DiY LED - Cree CXA3070

[SupraSPL]

I plan on using some of the luxeon M 12 watt for supplemental blues all the way thru flower (during flower just turn on more 3K cobs)
check out their 30mm copperpad mounting
http://shop.stevesleds.com/Philips-Luxeon-M-ROYAL-BLUE-12-Watt-LEDs-Luxeon-M-RB.htm

Steves once offerred the Luxeon M E5G bin, which was the the most efficient LED I have ever seen for sale (63% at 700mA). The only reason I dont use or recommend it is because there is really no application where we would need that much blue concentrated in such a small area. But maybe your experiment will prove otherwise and they will ooze out hash resin

It looks like they updated their stock with a lower bin, the C5G, 57.7% efficient 700mA. So that makes the Luxeon ES N4R currently the best option after all, 60.1% at 700mA.

 

[SupraSPL]

Where would be a good place to start a project of consensus? What format would be appropriate? Spreadsheets?

I might be thinking type far ahead but I would like to include 'popular' COBs, grow area, light patterns, passive versus Active cooling, veg and flower options, auxiliary equipment like power supplies. Trying to do a general solution but that maybe too much, at least in the start. Plant height, SOY yes or no, CO2. Just stream of conscience at this point.

Lots of great ideas there! I have been pondering that too. One idea was to start a thread showing a few simple example builds, Alpine 11 based CXA and Alpine 11 based Vero29, heatsinkUSA style active cooled setup and a passive setup. Efficiency based lamp vs cost based lamp. A passive cooled vegging lamp, maybe based on the Vero 5K. And a combo lamp like mentioned above vegging and flowering, warm white with a blue string. Maybe that would help people figure out which direction to go for starters. I know some are using 3500Ks and 4000Ks, some are using deep reds. So many options I guess the most important thing we need people to know is that LED works, especially when you DIY.

 

[PurpleBuz]

Steves once offerred the Luxeon M E5G bin, which was the the most efficient LED I have ever seen for sale (63% at 700mA). The only reason I dont use or recommend it is because there is really no application where we would need that much blue concentrated in such a small area. But maybe your experiment will prove otherwise and they will ooze out hash resin

It looks like they updated their stock with a lower bin, the C5G, 57.7% efficient 700mA. So that makes the Luxeon ES N4R currently the best option after all, 60.1% at 700mA.

Plan is 4 vero 18s 3K in a 1 foot square box with reflectors plus one naked luxeon M royal blue in the center. so its not really that much royal blue.

are the luxeon royal blues more efficient than the crees ?

 

[SupraSPL]

Yes the best deep blue Cree I have seen is the XTE Q04 group which is 50.6% efficient min at 700mA (2.2W).

If you run your V18s at 1.05A, they will dissipate 124W. They are 8% blue so that is 10W of blue, or 3.7 PAR W. If you add just 1 Luxeon M at 1.05A, that adds another 12.2W of blue or 6.46 PAR W. So that would almost triple your total blue.

 

[Kuifje76]

What are the best 630nm leds for the moment ?
and how do you go from W to PAR W ?

 

[SupraSPL]

The best 630s to my knowledge are the Cree XPE P3 bin. You can get them on 20mm stars from AC-RC.net, LEDGroupbuy, Digikey, and Cutter. At 700mA Tj 50C they are 40.8% min. The XPE2 is also a good choice but it is more expensive and has different characteristics (more sensitive to temp droop, less sensitive to current droop). At 700mA Tj 50C I estimate 38.5%. These reds are very sensitive to junction temp so if you can run them cool you will get a nice boost.

To go from dissipation W to PAR W you multiply by the efficiency. Of course the efficiency is just an estimation, we need to know junction temp, drive current, Vf, bin and output varies from LED to LED on top of all that (min vs typical etc). Then once you have PAR W (or PPF), you have to factor in reflector losses and distance to the canopy to get PPFD (actual photons in the canopy). And finally not all canopies are perfectly flat so that is quite a variable as well.

 

[indianajones]

construction is coming along pretty well, it got beaten up pretty
badly in shipping and a lot of screws are bent. been a PITA getting
the ends off so i can slide the driver in. ends on ezmates are
soldered then covered with liquid black tape. this is where i left
off yesterday.

 

[Blahman501]

I assume you are adding the 5K because you want to be able to veg with it? (Sorry I am clueless about hydro, I think in terms of plants that can be moved from room to room.) If so, maybe you could add some Luxeon ES blues on a string and just click them off after the first 2 weeks of flowering. That way you won't have worry about color mixing and you can use the CXA3070 3000K AB bin X 5 (unfortunately the CXA3070 is hard to get in high bin 5000K). Also the Luxeon ES N4R are impressively efficient (60% min @ 700mA). I tested them against the M4R bin and the vendor is being truthful about the bin increase.

http://shop.stevesleds.com/Philips-Luxeon-ES-ROYAL-BLUE-3-Watt-LEDs-Luxeon-ES-Royal-Blue.htm

Thanks for the fast reply! I was operating under the assumption that I would flower/veg in the same location. I guess I was under the impression that most people growing with COBs were doing the same, but is that incorrect? Should I reconsider this? I have another space that is about 6x3 that I could utilize some or all of if necessary. I'm growing purely for myself and basically want to end up with a perpetual stream of 2-3oz a month. If I had separate veg and flower locations, I could stagger my crops to have 2 flowering and 2 vegging at any given time, and that might make it easier to keep up a "perpetual" flow.

If I went that route, would 3-3000Ks for flowering and 3-5000ks for vegging be sufficient? (All 3070s) Also, I was under the impression the AB bins were impossible to find. Where is the best place right now to find them in small quantities?

 

[Blahman501]

my guess, roughly :
6x@1050ma, you get 5.610lm/sqf for 6sqft (total 33.658lm)
5x@1050ma, you get 4.675lm/sqf for 6sqft (total 28.048lm)
calculated all 3000k AB bin's though, it will be a bit higher with the 5k

but just looking @lumens/sqf, my bet would be to go for 6
drivers i don't have a clue, i'm always looking at meanwells for their efficienty, but 6 is a difficult number...

would be great to have a tool for people to just fill in a few numbers and combine everything (also driver eff., heatsink °C/W etc), maybe Supra can 'tune' my upload

Thanks a bunch for this! The spreadsheet idea is a great one, and I downloaded and am looking at it now. Would be curious to hear your thoughts on my reply to Supra, considering separate veg and flower locations for 2 plants each.

 

[SupraSPL]

I have always been of the mindset that I would be keeping clones of the best stuff I came across, so I needed a separate veg space right from the start and got used the that as "the way". So I have to remind myself to think from other peoples POV that run autos or don't keep clones. But having a separate veg space does allow you to get harvests every 2 months which is nice. For those who are impatient or testing a lot of varieties, if you set up a pair of flower rooms and flip flop the light schedule you get a harvest every month. That is great but may be hard to keep up on over a long period of time. Some growers will stagger the age of the plants in their flower room to get a harvest every month which is great idea, but the flip flop effectively cuts your heat in half.

You can veg a lot of cannabis with a small amount of light. I use 110W for veg and 1300W for flower. So if you have a separate flower room, the vast majority of the wattage would be in there. You could use 5000K Vero 10s for veg and 3000K CXA3070s or Vero 29s for the flower area. The ABs have been popping up here and there, Arrow and Ali, I can help you track some down.

 

[SupraSPL]

Build look good IJ, what screws got bent?

 

[Kuifje76]

Thanks a bunch for this! The spreadsheet idea is a great one, and I downloaded and am looking at it now. Would be curious to hear your thoughts on my reply to Supra, considering separate veg and flower locations for 2 plants each.

Glad i can help, most of the numbers are from Supra, some are 'guestimated' ex 600mA, so not everything is 100% correct, but it's good to get an idea.
Maybe we could make a sticky with files for calculation etc, it's to difficult to find them in a 114page thread (and counting )

2 seperate chambers are a must if you want a perpetual grow, i'd go for 1x5000K grow and 4x3000K flower (Epicfail uses 1cob/sqf mostly @700mA and he only goes full power week4-7 (1050mA)
i thought he used the HLG-120H-1050B (B is dimmable 0-100%, A 50-100% with internal pot (screw)) but i'm a bit confused since max voltage is 148V and i thought you needed 38,5Vx4 =154V, maybe he or someone else can comment on that.

i'm gonna go for 5cobs/4sqf with one HLG-185H-1050B, all voltage is used up (190V),( also a tip from Supra for max eff) and 94% efficienty , someone pointed out not to go under 20% with the B model, else efficienty takes a dive.

ps: heatsink °C/W in the sheet i get from BurminghamAluminium for 50cmx50cm 9730HS, don't have prices yet, could be too expensive afterall) but there you can get an idea of different heatsinks with different cooling capacity and then change some numbers in the sheet

 

[Kuifje76]

yo could always use a fake wall/board to lower your total footage, just an idea

 

[SupraSPL]

i thought he used the HLG-120H-1050B (B is dimmable 0-100%, A 50-100% with internal pot (screw)) but i'm a bit confused since max voltage is 148V and i thought you needed 38,5Vx4 =154V, maybe he or someone else can comment on that.

i'm gonna go for 5cobs/4sqf with one HLG-185H-1050B, all voltage is used up (190V),( also a tip from Supra for max eff) and 94% efficienty , someone pointed out not to go under 20% with the B model, else efficienty takes a dive.

Looks like it should be about 36.25X4=145V, just right. Same with your 185H, very close match Another tip from @Greengenes707 we might be able to get another point of efficiency by using 240V if you have access to it.

 

[Kuifje76]

I do have 240V, how's that an advantage ?

 

[SupraSPL]

I think what it comes down to, the higher voltage slightly reduces voltage drop in the wires, connections, joints and circuitry. Voltage drop is another way of saying that some of the current has turned into heat along the way of getting "used". The higher the voltage (pressure), the less the resistance of the circuit translates into heat because less current is flowing through it.

The same thing goes for wires. If we use too small of a wire for a given voltage/current combo, the wire heats up due to resistance and we get a voltage drop from one end of the wire to the other (waste heat). But if we increase the gauge of the wire, the heat and voltage drop go away. Another way to achieve the same effect is to use a higher voltage and lower current in the small wire.

The downside is, more expensive GFCI, plugs, timers, outlets, kill a watt etc. Less compatibiltiy with existing wiring. More dangerous if you managed to get shocked.

The upside, less potential for wires to heat up, fire safety. Smaller wiring can be used. Slight uptick in efficiency.

 

[alesh]

I think what it comes down to, the higher voltage slightly reduces voltage drop in the wires, connections, joints and circuitry. Voltage drop is another way of saying that some of the current has turned into heat along the way of getting "used". The higher the voltage (pressure), the less the resistance of the circuit translates into heat because less current is flowing through it.

The same thing goes for wires. If we use too small of a wire for a given voltage/current combo, the wire heats up due to resistance and we get a voltage drop from one end of the wire to the other (waste heat). But if we increase the gauge of the wire, the heat and voltage drop go away. Another way to achieve the same effect is to use a higher voltage and lower current in the small wire.

The downside is, more expensive GFCI, plugs, timers, outlets, kill a watt etc. Less compatibiltiy with existing wiring. More dangerous if you managed to get shocked.

The upside, less potential for wires to heat up, fire safety. Smaller wiring can be used. Slight uptick in efficiency.

There's also a fact that above 50% of load, HLG-XX-C drivers are simply more efficient when running off 240V.

 

[ballist]

Less dribble. P=I^2R

 

[indianajones]

Build look good IJ, what screws got bent?

the ones on the end that keep the driver area enclosed. i circled
which ones are bent in the picture. i was able to get all of them
but 3 out on the one that was dropped badly, of course it is on the
side with the power inlet so i have to do ninja work to wire the
driver. on the plus side- i learned to solder. it's pretty funny to watch
me do it because of my shakes, but i'm finding ways around it

 

[happy75]

A testament to the efficiency of the 3K AB, the 5K Z4 makes less lumens than the 3K AB because we don't have access to the top bin 5K.

I have access to top bin 5000K. I have here right now AD's and BB's in 5000K. AD's are 70 cri, the BB's 80 cri. Wanted to veg with them, but first want to finish my other lights. If I can make somebody happy with these....