The reason I decided on the 15V 4A meanwell is because that left the least amount of volts unused (I think it was .2V), which means that I could run the meanwell slightly above it's "rated" voltage, or run the leds very slightly under their "rated" voltage. Maybe I'm still thinking in resistances etc, but less wasted voltage=less resistance=less heat, which may or may not be negligible.
I was thinking this was a good fit, assuming parallels work
http://www.meanwelldirect.co.uk/pro...with-Dimming-Function/ELN-60-15-D/default.htm
I could go the 48v route, but I'm still wondering what makes parallels difficult? I've taken the electricity and magnetism course, but haven't had any hands on experience yet (or very very little, that can't really be applied here, since it was in a "lab" setting).
I'm also wanting to build a light for a 29 gallon and a 40B, so I'm wanting a somewhat upgradeable design, and preferably one driver maybe two I guess. The most I can run off of the 15V and 48V are 12 leds at 800ma. For the 24V, I can run 18 leds (3 strings of 6), but I've only been looking at this line of meanwells, as this is what was recommended to me by a guy on another forum. I think I'm wanting 1 led per maybe 12-18 in^2, and my 10 gallon has 200 in^2. My 29g has around 360 in^2 of surface, and the 40 has ~650 in^2. Assuming I do this at the high end of the "lighting spectrum", then I'll need 20 leds for the 29 and 36 for the 40B. For the 40B, it looks like the best solution is two 24V's, and for the 29, maybe I'll just go with 1 24V, and just go with slightly lower lighting.
This is looking good so far, and looking easier than I originally thought.
Do you have any ideas on a moving sun-type using an arduino and a rtclock? Or maybe some way to control the brightness/dimness of the leds with a timer? Maybe this will be too big of a hassle...