I am after some lighting recommendations to emulate daylight at a reasonable cost on a benchtop. My main requirement is to match the spectrum of daylight as closely as possible and then make it roughly uniform over an area of 30*30cm to 1m^2.
I have read that short arc xenons are very good for this but they seem fairly rare and also have the tendency to blow up; the latter part makes me a bit uneasy.
The next thing I've come across are car HID's that use a combination of xenon and metal halides, would anybody know what their output spectrum is like?
First, are you trying to just match spectrum or spectrum and intensity? A 1 m^2 surface would need upto 1kW of energy landing on it, and probably 2kw+ at the bulb. As far as short arc bulbs, you simply have a time totalizer hooked up with its power supply to measure running hours, and replace at 50-75% of the estimated service life.
If you are going for the above 2kw+ youre probably into long arc xenon teritory rather than short arc.
Thanks very much for the reply. Yeap, spectrum and intensity.
The high power consumption of the 1m˛ makes me want to move more towards the 30x30cm size and the 10x lower power use.
Does replacing at 50-75% of service life reduce the likelyhood of shattering by a significant amount? We would only be using it for 4-5 hours once in a while and I dont see this rig being used too heavily.
Would typical car headlamp HID lamps be roughly sunlight-like? they seem quite easy to come across on ebay for cheap.
I'm no short arc expert(someone else chime in, please!, or perhaps ask a mod to move this to spotlights and HID subforum as there isnt too much traffic though here) , but I believe that lamp explosions are and end of life behavior (obviously, not like its working after that ), not just something that happens mid service life, provided the lamp hasnt been abused by contamination on its surface, or running with excessive power, but thats the same as any other high power bulb. Hence retiring them with plenty of service life still left on the dial. Ive been working with a 75W short arc that went bang (counter intuitive controls on the 200W variable power supply and a single analog meter that read volts, amps and watts = bang, but that bulb was being severely over driven, and it took it for ~20 minutes) Your fixture design should still account for the possibility of something happening, to at least protect an operator, and hopefully other expensive lamp parts.
I cant find any real emissions spectra of an automotive type HID lamp, but searching it seems they have a CRI of 65-75 depending on models, which would make it seem reasonable to expect that the emission spectrum isnt so nice. vs 95+ for short arc lamps.
Do you care if your light is collimated or is diffuse fine? If you want collimated to, you're going to run into significant losses, it looks like newport optics is using a 1600w lamp (!) for a collimated solar simulator with an 8"x8" beam and getting ~1kw/m^2 with some filters in use. Which makes me suspect that even if their output spectrum was nice, 35w HID lamps may need to come by the gaggle if you were going to light up even a 30cm x 30 cm square with a 1 sun insolation.
Is your testing short duration? (load a panel, flash test it with it at ambient temps to check its power output, load next panel repeat)
http://www.luminyscorp.com/ has a lot of high-power, systems, some meant for crash tests etc, where they arent running constantly, using short arc and long arc xenon lamps and may be suited for your testing if more diffuse/non collimated light is acceptable. And they might even have something that can run for longer durations if thats what you need.
Aah thats a good idea, I'll try asking a mod to move it.
Sweet - thats somewhat of a relief. From the warnings I've seen scattered around about xenon lamps exploding; I would've thought they're like live hand grenades or something.
The application is to compare some solar cells / panels that we buy and to calculate efficiency. We also want to look into how cut cells (more square packing and higher voltage strings) differ in efficiency to uncut cells.
Since we will be testing mainly with one or two types of cells I'm now thinking that CRI will not be such a dealbreaker as it is not expected that the spectral response of the cells will change as they are cut and we are more interested in a comparison between cells.
Would you have any recommendations on where one might source such high power lamps on the cheap?
I guess if the price is too high we'll have to stick to natural sunlight with a reference cell!