Solar Radiation Simulator

not a chance dude

Get a light meter and measure the output from your bulbs. Convert lux to watts per square meter. (1 lux = 1 lumen/square meter)

http://www.rapidtables.com/calc/light/lumen-to-watt-calculator.htm

https://en.wikipedia.org/wiki/Solar_simulator

You should start by reading through MIL-STD-810, section 505.5 'Solar Radiation (Sunshine)'. It will give you some idea of what you are dealing with.

From my personal experience, two 500 Watt light bulbs won't be enough to generate the 1kW/m2 insolation (solar loading) you want. Certainly not at a distance of 0.55 meters.

One thing you will need to do, since these lamps probably radiate over nearly a full 4 pi steradian spherical region, is cover the interior of the room with as pure a white as you can get. The rule of thumb for interior reflections off a diffuse surface and then shining onto an opening is Gain = (R1/R2)⁴, where R1 is the reflectivity of surface #1 and R2 is the reflectivity of surface #2 (and 4 is the typical number of photon bounces). So if your walls have a reflectivity of 80%, painting them with a white paint having a reflectivity of 92% will give you a gain of (0.92/0.80)⁴ = 1.75, or a 75% increase in luminance at the opening. There are plastics sheet materials, such as the DRP material from the W L Gore company, that have a reflectivities of 97% or better. The higher the reflectivity, the better the net output of the lamps will be by this reflectance gain.

You might also want to google 'solar radiation test labs' and see what they offer in terms of testing and consulting work with what you need.

To get more help than this, hire an optical engineering consultant. It takes optics engineers a lot of learning and experience to know how to do this properly. You're not likely to find much useful help for free.

Luckily there appear to be no requirements concerning spectrum matching.

.

Your surface will need to be far less than 1 square meter because most of the combined 1000 watts will be leaving via convection from the hot lamps. A deep parabolic reflector is probably necessary.

.

The answer to question #3 is provided in imposed condition #4: 0.55 m.

YESSS!! We got a match!

NOPE!!!

What is the purpose of your simulation, to test solar panels, color rendition, resistance to fading, aging, lighting patterns, etc., etc.? If your goal is "...simulate solar radiation..." then clearly your choice of bulb color spectrums needs to be rethought since leaving out a third of the available color temperature range will leave your "simulation" lacking in credibility.

Follow the links from others and/or go to Google for information on how NASA does it for space probes, they've probably done more testing of this phenomena than anyone on the planet.

Contact General Electric Industrial Products. About 30 years ago they made "Solartron" wet tube solar panels and had a unit similar to what you want. It was comprised of a large panel of lights mounted on a 3 dimensional frame. It was for testing the panels and for R+D work. Those there might not remember the product let alone the light panel. It was not a item they sold but a one of a kind part of their solar panel testing system.

The whole product line and equipment was sold to a guy in Orange, New Jersey. He then sold it to a Chinese corp and it was shipped there for their use.

They would probably be the best ones to talk to about the equipment they use/used. Why spend time developing something that has already been developed?

Good Luck, Old Salt

Thank you for the overwhelming answers. The test is to analyse the performance of a product under direct solar radiation (simulated of course) and how does the measurement of that sensor(product) change with the intensity of light.

The main problem which I am encountering is that I am not able to find the 500 W bulbs. I can only find 400W or a 1000W. I plan to incorporate these bulbs in a cooled box so as to negate the effect of the temperature of the bulbs to the room temperature.

Any ideas as to where I can find these 500W bulbs. And as per one of the answers, I understood that the efficiency of these bulbs would not be so that receive 1kW/m2 on my surface which is definetly much less than 1m2 in area. And would painting the walls white help my purpose?

Also, I need to know whether using a parabolic reflector increases the intensity or rather the power on my product?

On that basis, either:

• The above advice isn't getting in, or
• The above advice is getting in and isn't taking root, or
• The above advice is taking root and isn't being heeded.

At least one of these needs to change before any headway can be made on this thread.

You could use simple arithmetic- Like say purchase a 100W, 5 units of it will be 500W. Isn't it?

Anyway bulb ratings are not lumens or light intensity - I believe they mean input or consumption of making light, heat included consequently.

Look, I don't think you should build the solar simulator. Getting the spectrum close is not going to be easy. If the spectrum is not close, of what valuable will the results be?

Aside from the spectrum issue, there are other problems. Metal Halides don't dim very well. Varying the intensity can still be done, but it is just another thing that is going to be an inconvenience at best.

.

This is what you should do: Your solar simulator would need to be calibrated to see what level of light it provides the test piece, right? Okay, take the calibration instrument and the test piece outside into sunlight (you will need to do this during the daylight hours) and compare the test piece response to the light intensity measured by the calibration instrument.

I could not find how much area you need illuminated and at what uniformity.

I can tell you that I had success using 4 General Electric ELH lamps for about a 4" x 4" area at a distance of about 3 feet, BUT it;s deficient in blue. I had to use DC power supplies. You also need a glass diffuser in front of the bulbs. Spectrum changes with the voltage applied to the lamps.

Intensity was adjusted by moving the lamp array. Uniformity, by tilting the lamps.

Because of the dichroic filter, the IR from the lamp goes out the back, so you can't block the back of the lamp. You will also need cooling (A fan is OK) and you need to get rid of the heat. We ducted it to the space above the ceiling tiles.

Eventually this was supplemented with a Oriel solar simulator and we were only concerned with 1 square inch of uniform light. It used a 1000 W arc lamp and spectral shaping filters. The lamp was rated for 1000 hours and was replaced at that interval because as the lamp ages, it becomes more deficient in blue light.

I was operating at the AM 1.5 Global spectrum at 100 mW/sqcm.

Maintaining the surface temperature at 25 degrees C was harder.

Daily calibrations were done using a rotating set of say 15 solar cells where 2 were used each day.

So, you used the one you used the day before and a new one to calibrate the intensity. For applications that were sensitive to blue light, we used a filtered solar cell. We were lucky to have procured a large number of 1 cm silicon solar cells without the anti-refection coating. Our calibrations were traceable to NREL.