Color LED VS White LED - Efficiency

[rentner]

A simple but difficult question:

Why are colored LEDs so inefficient? For example a blue 10W LED 450nm has 24lm/W, while a white LED 10W Modul easily archives efficiencies of over 80lm/W


And does anybody know the different effects of the LEDs to Plants? I read, you can use a combination of 450nm and 660nm LEDs to grow anything from Salad to huge Tomato plants. Many people use HID for "illegal" purpouse. But isn't that a waste of energy? I mean, is a plant to absorb all that light from HIDs?

"Professional" growlamps made of LEDs sometimes only use 450nm Blue and 660nm Red LEDs and growers claim, that they are efficient. But if I calculate that in, they have much less lumen per watt as white LEDs.

BUT: Aren't white LEDs just a combination of a few colors and it only seems as it is white? I mean, White always is a combination of different colors of light. How does that all work? It is a bit confusing and I cannot realy find detailed information.

What is the max Lumen per Watt ratio at 100% efficiency? Does this vary from color to color since shorter wavelengths have more energy? Or Am I completely getting this wrong?

Thanks for thousands of answers.

[mazurov]

Lumens are a measure of light emitted as seen by human eye. Since said eye is most sensitive to green, green LEDs are very efficient when efficiency is presented in lm/W. Take a look at Rebel ES Color datasheet, for example - they give 154-201 lm/W for LXML-PX02-0000 at 350mA.

[Zero999]

LED grow lights use red and blue LEDs because photosynthesis doesn't use green light, which is why plants are green: the green light is reflected because it isn't used.

Lumens per Watt is the incorrect unit for measuring the intensity of light, used for other purposes than illumination. In this case the gradient power,  over the bandwidth you're using is a more appropriate measure, since any emission in the yellow/green spectrum is wasted because the plants don't use it.

[rentner]

Ok, that seems logical! So this is, why Red and Blue is "inefficient" if measured in Lumens.


Well, how do I measure the "real" efficiency of an LED? And is there a mathematical way to compare HID lamps, regular incandescent lamps, Flouresent lamps and LEDs for use in photosynthesis? You sayd, a Plant can not use Green, because green surfaces reflect green light. Seems logical. Well, what if I would use orange for example instead of red and blue? It should not be reflected, of course. Can a plant use that? Well, sorry, to ask something that is more related to biology than to electronics. But I saw a "professional" grow Lamp using about 10% orange LEDs. Does that make sense?

Let's think in wavelengths: Green is a bit closer to blue than to red - would it make sense to just use red light? Because theoretically it gets reflected a bit less.

Well, also, I now understand, why some growers use green light, to work on a plant in its night cycle, because this does not interfere that much.



It is not that much about electronics, but very interesting. That would automatically mean, that LED grow lights with the same efficiency as HID lamps are still more efficient for plants. That makes clear, why LEDs are used more often recently. Also the costs are less, because a HID dies much faster and often is not that efficient.


Am I wrong with this?

[mazurov]

There is a thing called PAR ( http://en.wikipedia.org/wiki/Photosynthetically_active_radiation ), you may want to take a look. There is also a patent describing a lamp made of 660nm, 612nm, and non-critical blue emiitters claiming to make a big difference.  I built such a lamp and it works well. NASA also did some studies about growing plants under color LEDs.

However, emitting strictly within PAR could be expensive. Color LEDs are more efficient than white since they emit directly without using a phosphor, but they are also more expensive and generally not made with power rating higher than 3W. You will need ~150W-200W  (electrical) per sq.m for good growth; the lamp I mentioned was built from Rebel ES emitters and the cost alone was higher than commercially available LED lamps of comparable power rating (made from random Chinese emitters of unknown wavelength and efficiency). You will also spend more time putting the lamp together since you need more parts to mount. If you want this lamp to last you need to design it to run cold which implies running emitters at 1/2 power bolted to a pretty large heatsink or use 100K hr rated fans if you decide to go with forced air.

The "follow the PAR" theory can also be easily countered. For example, professionals still prefer using HID and HPS lamps, which emit most of its energy outside of PAR (especially HPS which is mostly yellow). If you ever come to the jungle, note that a) you won't be able to see the sky standing on the ground therefore most of the light on this level is green since the sunlight is filtered through leaves of higher plants and b) despite that, the ground level plant growth is quite vigorous.


I experiment with LED grow lamps for approx.3 years using high-power Cree modules, starting with CXA2011. I use white LEDs with two color temperatures - 6500K ones because they are more efficient, and 3500K ones because they emit closer to PAR.  I still can't tell which color temp. is better, after several harvests. I can only see that they work better than fluorescent (which has similar efficiency if you don't overdrive them).

[Rick Law]

LED grow lights use red and blue LEDs because photosynthesis doesn't use green light, which is why plants are green: the green light is reflected because it isn't used.

Lumens per Watt is the incorrect unit for measuring the intensity of light, used for other purposes than illumination. In this case the gradient power,  over the bandwidth you're using is a more appropriate measure, since any emission in the yellow/green spectrum is wasted because the plants don't use it.

Just to help complete this...Some plants however have secondary pigments - a second stage photosynthesis to use more of the unused spectrum...

[Rick Law]

Ok, that seems logical! So this is, why Red and Blue is "inefficient" if measured in Lumens.


Well, how do I measure the "real" efficiency of an LED? And is there a mathematical way to compare HID lamps, regular incandescent lamps, Flouresent lamps and LEDs for use in photosynthesis? ....

You must decide "Efficiency in doing what?"  If it is providing lighting for people, then it should measure what is most helpful to humans.  However, humans differ - some will not see some colors as well as other.  So the measurement is entirely subjective.

If one is merely to measure how much electrical energy is converted to light, then it would be meaningless.  For example, it could a 100% efficient in converting say to UV, but it wont help us see anything at all.  To typical human eyes, the 100% efficient light is giving zero light.

[rentner]

You will need ~150W-200W  (electrical) per sq.m for good growth;

Wait, what ammount of power would you need with HPS or MHL, to compare this? This would give us a resulting an aproximal "efficiency" for plants.

I experiment with LED grow lamps for approx.3 years using high-power Cree modules, starting with CXA2011. I use white LEDs with two color temperatures - 6500K ones because they are more efficient, and 3500K ones because they emit closer to PAR.  I still can't tell which color temp. is better, after several harvests. I can only see that they work better than fluorescent (which has similar efficiency if you don't overdrive them).

That is very interesting. Well, it is cheap to get such Power modules for a few bucks.

(5 bucks a 100W modul, 2 bucks a colored 3W modul)

Not overdrive a fluorescent? So let's say for a 100W FL a 60W ballast?

Ok, that seems logical! So this is, why Red and Blue is "inefficient" if measured in Lumens.


Well, how do I measure the "real" efficiency of an LED? And is there a mathematical way to compare HID lamps, regular incandescent lamps, Flouresent lamps and LEDs for use in photosynthesis? ....

You must decide "Efficiency in doing what?"  If it is providing lighting for people, then it should measure what is most helpful to humans.  However, humans differ - some will not see some colors as well as other.  So the measurement is entirely subjective.

If one is merely to measure how much electrical energy is converted to light, then it would be meaningless.  For example, it could a 100% efficient in converting say to UV, but it wont help us see anything at all.  To typical human eyes, the 100% efficient light is giving zero light.

Hm... I mean the efficiency of the emitted wavelengths, that can be used best in a plant.
But I don't mean the efficiency of the photosynthesis process itselfe.

So let's say, I use a 100W 6500k LED Module.

That means, how much energy is emitted in a way, so a green leaf can use it. I mean, Plants can use a tiny ammount of the green light, but it's just a few percents. It is a complex graph which makes it difficult to understand the efficiency.

[Rick Law]

Ok, that seems logical! So this is, why Red and Blue is "inefficient" if measured in Lumens.


Well, how do I measure the "real" efficiency of an LED? And is there a mathematical way to compare HID lamps, regular incandescent lamps, Flouresent lamps and LEDs for use in photosynthesis? ....

You must decide "Efficiency in doing what?"  If it is providing lighting for people, then it should measure what is most helpful to humans.  However, humans differ - some will not see some colors as well as other.  So the measurement is entirely subjective.

If one is merely to measure how much electrical energy is converted to light, then it would be meaningless.  For example, it could a 100% efficient in converting say to UV, but it wont help us see anything at all.  To typical human eyes, the 100% efficient light is giving zero light.

Hm... I mean the efficiency of the emitted wavelengths, that can be used best in a plant.
But I don't mean the efficiency of the photosynthesis process itselfe.

So let's say, I use a 100W 6500k LED Module.

That means, how much energy is emitted in a way, so a green leaf can use it. I mean, Plants can use a tiny ammount of the green light, but it's just a few percents. It is a complex graph which makes it difficult to understand the efficiency.

Easy to misread and the mind fills in what it thought it saw.  Read again and you will see it does not use green.  The unused light is reflected rather than used -  thus the colors you see are the colors it doesn't use.

It typically absorbs more than a few percent.  Those plants with 2 pigment cycles achieve a lot higher and well into the double digit percentage.  Cycles here is not in reference to which one goes first and which one goes second.  They occur at the same time but in different parts along the electron transport chain.

Plants differ.  Some plants absorb some spectrum well while other plants do not.  While photosynthesis occurs in all plants, but they don't all work the same way.  For example, 3C plants, 4C plants, and CAM plants (such as cactus) handles photosynthesis very differently.  All plants are evolved to best fit the environment.  CAM plants evolved special handling for dry weather such as closing the stomata for "part 1" to avoid loosing water, and do "part 2" later in the dark while it must open the stomata to take in carbon dioxide.

Those 3C/4C are Calvin cycle stuff belonging to the light-independent reactions (aka dark reactions) but they are clearest as an examples of how plants differ.  What you are interested in are the light-dependent reactions part where light energy is used to create ATP and NADPH.

To be most efficient (and not killing your plants in the process), you must research for your "plants of interest" what spectrum (pigments) do they have and in what percentage.  Also remember, there could be two absorption cycles where ATP and NADPH are created: the primary pigments and the secondary pigments each doing part of the work. 

Look up light-dependent reactions and pigments to understand everything before the C3/Calvin cycle, then you'll know what to search and what to focus on.  (The bold italic words are probably of interest...)

If that gets too complicated (and it is), just use white.    Let the plant takes what it takes and reject (reflect) the others.  Those light energy eventually converts to heat warming things up for the plants themselves.  If you don't mind testing (to see if your first crop goes south), use blue light.  Rare to see blue on leaf so that one probably is best "general go to one".  But do so at your own risk (don't blame me if your plant dies).

Good luck...

Rick

[mazurov]

All this is quite complex and there are many important factors other than light spectrum and intensity, as already stated by @Rick Law.  Besides, even the "efficiency" as a measure of dollars spent to produce certain amount of light usable for plant growth needs clarification. What is more important to you - capital or operational expenses and what is your maximum for each of those? It could very well be that you won't be able to afford the light source you're attempting to design; otherwise make 50W strings of every color available in Rebel ES line, mix them in the lamp  and grow your plants under this making observations and adjusting output of each string until happy. You'll get your ideal light in 3-5 years and if you run it cold it will last forever. Temperature directly affects efficiency expressed as a unit of light emitted per unit of current (cooler is better) therefore you need to factor in the cooling expense as well (it could be feasible to water-cool your LEDs when growing wasabi, for example, since you'll need plenty of cold water anyway).

Note that the efficiency of the emitter also goes down as current through it goes up. Max. and min. specified output are rarely practical, you'll still need to make sacrifices.

To Rick Law: blue makes plants grow taller, which is often undesirable for indoor grows.  You'll need red to counter that.

[Rick Law]

...
To Rick Law: blue makes plants grow taller, which is often undesirable for indoor grows.  You'll need red to counter that.

Ah... Interesting!  I didn't know that blue would make them grow taller!  Thanks!

[mazurov]

...
To Rick Law: blue makes plants grow taller, which is often undesirable for indoor grows.  You'll need red to counter that.

Ah... Interesting!  I didn't know that blue would make them grow taller!  Thanks!

I checked my sources once again and it's actually other way around - red makes plants leggy, not blue. Sorry about that. Strange that I can't see this under my lamps - maybe you need pure red to see the effect.

For all interested - this is a good paper (and a list of references) on a subject -> http://gravitationalandspacebiology.org/index.php/journal/article/download/2/2?origin=publication_detail

[rentner]

And what about UV-B light? I read, that this can damage plants, which seems logical.

But it also helps growing/flowering plants, which build up resins. Maybe the resins are for UV Protection.



Many people think, that the THC containing balls on the flowers protect from UV and it is shown, that they get larger with a small ammount of UV. This could possibly work with many plants. But what is to much?


I propably want to make an experiment with a box with many small boxes inside. Each small box with different light combinations. I want to test, which box has the best growth, most flowers and best fruits whith the different light combination. This will show, what realy is efficient.

[Rick Law]

And what about UV-B light? I read, that this can damage plants, which seems logical.

But it also helps growing/flowering plants, which build up resins. Maybe the resins are for UV Protection.



Many people think, that the THC containing balls on the flowers protect from UV and it is shown, that they get larger with a small ammount of UV. This could possibly work with many plants. But what is to much?


I propably want to make an experiment with a box with many small boxes inside. Each small box with different light combinations. I want to test, which box has the best growth, most flowers and best fruits whith the different light combination. This will show, what realy is efficient.

No doubt some plants will need protection against UV and some plants may want more UV.  There is a huge variation in plants.  You have to be plant specific when you get to that level.  Otherwise, it is like "find the best food for a 4 leg animal" - you wont find one that suits all.

I think you may be barking at the wrong tree.  Photosynthesis is controlled by (1) Amount of light (2) Temperature (3) Water (4) Carbon dioxide.  You may get more bang for the buck optimizing other factors first because they are easier to measure and to modify.

My person gut feel (no proof, just shall we say instinct) is not wave length but duration.  As all plants must deal with clouds, almost surely, their evolution equipped them with deal with less-then-full sun.  Perhaps at little as 20% of day-light-hours could be adequate if given good airflow (plenty of fresh CO2 and excess O2 carried away), good water flow, etc.

Why 20%?  I read from more than one source that typical human uses only about 20% of their lung capacity, so that seem a good number to use.  I think somewhere between 20%-30% is the magic number.

"Amount of exposure" can be controlled by a simple mechanical on/off switch, and/or controlling temperature which can be controlled by any thermostat, or air flow (which is simple fan connected to the light's timer)...

Plants in someway are far more complex compared to animals.  They have 4 to 5 times more genes than animals because they must have ways to deal with everything whereas most animals can just run away.  The "extra" genes allow them to code in more behavioral modifications so they can coup with changes in their environment.  Birds can fly north while the tree stood there naked lamenting if only I can move.

Rick