-
ICMag with help from Landrace Warden and The Vault is running a NEW contest in November! You can check it here. Prizes are seeds & forum premium access. Come join in!
Bueno Grows w/ DIY COB LEDs
- Thread starter Bueno Time
- Start date
Day 27 12/12 Update, pollination was successful, there are withering orange hairs on the selected buds. Other than that nothing new really.
Sugar Jones clones on right side, OXs front and back center and back left, PK Bubba in front left.
Sugar Jones clones on right side, OXs front and back center and back left, PK Bubba in front left.
wow u r gunna get a good harvest. nice job those light look like thety r working great for ya.
Soulrebel239
Member
That sog looks real good.
Day 35 12/12 Update
I took one of the SJ clones out of the tent and the damn thing was so heavy it fell over shortly after taking the pictures. The one I took out of the tent is just getting Floranova Bloom at ~1.7ec now, the one I didnt take out of the tent is also getting the Floralicious Plus along with the FNB.
SJ in the tent
SJ out of tent
The 3 OX and 1 PB, buds arent huge but they smell and look very good but the fan leaves look like hell. I can tell from the smells they are going to be some good stoney kush smoke, cant wait to sample and select the pheno(s) I want to run again in a proper soil mix. All 3 OX have a very similar smell with only a minor variation between the 3 so selection will come down to the smoke quality of the 3 however OX1 does seem to be a slightly larger yield-er to the eye so far.
I took one of the SJ clones out of the tent and the damn thing was so heavy it fell over shortly after taking the pictures. The one I took out of the tent is just getting Floranova Bloom at ~1.7ec now, the one I didnt take out of the tent is also getting the Floralicious Plus along with the FNB.
SJ in the tent
SJ out of tent
The 3 OX and 1 PB, buds arent huge but they smell and look very good but the fan leaves look like hell. I can tell from the smells they are going to be some good stoney kush smoke, cant wait to sample and select the pheno(s) I want to run again in a proper soil mix. All 3 OX have a very similar smell with only a minor variation between the 3 so selection will come down to the smoke quality of the 3 however OX1 does seem to be a slightly larger yield-er to the eye so far.
Closeups
OX1
OX2
OX3
PB3
SJ
OX1
OX2
OX3
PB3
SJ
Get the highest bin you can find and I would use the 80 CRI (thats what I chose to use). The 90 CRI was more for photography or when you would want to show the true color of the object being lit up by the light from the COB. There are a couple people who do use the 90 CRI though for growing. Either one will work, its up to you what you want to try out in your DIY fixture.
I like the spectrum curve of the 80 CRI personally and then when I added to it some 660nm, a couple 630nm and 450nm the spectrum is even better in my opinion.
Check out the chart below and see which you feel is better (there is no right or wrong answer, just whatever you decide to go with).
From ~500-650nm the 80 CRI wins by a long shot and from ~650-700nm the 90 CRI wins by a small amount but then after 700nm the 90 CRI wastes some energy on far red production.
So me using 80 CRI and a couple 630nm and 450nm and some 660nm to bump up the deep red coverage, I hit all the wavelengths of the 90 CRI with a large boost in the midrange ~50-650nm with the 80 CRI vs 90 CRI.
Thats not to say this is the only way or the best way or the right way, just the way I did it. If you want to try 90 CRI go for it and let us know how it works out, most DIY LED guys are still experimenting with spectrums or spectrum combinations.
cocogrower
Member
Awesome grow this time to Bueno! Your Buds already looks fat and frosty, cant wait to see how it turn out in the end! 
NOCTUS, what K's are you thinking about? I think you should look closely at the lumen output of the different bins and then see what kind of spectrum you get if you tweak it a bit! When you watch this diagrams, keep in mind that the curvs are not absolute, they are just showing the spectral distribution.. So the 93 cri lamps you cab score actually is "worse" than they look at the diagram since they have a lower lumen output. This means that a high binned 5000 K probably gives you more photons in 630 nm than a low binned 3000K...
You should take a look at my thread if you haven't, Its about building the lights you are talking about I finally updated it with a partial result..
Bueno, Cant you try to use the BML calculator( http://www.bmlcustom.com/custom-led-strip/ ) to estimate your spectrum and get a graph of it? You have yo estimate the percentage of watts in each spectrum, and compensate for differences in efficiency between your leds and the crappier leds they are using .. It would be really interesting to see what your secret formula looks like!
NOCTUS, what K's are you thinking about? I think you should look closely at the lumen output of the different bins and then see what kind of spectrum you get if you tweak it a bit! When you watch this diagrams, keep in mind that the curvs are not absolute, they are just showing the spectral distribution.. So the 93 cri lamps you cab score actually is "worse" than they look at the diagram since they have a lower lumen output. This means that a high binned 5000 K probably gives you more photons in 630 nm than a low binned 3000K...
You should take a look at my thread if you haven't, Its about building the lights you are talking about
I finally updated it with a partial result..Bueno, Cant you try to use the BML calculator( http://www.bmlcustom.com/custom-led-strip/ ) to estimate your spectrum and get a graph of it? You have yo estimate the percentage of watts in each spectrum, and compensate for differences in efficiency between your leds and the crappier leds they are using ..
It would be really interesting to see what your secret formula looks like!
cocogrower
Member
I went with the CXA2530 U2 4000K 70 cri:
Combined with 2 best binned 630's and 660's ( Xp-e p3 and.. forgot the exact names but Bueno uses them to i believe), the spectrum of my more flower oriented lights will look like this:
If this is good or very good, i don't know.. But hey, at least i did it on purpose, after reading a ton of threads about it. And i'm really enjoying seeing it shine, and eventually grow, hopefully something slightly similar to the Hierba-Bueno! Haha
Combined with 2 best binned 630's and 660's ( Xp-e p3 and.. forgot the exact names but Bueno uses them to i believe), the spectrum of my more flower oriented lights will look like this:
If this is good or very good, i don't know.. But hey, at least i did it on purpose, after reading a ton of threads about it. And i'm really enjoying seeing it shine, and eventually grow, hopefully something slightly similar to the Hierba-Bueno!
Haha
Awesome grow this time to Bueno! Your Buds already looks fat and frosty, cant wait to see how it turn out in the end!
NOCTUS, what K's are you thinking about? I think you should look closely at the lumen output of the different bins and then see what kind of spectrum you get if you tweak it a bit! When you watch this diagrams, keep in mind that the curvs are not absolute, they are just showing the spectral distribution.. So the 93 cri lamps you cab score actually is "worse" than they look at the diagram since they have a lower lumen output. This means that a high binned 5000 K probably gives you more photons in 630 nm than a low binned 3000K...
You should take a look at my thread if you haven't, Its about building the lights you are talking about
Bueno, Cant you try to use the BML calculator( http://www.bmlcustom.com/custom-led-strip/ ) to estimate your spectrum and get a graph of it? You have yo estimate the percentage of watts in each spectrum, and compensate for differences in efficiency between your leds and the crappier leds they are using ..
I did try to figure it out a few months back but I couldnt get the right ratios of 3000K WW, 630nm, 660nm, 450nm with the amount of slots available in the calculator and I gave up trying to figure it out after a short while.
I went with the CXA2530 U2 4000K 70 cri:
View Image
Combined with 2 best binned 630's and 660's ( Xp-e p3 and.. forgot the exact names but Bueno uses them to i believe), the spectrum of my more flower oriented lights will look like this:
View Image View Image
If this is good or very good, i don't know.. But hey, at least i did it on purpose, after reading a ton of threads about it. And i'm really enjoying seeing it shine, and eventually grow, hopefully something slightly similar to the Hierba-Bueno!
My 630nm are Cree XP-E P3 and my 660nm are Luxeon ES EX6. When I designed my bars they were going to supplement two COBs and I ended up adding a third COB so the ratio of spectra is different with >50% more 3000K WW.
Im not even sure yet what differences I am seeing from the supplemental LEDs this round. Lots of different combos to try out. I might stick with what I have now for a few runs at least though until I max out production with it then maybe build a new setup or add more COBs (along with a stronger exhaust fan setup to handle more heat).
cocogrower
Member
Hi Bueno!
Sorry for nagging on you about the using the BML calculator, i realize that you've given me the same explanation before.. I've just got this obsession, and can't help myself sorry. LOL (if you are curious you could give me watts and l/w in each spectrum and I could give it a try. But i can totally see that it will be difficult, especially with the small percentage of blue..)
I think your spectral distribution shouldn't be to different from mine, except that you probably have about the double amount of far red.. Trying to understand how important this is, and if 2-3 percent is enough. But will at least try one or two full runs with my new lights before changing anything!
Sorry for nagging on you about the using the BML calculator, i realize that you've given me the same explanation before.. I've just got this obsession, and can't help myself sorry. LOL (if you are curious you could give me watts and l/w in each spectrum and I could give it a try. But i can totally see that it will be difficult, especially with the small percentage of blue..)
I think your spectral distribution shouldn't be to different from mine, except that you probably have about the double amount of far red.. Trying to understand how important this is, and if 2-3 percent is enough. But will at least try one or two full runs with my new lights before changing anything!
No problem, you can give it a go yourself if you want to try.
Ive got 3 CXA3070 3000K AB bin @ ~1450mA each. Close to 160w at around 136lm/w (42% efficiency). The red/blue bars are ran at a little under 650mA and there are 4 XP-E P3 630nm ~5.6w at 42% efficiency, 4 Luxeon ES M4R 450nm ~7.2w at 56% efficiency, and 10 Luxeon ES EX6 660nm ~15.5w at 40% efficiency. I dont know of any lm/w ratings on the reds and blues.
I dont know if the math is flawed since I am a little baked and just started punching numbers into a calculator but I just took the wattage of each x efficiency of each and then added those together and divide each (wattage x efficiency) by total wattage x efficiency of all LEDs. I got 78.944 total, 630nm 2.352, 450nm 4.032, 660nm 6.2, and 3000K 66.36.
So each number divided by 78.944 should give a rough percentage of each. (Rounded numbers in parenthesis)
3000k 66.36/78.944 = 84% (84%)
630nm 2.352/78.944 = 2.9% (3%)
450nm 4.032/78.944 = 5.1% (5%)
660nm 6.2/78.944 = 7.8% (8%)
I know its not perfect but if this logic isnt totally flawed, those would be the rough percentages based on wattage and efficiency of each.
The 3000K is made up of a very broad spectrum of frequencies while the monochrome reds and blue target a very narrow frequency.
cocogrower
Member
Wow thats an insane amount of quality light!
That seems about right, in my stoned and tired eyes.
Now the problem is to put those numbers in to the calculator. Each led (1 of 15) represents 6,66% - an evil number, that makes the calculation kind of difficult..
The next, more delicate, problem is that your leds piss on BML's when it comes to efficiency. Some more than others. I don't really understand what they mean by "wall plug efficiency" One would like to think that it's the same numbers we are talking about, but XP-E P3 ≈ 100 l/w and their 630's gives 55 l/w, yet they claim 37% efficiency.
Your 3000k are 58% more efficient than theirs (136/86 =158,1)
your 630's are almost 100% more efficient but probably closer to 85 since you drive them harder than 350 mA)
Another problem is that your amount of 630 only equals half a dot. I won't make a proper calculation of this but i will tell how my estimation works.
So, this is a rough calculation/estimation, since there are so many limitations, and I kind of sucks at math.. I still think that its gonna give us a pretty good idea.
12 dots of 3000k equals 80,4% (this is less in your setup, but these leds are at least less lousy than their 630's and we need 3 dots to play with.
If we choose one 450 one 660 and one 630, we will get to a little bit to much 630 (not so much since they are so much weaker then yours), a little bit to much blue, and to little 660.
So when looking at the diagrams, we have to "ad some" 660 and think away a little bit of blue and a tiny amount of 630.
We can also exchange one of the 3000k dots to a 660, and instead "think away" a fair amount of 660. This isn't scientifically done, but i do believe that the "real" graphical representation, of real spectral situation in Buenos cabinet, isn't to far away from a mix of these two diagrams.
Puh.
Now i really have to sleep! tell me tomorrow if you don't agree on this.
Good night!
That seems about right, in my stoned and tired eyes.
Now the problem is to put those numbers in to the calculator. Each led (1 of 15) represents 6,66% - an evil number, that makes the calculation kind of difficult..
The next, more delicate, problem is that your leds piss on BML's when it comes to efficiency. Some more than others. I don't really understand what they mean by "wall plug efficiency" One would like to think that it's the same numbers we are talking about, but XP-E P3 ≈ 100 l/w and their 630's gives 55 l/w, yet they claim 37% efficiency.
Your 3000k are 58% more efficient than theirs (136/86 =158,1)
your 630's are almost 100% more efficient but probably closer to 85 since you drive them harder than 350 mA)
Another problem is that your amount of 630 only equals half a dot. I won't make a proper calculation of this but i will tell how my estimation works.
So, this is a rough calculation/estimation, since there are so many limitations, and I kind of sucks at math.. I still think that its gonna give us a pretty good idea.
12 dots of 3000k equals 80,4% (this is less in your setup, but these leds are at least less lousy than their 630's and we need 3 dots to play with.
If we choose one 450 one 660 and one 630, we will get to a little bit to much 630 (not so much since they are so much weaker then yours), a little bit to much blue, and to little 660.
So when looking at the diagrams, we have to "ad some" 660 and think away a little bit of blue and a tiny amount of 630.
We can also exchange one of the 3000k dots to a 660, and instead "think away" a fair amount of 660. This isn't scientifically done, but i do believe that the "real" graphical representation, of real spectral situation in Buenos cabinet, isn't to far away from a mix of these two diagrams.
Puh.
Now i really have to sleep! tell me tomorrow if you don't agree on this.
Good night!
Ya I think I can picture in my mind how the curve looks but not how big the spikes of "boosted" wavelengths 450nm 630nm 660nm.
stardustsailor did some very in depth mathematical analysis of the CXA3070 3000K AB bin COB and here was one of his breakdown graphs. My spectrum should be like this with a hair more 630nm, little peak at 440-450nm, and a little bit bigger boost at 660nm.
Somehow he broke it down into umol/sec of many different wavelengths of output and then made a bunch of neat charts and data tables with the info. The pic above was one part of the analysis but I like how it shows the data pretty simply and effectively.
cocogrower
Member
I'm also a bit astonished that a few extra watts in a sertain wavelength makes such a difference. I think the 630 peak get so accentuated because you already have a peak close that region from your 3000k. That a few extra watts gives a big peak also makes more sense when when i'm looking at the diagram stardust made. It's like you get an exponential increase in lumen flux compared to actual diff in w. With that said i admit that there is some insecurity about excakt height of those tops, since the calculations were a bit fussy in the edges.
Are you thinking/hoping that your peaks are less accentuated? I think it looks really good! Following the discussion on this other forum it seems as if many people wants to mimic sunlight.. But at the same time it seams as if many people are getting fat buds with more red.. Possibly strain dependent... who really knows? I would't be surprised if the genetics of MJ already has adopted to artificial light to some degree. Anyhow I think you are pretty set! You have the cobs everybody is raving about, in the highest binning. And you have added the wavelenghts that are absolutely essential to MJ. And the amount of lumen you have is so high that your plants probably are saturated in every wavelength! I asked you before about of lumen per square foot.. I don't remember the excact size of your tent.. was it 2x3? you must have much more than 10000 lumen per f2!!! Some claim that the proof is in the pudding, and I'd most def say that you have the pudding!
My spectrum is similar to yours except that i lack some far red and orange. The only thing i'd like to change in the future, mostly for the fun of it, is to ad some far red in "dusk/dawn" and a bit of 10000k in mid day to mimic the brightest day light, with added UV for extra resin..
Your set up rock Bueno! If i hadn't seen your fingers in the pics, i would also have said that they are VERY GREEN!
Are you thinking/hoping that your peaks are less accentuated? I think it looks really good! Following the discussion on this other forum it seems as if many people wants to mimic sunlight.. But at the same time it seams as if many people are getting fat buds with more red.. Possibly strain dependent... who really knows? I would't be surprised if the genetics of MJ already has adopted to artificial light to some degree. Anyhow I think you are pretty set! You have the cobs everybody is raving about, in the highest binning. And you have added the wavelenghts that are absolutely essential to MJ. And the amount of lumen you have is so high that your plants probably are saturated in every wavelength! I asked you before about of lumen per square foot.. I don't remember the excact size of your tent.. was it 2x3? you must have much more than 10000 lumen per f2!!! Some claim that the proof is in the pudding, and I'd most def say that you have the pudding!
My spectrum is similar to yours except that i lack some far red and orange. The only thing i'd like to change in the future, mostly for the fun of it, is to ad some far red in "dusk/dawn" and a bit of 10000k in mid day to mimic the brightest day light, with added UV for extra resin..
Your set up rock Bueno! If i hadn't seen your fingers in the pics, i would also have said that they are VERY GREEN!
~160w 3000K dissipation at ~136lm/w = ~21760lm from the 3000K COBs in a 6 sq ft tent, so 21750lm/6 sq ft = 3625 lm/sq ft plus the light from the reds and blues. My LED dissipation wattage per sq foot is only ~32w/sq ft, though my particular LED bins/drive current combos result in a higher efficiency in output per watt used than HPS and also better PAR efficiency as well.
cocogrower
Member
Hm yeah those last numbers probably were a bit flawed.. I think that you accidentally took your total 3000k lumen number x 3.. It was 3 am in my time zone, sorry for not seeing that right away!
But like you say, 3650 l/ sq ft should be more than enough. Less would probably be beneficial if you only chased results in g/w..
if we look at your light spectrum in percentage of watts, it will look like this
3000K 160w 84,9%
630 nm 5,6w 2,9%
450 nm 7,2w 3,8%
660 nm 15,5w 8,2%
total wattage 188,3
Oddly enough not so different from the results you had. LOL, anyways the blue spike should definitely be less pronounces than in the diagrams i made. Anyways you do have a killer light, and these numbers might be interesting for anyone that likes to do the woodoo that you do!
But like you say, 3650 l/ sq ft should be more than enough. Less would probably be beneficial if you only chased results in g/w..
if we look at your light spectrum in percentage of watts, it will look like this
3000K 160w 84,9%
630 nm 5,6w 2,9%
450 nm 7,2w 3,8%
660 nm 15,5w 8,2%
total wattage 188,3
Oddly enough not so different from the results you had. LOL, anyways the blue spike should definitely be less pronounces than in the diagrams i made. Anyways you do have a killer light, and these numbers might be interesting for anyone that likes to do the woodoo that you do!
