Not sure if you have seen this AP, but what this graph is showing is actually even worse in practice because we have access to top bin 3000K 80 CRi but no access to any high bin high CRi COBs. Although the high CRi or 2700K COBs may have a better spectrum than the 3000K 80 CRi, you can see the sacrifice that is made to achieve that, it has been described as putting a filter over a 3000K 80 CRi, many fewer photons are lost. If you are interested in a better spectrum than the 3000K 80 CRi alone, a very good approach may be to supplement with 660nm Cree photo reds which are over 42% efficient and could add a large 660nm spike to this graph with no efficiency sacrifice. You could even widen the 660 spike by mixing photo reds of 650, 660 and 670nm.Based on the diagrams, I would choose the CRI 93 (3k) but after comparing the four analyses, I'd have to ask you for insight.
With the 93 CRI, you get less green and more blue and infrared (~3-5%), which is good in my book. Also, the 93 CRI peaks at 630nm, while the 80CRI spikes around 600nm. However, the different estimates lead me to believe that the 80 CRI performs at a higher efficiency, thus begs the question: what do you make of all of this?
Not sure if you have seen this AP, but what this graph is showing is actually even worse in practice because we have access to top bin 3000K 80 CRi but no access to any high bin high CRi COBs. Although the high CRi or 2700K COBs may have a better spectrum than the 3000K 80 CRi, you can see the sacrifice that is made to achieve that, it has been described as putting a filter over a 3000K 80 CRi, many fewer photons are lost. If you are interested in a better spectrum than the 3000K 80 CRi alone, a very good approach may be to supplement with 660nm Cree photo reds which are over 42% efficient and could add a large 660nm spike to this graph with no efficiency sacrifice. You could even widen the 660 spike by mixing photo reds of 650, 660 and 670nm.
SDS we need a hybrid. It's the only way we are gonna get the high CRI, broader peaks that growers want from our "whites". There's so many blends companies are using with COBs we're never going to know what blend works best until we build our own grower centric white. At least that's how I'm looking at it nowadays. I'm telling you guyz, COBs are dead .
http://news4led.blogspot.fr/2014/03/design-approaches-for-high-cri-leds.html?m=1
Design Approaches for High-CRI LEDs
Ninety-CRI LEDs have been about 20% less efficient than 80-CRI, limiting adoption. Here are two approaches to closing that gap.
White LEDs are typically fabricated by combining blue LEDs with yellow, green, and red phosphors. Historically, cool white LEDs were the first to reach the market, for applications in which high CRI was not a requirement. These LEDs were made with only yellow phosphors and had a bluish-white tinge, in the CCT range of 5,000 to 6,500 K. As illumination applications became more prevalent, LEDs with a warmer white appearance in the CCT range of 2,700 to 4,000 K became available. These warm white LEDs used a combination of yellow and red phosphors to achieve the desired CCT and minimum 80 CRI but with an efficiency penalty when compared to the cool white LEDs.
Figure 1. Lumen loss of current red phosphors in warm white LEDs
Figure 1 shows the spectrum for a typical 2,700 K, 80-CRI white LED (green line) and the human photopic response (blue dashed line). Although the addition of conventional red phosphors enables warmer CCTs and higher CRI than yellow phosphors alone, there is significant lumen loss since much of the red phosphor emission occurs outside the human eye's sensitivity. To make a 90-CRI LED, the typical approach is to use an even redder phosphor to reach the higher CRI, but with even less emission in the eye's sensitive region. As a result, a 90-CRI LED at 100 lm/W is about 20% less efficient than an equivalent 80-CRI LED.
One approach to improve the efficiency of the 90-CRI LED is to focus on phosphor improvements such as the introduction of a narrow (< 50nm FWHM) red phosphor. Figure 1 also shows the modeled spectra of a 2,700 K, 90-CRI white LED (red line) that includes a narrow (30 nm FWHM) red phosphor. The narrower red phosphor increases the lumen efficiency by substantially reducing the emission outside the eye's sensitivity range. By optimizing the material quality, the wavelength peak, and the width of the red phosphor's emission, up to 20% gains in LED efficiency may be achieved.1 This approach is expected to benefit all L1 LEDs including single emitters such as LUXEON TX, LUXEON 3535L, or LUXEON COB.
Another approach is the to use direct red LEDs in combination with white LEDs to create a hybrid LED module with high efficiency and high CRI. A direct red LED has a significantly narrower emission than a red phosphor, leading to a similar reduction to the lumen loss shown in Figure 1. In 2013, as part of a DOE SSL Product Development project, a hybrid LED module was demonstrated by Philips Lumileds with 712 lm, 140 lm/W, CRI = 91, and R9 = 75 operating at 85°C.2 The direct red wavelength (610 to 615 nm) was selected for optimal efficiency and R9 impact, and an off-white LED was used to provide the green part of the spectrum. The combination of red and white LEDs does result in more complicated system designs, but it can provide the added benefit of color tunability or integrated color control.
Approaches for efficient 90-CRI LEDs
High-CRI technology
Examples
Typical specs at 85°C
High-CRI phosphor
Luxeon TX, MP 3535L, COB
95–100 lm/W, CRI = 95, R9 = 80
- Simple configuration
- Narrow red phosphor improves efficiency
- Large efficiency penalty w/ conventional red phosphors
Direct red and off-white
Hybrid LED for DOE SSL Product Development project
140 lm/W, CRI = 91, R9 = 76
The table summarizes these two technical approaches towards enabling high-efficiency and high-CRI LED solutions for solid-state lighting. As these technologies mature, we expect to see increasing adoption of 90-CRI solutions as the color quality benefit begins to exceed the efficiency tradeoff.
- High efficiency w/ high CRI
- Color tunable
- High LED and system cost
References
- O. Shchekin, J. Bhardwaj, "LED technologies for Breakthrough in performance and cost," China SSL 2012
- G. Huang, W. Soer, G. Vu, F. Jin, "High Power Hybrid Warm White LEDs for Illumination," poster presented at DOE SSL R&D Workshop, Tampa, FL (2014)
I didn't initially see that but good point to mention. It appears we have the technology, publicly, to replicate a good portion of the vegetative states for plants*, with the exception of "mono" UV and (IR) stars. I suppose if you had one/tenth the knowledge of StarSailor over this matter, you could come up with a mathematical scheme to mirror the sun's spectral qualities, during the Spring, using specialized monos.Rich info Stardust!
Look at the Cool LED numbers compared to the sun numbers in the visible percentages. Almost a perfect match.
Not sure if you have seen this AP, but what this graph is showing is actually even worse in practice because we have access to top bin 3000K 80 CRi but no access to any high bin high CRi COBs.
If you asked me whether more or less available phosphor on a given COB was good, I'd be found dumbfounded .2700K leds differ from 3000K ones only in (of same type ) phosphor load.
Different CRI 2700 & 3000K CCTs ,use different phosphor types .
I.e .( Random values ,just for comparison )
3000K 80 CRI : Blend of 25% Green silicate phosphor & 75% Amber silicate phosphor .
Total phosphor content : 100mg /sq. cm of LES
2700K 80 CRI : Blend of 25% Green silicate phosphor & 75% Amber silicate phosphor .(same as above )
Total phosphor content : 130mg /sq. cm of LES
3000K 95 CRI :Blend of 30% Green silicate phosphor & 70% Red nitride phosphor .
Total phosphor content : 100mg /sq. cm of LES
2700K 95 CRI :Blend of 30% Green silicate phosphor & 70% Red nitride phosphor .
Total phosphor content : 130mg /sq. cm of LES
So,actually 2700 & 3000K of same CRI series ,differ mainly (spectrum wise ) on the amount of blue wls
they output.2700K output less blue excitation wls and a tad more the phosphor emitted wls ,than 3000K.
" Although the high CRi or 2700K COBs may have a better spectrum than the 3000K 80 CRi "
Regarding the high CRi 2700K and 3000K, the graphs I have seen show a significant peak wavelength shift toward deep red. But regarding 2700K 80 CRi, I see your point and it is a good one. Unfortunately I rarely get to see 2700K graphs compared to 3000K and I have always wondered why they are not included. My assumption was that 2700K has a slightly longer wavelength peak into the red, but if they are simply sacrificing some of the blue for more of the same exact phosphor, the peak would not shift, which would make 2700 K 80 CRi flat out inferior to 3000K 80 CRi. I will have to explore that issue further thank you for pointing it out.a point that I kinda disagree with you...actually is the other way around ,the opposite...actually 2700 & 3000K of same CRI series ,differ mainly (spectrum wise ) on the amount of blue wls
they output
I absolutley agree. When I supplement with deep reds I add blues as well to maintain the blue ratio. I did not mention that in my post because I am trying to be concise but it is a very important point for new DIYers to understand. The blue ratio is very important during flowering. For starters and very noticeable, insufficient blue decreases frosting, flavor and potency. It will also affect stretch and internode distance.My doubts or concerns about mixing reds & deep reds with the CXA 3000K 80 CRI warm white light ...Adding reds-deep reds ,will alter (decrease) the 10.6 % ,which is pretty close to 7% .
Not good during vegetative growth,at least.To compensate a red led enhancing ,blue leds might have to be used.
Here is what I have noticed so far in in practice. Switching from HPS to LEDs resulted in a noticeable decrease in flowering time and a cleaner finish. Maybe it is due to cooler canopy temps, maybe it is due to the infrared glow of large heatsinks cooling off after lights out. Maybe it is both, but in my experience, canopy temps and maybe even night time temp is much more important when it comes to flowering duration.Adding reds -deep reds will (affect deep red/far red ratio, possibly causing) prolonged flowering duration...reproductive stage duration and maturing process .
Prolonged flowering periods do not mean increased yields ,necessarily .
So I take it that Z2 bin, from either DigiKey or Mouser, is the common, average grade quality whereas AB bin is like the top selection, the cat's meow as one of my old but whimsical professors would say, yes?As of just a dew days ago, we were able to buy 20 packs of the CXA3070 3000K AB bin from Arrow. Unfortunately they just sold out and now all we have is the Z2 bin from Digikey or the Z2 bin in 20 packs from Arrow. However, there may be a few RIU members that have access to the AB bins but I am not sure what is currently available. Hopefully more ABs will become available for a fair price.
I agree.............with some "tweaking" they could be the best WW cob on the market........if the egg stays sunny-side upThose Alta leds have pretty impressive numbers. They may be in my future.
267 lumens/watt @93-95 cri is very impressive indeed. If this is for real they could become the leader in horticultural led tech.I agree.............with some "tweaking" they could be the best WW cob on the market........if the egg stays sunny-side up
Well ,due to shortage of time ...That is why it is more efficient to supplement warm whites with deep reds rather than using high CRI to get more red. In other words, using a blue LED to make a high CRI warm white LED is kind of a fools errand, but it is what we have to work with at the moment.
You can take that line of thinking a step further and use cool whites and supplement with reds and deep reds, but in the case of the CXA3070, our warm whire were actually more efficienct than the cool whites because of the bin selection at the time. Also, the cool whites are dominant in the yellow/green range rather than the red range, which may have some undesirable photomorphogenic effects during the flowering phase (slower maturation, fluffier buds/leafier buds/foxtailing?)
I ran a room with almost entirely 3000K warm white 80CRi during the summer the canopy temps were slightly elevated and I noticed a longer flowering duration, increase in yield and slight decrease in resin quality. Then I added a bunch of blues/deep reds and ran the room again, with slightly cooler canopy temps and the ladies finished on time and much better outcome (slightly decreased yield, better quality resin and buds). As you know I run and re-run some genos over and over so I can notice when they are expressing different phenos.