DIY LED array

gconcepcion · Post by **gconcepcion** » Fri Mar 12, 2010 2:16 pm

LED prices have almost dropped into the affordability range recently...

RapidLED.com sells reasonably priced DIY arrays w/ 3w Cree XR-E LEDs.

How many 3w LEDs do you think would be sufficient to light a 29 gallon (30x12x18). and put out enough light to grow any plants I want? Currently I have a 150 watt HQI (6700k) on my 29 that was just set up, but i'm thinking that I want to make the jump to LEDs...

Anyone have an LED array set up on a tank around 29 gallons? or with a similar foot print or depth?

Post by **krisw** » Fri Mar 12, 2010 2:23 pm

Greg, are you sure that these LEDs emit the proper wavelengths to encourage photosynthesis? I imagine that the best way to figure out how many LEDs you'll need is to find out how many lumens each emits, and compare that to an existing MH or T5 fixture. I think you're paving new ground, but if you do go forward with this project, a lot of us would be interested in your results.

gconcepcion · Post by **gconcepcion** » Fri Mar 12, 2010 3:14 pm

krisw wrote:Greg, are you sure that these LEDs emit the proper wavelengths to encourage photosynthesis? I imagine that the best way to figure out how many LEDs you'll need is to find out how many lumens each emits, and compare that to an existing MH or T5 fixture. I think you're paving new ground, but if you do go forward with this project, a lot of us would be interested in your results.

The CREE XR-E Q5 cool whites burn in around ~6500K while the Royal Blue CREE XR-E's emit light approx the 450-460nm range (almost perfect for chlorophyll-B which has an absorption peak around 453nm) but maybe slightly shifted out of the range of the major chlorophyll-a absorption peak of (430nm)

CREE XR-E's are the LEDs that most of the reefers are using now a days to grow all of their scleractinians (SPS corals)...

Unfortunately the typical LUMENS or watts per gallon rule isn't effective for quantifying the amount of actual photons the plants are receiving. They are measures for our visible spectrum, not for the spectrum at which photosynthesis occurs and I'm worried that any decisions based on these metrics would not be accurate, and as a nerdy scientist, I like to be as accurate as I can... however the concept of LED's is still pretty new to me.

Ideally i'd like to take quantitative measurements of photons being emitted with a photosynthetically active radiation sensor underwater at the substrate level with both the metal halide I have set up, and some arbitrary number of LED's to test... unfortunately I don't have an LED set up to test on yet, nor do I know anyone who does... (within a 20 mile range) which is why i'm politely asking if anyone has an LED array set up at home that might let me come and take some measurements?

thanks for the input! i'll make sure to keep you guys updated.

SCMurphy · Post by **SCMurphy** » Fri Mar 12, 2010 4:52 pm

I have LED bars that only emit light in the two chlorophyll activation spectrum. Ugly light to the eye, tasty light to the plants.

gconcepcion · Post by **gconcepcion** » Fri Mar 12, 2010 9:39 pm

What brand LED bars are they & what wattage are the LEDs?

I went ahead and pulled the trigger on a 12 x 3 watt (6x CoolWhite; 6x Royal Blue)
dimmable DIY retrofit kit for $145 for a total of 36watt LED. This would give me a 1.24WPG estimate for a 29 gallon, but I don't trust that so I will set the array up, get some quantitative PAR measurements of the number of photons being emitted compared to the 150 watt HQI already on the tank, and from there decide if and how many more I want to add to the array, or if maybe some interesting T5HO/LED combo would be better or more economical.

I also need to pick up a heatsink to mount the LEDs which should probably run me in the $40 range.

It might be a week or two (or three) before I figure out how to wire this thing properly but I will report back w/ results

Conagher · Post by **Conagher** » Sat Mar 13, 2010 8:36 am

Cree separates their LEDs into bins (and sub-bins) that you can specify to obtain a very specific wavelength of light.
I have been looking into this as well for close to a year, and the price has been still a little bit more than I wanted to spend; and I have not been able to nail down the actual number of LEDs I would need. This seems to be the major sticking point on a most DIY LED lighting projects, as (most) people do not have access to required instruments that will give a good PAR measurement for their target.

I will be watching this project closely.

gconcepcion · Post by **gconcepcion** » Sat Mar 13, 2010 4:55 pm

Conagher wrote:Cree separates their LEDs into bins (and sub-bins) that you can specify to obtain a very specific wavelength of light.
I have been looking into this as well for close to a year, and the price has been still a little bit more than I wanted to spend; and I have not been able to nail down the actual number of LEDs I would need. This seems to be the major sticking point on a most DIY LED lighting projects, as (most) people do not have access to required instruments that will give a good PAR measurement for their target.

I will be watching this project closely.

LED configurability options are unparalleled in lighting technology, I haven't even begun to get that in depth, i'm so overwhelmed by all of the choices! But it sure would be great to be able to specify the exact wave length for every single 1w (or 3w) LED!

I'll just stick with the basics for this first fixture, cool white 6700k LEDs and royal blue 450-460nm LEDs... maybe i'll get more adventurous after I gain some confidence from building this first array.

I stopped by the lab this morning for some quick and dirty PAR readings measured in micromoles of photons/m^2/sec under my HQI and another 90W LED array we ordered for a research grow chamber that was delivered yesterday that I had completely forgotten about. Our PAR sensor is calibrated to detect photosynthetically active radiation between 400nm - 700 nm exclusively.

The 150W HQI (temporarily 14,000K) was giving readings of ~65-75 umol at the substrate level with the 120W LED clocking in at a whopping 450-460 umol !

Now this was FAR from rigorous testing, and not exactly a scientific comparison because the HQI is positioned 12" above the water surface, whereas the LED array was resting on the tank rim, but they are the positions where the lights would be if they were permanently installed on the tanks.

At any rate, i'm definitely excited to see how a 36watt LED array compares to my 150W HQI. It may not be as far behind the HQI in terms of PAR output as I was expecting...

more later...

oh, and btw, you can watch the progress of the tank and updates @
http://www.phototrophic.net

gconcepcion · Post by **gconcepcion** » Mon Mar 15, 2010 8:06 pm

Gotta hand it to rapidled.com.

Excellent customer service and super fast shipping. I ordered the kit on Friday, and it was delivered monday morning.

So fast that I haven't even decided on/ordered a heat sink to mount them to yet.

I'm psyched to wire it up. I think I will actually get some decent PAR readings with the current number of diodes. Maybe only need to add another half dozen maximum.

more updates soon

SCMurphy · Post by **SCMurphy** » Tue Mar 16, 2010 10:04 am

This is one of the arrays I have. I would love to add some white light LED's to it. Unfortunately I know nothing about electronics.

http://www.groworganic.com/item_GP834_S ... ro_Ba.html

JSnyder · Post by **JSnyder** » Tue Mar 16, 2010 4:56 pm

gconcepcion wrote: I'm psyched to wire it up. I think I will actually get some decent PAR readings with the current number of diodes. Maybe only need to add another half dozen maximum.

What size tank are you going to be using this on? 29g? What degree optics are you using? The problem with led's is coverage. The tighter the optics, (better par in deeper tanks) the closer the led's have to be to avoid spotlighting. I think you'll more than happy with the par readings, depending on the depth of the tank (and optic choice) you might find it to be too much. This was the case with my led fixture, I run them at no more than 600ma.

DIY LED array

DIY LED array

Sponsors