The Ideal spectrum.

[Rahz]

If we see that the various photoreceptors are at work creating the Mcree curve it would not be difficult to explain the shape which is a mix of symmetry and asymmetry. The 580-680 range of the Mcree curve isn't exactly smooth, but the relative deviation is minor, even for the 660-680 range though it does poke out a little bit. Beyond that it falls back in line. There's also an odd linear region between 475 and 525.

If we superimpose a random plant photoreceptor response graph onto the Mcree curve we get some interesting results. I'm not suggesting either of the two data sets are perfect for a particular plant, but to some degree they should be close right?

This might suggest that all other things being equal it would be useful to target a handful of wavelengths. The individual receptor response curves explain the Emerson effect. Since LED contains almost no 400nm light there's especially good reason to add light in the 660-690 region, 675 specifically so chlorophyll A is active. There are also 4 phosphor curves in the pic, 4000K 80CRI, 4000K 70CRI, 3000K 70CRI and 2700K 80CRI. The 4K SPDs ave been compressed so the 400-500nm band is in the 10-12% range. The other two samples were compressed at the same ratio for comparison but might benefit from some additional blue.

For a simple 2 part solution (cob+660) I would probably go with 5000K 70CRI or 6500K 80CRI. Those SPDs would have been more difficult to carve out of their respective graphs so I haven't added them. I think 4000K 70CRI would work well enough using 660nm to lower the K to 3000. If we use 660nm light to lower the 3000K sample to 2700K that would provide about 5% blue, which is similar to traditional HPS spectrums.

Another approach would be to target specific colors and improve on red/blue, which might prove to be more efficacious if done in the correct ratios. The target wavelengths, going by the chart below would be 400 (or 675), 440, 575 and 625. I would hesitate to guess on the desired ratios. I'm also not clear on why chlorophyll A is showing a stronger 400nm response as an individual photoreceptor when it seems to create more photosynthesis in the Mcree curve at 675, or whether 400nm light would invoke the Emerson effect the same as 675nm light does. Any ideas or consideration on the subject?

.

 

[trojanvirus]

here's a read you might enjoy.

 

[Dave455]

here's a read you might enjoy.

??

 

[Rocket Soul]

If we see that the various photoreceptors are at work creating the Mcree curve it would not be difficult to explain the shape which is a mix of symmetry and asymmetry. The 580-680 range of the Mcree curve isn't exactly smooth, but the relative deviation is minor, even for the 660-680 range though it does poke out a little bit. Beyond that it falls back in line. There's also an odd linear region between 475 and 525.

If we superimpose a random plant photoreceptor response graph onto the Mcree curve we get some interesting results. I'm not suggesting either of the two data sets are perfect for a particular plant, but to some degree they should be close right?

This might suggest that all other things being equal it would be useful to target a handful of wavelengths. The individual receptor response curves explain the Emerson effect. Since LED contains almost no 400nm light there's especially good reason to add light in the 660-690 region, 675 specifically so chlorophyll A is active. There are also 4 phosphor curves in the pic, 4000K 80CRI, 4000K 70CRI, 3000K 70CRI and 2700K 80CRI. The 4K SPDs ave been compressed so the 400-500nm band is in the 10-12% range. The other two samples were compressed at the same ratio for comparison but might benefit from some additional blue.

For a simple 2 part solution (cob+660) I would probably go with 5000K 70CRI or 6500K 80CRI. Those SPDs would have been more difficult to carve out of their respective graphs so I haven't added them. I think 4000K 70CRI would work well enough using 660nm to lower the K to 3000. If we use 660nm light to lower the 3000K sample to 2700K that would provide about 5% blue, which is similar to traditional HPS spectrums.

Another approach would be to target specific colors and improve on red/blue, which might prove to be more efficacious if done in the correct ratios. The target wavelengths, going by the chart below would be 400 (or 675), 440, 575 and 625. I would hesitate to guess on the desired ratios. I'm also not clear on why chlorophyll A is showing a stronger 400nm response as an individual photoreceptor when it seems to create more photosynthesis in the Mcree curve at 675, or whether 400nm light would invoke the Emerson effect the same as 675nm light does. Any ideas or consideration on the subject?

View attachment 4248849 .

Whats your opinion on 3000 or even 2700k 90 cri plus 660nm for flowering? Im Ioping to have a spectrum that peaks in 630 and 660 quite evenly, with a fat tail of far red.

 

[Rahz]

I don't really have an opinion either way. I've been playing around with 4000K 70/80 CRI +660 because it can be adjusted down to 3000K, but some 3000K variant adjusted down to 2500-2700K might work well too. Going by the photoreceptor response curves my only concern would be that phycoerythrin wouldn't be getting much action on the high end or the low end. How important that is I'm not sure but If a lower CRI + 660nm is better, that might be the reason. I thought 80 CRI would split the difference best in that regard, but when looking at the two 4000K samples in the chart I posted it's kind of a toss up between 70 and 80.

To address @trojanvirus

One thing to keep in mind is that the main historical problem with blurple AFAIK is cost and/or efficiency. Someone recently cited Bugbee (?) suggesting blurple being most efficient and there's a US based LED manufacturer (name escapes me ATM) that produced a chart from their study indicating the same, blurple most efficient followed by warm white then cool white.

So while I don't doubt the results in the link I do question the efficacy of green light in real world indoor growing conditions. The Mcree curve was based on whole leaf tissue, so if green can gain efficiency I would expect to see it show up in whole plant response. However, indoor plants aren't getting hit with daytime light levels. Typically a person will use half that or less and satisfy the DLI by keeping that intensity solid for 12 hours. Plants are hopefully going to be on the shorter side as well so how much benefit the extra penetration will provide is in doubt.

Finally, I can offer results from my personal studies using 3000K 70CRI -vs- 3000K 90CRI. Yield is roughly equal yet the 70CRI sample has higher green levels and about 10% more PAR than the 90CRI sample. It's not an open and shut case, but does appear to favor red. One could theorize that 90CRI at 90% PAR is able to barely edge out 70CRI due to the Emerson effect alone and not because 90CRI contains more 630nm light... suggesting that perhaps green/deep red would be the killer combination... but I would need to see that happen real world to have any confidence in the idea.

 

[Rahz]

??

Doesn't show up until you hover but the link is in the text.

 

[trojanvirus]

Doesn't show up until you hover but the link is in the text.

Yes, I hyperlinked the text. I tried to change the text color, but it never works.

My last plant was grown under 2700K light and didn't stretch at all until transition, but the entire plant was pretty well rock bud. I don't know if I want to add color temps much higher than 3500K, although many claim terpene production is increased.

 

[Rocket Soul]

I don't really have an opinion either way. I've been playing around with 4000K 70/80 CRI +660 because it can be adjusted down to 3000K, but some 3000K variant adjusted down to 2500-2700K might work well too. Going by the photoreceptor response curves my only concern would be that phycoerythrin wouldn't be getting much action on the high end or the low end. How important that is I'm not sure but If a lower CRI + 660nm is better, that might be the reason. I thought 80 CRI would split the difference best in that regard, but when looking at the two 4000K samples in the chart I posted it's kind of a toss up between 70 and 80.

To address @trojanvirus

One thing to keep in mind is that the main historical problem with blurple AFAIK is cost and/or efficiency. Someone recently cited Bugbee (?) suggesting blurple being most efficient and there's a US based LED manufacturer (name escapes me ATM) that produced a chart from their study indicating the same, blurple most efficient followed by warm white then cool white.

So while I don't doubt the results in the link I do question the efficacy of green light in real world indoor growing conditions. The Mcree curve was based on whole leaf tissue, so if green can gain efficiency I would expect to see it show up in whole plant response. However, indoor plants aren't getting hit with daytime light levels. Typically a person will use half that or less and satisfy the DLI by keeping that intensity solid for 12 hours. Plants are hopefully going to be on the shorter side as well so how much benefit the extra penetration will provide is in doubt.

Finally, I can offer results from my personal studies using 3000K 70CRI -vs- 3000K 90CRI. Yield is roughly equal yet the 70CRI sample has higher green levels and about 10% more PAR than the 90CRI sample. It's not an open and shut case, but does appear to favor red. One could theorize that 90CRI at 90% PAR is able to barely edge out 70CRI due to the Emerson effect alone and not because 90CRI contains more 630nm light... suggesting that perhaps green/deep red would be the killer combination... but I would need to see that happen real world to have any confidence in the idea.

On cris: i think you have to consider finishing times aswell, if i remember right the 70 cri finished way slower than the 90cri in that test of yours to end up yielding the same. To me that a win for 90cri, same yield faster. Have you considered 3000k 70 cri, 660 +730? Or even only 730?

On blurple: to me it seems quite obvious that cannabis quite like blurple, but not as much as light intensity. If i i were to add blurple id probably pass on straight 450/660 nm monos but i would love to try to add some phosphor based blurple. The kind that look like youre standard 90 cri spectrum but having the red peak at 660 instead of 630, with a long fat tail of far red. If my memory serves me this was the blurple spectrum cited in that article.

Other spectrum: sunshine like spectrums like that 5600 97cri that cutter has? Ive also thought a lot of about adding infra red. The setup where my light runs has big problems with cold winters. It might actually be more efficient heating the cannopy with infra red than heating the whole room. Id Id a like to explore infrareds on lights out to increase metabolism at lights out, when the plant does most of its growing. But not sure if it might f up the plant.

 

[Dave455]

I don't really have an opinion either way. I've been playing around with 4000K 70/80 CRI +660 because it can be adjusted down to 3000K, but some 3000K variant adjusted down to 2500-2700K might work well too. Going by the photoreceptor response curves my only concern would be that phycoerythrin wouldn't be getting much action on the high end or the low end. How important that is I'm not sure but If a lower CRI + 660nm is better, that might be the reason. I thought 80 CRI would split the difference best in that regard, but when looking at the two 4000K samples in the chart I posted it's kind of a toss up between 70 and 80.

To address @trojanvirus

One thing to keep in mind is that the main historical problem with blurple AFAIK is cost and/or efficiency. Someone recently cited Bugbee (?) suggesting blurple being most efficient and there's a US based LED manufacturer (name escapes me ATM) that produced a chart from their study indicating the same, blurple most efficient followed by warm white then cool white.

So while I don't doubt the results in the link I do question the efficacy of green light in real world indoor growing conditions. The Mcree curve was based on whole leaf tissue, so if green can gain efficiency I would expect to see it show up in whole plant response. However, indoor plants aren't getting hit with daytime light levels. Typically a person will use half that or less and satisfy the DLI by keeping that intensity solid for 12 hours. Plants are hopefully going to be on the shorter side as well so how much benefit the extra penetration will provide is in doubt.

Finally, I can offer results from my personal studies using 3000K 70CRI -vs- 3000K 90CRI. Yield is roughly equal yet the 70CRI sample has higher green levels and about 10% more PAR than the 90CRI sample. It's not an open and shut case, but does appear to favor red. One could theorize that 90CRI at 90% PAR is able to barely edge out 70CRI due to the Emerson effect alone and not because 90CRI contains more 630nm light... suggesting that perhaps green/deep red would be the killer combination... but I would need to see that happen real world to have any confidence in the idea.

How about 80cri 3000k cree cob with added arcadia 13% reptile. seems good combo.

 

[Rahz]

How about 80cri 3000k cree cob with added arcadia 13% reptile. seems good combo.

I don't have enough experience with UV to comment directly, but if adding it to a phosphor spectrum I might try 3000K 90CRI. Small differences in these spectrums so I suspect 3000K 80CRI + UV would work well. Would it be the best? I dunno.

 

[Rahz]

On cris: i think you have to consider finishing times aswell, if i remember right the 70 cri finished way slower than the 90cri in that test of yours to end up yielding the same. To me that a win for 90cri, same yield faster. Have you considered 3000k 70 cri, 660 +730? Or even only 730?

On blurple: to me it seems quite obvious that cannabis quite like blurple, but not as much as light intensity. If i i were to add blurple id probably pass on straight 450/660 nm monos but i would love to try to add some phosphor based blurple. The kind that look like youre standard 90 cri spectrum but having the red peak at 660 instead of 630, with a long fat tail of far red. If my memory serves me this was the blurple spectrum cited in that article.

Other spectrum: sunshine like spectrums like that 5600 97cri that cutter has? Ive also thought a lot of about adding infra red. The setup where my light runs has big problems with cold winters. It might actually be more efficient heating the cannopy with infra red than heating the whole room. Id Id a like to explore infrareds on lights out to increase metabolism at lights out, when the plant does most of its growing. But not sure if it might f up the plant.

In my experience 3000K 70CRI provided the best veg. Although I didn't use any 4000K spectrums I thought the 3000K 70CRI sample looked better than 3500K 80CRI. So I think 3000K 70CRI with added 660 would work quite well.

I didn't see the spectrum cited in the article. If you're right that would be some food for thought. Hopefully this thread will take off and we'll get citations with links and other relevant info.

What would be bad ass is a pre-built lamp with for instance 650 PPF of 4000K 80CRI and a separate channel with 350 PPF of 660nm and a third channel with 730. Onboard circuit automatically keeps the 730 on for 15 minutes past lights out. Manually select whether phosphor only or phosphor + monos.

Whether these sunlight spectrums will prove themselves in the long run I don't know. I don't suspect sunlight is the best just because plants evolved under it. Nature does the best with what it's got, but it would be a mistake to assume the spectrum can't be improved artificially especially from an efficiency perspective. The data seems to be indicating sunlight isn't the best in that regard, but I'm open to seeing results that suggest otherwise.

 

[Rocket Soul]

Im actually not sure about the article now as i had another look. Tried chasing down the spectrum this pic is the best i found, but it would only represent the red peak, missing a 450 blue peak. Peak in 660 with nice juicy far red tail going all the way out to 780.

 

[wietefras]

I don't suspect sunlight is the best just because plants evolved under it. Nature does the best with what it's got, but it would be a mistake to assume the spectrum can't be improved artificially especially from an efficiency perspective. The data seems to be indicating sunlight isn't the best in that regard, but I'm open to seeing results that suggest otherwise.

Plants are hugely inefficient anyway. Only a few percent of the light they receive ends up as an increase in solid matter. Somewhere between 2% and 8% depending on circumstances and who made the estimate)

Also, plants might have their reasons not to absorb certain wavelengths. like too "communicate" with animals around them. There is a reason flowers are so brightly colored. Perhaps there is a reason for them not taking up green light so readily has a purpose too?

 

[Rahz]

Anything is possible. Some plants have red leaves and do appear to at least be absorbing more green than (certain wavelengths of) red. The simple answer is that by some mechanism green plants aren't as efficient at retaining green light. If this was an effect of saturation and clever use of energy potential (allow green light into canopy where it bounces until absorbed) one might expect to see less green in shade loving plants but that doesn't seem to be the case. Or perhaps plants are clever and we don't see the same benefit at sub solar light levels. One would think the evidence would show high K and low CRI providing enhanced yield if green light was on par with blue and red, yet yield results for these various spectrums don't seem to indicate that correlation. This could be due to other inefficiencies with low K high CRI. Hard to say for sure. A 550/680 -vs- 620/680 test might shed some light on the idea. One might be able to show that green light in full spectrum is more efficient than expected (Mcree) yet still not as effective as putting that energy into the 600-620 band, among other possible results.

There does seem to be some evidence that red -w- blue supplement is more effective than full spectrum, but I guess that doesn't conclusively prove anything. I just hate to guess unless it's in the name of science (someone else is paying for it).

 

[Old school grower]

Great info here. I believe that the spectrum should change throughout the entire grow. Not just veg and flower. Ex: adding 450nm last week or so of flower to increase terp or to slow stretch if desired.

 

[Rahz]

Plants are hugely inefficient anyway. Only a few percent of the light they receive ends up as an increase in solid matter. Somewhere between 2% and 8% depending on circumstances and who made the estimate)

Also, plants might have their reasons not to absorb certain wavelengths. like too "communicate" with animals around them. There is a reason flowers are so brightly colored. Perhaps there is a reason for them not taking up green light so readily has a purpose too?

I don't want to ignore that potential but let's disregard these mysterious variables for a moment and just consider the individual photoreceptor peaks. If we examine for instance the efficacy of 675 -vs- 650 in activating chlorophyll A the difference is huge. Unfortunately we can't just pick a frequency and order our LEDs in that frequency (yet).

But for instance, two different photoreceptors show peaks at 570nm and 625nm respectively. Their performance curves intersect at around 595. Putting X radiance at 595 would stimulate both photoreceptors, but if we put .5X at 570 and .5X at 625 should the results be better? Going by receptor response alone maybe they should be. If the effect is as pronounced as the chart in the OP seems to indicate the potential for such pinpointed spectrums could be huge.

 

[Old school grower]

If your interested my flower light is using 4-3000k 90Cri and 4-3500k 90Cri plus a ton of monos individually controlled by spectrum.

 

[Rahz]

If your interested my flower light is using 4-3000k 90Cri and 4-3500k 90Cri plus a ton of monos individually controlled by spectrum.

What kind of monos are you adding in?

 

[Old school grower]

4-Xpe per star 10w 12v. Here is a pic 385nm.
I have 385, 450, 620, 630, 660 and 730. Many of each. Lol

 

[Old school grower]

What kind of monos are you adding in?

Using a couple 4ch sonoff pros great for dc up to 10A. You can set individual timers ex: 5 minutes on every hour during lights on or on for 40 minutes mid day. Wifi app check from anywhere.