Nice.
Couple questions/observations.
Is there something between the leds and water to keep them from being splashed? or getting condensation/salt spray on them??
It looks like you have a clear cover over the back of the heatsinks? this will prevent them from cooling and they will burn up without a fan forcing air into the unit. (especially if you are using "ebay special" button type leds (anything other than cree or luxeon).?
With the LEDs spaced that close on the channel (assuming 1" channel as you said and not the more common inch and a quarter) you will not be able to drive the LEDs much over 350mA without a fan for cooling or they will over heat and burn, plus you won't be able to touch the heatsink without possibly getting burned. However if you have a fan to blow even a small amount of air over the heatsinks I'd have no reservation running them up to 500mA. (700mA is really streaching it for those type LEDs) As close as they are to the screens I don't think you'll need more than 100mA to produce sufficient PAR for a good mat of algea.
In the order of your questions.The LEDs getting way hotter than the heatsink makes sense, especially after taking a close look at them. They do look like buttons, must be what I have. I wonder if a tiny bit of thermal grease at the button/star junction would help.
It shouldn't be too much to get fans in position. Does a pair of 60mm fans centered perpendicular over each scrubber sound good? The fans I was looking at are 12V. Would they need a separate power supply? Could the magic of the arduino use the same timer for the fans as the lights?
I'll have to re-think lenses. I was thinking that since the LEDs are so close, the best thing would be no lenses....One good thing is the light is close to 100% PAR. Should be anyway.
How fast are LEDs advancing? Is what was best last year what gets sold on eBay for cheap?
I found some 5V fans I think would do the trick. They have a 2 pin connection, is that what I want? Some fans have 3 pin connection. Is it possible to connect the fans to the same dimmer that controls the LEDs? If that adds much complexity, I'll run the fans wide open all the time. I have another 5v wall wart with plenty of amps.
I diagrammed what 60 degree lenses would do with this setup and was quite surprised at the nice overlap. I would not have considered lenses, great suggestion-they are on the way.
Thanks guys!
A 3 pin fan is usually still just a simple DC motor and can be powered without the 3rd wire (which is usually a rpm sensor). Fans vary greatly in amperage, ones i use use something like 80mA on a 120mm fan, while others i have had use as much as 500mA. They should say what they use on the sticker over the shaft on the fan frame.
I was thinking of nutrients, but not as a limiting factor- that's a good point. I was thinking about adding a measured amount of flake food every day to try to keep things consistent. If it does end up nutrient limited, it should be easy to see as the O2 production will fail to spike.
This article is what gave me the idea for this scrubber. The 5th page down has a graph of O2 production. It looks like switching the scrubber lights somewhere in the 4 hr. range might be best for max production, assuming a 4 hr. rest is sufficient. I was surprised that in low light, algae can be a net consumer of O2.
http://www.int-res.com/articles/meps/134/m134p207.pdf
The parts are getting here, it's still gonna be a couple three weeks or so until everything is here.