LED based solar simulator testing method

[jishnup]

Sir,
I'm an undergraduate in 'Electronics'.Intending to develop a solar simulator with LED as the light source.
It is understood that the spectral irradiance of sun can not be generated with the LED.So my plan is to light up a number of high power LED arrays of different color ( at present we have chosen three colors ), with a proportionate irradiance that of solar spectrum. This would be achieved by adjusting the current through the arrays.For this LM317 is to be configured as the current regulator.Up to here i think am OK.
But with this project, i need to show the solar cell testing too. Searched hardly in the web. Couldn't find what am longing for.Many times some in built devices are found.But none of them are mean for LED based simulators.However i need to implement the circuit with discrete components. Some of the read papers says that since the radiant intensity of sun can not be reproduced ,some special method are needed for determining the spectral response.And they gave one equation for short circuited photovoltaic current in terms of absolute spectral response and number of incident photons. But i couldn't understood nothing more. That paper is attached here.
Can you suggest any circuit or method for the testing of LED based solar simulator..
I'm actually in a situation to leave this project here if couldn't show the testing .Please help me...
Faithfully
<< personal e-mail address deleted by the Moderators >>

 

[berkeman]

Welcome to the PF.

My initial reaction is that 3 LED wavelengths would provide an inadequate simulation. Can you comment on why I say that? And can you comment on what would be required for an adequate simulation within some tolerance?

 

[sophiecentaur]

Welcome to the PF.

My initial reaction is that 3 LED wavelengths would provide an inadequate simulation. Can you comment on why I say that? And can you comment on what would be required for an adequate simulation within some tolerance?

My thoughts too.
A Halogen lamp is a pretty 'danged' good simulation for Sunlight as it has no holes in its spectrum (even better with the appropriate filter over it). This can't be said for RGB LEDs. If you are contemplating an LED source, you will probably not be thinking in terms of high powers so any energy savings wouldn't be very relevant.

 

[jishnup]

My thoughts too.
A Halogen lamp is a pretty 'danged' good simulation for Sunlight as it has no holes in its spectrum (even better with the appropriate filter over it). This can't be said for RGB LEDs. If you are contemplating an LED source, you will probably not be thinking in terms of high powers so any energy savings wouldn't be very relevant.

Sir,
Thank you for the reply. The actual project is meant for partial implementation. We intend to use this to measure the spectral response charecteristic of a solar cell. From what I understand, RGB leds and white would be suffcient for this purpose. Can someone point me to a reliable method for testing of spectral response of a solar cell?

 

[Bobbywhy]

This National Laboratory tests PV performance and offers “more information”:

“Spectral responsivity (SR) measurement is an important part of the National Renewable Energy Laboratory (NREL) photovoltaic (PV) device performance assessment process. Spectral responsivity systems measure how a device responds to selected narrow (spectral) bands of irradiance. Responsivity is measured in units of amps per watt versus wavelength and reported in terms of quantum efficiency (QE) — a measure of how efficiently a device converts incoming photons to charge carriers in an external circuit. SR systems measure the spectral response at different temperatures (10° to 80°C) voltages (±15 V), light levels (0 to few suns), and different chopping frequencies (<0.2 to 400 Hz).”

http://www.nrel.gov/pv/measurements/spectral_responsivity.html

 

[sophiecentaur]

Sir,
Thank you for the reply. The actual project is meant for partial implementation. We intend to use this to measure the spectral response charecteristic of a solar cell. From what I understand, RGB leds and white would be suffcient for this purpose. Can someone point me to a reliable method for testing of spectral response of a solar cell?

RGB LEDs will not have a continuous spectrum and cannot produce a specific wavelength to order. The very fact that they are described in this way implies that we're talking Colourimetry and not Spectroscopy. To find spectral sensitivity of a PV cell, there is no point in producing illumination with a combination of R,G and B LEDs which 'looks' a particular colour. The apparent colour tells you nothing about the details of the actual spectrum of the light that's being produced.

Whatever source you use (and a continuous source like a filament lamp would be best), you will need a prism or diffraction grating to isolate narrow bands of wavelengths in order to find the PV sensitivity over the whole spectrum, one band at a time. Some 'optics' is necessary for this and it may be worth while getting hold of a basic spectrometer rather than DIY.

Google Student Spectrometer Sale. A few $100 would get you something suitable, I'm sure- even less on the second hand market.