CRI and ANSI White bin questions

SSGrower

Well-Known Member
So I get that a higher CRI value means the led is capable of creating a more uniform and pleasing lighting aesthetic for humans but does it really mean anything to the plant so long as the minimum spectral power needs are met? The CREE literature shows graphs of relative power output at a given frequency for various CRI values but I interpret this as an intensity related to the overall quantity of light generated by that LED. Meaning that although the 93 CRI 3000k leds show more area under the graph at a given frequency it may not actually produce more photons in that region that an 80 or even 70 CRI?

For whatever reason I think the plant can handle a slight variation of color so long as it meets some minimum level. A little extra blue or a little extra red and the light looks blurple to us and is annoying but doesn't have a huge adverse affect as long as we stay away from the extremes.

On ANSI color bins I think my question is at least slightly related to the above thought train but it seems to my uneducated brain that these focus on getting accurate color at the light temperatures that are pleasing to humans and less effort is placed on manufacturing leds optimised to create a photoactive environment. Basically it looks like these might be good option if high CRI values are desired in the 5000-6500k color range? There must be a reason why these are so much less expensive but I wonder if it is that CREE dosen't recognise them as meeting the needs of a specialty market and therefore could "justify" asking a premium price for them??? or are they total crap for plants??

I am new to LEDs and due to a poorly researched purchase have been thrown into this world. So I beg your forgiveness in advance, and appreciate any insights.
 

welight

Well-Known Member
Couple of points, standards are omnipresent in everything including lighting, one of the first related to colour consistency was ansi binning that referred to cct within what Cree would define as approx 4-8 bins, so relatively wide meaning you could get leds that vary greatly in colour, led to led, for that reason they were lower cost, in more recent times cree has gone from ANSI BINS to MACADAMS STEPS, so much narrower colour ranges, this is really driven by lighting spec,when architects just dont want this variation for obvious reasons. Hi CRI is a good thing but the penalty you pay is flux, SPD, Phosphors that create HI CRI just dont flux out as well as lower CRI
Cheers
Mark
 

SSGrower

Well-Known Member
Couple of points, standards are omnipresent in everything including lighting, one of the first related to colour consistency was ansi binning that referred to cct within what Cree would define as approx 4-8 bins, so relatively wide meaning you could get leds that vary greatly in colour, led to led, for that reason they were lower cost, in more recent times cree has gone from ANSI BINS to MACADAMS STEPS, so much narrower colour ranges, this is really driven by lighting spec,when architects just dont want this variation for obvious reasons. Hi CRI is a good thing but the penalty you pay is flux, SPD, Phosphors that create HI CRI just dont flux out as well as lower CRI
Cheers
Mark
Thanks for tracking my thought process I had made the assumption that the thickness and consistency of the coating was a large part of the manufacturing variances in actual color production and flux, more so than the actual semi-conductor manufacturing. Ive posted this question in the noob area and didnt get any traction, thanks and I'll put a summary there.
 

nevergoodenuf

Well-Known Member
I just cut down a non-scientific side-by-side with 500 watts of 2700K 90+cri and 4000K 80cri and I could not see any difference. Everything looked the same, as I cut them down. I expected to see some difference. I will try it again eventually, but try to do a more controlled test.
 

grouch

Well-Known Member
I just cut down a non-scientific side-by-side with 500 watts of 2700K 90+cri and 4000K 80cri and I could not see any difference. Everything looked the same, as I cut them down. I expected to see some difference. I will try it again eventually, but try to do a more controlled test.
Can't wait for the smoke report
 

grouch

Well-Known Member
as a takeaway reading some of these, seems higher CRI gives a closer map on to MCREE's
Cheers
Mark
Would you be able to run tests on the cxb-3590 3000k 80 cri and the 90 cri that you carry? I would be interested to see how the spectrum compares within the same bin.
 

SSGrower

Well-Known Member
I just cut down a non-scientific side-by-side with 500 watts of 2700K 90+cri and 4000K 80cri and I could not see any difference. Everything looked the same, as I cut them down. I expected to see some difference. I will try it again eventually, but try to do a more controlled test.
How non scientific are we talkkin? Do you mean they were totally separated or 500 watts of 2700 and 4000 split in one room?
 

SSGrower

Well-Known Member
So what I'm takin away is since I plan on mixing color Temps on fewer than 10 leds anyway the lower Flux on the higher cri chips is more of a downfall than the color accuracy and frequency distribution advantages. Especially when focusing heavily on flux?
 

Greengenes707

Well-Known Member
as a takeaway reading some of these, seems higher CRI gives a closer map on to MCREE's
Cheers
Mark
Actually the opposite.
When the McCree curve is factored into each spectrum, the 80CRI has less of a loss.
SDS has it as PER, but the more commonly known as YPF.

The high CRI "waste" it's load above 700nm. Though it is used, we all know the steep drop off at 680 to basically nothing by 740nm in the RQE(mccree) isn't contributing to the YPF very much.

Stardust is using the whole spectrum ~380-780nm so they are counted and still the higher CRI falls short in it's fitting of the RQE/mcree curve.
80cri=94,848-85,960=8888µmols÷94,848= ~9% loss to RQE
90cri=81,870-73,332=8538µmols...8538÷81,870= ~10% loss to RQE
That is just speaking spectrum potential, not counting output differences.
So what I'm takin away is since I plan on mixing color Temps on fewer than 10 leds anyway the lower Flux on the higher cri chips is more of a downfall than the color accuracy and frequency distribution advantages. Especially when focusing heavily on flux?
Output/flux is the 90% of the goal. The rest is spectrum, and when deciding between whites, leads us to 3500K giving the highest output while still hitting the somewhat accepted values of nm range ratios like blue, red, and r:fr.
 

grouch

Well-Known Member
Actually the opposite.
When the McCree curve is factored into each spectrum, the 80CRI has less of a loss.
SDS has it as PER, but the more commonly known as YPF.

The high CRI "waste" it's load above 700nm. Though it is used, we all know the steep drop off at 680 to basically nothing by 740nm in the RQE(mccree) isn't contributing to the YPF very much.

Stardust is using the whole spectrum ~380-780nm so they are counted and still the higher CRI falls short in it's fitting of the RQE/mcree curve.
80cri=94,848-85,960=8888µmols÷94,848= ~9% loss to RQE
90cri=81,870-73,332=8538µmols...8538÷81,870= ~10% loss to RQE
That is just speaking spectrum potential, not counting output differences.

Output/flux is the 90% of the goal. The rest is spectrum, and when deciding between whites, leads us to 3500K giving the highest output while still hitting the somewhat accepted values of nm range ratios like blue, red, and r:fr.
I wouldn't go as far as to call the excess red/far red waste in the 93cri. Matching the McCree Curve isn't the only goal with or lighting. We are growing plants that adapt to what we give them in our artificial environments. I would like to try a couple of the higher cri and see the difference before I rule it out on paper.
 

ttystikk

Well-Known Member
While this isn't an LED example, I noticed much better frost but less yield when running a 93CRI 860W CDM Allstart lamp (at 1kW) vs a 22CRI HPS 1kW lamp. This result led me away from thinking high CRI was a panacea for performance.
 

Greengenes707

Well-Known Member
I wouldn't go as far as to call the excess red/far red waste in the 93cri. Matching the McCree Curve isn't the only goal with or lighting. We are growing plants that adapt to what we give them in our artificial environments. I would like to try a couple of the higher cri and see the difference before I rule it out on paper.
That's cool. Options are great. And sometimes if motivated enough by them to be a true believe, can actually drive new scientifically valid and accepted pricipals.
That said...your disagreement is with accepted science, not my opinion.

You also imply I said that the 80cri has nothing over 700nm. It does and is more in line with sun like and ideal to the higher CRI counterpart. As where too much is of the high CRI is devoted up there.
 

welight

Well-Known Member
Rendering has always been more about the appearance of objects under light or via a camera, realising natural colours as best we can, it was not really about the resultant effect of its relationship with matter in this case things what grow
cheers
Mark
 
Top