Which wavelengths to choose

Hello! I am new here, but not new to growing... Been doin it outdoor for over 15 years now. Thinking of building myself an LED grow light, but would like some input as to which wavelengths to use.

From what I've researched, here's what I've gathered:

BLUE - 450nm
BLUE - 465-475nm
RED - 625nm
RED - 660nm
Infrared - 730-740nm

These are the most important wavelengths to hit.. But what ratios of each should I use?
Here's my best estimated guess:

Flowering
Red:Blue ratio 4:1
RED : 25% 625nm, 75% 660nm
BLUE : 50% 450nm 50% 465nm-475nm
Infrared : 730nm - 50W in a 1000W system

Veg
Red:Blue ratio 1:1
RED : 25% 625nm, 75% 660nm
BLUE : 50% 450nm 50% 465nm-475nm
Infrared : 730nm - 50W in a 1000W system

Does this sound about right?
I am most unsure about the flowering RED:BLUE ratio..

Forget about wavelengths, start reading about white COB leds. Cree or Bridgelux.

I agree. Some members use supplemental color leds but for the most part it is white COBs. I have both Bridgelux and Cree myself.
(1) Cree cxa3070 3k
(4) Cree cxb3070 3k
(4) Bridgelux vero 10 5k

Yes, I will be using a smaller percentage of cool white and warm white leds to add in some of the missing wavelengths needed for the less prominent photoreceptors... but to get the most growth for the amount of electricity used, I really believe it's got to be specific wavelength technology. Besides... where's the fun in making things so simplified?? Those cree white LEDs use up a lot of their energy producing green light (500- 600nm) which is almost useless to plants.. even if the plants can use it a small bit, they really can't use it efficiently at all.

And you know this because..

I think that theory has been busted by the impressive performance of HPS, which is mostly green, yellow and orange. Many LEDs growers have abandoned worrying about wavelength in terms of photosynthetic efficiency. Not saying that red/white/blue doesnt work, but straight white avoids color mixing issues, easier to lens/reflect, higher emitter efficiency, cheaper/PAR W and less labor to assemble.

One word...McCree

The ONE and ONLY scientifically accepted action spectrum in the world.

All about dat white light.

Can't get enough of those white CoB's baby.

@boneclone
One of the most popular COB's that really is the workhorse round here....the Cree CXA 3000k 80 CRI is not full of green actually...Thanks SDS!

boneclone said:

Yes, I will be using a smaller percentage of cool white and warm white leds to add in some of the missing wavelengths needed for the less prominent photoreceptors... but to get the most growth for the amount of electricity used, I really believe it's got to be specific wavelength technology. Besides... where's the fun in making things so simplified?? Those cree white LEDs use up a lot of their energy producing green light (500- 600nm) which is almost useless to plants.. even if the plants can use it a small bit, they really can't use it efficiently at all.

Click to expand...

Well, I posted the graph ^ courtesy of StarDustSailor and you can ascertain your own answer about the amount of green...

From what I understand, just a blip...at low levels of intensity, red is used first due to its photon intensity and then as power is increased blue comes on and then green....BUT green can increase while blue and red "peak"....green is not useless, one of the biggest myths in the entire gardening world with small very limited exceptions...

Like Supra said above, green or green-weighted can and will grow Big plants...aka the HPS


This one should be a starting point for any further discussion:
https://www.rollitup.org/t/photosynthesis-under-solid-state-light-setting-the-standards.833449/

And I really enjoy reading you guys argue...I mean discuss the (fact?) that green output in LED's helps with their penetration issues.

As in some of the green wavelengths bypass the red/blue saturated canopy and feed the lower leaves a little bit

Abiqua said:

@boneclone
One of the most popular COB's that really is the workhorse round here....the Cree CXA 3000k 80 CRI is not full of green actually...Thanks SDS!
View attachment 3384058

Click to expand...

Interesting...I produced pretty different numbers for 3000K 80CRI CXA's.
First column is distribution of power, second one is distribution of photons.

It's simply integral of respective region divided by the integral of the whole curve.
@stardustsailor Any idea where the discrepancy came from?

Greengenes707 said:

One word...McCree

The ONE and ONLY scientifically accepted action spectrum in the world.

Click to expand...

You mean social acceptable? Just kidding. Inside joke among friends.

I don't know what is an action spectrum...or how that got accepted in Science as a term.

The idea he is speaking to and the only reason I am here, building a light is these new Chip on Board, integrated arrays with 100s of tiny ROYAL BLUE emitters.

They are embedded in a phosphor layer that downshifts and spreads the light into full spectrum, white light. But, you can, choose different phosphors to give different color temperatures just like CFL but, 10 times more efficiently.

So, it is rather passe, (all not actual science is passe) to think that discrete frequencies, 4 in your case can grow good pot. No. The plant uses every single frequency within 290-710 nm for photosynthesis. That is science. Some frequencies are more important at different times. The total Sunlight color temp is somewhat cooler in spring.

Therefore, no one is recommending a bunch of small LEDs at discrete frequencies anymore. Sure they still sell them and in Romania they still sell Snake Oil.

The point is, if you get 3500K white COBs and run them at 1/3 power, you will get the most efficient production of light we now know of and at the most efficacious spectrum spread for Veg and Bloom.

However, one could use 3000K for Bloom and 5000K for Veg.

There are several important COB build threads right now. Plus even I have one.

I screwed it up at first, didn't know what I didn't know, so here is my advice. Read everything in the last year by SupraSPL, alesh, and StardusterS. Also bict and Abiqua are helpful.

Some of these guys are worse than useless. Pick your poison.

Doer said:

You mean social acceptable? Just kidding. Inside joke among friends.

I don't know what is an action spectrum...or how that got accepted in Science as a term.

The idea he is speaking to and the only reason I am here, building a light is these new Chip on Board, integrated arrays with 100s of tiny ROYAL BLUE emitters.

They are embedded in a phosphor layer that downshifts and spreads the light into full spectrum, white light. But, you can, choose different phosphors to give different color temperatures just like CFL but, 10 times more efficiently.

So, it is rather passe, (all not actual science is passe) to think that discrete frequencies, 4 in your case can grow good pot. No. The plant uses every single frequency within 290-710 nm for photosynthesis. That is science. Some frequencies are more important at different times. The total Sunlight color temp is somewhat cooler in spring.

Therefore, no one is recommending a bunch of small LEDs at discrete frequencies anymore. Sure they still sell them and in Romania they still sell Snake Oil.

The point is, if you get 3500K white COBs and run them at 1/3 power, you will get the most efficient production of light we now know of and at the most efficacious spectrum spread for Veg and Bloom.

However, one could use 3000K for Bloom and 5000K for Veg.

There are several important COB build threads right now. Plus even I have one.

I screwed it up at first, didn't know what I didn't know, so here is my advice. Read everything in the last year by SupraSPL, alesh, and StardusterS.

Click to expand...

Stop quoting me in things you don't understand.
You haven't done shit. You don't have a single led grow under your belt. You don't even have cobs I hand yet. Your a tool that is jumping the gun.

Shut the fuck up...build your light...use it...then start preaching.

ReeferDance said:

And I really enjoy reading you guys argue...I mean discuss the (fact?) that green output in LED's helps with their penetration issues.

As in some of the green wavelengths bypass the red/blue saturated canopy and feed the lower leaves a little bit

Click to expand...

Well, again penetration is a ganja myth to me.

The leaves are semi-transparent to PAR frequencies.

Think of a leaf as a stylized flat sheet of glass. And across that span of glass floor are millions of tiny factories. The photons that hit a factory are used by that factory. This is all submicroscopic, in fact, right? Photons are tiny, and factories are big clump of chlorophyll molecules..

The rest of the photons pass thru the glass floor with a loss of say 20% And it is also scattered by the floor and spread. Next the photon stream -20% of it, falls on a plate of glass (leave model) that is 20% BIGGER. The leave after that is BIGGER. No one get this obvious fact for some reason. The leaves are bigger to function in the shade of the leaves above when necessary.

The photons that miss factories above will hit factories on the larger leaf below, and so on.

It is just not correct this idea that you need "penetration."

Greengenes707 said:

Stop quoting me in things you don't understand.
You haven't done shit. You don't have a single led grow under your belt. You don't even have cobs I hand yet. Your a tool that is jumping the gun.

Shut the fuck up...build your light...use it...then start preaching.

Click to expand...

No one is preaching but you. Commercial guy. So competitive with your nothingness.

And you spoke the usual not helpful that I continue to correct until you get real, bonky.

I will shut you up, before you succeed with me as you know quite well by now.

Don't invoke science, when selling snake oil. You are lacking.

Doer said:

Well, again penetration is a ganja myth to me.

The leaves are semi-transparent to PAR frequencies.

Think of a leaf as a stylized flat sheet of glass. And across that span of glass floor are millions of tiny factories. The photons that hit a factory are used by that factory. This is all submicroscopic, in fact, right? Photons are tiny, and factories are big clump of chlorophyll molecules..

The rest of the photos pass thru the glass floor with a loss of say 20% And it is also scattered by the floor and spread. Next the photon stream -20% of it, fall on a plate of glass (leave model) that is 20% BIGGER.

The photons that miss factories above will hit factories on the larger leaf below, and so on.

It is just not correct this idea that you need "penetration."

Click to expand...

Why again is the presumption of needing penetration incorrect?

AquariusPanta said:

Why again is the presumption of needing penetration incorrect?

Click to expand...

I just said. In terms of frequency it is not a correct terminology.

This is all about the number of photons and how well packed. Bigger watts pack more photons per area. Even still some will be used by a leaf and most will pass.

The very idea that green penetrates is wrong. Green is reflected. It is never used.

Doer said:

I just said. In terms of frequency it is not a correct terminology.

This is all about the number of photons and how well packed. Bigger watts pack more photons per area. Even still some will be used by a leaf and most will pass.

The very idea that green penetrates is wrong. Green is reflected. It is never used.

Click to expand...

Are you stating that green "frequencies" are useless for plant growth, by those "frequencies" being reflected and not absorbed?

There have been lots of study on this. Ganja is a very run of the mill type plant. It is extremely hardy and does very well without humans. All these God Myths of Ganga needs are just not true. This is not an orchid and is easy to grow.

This is not my first rodeo. I have grown for a couple of years with 1000w HPS and CFL for Veg.

The best study and really one of the first after they invented quantum PAR meters was to investigate strawberries. Talk about a dense canopy. And there was the usual myth debate about breeding strawberries with smaller leaves, "to let the light in."

Wrong assumption. They put the PAR sensor under there and were shocked.
There is still plenty of light even though strawberry can have 3-4leaves stacked one above the other. This is how we know that leaves do not block sunlight. They absorb some and pass the rest downward.