Quote Originally Posted by thepaan
So, I've been toiling over some math for the last few days and I think I have come up with something epic. From the top even though a lot is going to be repeat:

Photosynthetically active radiation (PAR) is the wavelength of light which plants respond to 400-700 nm. The daily light integral (DLI) measured in moles per square meter per day (mol/m2/d) says you need 22 or more to grow excellent crop plants. I couldn't find a listing for MJ but here is a chart (down on page 5) for a bunch of other plants. LEDs have light output measured in watts (W). 1 watt = 1 joule per second. I have calculated mol/J (and µmol/J) for every 10 nm between 400 and 700, inclusive and found the average. I then converted DLI (mol/m2/d) to µmol/m2/s for 12 hours of light and divided by the average µmol/J to get W/m2. Lastly, I multiplied by 60% which is the average efficiency of all wavelengths between 400 and 700 nm at driving photosynthesis (the idea being that deep red is 100% efficient).

This is the spreadsheet I came up with. It tells you how many red LEDs and from that you can figure out how many blue per square meter you should be using. What do you guys think?
I think that is a good approach and is great data - allow me to toss out another approach.

First, I kept in mind that plants pretty much adapted to their lighting environment. Light levels and spectral balance is somewhat important. Next, for the boundary between spring and summer, I checked with NASA records to check the insolation levels recorded at the edge of the atmosphere and at the edge of sea level - I noticed the balance, at least in terms of radiometric power, were almost even - blue was in the lead. Then I checked again for the beginning of September (bear all of this is for my old area, in Tennessee, I haven't checked to see if there has been any data for my area of Cali.) The ratio had changed , with red slightly in the lead. So I figured a 60:40 blue:red mix for vegetation, and a 60:40 red;blue mix for blooming. I might modify the panel once it arrives, see if I can't drop a dimmer on the blues so I can drop the blue down to help trigger flowering, and then bump it back up a little to provide some extra boost. I've always had poor yield results, especially with tomatoes and peppers, with those 7:1:1 panels. More blue is needed. Other lighting companies realized this and are starting to offer blue-dominant panels to supplement their lacking UFO panels. NASA realized their folly and some of the latest panel design I've seen from them is pretty much 1:2 blue:red (looks like 465 and 670nm.)
khyberkitsune Reviewed by khyberkitsune on . Calling out to Weezard for LED advice My friend, I've read your stuff here and at Steve F.'s site. I want to build a dedicated flowering lamp using Ledengin 15 watters. (4 reds, 1 blue, and possibly 1 warm white; LM 317's so I can tweak the ratios.) Essentially, a copy of your 4+1 cake pan lamp but built on a 4 x 18" finned heatsink so I can see what's going on under it.) If I were to post my picks for power supply, resistor ratings / values, etc., would you critique and perhaps offer some guidance? I have already purchased Rating: 5