Concerntrating Sun Light

  • Thread starter Kamilan
  • Start date
Hi All,

Not sure where this topic should be placed but it involves optical and electronics.

I want to design a solar simulator that can concentrate the light onto a photovoltaic of area of 3600mm2. The concentration of sunlight target is 500suns to 1000suns. The goal is to be able to expose the photovoltaic to concentrated sun light for long periods of time simulation weather conditions. (weeks, months).

There are pieces of equipment like a "Flash Tester" but these can only be operated for short periods of time.

Now.... a possible off the shelf solution is to purchase this solar simulation chamber.

My plan is to buy a 75cm satellite dish and spray paint it with reflective chrome paint. Then determine the best focal area to concentration level using sunlight. Then determine if this would work within the constraints of the chamber.

My question is:

1) Is the above plausible.

2) What is a good resource for determining the dimensions of the parabolic dish so that I can ensure I will get a focal area of 3600mm2.

Any help or focus would be a great help.




Gold Member
I can offer only this suggestion: consider not using "spray paint with reflective chrome paint" and instead, use this:

"ReflecTech® Mirror Film has high reflectance in the wavelength range important for sunlight. In the graph below, hemispherical reflectance of the film is plotted on the left axis while the irradiance of sunlight is plotted on the right axis, both as functions of wavelength. The solar weighted average of hemispherical reflectance is 93%."


Gold Member
A concentrator such as you plan will melt the cells you wish to test unless cooling is provided.
Fortunately it should be possible to water cool a 6cm square cell with garden hose level flow rates.
Hi Bobbywhy and Etudiant,

As it turns out the problem gets more and more complicated.

The parabolic dish idea is still in. The light source needs to be collimated.

CR = D^2/Fa^2

Concentration Ratio = Diameter of the Dish squared / Focus Area squared.

This means to achieve my target of 500 suns the dish would need to be 1.5m in diameter.

This changes the strategy. The goal now is to project 1000W/m2 onto an area of ~1.8m^2 considering it is a paraboloid shape.

The light source(s) not only need to be to supply ~1Sun onto the area, but also be collimated as well which means losses in the collimation lens as well.

More to think about. But that's for the information.




Gold Member
I'm confused about your requirements.
Noon sun gives about 1000 watts/m**2, which you now state is your goal.
So why does that change the strategy, so you don't need 500 or 1000 suns?

More generally, if the problem is to reliably get 1000W/m**2, a simple corner reflector should be adequate.
Hi Etudiant,

If I can successful project collimated light onto a parabolic surface of 1.8m^2 with an intensity of 1000W/m^2 at the surface, I can then concentrate that light to hopefully 500 suns onto my target area of 3600mm2.


Gold Member
Forget lenses and collimation, just make the capture dish bigger and place your sample outside of the focus point, closer to the concentrator surface.
That way you have a zone of adequate intensity that remains even as the sun moves. A trough concentrator would be a more workable solution than a dish.


Gold Member
You’ve not responded to etudiant’s question about heating of your PV cell. At 500 suns what quantity of heat will be absorbed by your cell? What will result from that? Does your cell’s manufacturer give data on temperature vs current output, for example? Or operating temperature range limits?

Instead of trying to invent a new solar concentrator to focus solar energy onto a PV cell you may find useful ideas from those who have struggled with this issue and who have achieved satisfactory solutions. Oftentimes we can discover useful techniques for our own projects from what others have already done.

Here are two technical papers that address your proposed project:

Comparison of Solar Concentrators Concentrator Models/V Trough/RAB76.pdf

Solar Concentrators

Here is a source of a variety of Compound Parabolic Collectors:

Here is a string of search terms one could enter into “Google Search” that result in hundreds of examples: “photovoltaic compound parabolic solar collector”

Hi Bobbywhy,

Looking back, I have been vague on the details. I want to make a solar simulator. The problem is the sun is not consistent. Therefore testing cannot be repeated, and must wait for the next day with appropriate DNI to achieve test results.

The aim of this research is to find a feasible way to simulate 500 suns onto the target area to gather data on the Concentrated PV system.

These CPV systems require 500 to 1000 suns to achieve optimal performance. You cannot just go out and buy a lamp.

However, I am trying to design a method to achieve this without breaking the bank.



Gold Member
Is the daily variation in the sun really so serious an issue when you are looking for week or month long periods of exposure?
Obviously if you live in some region such as Seattle, where sunshine is intermittent, there would be an issue, but assuming you are living in a more favorable environment, the variation should not be a deal breaker.
Moreover, you could set up a linear trough concentrator which would illuminate several samples concurrently.
That way, your samples are uniformly exposed even if the exposure intensity fluctuates.


Gold Member
Suddenly the project you are describing has become more clear! It’s easy to see why you cannot use the actual sun for your tests. From Wikipedia:

“Solar spectrum at the Earth surface changes constantly depending on the weather and sun position. This results in the variation of φ(λ), QE(λ), α(λ) and thus the short-circuit currents JSCi.”

“Light concentrators increase efficiencies and reduce the cost/efficiency ratio. The three types of light concentrators in use are refractive lenses like Fresnel lenses, reflective dishes (parabolic or cassegrain), and light guide optics.”

Our Federal Government works in this area:

"NREL Solar Cell Sets World Efficiency Record at 40.8 Percent"
August 13, 2008 [Broken]

"Large-Area Continuous Solar Simulator (LACSS)
The LACSS is a custom-designed system that is used for measuring module performance. It uses a Spectrolab model X200 continuous 20-kW xenon short-arc lamp to provide a filtered light source to simulate the global spectrum (IEC 60904-3, class A). It is capable of measuring modules as large as 152 cm × 122 cm with spatial nonuniformity of ±3%. Sample temperature is measured with a spring-loaded, type T thermocouple at the back of the module." [Broken]


Here is an opportunity for you to tap the knowledge of our Federal program, presumably for free:

“We also serve as an independent facility for verifying device performance for the entire PV community. We help the PV community solve its special measurement problems, giving advice on solar simulation, instrumentation for I-V measurements, reference cells, measurement procedures, and anomalous results. And we collaborate with researchers to analyze devices and materials.”

Here are three companies that build highly concentrated Solar Simulators:

“System allows for testing Terrestrial Solar Cells at fluence levels between 625 and 1175 suns.”

“The fully reflective design utilizes a series of mirrors, including a special folding mirror that bends the light source without the drawbacks of refractive optics. This allows the simulator to direct the light beam from the arc lamp source to the target plane, producing a higher intensity, uniform illumination at the target plane than a competitor’s simulators. In Sciencetech’s line of fully reflective solar simulators the power output will typically be 1.3 times more powerful when compared to simulators that use the same wattage of arc lamp with diffusers to make the light field uniform at the target.”

“For applications where spectral match and control is critical, such as multi-junction solar cell testing, our close-match technology is invaluable.”
Last edited by a moderator:


Gold Member
A simulator that provides spectral control, uniform illumination and 500-1000 sun intensity is a tall order and certainly not cheap. Fortunately there is no indication that this is what is actually needed.
The requirement:
The goal is to be able to expose the photovoltaic to concentrated sun light for long periods of time simulation weather conditions. (weeks, months).
To me, that means outdoor exposure and accepting fluctuating solar illumination intensity, because that is the real world.
A more controlled environment does have the huge advantage that tests are comparable even if run at different times, but the price will be high for that benefit.

Physics Forums Values

We Value Quality
• Topics based on mainstream science
• Proper English grammar and spelling
We Value Civility
• Positive and compassionate attitudes
• Patience while debating
We Value Productivity
• Disciplined to remain on-topic
• Recognition of own weaknesses
• Solo and co-op problem solving

Hot Threads