Cree recommended Xlamp LED DIY Build

Stephenj37826 said:

As far as the Cree reference vs the Gavita this could very easily have a different outcome if they installed the SR reflector insert into the Gavita. Or one could check out Nanolux. They have a square reflector made to target a 4X4 area.

As far as quantum boards vs the reference design. I tend to think if the same amounts of money was invested in a Quantum boards build they would be superior.

Click to expand...

IDK, I think those 5w diodes w/ optics will hold their par value at greater depths. Plus they can add a few key colors for enhancement. Maybe the QB would be a cheaper build but I'm not sure it will outperform the Refferrnce design.
Guess time will tell.
DIY QB are pretty affordable I hear. What do you plan on selling yours for manufactured? Will you offer it on different currents or just make it powerful w/ a dimmer? Can those QB be enhanced? Any optics chosen yet?
They sound great. Have you chosen the diodes you want yet?

Plant Lobbyist said:

Multiple light sources are better than single ones. Redirection of light destined for outside the growing area will also increase PPFD within the growing area. Focusing will accomplish the same thing. The reference design includes technologies used in LED lighting that waste less energy on heat and producing photons that aren't used for growing. Same thing with QBs, strips, etc.

Reference design target audience is commercial operations. Makes it easy to anticipate operational costs over the expected lifespan and allows the entity to have whoever they want assemble using an already vetted design. When you are looking to have 10k fixtures of this size built, you will probably be going with a fairly well established vendor to fulfill the order.

Of course that isn't to say we can't use them for hobbyist purposes. Effective growing area is what we all want to know from a light and this provides enough information for us to determine how it would perform.

Click to expand...

What photons that aren't used for growing? They all are. Every single nm.

Hybridway said:

IDK, I think those 5w diodes w/ optics will hold their par value at greater depths. Plus they can add a few key colors for enhancement. Maybe the QB would be a cheaper build but I'm not sure it will outperform the Refferrnce design.
Guess time will tell.
DIY QB are pretty affordable I hear.

Click to expand...

You could use this as your starting point for the reference design fabrication. It may not be exactly like the reference but 10 of these will get the 144 white XPG3 part and give you 6 extra. And save you a ton of soldering.

At 1.4 amps 10 of these are 630 watts. The design calls for 1.05 amps though.

ichabod crane said:

You could use this as your starting point for the reference design fabrication. It may not be exactly like the reference but 10 of these will get the 144 white XPG3 part and give you 6 extra. And save you a ton of soldering.

At 1.4 amps 10 of these are 630 watts. The design calls for 1.05 amps though.

Click to expand...

I like it!

I would like to possible use these for a build but still trying to get info on them.

CobKits said:

that the gavita casts a lot of light outside of the 4x4 area which is completely discarded and they compare it with highly focused lenses in an area it perfectly covers

that they claim a light that is 1.82 umol/J in a sphere is twice as efficient as one that is 1.72

its maybe one step above that dude who measures oohmoles at single point, with some fancy graphs and a cree logo on it.

Click to expand...

Simple answer, optics.

sixstring2112 said:

I would bet 800w of low driven quantum boards would rival this cree experimental lamp.you could do like 4 boards 3000k ,2 boards 4000k and 2 boards 5000k.all the colors of the rainbow driven super soft like 50w per board i think they are pushing close to 200lpw

Click to expand...

But they'd never spectral mix & would only hold #'s worth growing for about 6" if that,no?

I believe the design runs XPE far red but I could be wrong. But if it does I would run these to save on soldiering.

http://www.rapidled.com/solderless-cree-xp-e-far-red-led/

I like that cutter has them on the same board as posted above but I think they should be spread out more and these would allow that. If cutter could make the above board with these mixed in then you would need to spread them out. But I think they are run at different maximum amps. But not sure with out looking it up.

Hybridway said:

What photons that aren't used for growing? They all are. Every single nm.

Click to expand...

Doesn't look like english is your native language. The reference to producing photons not used for growing are for those landing outside of the growing area, absorbed by other material, etc and has nothing to do with light characteristics.

Besides, all photons regardless of wavelength hitting a plant surface will do something. Happens everyday.

So I checked the first post and I was wrong. I post above that it was XPE far red but it was not. It is instead the photo red. Here is a link to that instead.

http://www.rapidled.com/solderless-cree-xp-e-photo-red-660nm-high-bin-led/

I also checked the max drive amperage and the XPE is 1 amp and the XPG3 is 2 amp. So if they were mixed the XPG3s would only be able to run at 1 amp instead because of the XPEs.

Plant Lobbyist said:

Doesn't look like english is your native language. The reference to producing photons not used for growing are for those landing outside of the growing area, absorbed by other material, etc and has nothing to do with light characteristics.

Besides, all photons regardless of wavelength hitting a plant surface will do something. Happens everyday.

Click to expand...

Understood.

The Aavid heatsink is $800 for single 12" piece :/ I also doubt the 2.12" tall fins conduct heat to the tips. I'd say heatsinks USA 7.28" extrusion works at least as well, for $31.56 usd per foot. The beauty of discreet LED is distributed heat footprint (COB concentrate heat in a relatively small area). Having (more) fins right to the edge better manages heat. The only advantage to aavid is the neat shape that allows a clean 80/20 fixture.

My thought was to eliminate the photo red LED, and use all 3500K 70CRI. Coloured LED are much less efficient than the white LED, so using all warmer LED will exceed the reference light's output (and much more in the 550-650 range).

The reference light is ~10% more efficient than the Gavita, not including the 20% that's above 700 nm (15% of pure IR heat). So 700 W LED probably equals a 1000W Gavita (plus A/C watts).

fwiw, while the reference light has a very small footprint, a single light engine works very well in a 2' x 2' tent (while keeping it 18" above the canopy). I currently have four COB @ 1.4 A (hate the flimsy pin heatsinks, btw) and plan on trying the light engine soon.