Altermatt Lecture:   The Solar Spectrum

 
 

3.5:  Why is a standardised AM0 spectrum necessary?

There are international standards defined for the AM0 spectrum [1][2][9]. Why is this necessary? Isn't the spectrum clearly defined and measurable?

Parts of the spectrum vary with solar activity (especially in the UV, and the intensity of some spectral lines). These relative variations can be ignored in photovoltaics, as these changes hardly influence solar cell performance. However, there is one main reason why we need a standard AM0 spectrum in PV:

The AM0 spectrum, integrated over its entire range, must equal the solar constant S, and this is not necessarily the case in real measurements due to daily variations (see the figure on the page "Measurement of the solar constant") and due to uncertainties in calibration. Hence, the measured data needs to be scaled.

Another reason is not so relevant to PV, but apparent when comparing different am0 spectra: the spectrum depends strongly on the resolution of the measuring instrument. The spectrum is composed of about 25,000 spectral lines. The figure below shows the theoretically calculated high-resolution spectrum (yellow) and a measurement one (black line); the latter is measured of course with limited resolution. With a smaller resolution of the spectrometer, more and more lines are smoothed out. The higher the resolution of the spectrometer, the more of these lines appear.

→ A list of obsolete AM0 spectra and solar constants.

Solar irradiance vs wavelength (two resolutions)

Figure: Yellow: calculated high-resolution spectrum, containing about 25,000 spectral lines [10]. Black: measurement with realistic resolution [1].

 

Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Refs

Please email corrections, comments or suggestions to support@pvlighthouse.com.au.

Advertisements