banner image
No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
Illumination-induced errors associated with suns- measurements of silicon solar cells
Rent this article for


Image of FIG. 1.
FIG. 1.

Homogeneity measurement of the suns- measurement chuck for a flash height of 87 cm. An array of solar cells in a 2 cm pitch was used in order to perform the whole measurement during the decaying edge of one single flash. The intensity deviation from the brightest to the darkest spot is approximately 7.5%. The deviations in the local open-circuit voltage have been calculated using the one-diode model [see Eq. (1)]. The reference cell is located approximately 2.5 cm outside the measurement chuck.

Image of FIG. 2.
FIG. 2.

Schematic of the symmetry element that has been used for the 3D network simulations. Please note that the solar cell grid illustration does not represent the actual simulated dimensions. The measured flash intensity data have been averaged between the four corresponding quarters.

Image of FIG. 3.
FIG. 3.

Schematic of the components used for the network simulations. The graph shows the components underneath two fingers, as indicated by the orange line in Fig. 2. An analogous layout is used for the axis along the fingers. curves for the illuminated and dark nodes were generated with PC1D (Ref. 10) featuring parameters for a standard industrial-type silicon based solar cell. The laterally inhomogeneous illumination intensity of the flash light was taken into account by shifting the curve, which was generated with PC1D for an illumination intensity of one sun, along the current axis, assuming a linear dependence of generated current and illumination intensity.

Image of FIG. 4.
FIG. 4.

Measured relative spectrum of the xenon photoflash (including two diffusers), which is used for illumination in the suns- measurement setup. The spectrum was measured in the timeframe from 1.2 to 1.8 ms after the peak flash intensity. The spectrum is compared with the AM 1.5G spectrum (Ref. 7).

Image of FIG. 5.
FIG. 5.

Normalized responsivity [see Eq. (2)] for an assumed solar cell with , calculated for the AM 1.5G and three of the measured flash spectra. The flash spectrum does not only differ from the standard AM 1.5G spectrum but also changes with flash intensity (Measurement 3 relates to a flash intensity of approximately 1 sun, measurement 8 to 0.2 suns and measurement 13 to 0.05 suns).

Image of FIG. 6.
FIG. 6.

Measured spectral responses of the crystalline silicon thin-film solar cell (solid line) and the high-efficiency silicon reference cell (with and without protection cover) (dashed lines).

Image of FIG. 7.
FIG. 7.

Calculated spectral mismatch factors of the investigated crystalline silicon thin-film solar cell for different thicknesses of a KG2 short-pass filter. Typical measurement regions for and are shaded. The voltage measurement error is calculated using the one-diode model [see Eq. (1)]. The flash spectra were measured consecutively with decreasing light intensity. Every flash spectrum measurement took approximately 0.6 ms.

Image of FIG. 8.
FIG. 8.

Normalized responsivity for an assumed cell with , calculated for the AM 1.5G and one of the measured flash spectra, with and without a 1 mm thick KG2 short-pass filter.

Image of FIG. 9.
FIG. 9.

Dependence of the illumination intensity on the height of the sample related to the height of the reference cell. The distance of the flash to the reference cell was 87 cm. The measurement data were once fitted using a theoretical model assuming a pointlike light source [see Eq. (6)] and once using a polynomial, whose equation is shown in the inset.


Generic image for table
Table I.

Parameters used in the distributed circuit simulations.

Generic image for table
Table II.

Averaged flash illumination densities for various measurement areas. Based on these averaged illumination densities, the measurement error has been calculated using the one-diode model. In addition, the measurement error calculated with the 3D network simulation for a metallized solar cell, utilizing the measured illumination density profile, is shown.

Generic image for table
Table III.

Measurement results for the investigated crystalline silicon thin-film solar cell. The area of the cell is . Calibrated steady-state measurements were carried out at the Fraunhofer ISE CalLab. Results for both correction methods are presented, showing the excellent results, which were obtained using the easy to apply KG2 short-pass filter. Please note that the pFF is expected to be higher than the FF from the steady-state measurement since the suns- does not account for series resistance effects.

Generic image for table
Table IV.

Investigated sources for inaccurate voltage measurements due to inappropriate illumination. All data are related to a flash height of 87 cm above the measurement chuck. While the spectral mismatch error significantly influences the suns- measurements for some kinds of silicon based solar cells, the error due to an inhomogeneous illumination can be neglected in most cases. The error due to a height difference of the sample and the monitor cell can be significant, making a correction necessary.


Article metrics loading...


Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Illumination-induced errors associated with suns-VOC measurements of silicon solar cells