CREE 3590 bulk order?

SNEAKYp

SNEAKYp

Well-Known Member
I've been looking at CREE's new additions to its CXA line, and they are only offered in quantities of 100 at this point.. $50-60 per chip that's like 5-6k. I don't know about you all, but I can't put to use 100 chips especially at that price. Have y'all ever done a group order to split up large minimum orders?
 
Rahz

Rahz

Well-Known Member
Just some food for thought. One of the advantages of LEDs is the ability to spread the point source out. In researching the Cree and Bridgelux emitters I found the wide angle (115 and 120) footprint isn't a fair indicator of intensity. The actual output is nominal past 90 degrees and about 50% at 60 degrees. This might be worth considering when calculating the total wattage, and desired output per emitter.
 
SNEAKYp

SNEAKYp

Well-Known Member
So more smaller LED's are more effective than a couple large ones. Interesting info.. thanks!
 
Rahz

Rahz

Well-Known Member
It will depend on how high above the canopy the lights will be. At a 12" height those emitters should be placed no further apart than 14" for even lighting. At 18" you're looking at around 21" max distance between emitters. You can place them closer together depending on how many watts per foot you're shooting for. I'm not necessarily saying the 3590s would be a bad choice.
 
SNEAKYp

SNEAKYp

Well-Known Member
Yea, but the only reason to have that amount of power would be for better penetration right? Using 150w led's as opposed to 60w led's 21" apart would just mean I could have more light reach lower portions of the grow and not extend my growing area. Am I making any sense?
 
Rahz

Rahz

Well-Known Member
Sorry for the short reply. I had Christmas stuff going on. I'm in the final planning phase of a Vero 18 panel but started out with the CXA family in mind. I first took the desired wattage and divided it by the watts each emitter would use. I then took the resulting numbers rounded up or down and plotted out a node pattern in my grow area for 3 of the CXA variants. I didn't plot the 3590s so I can't say what the node distance would have been but I did plot a 3050 running at 66 watts. Node distance was about 13".

I ended up going middle of the road choosing 2530s, but then switched to the Vero 18 using the same grid. The node distance of that grid was 10" which in theory means better light mixing and light angle coverage (at a height of 12").

Based on that I'm guessing you could get a nice spread with 100 watt emitters at 16-18" or so. Define the base distance of a 60 degree isosceles triangle at various heights to get an approximate distance for nodes.

I picked 12" just cause it seemed like a good idea, and I'm not sure how to quantify intensity at various distances, and am working on the assumption that closer is better. Playing with a par meter and a few test subjects would be ideal.
 
SNEAKYp

SNEAKYp

Well-Known Member
Rahz said:
Sorry for the short reply. I had Christmas stuff going on. I'm in the final planning phase of a Vero 18 panel but started out with the CXA family in mind. I first took the desired wattage and divided it by the watts each emitter would use. I then took the resulting numbers rounded up or down and plotted out a node pattern in my grow area for 3 of the CXA variants. I didn't plot the 3590s so I can't say what the node distance would have been but I did plot a 3050 running at 66 watts. Node distance was about 13".

I ended up going middle of the road choosing 2530s, but then switched to the Vero 18 using the same grid. The node distance of that grid was 10" which in theory means better light mixing and light angle coverage (at a height of 12").

Based on that I'm guessing you could get a nice spread with 100 watt emitters at 16-18" or so. Define the base distance of a 60 degree isosceles triangle at various heights to get an approximate distance for nodes.

I picked 12" just cause it seemed like a good idea, and I'm not sure how to quantify intensity at various distances, and am working on the assumption that closer is better. Playing with a par meter and a few test subjects would be ideal.
Click to expand...
It's all good I've been running around buying last minute gifts too! so a couple 100w's would work for a 32"-36" long area (if width <30") I know the Area 51 newest light covers a min of 24" x24" and that's 190w spread over 72 LED's. I actually just bought a 2530 a couple days ago, why the switch to Vero 18?
 
Rahz

Rahz

Well-Known Member
Rather than running the 2530 at 65% I can run the Vero 18 at just over 50% and get the same output, plus they're substantially cheaper and mounting holes are built in saving an additional expense. As a bonus the drivers are a couple bucks cheaper than the ones I had picked for the CXA.
 
SNEAKYp

SNEAKYp

Well-Known Member
I should research more into what I'm buying more haha. Sounds like you did yours. Thanks for all the advice, and insight man.
 
D

darkbug

Active Member
Rahz said:
Rather than running the 2530 at 65% I can run the Vero 18 at just over 50% and get the same output, plus they're substantially cheaper and mounting holes are built in saving an additional expense. As a bonus the drivers are a couple bucks cheaper than the ones I had picked for the CXA.
Click to expand...
hi which drivers did you pick out?
Im looking into building a led array for a 2x4, or 2x3 after seeing what bbspills did
Thanks.
 
SupraSPL

SupraSPL

Well-Known Member
Thanks OP for pointing this new LED out. This might help choosing how hard to drive it and how it compares against the Vero29. If you match the dissipation wattage instead of the drive current, you get a good picture how much more efficient the CXA3590 is.

Cree CXA 3590
assuming: 3000K top bin (BD), Tj ~50c
.35A 24W 47%
.7A 51W 40.8%
1.05A 80W 35.9%
1.4A 112W 32.3%
1.8A 150.3W 27.6%

Vero29 3000K Tj ~50c
.7A 24W 37.6%
1.4A 51W 34%
2.1A 80W 31.2%
2.8A 110W 27.8%
3.15A 126W 26.4%
 
SNEAKYp

SNEAKYp

Well-Known Member
Supra- The percentages are how much of the power is being lost to heat dissipation at various wattage? Am I interpreting that data right?
 
SupraSPL

SupraSPL

Well-Known Member
Yes the percentages shown are photons emitted and the remainder is heat that has to dissipate through the thermal path. So at 350mA the CXA3590 dissipates a total of 24 watts, 11.3 watts of photons and 12.7 watts of heat.

So if the top bin was available for $63, running them that soft would cost about $2.60/watt and 47% efficient versus the best XML2 3000K T5 which @700mA cost $3.25/watt and is 40.8% efficient. In other words the CXA3590 is by far the most efficient warm white so far, 15% more efficient than the XML2.
 
SNEAKYp

SNEAKYp

Well-Known Member
I'm already envisioning great things for this beast. Maybe I'll hold off until they give us DIYers a chance.
 
SupraSPL

SupraSPL

Well-Known Member
Still waiting for these to become available but if they don't for whatever reason, the CXA 3070 is not far off the mark in terms of efficiency and price.

CXA3070 3000K bin Z2 is $40 at Digikey. At .7A Tj 50C it will dissipate 25W and efficiency would be 42.5%.
 
M

Michael9999

New Member
Hello Everyone, I'm new here, and looking forward to learning more about driving LEDs for my grow room. I just received a batch of CXA3590, 5000K, 70 CRI, Bin CD from Newark. I mounted one of the arrays on a Noctua NH-U14S and connected a Meanwell HLG-120H-C running at 1400mA. Welding goggles are not enough. Here are some pics of the chip running at 28C after a couple of hours:
 

Attachments

You must log in or register to reply here.
Top