‘bins’, each containing approximately the same level of solar irradiance (older standards still use 6 bins, but the size of each bin was based on wavelength intervals rather than percentage irradiance each contained).

Bin	Start Wavelength (nm)	End Wavelength (nm)	Percentage of Integrated Irradiance
1	300	470	16.61%
2	470	561	16.74%
3	561	657	16.67%
4	657	772	16.63%
5	772	919	16.66%
6	919	1200	16.69%

Classification	Spectral Match of All Bins
A+	0.875 – 1.125
A	0.75 – 1.25
B	0.6 – 1.4
C	0.4 – 2.0

All bins must be within the range given to achieve the corresponding classification e.g., if bins 1 – 5 have a spectral match of 1.12 (A+), but bin 6 has a spectral match of 0.65 (class B), the measurement would be class B for spectral match.

The spectral match calculation must be performed in at least four locations within a specified ‘test area’ in the solar simulator test plane. This test area is the area over which the classification is measured and valid. This will be the area where you should place your device to be tested. These four locations are generally chosen at the outer edges of the test area. Each location is assigned a classification, and the overall solar simulator classification for spectral match decided by the lowest performing location. This ensures that the spectral irradiance across the test area does not vary by over the test area. Therefore, lateral device location should not affect your results, as long as your device is within the test area.

There are no requirements for the total irradiance of a solar simulator given in the standards. However, the standardised AM1.5G solar spectrum has total integrated irradiance of 1000 W/m² which is commonly referred to as 1 sun intensity and most solar simulators are able to achieve at least this irradiance.

Spatial Non-uniformity of Irradiance

---

The light received from the sun is of uniform intensity on a flat plane (neglecting local effects such as shadows). In other words, over a small area sunlight is evenly distributed. A solar simulator should also produce a uniform light output. Homogenising optics are often used to achieve this. High spatial uniformity is important for solar simulators, as non-uniform light intensity would make measurements location dependent which we want to avoid.

The non-uniformity of irradiance is evaluated by moving a light intensity measuring device, e.g., a reference solar cell or a photodiode, across the test plane of the solar simulator in order to build a 2-dimensional grid of the light intensity it measures. If the light is distributed evenly, this measurement should not change much across the test area.

The required detector properties and scan parameters are dependent upon the size of the test area and the intended use of the solar simulator (e.g., for photovoltaic cells or modules, single or multijunction devices). The points of maximum irradiance (I_max) and minimum irradiance (I_min) are taken from the grid, and the spatial non-uniformity is calculated using the following equation.

 

 

This calculation yields a percentage. This percentage is then compared to the values in the table below, to determine the spatial-non uniformity classification.

Classification	Spatial Non-uniformity (%), S
A+	S ≤ 1
A	1 ≤ S ≤ 2
B	2 ≤ S ≤ 5
C	5 ≤ S ≤ 10

Temporal Instability

---

Temporal instability represents how much the light output intensity of the solar simulator changes over time. A high temporal instability will make measurements less repeatable and can make results appear noisy. Temporal instability can result from number of factors relating to the hardware of the solar simulator, for example noise in power supplies or arc flutter in discharge lamps. This flickering can appear over very short timescales (less than a second) or as intensity drift over longer timescales (days or weeks). For this reason, temporal instability is classed by either short-term instability (STI) or long-term instability (LTI).

STI relates to any intensity variations that occur between the collection of data points e.g., the points on an I-V curve. This would generally be on the order of seconds to minutes. The maximum and minimum illuminance recorded during the I-V curve measurement (I_max and I_min respectively) are often used to calculate the STI. Determining a solar simulator’s STI relies on the ability to track illuminance during a measurement. However, this is often not possible for I-V measurement systems.

LTI is associated with timescales long enough to acquire a complete dataset. For example, the time needed to take a full I-V curve or the duration of a radiation exposure experiment. Depending on how the solar simulator, this can be anywhere from seconds (flash measurements) to months (such as for solar cell stability experiment). An illuminance measurement is taken before and after the complete measurement, and these two values (the highest being I_max and the lowest being I_min) are used to calculate LTI.

 

 

The calculated STI and LTI values are then used to assign a classification according to the table below.

Classification	Short term Temporal Instability (%), STI	Long term Temporal Instability (%), LTI
A+	STI ≤ 0.25	LTI ≤ 1
A	0.25 ≤ STI ≤ 0.5	1 ≤ LTI ≤ 2
B	0.5 ≤ STI ≤ 2	2 ≤ LTI ≤ 5
C	2 ≤ STI ≤ 10	5 ≤ LTI ≤ 10

Spectral Coverage and Deviation

---

Spectral coverage and spectral deviation from the standard AM1.5G spectrum are two new values that should be assigned to a solar simulator. They are not used in the classification but give an indication of how well the spectral shape of the solar simulator matches the solar spectrum. They have been introduced as new solar simulator technology (especially LED technology) has changed the overall shape of the irradiance.

Spectral coverage (SPC) is a measure of how much of the solar spectrum is represented in the solar simulator emission. It is the percentage of the irradiance within the 300 nm – 1200 nm region where the irradiance from the solar simulator (IS) is at least 10% of the irradiance of the AM1.5G spectrum (IR). A SPC of 100% indicates that the entire spectral range has at least 10% of the ideal irradiance. A small SPC indicates that the solar simulator spectrum has gaps in the irradiance.

 

 

Spectral deviation (SPD) is a measure of how closely the irradiance from a solar simulator follows the shape of the AM1.5G spectrum. Solar simulator irradiance above and below the standardised spectrum both contribute to a higher SPD. An SPD of 0% indicates a perfect match.

 

Solar Simulator

Browse now

Learn More

---