Tonight I took some pics of them while watering and around day 20-24 from seed. Might top the taller of the Starkillers and TPOGs but not going to top the OX or Sour Bubbles since they shouldnt stretch much.
-
Happy Birthday ICMag! Been 20 years since Gypsy Nirvana created the forum! We are celebrating with a 4/20 Giveaway and by launching a new Patreon tier called "420club". You can read more here.
-
Important notice: ICMag's T.O.U. has been updated. Please review it here. For your convenience, it is also available in the main forum menu, under 'Quick Links"!
Bueno Grows w/ DIY COB LEDs
- Thread starter Bueno Time
- Start date
Dion
Active member
hi, sorry im late. lol I was just researching which colour to use when i found this, VERY NAICE!!! I LIKE!
the last led i built i used surexi chips by illumitex but ive been very interested in trying the cobs from cree, the CRI looks fantastic.
may I ask why you are moving away from the cree cobs now?
thumbs up, big love
the last led i built i used surexi chips by illumitex but ive been very interested in trying the cobs from cree, the CRI looks fantastic.
may I ask why you are moving away from the cree cobs now?
thumbs up, big love
I have noticed a few issues with my CREE setup and only 3 CXA3070 COBs in a 2x3 tent the light is extremely bright and intense in the middle under the cobs to the point where it is hard to keep the plants form yellowing out in the center and the sides of the tent get much less light and lots of shadows due to the way the light is emitted in a cone shape. I am countering this by spreading the intensity and power out more evenly over the canopy using many more (>11 times more COBs, 35 vs 3) to eliminate most shadowing and the issue with the extreme light intensity in middle and lack of light intensity on the sides. Better spread, more even light output and I designed it so that it will be roughly the same efficiency at max power on the dimmable driver and actually more efficient than my AB bin CXA3070s when dimmed down since they are running at such a low drive current 200-215mA max in my setup that I am working on.
Also the CXA3070s need a decent gap above the canopy due to the intensity "hotspots" directly under them and to allow the light "cone" to spread out and cover the canopy and I run out of height and plants grow up too close the the COBs causing the yellowing/burning issues and lack of lighting off to the sides since the emitted light doesnt get to spread out to cover the full canopy when its so close to the emitters. The new setup Im working on should eliminate these issues almost entirely if not all together.
Im going to smaller Veros because they are better efficiency than the smaller CREEs and the price is good on them.
I also picked a spectrum with more blue in it this time too so I can veg better with the Veros and then switch on 10 660nm and 4 630nm reds for red boost in flowering. I eliminated my 4 450nm blues from the red/blue string Ive been running with my 3 3070s setup, since there is plenty of blue in the Veros I went with.
Vero 29 wont be as efficient at 1400mA but if driven harder will maintain a higher efficiency than CXA3070s, the Veros shine more at higher currents whereas the CXA3070s shine (do well/best) at 50% or less of max current (1400mA or less) for efficiency purposes. That being said Vero 29s are very nice and the price is very good as well. 2700K Vero is very red/amber/orange if thats what you are looking for. I would recommend Vero 3500K as a replacment for CXA 3000K. Some people even say Vero 4000K is about equivalent to CXA 3000K but I disagree since the Vero 4000K has more blue in it but the LER is the same between the two spectrums.
I went with Vero 4000K 80CRI for my new build, I wanted the extra blue vs my CXA 3000K 80CRI COB setup to reduce stretch a bit hopefully and also since I finish veg for 2-3 weeks before flowering in my flower tent due to a tiny veg area.
Dion
Active member
thank you so much for taking the time to answer
i have been looking at the cxa 2540 for exactly the reason of light spread, ie 1 fixture every 30cm or so, was not plannning on using a lens at all (115 degree angle off the cob) those babies run up to 2.1 mA at 37v was also going to put a dimmer on it for cloning/seedlings and also to control the internodes spacing
another huge thing is the heat sinks, i tend to prefer passive heatsinks over that fan shiznit...i guess i like to make all my fixtures ip65 just incase(when) i drop one in the tank lol also its cheaper to get those star looking heatsinks where i am.
agian thank you for your reply, i just wanted to make sure it wasnt anything to do with the light that you were switching, if i understood correctly its more a matter of economics (seen as i am looking at the smaller ones and more of them the light spread issue shoudl be moot )
how about the CRI? looking at the cree pdf the 3000k 93 CRI seems to hit the best PAR...although i cant seem to find any spectral graphs for the 5000k/6500k 93 CRI(although i can buy them)
and the binning
anyway sorry if this is taking from your glog if you have any tips would appreciate them, its my first time working with cree chips(still cant find those 303lumens per watt chips i heard about)
keep up the good work, im still reading through it from the start
thank you so much for taking the time to answer
i have been looking at the cxa 2540 for exactly the reason of light spread, ie 1 fixture every 30cm or so, was not plannning on using a lens at all (115 degree angle off the cob) those babies run up to 2.1 mA at 37v was also going to put a dimmer on it for cloning/seedlings and also to control the internodes spacing
another huge thing is the heat sinks, i tend to prefer passive heatsinks over that fan shiznit...i guess i like to make all my fixtures ip65 just incase(when) i drop one in the tank lol also its cheaper to get those star looking heatsinks where i am.
agian thank you for your reply, i just wanted to make sure it wasnt anything to do with the light that you were switching, if i understood correctly its more a matter of economics (seen as i am looking at the smaller ones and more of them the light spread issue shoudl be moot )
how about the CRI? looking at the cree pdf the 3000k 93 CRI seems to hit the best PAR...although i cant seem to find any spectral graphs for the 5000k/6500k 93 CRI(although i can buy them)
and the binning
anyway sorry if this is taking from your glog if you have any tips would appreciate them, its my first time working with cree chips(still cant find those 303lumens per watt chips i heard about)
keep up the good work, im still reading through it from the start
No problem man always willing to try to help or at least steer someone in the right direction.
Im not sure on the efficiency numbers of the CXA2540s, actually only really knowledgable on the CXA3070s as far as CREE cobs go. I had my 3 CXA3070 spaced roughly 12" (~30cm) apart down the center of the tent and the coverage between them was plenty good just not so much on the far sides, front and back walls, away from the center line of the tent. In hindsight 6 CXA3070s would have been optimal in my 2'x3' footprint tent 1 COB per sq ft. On my new fixture the Vero 10s are spaced ~4" (~10cm) apart front to back and ~5" (~12.5cm) apart side to side, coverage should be excellent I think.
The binning, CREE bins their COBs but Vero doesnt so Vero doesnt separate the COBs based on quality and output/specs of the particular batch that is made. CREE bins them and the higher bins have higher efficiency/output basically. Getting the highest bin you can find for CREEs is important if you are very concerned with efficiency (which I think you should be when building or even buying a LED fixture from a company, otherwise HID is much cheaper per watt, however you can build a more efficient LED setup if properly designed with quality and efficient components, also being able to tailor your spectrum to your liking with LED is an obvious win vs HID).
Passive heatsinking generally requires about 3x the amount of aluminum mass vs active (fan) so it costs about 3x more but is no doubt more reliable, efficient (no power used on fans), potentially safer like you said (water).
As far as CRI, I am pretty much of the agreement with some others that the 93CRI "wastes" a significant portion of the power on far reds but there are about as many ways of thinking as there are growers. If you want to try it dont let me stop you but I like the 80CRI for warm whites, its what I have been using and works well, most others are using it too that I have seen.
edit:
Here is a chart I had saved that shows CREE 3000K 80CRI vs 93CRI, the 93CRI is missing a lot in the yellow/orange/red and has some "wasted" power in the beyond 700nm range. The 93CRI does have a small but nice boost in deep red range though over the 80CRI.
Also dimmable drivers are awesome and I highly recommend them though they are generally more expensive they are worth the extra cost IMO for adjusting the amount of light and heat in the grow area making it easier to keep your temps in check during different seasons.
I was looking back at some spreadsheets I save SupraSPL posted up. Comparing the CXA3070 AB bin (top bin 3000K, for now), Z4 bin (a step down from AB bin in 3000K) and Vero 29 "typical" figures from Vero and CREE CXA spreadsheet data. The Vero 29 stacks up well, especially considering Z2 is the most commonly available bin for CXA3070 3000K and its harder to source Z4 and AB bin though it can be done with some footwork.
Heres the spreadsheets to compare and you can then also see how the drive current you choose determines total wattage of each COB and effects efficiency, etc.
Also when considering price of Vero 29s they just went down to about $28 each from digikey (great company BTW, very fast shipping) vs $40+ for CXA3070s depending on bin, I have seen people selling AB bins for $60 each, yikes. (I paid $50 each for my AB bins when I found them). Veros start to look pretty good...
Dion
Active member
i think the last leds i built are like the ones you made later(im only on page 7) 2 bars with no fans, i got my drivers from china(i have the equipment to test them so wasnt worried about frying stuff) with a pentameter (dimmer thing) its only an extra 2 bucks from my guy, or i can have him adjust the mA to exactly what i want(last time i needed 580mA, he does that for free.. so i will end up paying about 10 bucks for dimmable drivers for these chips.
btw the specs for the 2540s are the same as the cobs you have bro, just smaller ones, i find it super interesting people are using the 80 CRIs as they apear to peak at about 605 in the red at about 65%@662 and are below 50% in the blue at aprox 455 vs the 93CRI peak at 630 (75% at 662) and 75% ish in the blue- not debating you just wondering if im reading this wrong or there is another cree doc i need to look at...
great thread, im new so dont know how to like your posts yet but as soon as i figure that out lol
btw the specs for the 2540s are the same as the cobs you have bro, just smaller ones, i find it super interesting people are using the 80 CRIs as they apear to peak at about 605 in the red at about 65%@662 and are below 50% in the blue at aprox 455 vs the 93CRI peak at 630 (75% at 662) and 75% ish in the blue- not debating you just wondering if im reading this wrong or there is another cree doc i need to look at...
great thread, im new so dont know how to like your posts yet but as soon as i figure that out lol
Yeah those charts look a lot different but I think it is because of the way the charts were made. The one I posted is umols/s/nm vs wavelength in nm and the CREE chart is relative output in % vs wavelength in nm. umols/s/nm is supposed to be an excellent if not the best way to measure output of a light source since its the quantity of actual photons emitted or something of that nature as far as my knowledge goes. I have read a little on it but havent gotten quite that deep into it as some others have. I am not positive but I think the chart I posted may have been from MrFlux on RIU from his calculations when analyzing different spectrums. I know I got that chart on RIU somewhere in the LED subforum but could never find it now theres way too much stuff over there to go through and find it again haha. Tons of LED knowledge over there I recommend everyone to check out that is interested in LEDs and DIY particularly, excellent people over there sharing knowledge and ideas.
As far as looking at the CREE chart the 93CRI would be an easy winner I agree with what you are seeing there.
And what I meant about the 2540 specs was I really know nothing about the typical vF they require to run, CREE test current, output and efficiency at test current, and efficiency/output at the various drive currents within its operating specs (which I also dont know haha). Basically I know nothing about the specifics of that model. Ive never looked up spreadsheets or anything about any other series than CXA3070 and Vero 10.
As far as looking at the CREE chart the 93CRI would be an easy winner I agree with what you are seeing there.
And what I meant about the 2540 specs was I really know nothing about the typical vF they require to run, CREE test current, output and efficiency at test current, and efficiency/output at the various drive currents within its operating specs (which I also dont know haha). Basically I know nothing about the specifics of that model. Ive never looked up spreadsheets or anything about any other series than CXA3070 and Vero 10.
Dion
Active member
yup, tnx bro agian, just for your information they are the little brother of the ones you have(incase you decide to return to the cree family same specs same evrything just smaller (37v) lol)
I'm really looking forward to this build now, I also am looking forward to having multiple lights that I can move around rather than a fixed bar. im slow reader, on page 9 now you do some good work man, those colas are massive =D
Dion
Active member
am i wrong in thinking that 662 and 453 are the most yummie for the plants? peaks for chlorophyl a and be respectively? in that case the 93CRI still wins...although the 500ok kills it for chlor b
Drive current is up to you, what I was running the CXA3070s at, 1400ma, isnt a bad drive current for Vero 29 should be around 40% efficient typical for 3000K 80CRI version. Veros dont get quite as large of an efficiency boost by running at lower currents like the CXAs do but you can still run more quantity of COBs at lower currents for a higher efficiency, its up to you (and any DIYer) to balance how much you want to spend and how efficient you want your setup to be when you decide drive currents. You can look up a few posts at the spreadsheets posted for Vero 29 3000K efficiency at various drive currents and decide what you think seems reasonable.
The Alpine 11s that I was using for the CXA3070s would cool a Vero 29 easily at 1400mA with the fan at 5v (same with the 3070s). They would cool at a higher amperage (= higher wattage) with the fan run faster than 5v up to a point. The bevel on the base of the heatsink is odd though and the Vero 29 plastic mounting ring would hang slightly off two sides of the mounting face though but when oriented correctly the actual thermal surface of back of the COB will fully fit on the heatsink face, just the plastic will hang off the sides from what I have seen others post pictures of.
Thanks Bueno, I seem to have a little more vertical height than you. with that being the case, i think the higher power COBs, like your 3070 or a Vero 29 will work good for me. I have a 3'x4' flower room with {6} 10 gallon NOtill geopots. Flowering 10 clones or seedlings per bag. I am thinking of {6} Vero 29 4000k driven at 1750ma. What are your thoughts? I currently use a 600 hps cool tube.
wow dude they r looking great with those white led lighting.
I have a 2'x3' footprint, 5' max height in flower floor to ceiling.
Which spectrum (K) are you going for with the Vero 29s? At 2100mA each they should be dissipating approx 79w each (474w total for 6) at ~37-38% efficiency depending on which you choose 3000K, 3500k, 4000K listed in order of least to most efficient. Not bad but also not record breaking efficiency, pretty comparable to HPS as far as efficiency, I would think you would have similar results as 600HPS possibly better but hard to say, if you designed it more efficient overall as a system then obviously it will be even better.
Just for example my CXA3070 3000K AB bins are 42% efficient at 1400mA and my new Vero setup will be ~42% efficient at max current and even a little more efficient when dimmed. Should be sweet, I always encourage people to build the most efficient setup they are willing to pay for (more efficiency = more $$) because then you are getting one of the main advantages of even using LED in the first place IMO (possibility of higher efficiency, better spectrum choices, etc. vs HID).
Also I want to mention the difference between 38% efficiency and 42% efficiency isnt just 4% more light, it is more than that, 42/38 = 1.105 or 110.5% an increase of just over 10% more light for an increase of 4% efficiency! That is 10% more light and should theoretically mean 10% more bud for the same wattage/power draw used when all else is equal.
Most guys run their Vero 29s and CXA3070s on the Alpine 11s at 1400mA (~52w) with the fan at 5v it cools them perfectly fine, I am not sure what the coolers can cool properly with the fan at the full 12v speed I would have to ask someone like SupraSPL on RIU about that he does lots of testing of current droops and light looses/gains due to chip temperatures from slower and fast fan speeds, etc he would probably be able to tell us if the Alpine 11s can effectively cool Vero 29s at 2100mA. (Im going to ask him and see.)
I have and am going to be using a Meanwell HLG-185H-C1400A to run all 35 Vero 10s 5 per string in series (vF adds together), 7 strings of 5 in parallel (current divides by 7). So approx 25.5 vF per Vero 10 x 5 = 127.5 vF total and 1400mA (have heard will put out ~1500mA when vF isnt maxed to give as close to 200w of output.) divided by 7 strings = 200mA per Vero 10 (215mA if at 1500mA total). Im just waiting on a cheapo driver I ordered to run single Vero 10s so I can run them individually and test vF of each to match strings of 5 as close as possible to same total vF so one string isnt hogging more current than another and everything should balance more evenly.
Thanks cheese good to see your still around.
http://www.jameco.com/webapp/wcs/st...kdYPExcbxwTv8MOvJIoyuT68Px9842DJywaAv3H8P8HAQ
BT, your the man! Thanks for the reply. The link above is for a 1750ma diver, would this be as efficient as the 1450 driver. I plan on using the Vero 29 400k. Will 6 of theseedling be good for a 3x4 footprint? What is the closest you can get your 3070 to you tops without damaging them?
BT, your the man! Thanks for the reply. The link above is for a 1750ma diver, would this be as efficient as the 1450 driver. I plan on using the Vero 29 400k. Will 6 of theseedling be good for a 3x4 footprint? What is the closest you can get your 3070 to you tops without damaging them?
