Tyleb173rd
Well-Known Member
The DIY COB builds fascinate me and the very easy tutorials by members here have boosted my confidence in making one. But if they are so good does that make the blue/reds obsolete?
Are COBs better than the blue and red LEDs?
- Thread starter Tyleb173rd
- Start date
BOBBY_G
Well-Known Member
search for 'blurple' in this forum and you can read all kinds of opinions
while reds and blues give a more tailored spectrum, COBs are available in so many different colors (3000-6500K) that also match the mc cree curve pretty well. The higher efficiency of cobs in terms of lumens/watt and heat output put them above red and blues at the current time. and all the research is in cobs right now so theyre only getting better...and cheaper....quickly
while reds and blues give a more tailored spectrum, COBs are available in so many different colors (3000-6500K) that also match the mc cree curve pretty well. The higher efficiency of cobs in terms of lumens/watt and heat output put them above red and blues at the current time. and all the research is in cobs right now so theyre only getting better...and cheaper....quickly
Abiqua
Well-Known Member
red and blue ..........is fine if green is added, which is essentially what white light is....aka r.g.b!
Green is a photosynthetic driver as well and one reason why white light is good...but specifically tailored red nm, green nm and blue nm diodes may be the future of high effic white light??
who know's every week, something new is revealed...white light works mucho
Green is a photosynthetic driver as well and one reason why white light is good...but specifically tailored red nm, green nm and blue nm diodes may be the future of high effic white light??
who know's every week, something new is revealed...white light works mucho
SupraSPL
Well-Known Member
Comparing them in terms of conversion efficiency and cost/PAR W, the monos seem to be lower efficiency and cost more. The exception is the blue, but we typically only use 10-20% blue so that doesnt help mono's case much. When you consider the labor involved and the cost of lens/reflectors, the COBs are even more appealing.
Luxeon ES deep blue 450nm, N4R bin, $2.50 ea. @700mA (2W) 62.6% efficient typical - cost $2.04/PAR W
Cree XPE2 red 630nm P4 bin, $2.75 ea @700mA (1.7W) 43.7% efficient typical - cost $3.75/PAR W
Cree XPE photo red 660nm bin 14 $2.81 ea @ 700mA (1.5W) 45.4% efficient typical - cost $4.03/PAR W
------------------------------------------------------
Cree CXB3590 3500K 80CRi CD bin $43 ea @1.4A (50W) 56% efficient typical - cost $1.56/PAR W
Cree CXB3590 4000K 70CRi DB bin $43 ea @1.4A (50W) 61% efficient typical - cost $1.44/PAR W
Cree CXB3590 3500K 80CRi CD bin $43 ea @.7A (23W) 64% efficient typical - cost $2.93/PAR W
Cree CXB3590 4000K 70CRi DB bin $43 ea @.7A (23W) 69% efficient typical - cost $2.64/PAR W
Luxeon ES deep blue 450nm, N4R bin, $2.50 ea. @700mA (2W) 62.6% efficient typical - cost $2.04/PAR W
Cree XPE2 red 630nm P4 bin, $2.75 ea @700mA (1.7W) 43.7% efficient typical - cost $3.75/PAR W
Cree XPE photo red 660nm bin 14 $2.81 ea @ 700mA (1.5W) 45.4% efficient typical - cost $4.03/PAR W
------------------------------------------------------
Cree CXB3590 3500K 80CRi CD bin $43 ea @1.4A (50W) 56% efficient typical - cost $1.56/PAR W
Cree CXB3590 4000K 70CRi DB bin $43 ea @1.4A (50W) 61% efficient typical - cost $1.44/PAR W
Cree CXB3590 3500K 80CRi CD bin $43 ea @.7A (23W) 64% efficient typical - cost $2.93/PAR W
Cree CXB3590 4000K 70CRi DB bin $43 ea @.7A (23W) 69% efficient typical - cost $2.64/PAR W
superbak3d
Well-Known Member
cobs are just multiple leds chips mashed together to form a single chip.
With that, cobs produce higher light intensity over diodes, and you can run more cobs in less space.
If 1 cxb cob = 2 chinese led panels, then it would take 8 panels to match 4 cobs.
So imagine 12 cobs, or 24 cobs. Panels can't touch those
With that, cobs produce higher light intensity over diodes, and you can run more cobs in less space.
If 1 cxb cob = 2 chinese led panels, then it would take 8 panels to match 4 cobs.
So imagine 12 cobs, or 24 cobs. Panels can't touch those
Tyleb173rd
Well-Known Member
Thanks for the quick responses.
dandyrandy
Well-Known Member
They work well. As you can see.
Fauxton
Active Member
SupraSPL, where are you getting those efficiency numbers? I can only seem to find luminous flux ratings, not radiant for CXBs. Cree's XTE series royal blue is 53% efficient (http://www.cree.com/~/media/Files/Cree/LED-Components-and-Modules/XLamp/Data-and-Binning/XLampXTE.pdf) and it makes it seem very unlikely CXBs would reach an efficiency of 69%. I'm very open to being corrected (and really I want to be, since I'd considering using CXB for my next grow light) but I need to see links to support your statements.
FranJan
Well-Known Member
Try this to find out where the numbers Supra uses for the CXB efficiency come from. Some people have their doubts to their validity but regardless it's an excellent way to get a grip on how much light one needs when growing with COBs and whites in general.
http://rollitup.org/t/math-behind.868988/
I don't think Supra did the math for the latest Osram or Philips Reds but regardless COBs just make getting a good spectrum that much easier than monos. Matching, mixing, and measuring monos is practically a guessing game unless you're a stone cold engineering stud or have a sphere.
I used cheap red/blue panels 3 or 4 years ago. They were only about 28 actual watts (as measured with a watt meter). I had 4 of these - they worked well for seedlings but just didn't put out enough light for anything but small plants. LED flood lights became reasonable and I decided to try a couple. They work remarkably well for vegging having excellent penetration and I've added several more. They are cheap, Chinese made (what isn't) LED flood lights and reliability is an issue. I always keep backups and I have no problem sourcing ($5 per 50 watter) and installing the COBs when they fail.
50 watters are $20-30 and 100 watters about $40-60. I use daylight LEDs but if they came in 50 watt blue or red that might be interesting to experiment with.
Fauxton
Active Member
Very interesting to think through that math. I reserve the right to be skeptical, but nevertheless I value their hard work as well as your reply. Thanks so much for taking the time to respond. I did several searches on RIU and the web but found it difficult to locate any good info.
alesh
Well-Known Member
If Cree provides us with correct charts and numbers, Supra's calculations are very close to reality. He's using luminous flux values halfway to the next bin, though.
Top bin XT-E RB is over 70% efficient at 100mA. The single dies in CXB3590 are driven at similar power levels.
BOBBY_G
Well-Known Member
well, crees bins are MINIMUM flux, so they would generally be expected to flux somewhere in the middle of that range
ttystikk
Well-Known Member
Speaking of new bins, there's one coming for CXB3590 chips, among others. I'm sooooooooo excited!
SupraSPL
Well-Known Member
Yes, that highlights the reason COBs are reaching such high efficiency compared to single diodes, reducing current droop and to some extent temp droop.
If you run a CXB3590 @ 25W, each die is running at only .15 watts yet $2.75/PAR W is still reasonable. If we ran an XTE at .15W, we would be at $15/PAR W or more.
The dies in both 3070 and 3590 appear to be the same size, but the 3590 has more of them (168 vs 132) and they are spaced further apart in the 3590. The 3070 has 11 strings of 12 dies in parallel and the 3590 36V has 14 strings of 12 dies in parallel.
Last edited:
Take a look at the McCree curve. While it is completely true that certain parts of the spectrum lead to faster growth, it is also true that warm white does a very good job of replicating the curve. Most LED companies omit tons of the yellows and greens and your grow room looks like you're on Mars. Whatever slight decrease in optimality you may receive by not using just reds for example (where the peak of the spectrum occurs) is more than made up for by being able to actually see your flowers fairly accurately (high CRI) and appreciate them in their natural glory.
Fauxton
Active Member
A great point. The studies I have looked at suggest faster growth rate for blurple (at least in the species studied), but there is much to be said for COB PPFD power, convenience, cost, and, like you said, being able to see what you love. That and my wife won't let me have a purple light without a grow tent.
BOBBY_G
Well-Known Member
"baby, this is the red light district, no rules here, ya dig?"
SSGrower
Well-Known Member
Can you guys read minds? I am trying to sort some of this out. I have taken some pictures and used the whitbalance on the camera to quasitatatively measure the colortemp and am finding Flourescents and incandescents the camera can compensate for and even for single color leds but arrays seem to offer a different challenge for the camera and cant be manually white balanced. Also many cameras make histograms but this only covers visible light.
SSGrower
Well-Known Member
Sorry, got pulled away basically my question is can the camera be a useful tool in this by showing where the photons are in the spectrum and then what to do with that info? I can try to tune to the same basic curves as the photo active spectrum but the problem with every camera and sensor are unique and to do true color must be calibrated with all the electronics.