I think most commercial sun simulators use Xe lamps but basically the only criterion is that you need to get near the AM1.5G solar spectrum. So you can try to find the spectrum of your lamp, the AM1.5G spectrum can be found on the NREL website, and you can compare.
I used a small battery operated laser pointer to test optical alignments for a solar parabolic dish system. It is quite effective and helps to see the reflection angles for the various parts of the solar reflector and dish mirrors. I would use the car xenon lamp if it is optical aspects you want to test.
I used to test my DSSCs in actual sunlight at around 12noon which gives nearly AM1.5 G conditions. Which gives nearly the same results as that of solar simulator. Thus either you test solar cells in actual sunlight if you dont have Xenon lamp. and then finally once you make a proper device then you can test at actual solar simulator available nearby.
Any comments by colleague friends on this answer are most welcome.
Dear Suhail, I do not agree with your suggestion since it is not stable and your measurements will befriend big errors. While measurement accuracy relatively should be high (micro) even tiny changes in your environment, truly will affect the results.
I think providing a handmade simulator is a good and simple way and simpler but not better way is using your car's xenon lamp. however you should do your experiment in a close dark place like a garage. and it's crucial to fix your solar cell's position during the experiment. and once you be able to get appropriate lamp (at AM1.5G solar spectrum) you can repeat your test and compare them with your previous experiments and then share the consequences with us.
I mean by using Xenon lamp that used in cars, is putting the lamp inside a reflector and making a housing for this simple system in order to perform my test in completely dark. the reflector as the one used for the fog lighting as shown in below image
metal halide lamps have significantly different spectrum than xenon arc ones. I think the best choice is looking for a low cost commercially available solar simulator or buy a suitable xenon arc lamp.
Dear Sabah Gaznaghi, I agree with your answer. But the answer which i had added was just to test the cells in actual sunlight when solar simulator is not available to optimize the process. Once, the process is optimized then we can test in near by labs. with solar simulator.
It is based on your application sensitivity to spectrum. If you are regarding PV applications, it is more sensitive to incident spectrum (as the case in the following standards: IEC 60904-9, ASTM E927, ASTM E948, ASTM E1036, ASTM E1362, JIS C 8912 and EN 60904). In this case, Xe arc lamp is the best then the metal halide lamp is the next cheaper choice. On the other hand, if you are carrying out thermal tests only, it is not important to cover the perfect spectrum, according to the following standards (ISO 9806, ISO 19467, EN 12975-1, EN 12975-2, EN 12976-1 and EN 12976-2). If this the case, I would recommend tungsten-halogen lamp to be used for its simplicity and to give you a large amount of heat (although, may be larger than real sun due to the larger power content in the IR bandwidth of its spectrum compared to the solar spectrum)
I just received the following sharing link (valid for 50 days from 24/04) for my new review paper in "Renewable and Sustainable Energy Reviews" on sun simulators light source selection: