Discussion of the pragmatic application of LED's

[guest456mpy]

Indeed!

"Shards of pure color
Grew large buds with magic rays
Tonight, we smoke bliss." -W. Zard





H.G.

 

[joe4444]

Some soil growers that passed to LEDs has changed to coco coir after a pair of grows. Soil is way more sensitive to keep the right moisture in it, and reduced transpiration and water uptake of LED grows makes more difficult to achieve the right dry/wet cycle. Hydro mediums are way more tolerant to it (I keep coco permanently wet, except for seedling and clones just rooted) thus the average grower get better results on hydro.

But I feel that its possible to do it well on soil too, but soil grower needs to find the right way of achieving it.

Is the problem with soil under LED that it stays too wet because the LED light does not evaporate water directly from the soil the way HID lights do? If that is the case, could you simply water a little less? You could adjust nutrients by, for example, increasing the concentration 25% and then water 20% less to achieve the same overall dose. I have a feeling it's not quite so simple.

Im only afraid that the increased water use efficiency with CO2 enrichment coupled with the reduced water uptake due lower overall temps and lack of IR result on the need of use too high ECs and possible deff of calcium (I would say that nutrient solutions with enhanced Ca content is going to be required on this scenario).

That sounds like hydro speak. Is the calcium deficiency a direct result of LED vs HID or an indirect result of adjusting ECs? In other words, is calcium deficiency something a soil grower needs to consider when using LED?

 

[joe4444]

Weezard, your UV light affecting trike maturity theory seems very sound...and promising! I'm getting even more excited about making my next grow all LED. You, knna, and Hempyguy are true scientists.

 

[joe4444]

At the risk of thoroughly dumbing down the general conversation here I have to ask...instead of arranging 20-50 LEDs in a single compact light, why not make multiple smaller lights with five to ten 5W LEDs each? That would allow you to spread them out (perhaps one per plant in a SOG scenario), which should also allow heatsinks to work more efficiently, right? You could also keep the lights closer to the canopy because there is less need to raise the light to increase the overall footprint. They don't necessarily have to be separate lights, but they could be more spread out inside a larger housing.

 

[SOTF420]

Joe, they do have those they are screw-in LED array light bulbs they have some new ones from Lumigrow you should check them out on their website. Several other companies make them as well for supplemental lighting.

Weezard, pure gold you posted my brother! Fantastic thread so far very informative.

 

[guest456mpy]

Joe,
If you are doing SOG, SCOG, LST,or other short height grow methods the 5 watt emitters are fine. Don't expect to grow taller than 3 ft plants without side lighting. Keep this in mind and you are golden. The way lights are packaged is a matter of convention or imagination, not driven by technology. Keep in mind that the LED's that we have used in our builds are getting "long in the tooth" and will be soon replaced by newer, much more powerful ones. The 5 and 15 watt emitters will soon be replaced by ones which will easily allow traditional "trees" instead of having to resort to "short height grow" methods. China is already installing LED street light using white 50 watt emitters so this is a glimpse into what is in store for us. It's not much of a jump of logic to expect individual lamps of this power range that are cheap and easy to use exactly what you propose in a couple years.

 

[joe4444]

So in a way it is analogous to HID? Growing a 5ft plant with a 150W HPS might yield some nice buds, but there won't be many in the first 4 feet...only at the very top, whereas with a 400W+ HPS you can get great buds from top to bottom. Is this why some people have success with a certain unnamed LED light utilizing 1W emitters because they are micro grows and/or training their plants to stay short? (I know, there are other problems with those lights.) It seems like you guys are quickly approaching the ideal spectrum and now your focus is shifting towards delivering that spectrum with more power to achieve penetration comparable to a 1kW HID. Or am I lost?

 

[guest456mpy]

Mine has adjustable color spectrum, since individual strains like different ratio.
You are, generally speaking, correct in your statement about power and penetration.

I hope "that certain company" is reading this thread. I would not mind if they took the info from here and made their next generation of lamps a viable alternative to the 1kw HID. I think we all just want the best to grow with to further the mutual cause. That's what the open exchange of truthful information does for us all. It's sets us all free.



H.G.

 

[guest456mpy]

I would like to wait for knna to weigh in on this first point before we wrap it up, postulate our first rule for LED application and move on to the next. Remember there are no absolute right or wrong answers here, just differing points of view which will allow us to make some sense of the seemingly different approaches.

Until then, peace and productivity to all!

H.G.

 

[guest456mpy]

Since knna has not posted I'm just gonna wrap up this point and bring it home.

I put this to the group for discussion:


Rule number 1
When choosing which style of LED to use or buy, the style of grow or strain should be considered. Grows that will up to 3 feet tall are fine with lower power emitters, taller grows need more watt per emitters even with narrowed beam patterns using focusing lenses.

H.G.

 

[SOTF420]

That is Rule #2.

Rule #1 is "Do not tell anyone whatsoever about your LED grow, even your mom."

Don't forget Rule #1!

 

[guest456mpy]

Originally posted by SOTF420

Rule #1 is "Do not tell anyone whatsoever about your LED grow, even your mom."

LOL, absolutely. But quoted more properly

Rule 1: No tell
Rule 2: No sell
Rule 3: No smell

I thought this particular discussion was how to apply LEDs to grow.

Therefore in the context of this thread:

Rule number 1
When choosing which style of LED to use or buy, the style of grow or strain should be considered. Grows that will up to 3 feet tall are fine with lower power emitters, taller grows need more watt per emitters even with narrowed beam patterns using focusing lenses.

H.G.

 

[guest456mpy]

I might add that any rules we mutually come of with are not numbered by order of importance. I'm just trying to foster the formation of an ad hoc committee with the purpose of creating:

1) A discussion of LED working practices
2) A working guide for neophytes to use in selection of equipment
3) LED specific grow enhancements

Maybe more.

I'm also trying to steer things away from "hot button" topics.

SOTF420, I know your an experienced grower so I personally hold your opinion in high regard.

H.G.

 

[knna]

First, I would like to point up that although I mainly work with wide angle beam LEDs, I respect designs using narrow angles. Both approachs have their own advantages.

Light distribution is a very complex topic, and somewhat difficult to explain. I would like to post some graphs of simulations to show the differences between the light densities obtained by same amount of light (micromols of photons per second, lm, PAR W, whatever) at given distances of the lamp. I believe that itsthe only way to show some things. Unfortunatelly, right now I dont have the time to do it, but hope some day I'll can (maybe I will edit the post to add them, so people can see on actual examples what I mean).

So first, I would like to dispell some myths about the inverse square law, that perpetuates the misunderstanding of how light is distributed on a grow area.

Reason of the inverse square law is geometrical. Its not related to any magic caracteristic of light. Light density is amount of light (from now on, i will use the keyboard friendly abreviation of micromols of photons per second, uE as unit of light; if you understand it better putting PAR Watts or lm instead, no problem, just exchange it in your mind) per unit area. Typically, uE/m2 (micromols of photons per second per square meter).

Inverse square law is related to the surface area of a sphere increasing at a rate of the square of the radius, as shown in the graph posted some posts ago. So its only valid to a single point of light spreading on all directions. That is something that only happen when using a unreflectorized HID on vertical setups. On the other hand, LEDs are directional, emiting all the light on a beam angle, as max of half a sphere and usually way smaller, and not with same intensity on all directions. Thus, inverse square law dont apply here.

To see it, imagine a LED with a lens that emits the light on a pattern that not spreads, for example, a square of 1" (gun of light style). It emits 10uE. At 1" of the LED, it lights a square 1" side, so 1sq inc. Thus light density is 10uE/sq inch. At 10", surface being lit continues being 1 sq inch, so the density is too 10 uE/sq inch. In this hipothetical case, light density is not affected at all by distance.

So if light density drops as distance increase, its just because the area being lit is larger. This is obvious, but little people is aware of it. The light density drop is directly correlated to the increase on the area to being lit.

For a single point of light, where all the light comes from an small point and spread to all the area lit by the lamp, light density is very high close to the lamp, because the footprint (area lit) is small. As distance increases, light density drops because footprint increases. But at the end, what we want is to lit enterely a given grow area, thus it is required some distance from plants to the lamp so light fill all the area. The light density we should look at when using a single point light source is that achieved for a footprint equal to the size of the grow, and that is achieved at a given distance of the lamp. In average, its uE emitted/m2 of grow area. (actually, there is large differences between the center and the edges, but I will enter on this later).

Say its 800uE/m2. If you measure the light density below that distance at which the lamp lits all the grow area, you will realize that it will vary very little whatever the distance. It will keep an average very close to those 800uE/m2. Simply because its the same amount of photons spreaded on the same surface area. What you will measure then is that average density varies little (only drops because the increased light loss due more light hits the walls), but as farer from the lamp, distribution in evenly (difference between max and min local light density is lower).

So that at the bottom areas of a grow light density is lower than on the upper parts is caused just because plants absorb light hitting them, thus only some light reaches the bottom areas, less as larger is the distance to the lamp and smaller the holes in the canopy. (Im talking of average values here. Analysis of how the way of deliver light (single point/multipoint or narrow/wide beam) affect locally, later).

One of the main factors that may affect all this is the reflectivity of walls. If walls were 100% reflective, from a given distance will not happen any further reduction on light density. As walls absorb part of the light that hit them, there is some loss, larger as less reflective are walls.

When we translate all this to LED lighting, many things change, but not one: at the distance at which the footprint of the lamp equals to the grow area, there wont be further reduction on light density (or only minimally, related to light loss at the walls). When using wide angle beams and LEDs distributed along all the grow area, the distance at which this occurs is very small, as small as few inches, less than 10cm. When using a crowded LED panel in the center and narrow angles, it may require 15" (or more, all depends of the shape and size of the grow area), but eventually is the same result. The type of refectivity affects too: its not the same to use aluminized mylar, with specular reflection, than flat white paint with near lambertian, diffuse, one (I will enter deeper on it on next posts).

The main difference between LEDs and HIDs in this regard, and that makes them more similar to the lighting obtained by fluorescents, is that lighting achieved is way more evenly, meaning that there is small differences between the areas with max and min light density. Local light densities are closer to the average light density in LED grows.

This is a great advantage of LED lighting for growing, often overlooked. Over given light densities, which almost always we surpass by large on our bloom areas, plants goes reducing its efficacy using the light as the light density increases.

Thus for two grows with an average light density of 800uE/m2, one very uneven (HID) and one very even (LED), plants grows more for the even distribution. Because on first case one leaf is exposed to 200uE/m2, which is below the capacity of the leaf of process light optimally, and other at 1400uE/m2, at which efficacy processing light can be half of that at 800uE/m2. While the other setup with 600uE on one leaf and 1000 on other have higher efficacy, because less light is used below its potential. This effect is especially important if we manage to grow the plants using irradiances (local light densities) not over the point that plant process optimally (300-400uE/m2 for our loved plant).

So in order to finish this post, one of the main advantages of LED lighting is the possibility of produce an even lighting.

In a grow space without plants, light density still at large distance of the lamp is close to that obtained at 10". When we consider what happen with plants in it, we must forget inverse square law as it dont explain at all why light density is lower on the bottom areas. Light being absorbed and reflected by plants explain it.

So the other myth is that of Penetration. Next post I will concentrate on it, as the differences between narrow and wide beam LEDs and how they are distributed affect to this, but very little the average light density.

 

[Weezard]

Mahalo Knna

Mahalo Knna

Whenever I read your posts, I always feel like I learned something.

Aloha,
Weeze

 

[guest456mpy]

Excellent post knna.
Rule number 1 is, of course, open for modification pending your input. We want to suggest factual information to new users, not the most popular post. I have read volumes of your posts on the internet wherever I have found them. Every time I read one of your posts I see things in a new and different light, no pun intended.

Thank you so much for your continued involvement on the behalf of all of the cannabis grower in the future.

H.G.

 

[SOTF420]

knna, some excellent points and very well put into words as usual

 

[knna]

Thanks, guys. I hope we can work improving our setups by eliminating the wrong concepts that lead us to work on wrong directions.

In this sense, lets examine Penetration (BTW, if you want to find thousands of excelent botanists articles about this topic, search for "extinction coefficient k" (add "canopy" to eliminate results of other areas apart of botany).

The main thing I would like to dispell about Penetration is that the light source caracteristics, although affect on some degree, dont rule out it, but is one of the factors that affect less penetration. At least, if we consider just top lighting.

Penetration is mainly ruled out by how crowded is the grow area (how many leaves) and the angle of leaves. Light density achieved at bottom areas mainly depends of them, and just minimally of the caracteristics of the lighting, especially if its a fixed one.

Some botanist concepts are usefull in this context. Leaf Area Index (LAI) is an adimensional figure that informs of how many leaves there are for surface unit area. It sums the area of upper part of all leaves and divide it by the surface area of the grow. Thus a LAI=1 means that on a 1 sq meter of grow area, there is 1 sq meter of leaves (counting only upper part).

In general LAI is over 1, mostly between 2-4, but it may reach 6. Notice that growing techniques often are intended to strongly affect this parameter, thus the impact of SOG, SCROG, LST, defoliating, etc on "penetration".

Extinction coefficient (k) describes the attenuation of light density with vertical height. It depends strongly on the orientation of leaves. Planophile plants (most leaves are horizontal) have way lower K than erectophile plants (most leaves vertical). Especially when incident light comes from the zenit (the vertical).

The higher the LAI and k, the less light that reach the bottom areas.

In general, when using indoor just top lighting, it dont worth to have a LAI over 3. Usually a little over 2 is the max optimally useable indoors.

An ideal lighting must be designed according to LAI and k of plants being grown. And given the lighting setup is done, we must adapt LAI and k to the best values to get the best of the grow. Tyeing and pruning are great tools that allows us to strongly affect those parameters. Of course, LAI itself is very affected by how large is the plant, thus choosing the right vegetative time is critical to obtain the best results, as any mynimally experienced grower knows.

Therefore, its very difficult to generalize. What is good for 5ft plants probably isnt for 1ft ones. Different strains have different leaf angles, and more yet, plants have some ability to adapt it to the lighting environment. At the end, growers sharing their experience on their own conditions is the only key to improve on the long term.

There is no rules valid for all situations in this sense. Possible combinations of LAI and K are too many to fit a solution optimal for all. Aditionally, we can strongly affect them using growing techniques. Recently many people has discovered how defoliating plants may improve yields.

So we should forget to find an universal valid solution, but concentrate on understanding factors that affect on each situation and try to get the best for that concrete situation, either by manipulating the light setup, either by using general growing techniques.

But the main thing we must have in mind is that is not lighting that rules penetration, but plant's caracteristics.

Only understanding the given plant caracteristics we can design a lighting that works better for them. And on a opposite way, we can train our plants to use better a given lighting setup. Improvements on LED growing involve those two factors. Lighting and plants pattern must be considered together.

 

[guest456mpy]

knna,
This is indeed a great post. It contains many facts, but I believe that it also makes some assumptions.

First of all by penetration I think that you are considering a continuous canopy, which I will further refer to in this post as CP. CP's are the most efficient use of energy when possible. These are several cases where this may not be the best case solution. Let me name just a a few.

Large scale cultivation is one case where this may not be practical. SOG and SCROG are the most common means of obtaining a CP. These are relatively time and effort consuming and at some point in scale the gain in energy efficiency is offset by the loss of efficiency in labor.

The other case might be when multiple strains are being cultivated. The different rates of growth complicate the maintenance of CP. Multiple strains are cultivated in order to minimize the effects of resistance buildup (in human consumption), to provide for different cannibanoid profiles (for human consumption) and to provide variety.

Sativa strain plants have smaller leaves which, by their very nature, block off less light to parts of the plants below that particular node.

So in my estimation, these are the situations which can benefit by the higher beam concentration to enable a taller growing plant. In my humble opininion technology should not dictate the style of cultivation, by the nature of the strain and the style of the grow accommodated by the choice of supporting technology.

What do you think?

The only other point I wish to comment on is from the previous post:

Inverse square law is related to the surface area of a sphere increasing at a rate of the square of the radius, as shown in the graph posted some posts ago. So its only valid to a single point of light spreading on all directions. That is something that only happen when using a unreflectorized HID on vertical setups. On the other hand, LEDs are directional, emiting all the light on a beam angle, as max of half a sphere and usually way smaller, and not with same intensity on all directions. Thus, inverse square law dont apply here

.

LED light is not actually directional, it only appears to be as an artifact of the way it is emitted. Light is emanated from the face of the device so it is a quasi-collimated light source. The light is not coherent like a laser, nor is it actually collimated, so it does spread out with distance. My empirical measurements with a meter have shown that the concentration or strength of the beam falls nicely with the inverse square law.

Discussion?

Again, I don't dispel what you say, merely bringing out some finer points for practical application. As a retired engineer specializing in X-ray and imaging design and implementation for a very large German engineering firm I have had a fair amount of both theoretical and practical applications. I very much enjoy this type of interchange with you as I seldom have an opportunity to converse with someone with the same background and education.

None the less, we must bring these things down to a language and context where "everyman" can understand and use the fruits of our discussion.

Once again, thanks for your timely and valuable input, please continue to do so, my friend.

H.G.

 

[knna]

knna,
This is indeed a great post. It contains many facts, but I believe that it also makes some assumptions.

First of all by penetration I think that you are considering a continuous canopy, which I will further refer to in this post as CP. CP's are the most efficient use of energy when possible. These are several cases where this may not be the best case solution. Let me name just a a few.

Large scale cultivation is one case where this may not be practical. SOG and SCROG are the most common means of obtaining a CP. These are relatively time and effort consuming and at some point in scale the gain in energy efficiency is offset by the loss of efficiency in labor.

The other case might be when multiple strains are being cultivated. The different rates of growth complicate the maintenance of CP. Multiple strains are cultivated in order to minimize the effects of resistance buildup (in human consumption), to provide for different cannibanoid profiles (for human consumption) and to provide variety.

Sativa strain plants have smaller leaves which, by their very nature, block off less light to parts of the plants below that particular node.

I think you understood bad my last post. I didnt enter of the topic of how to optimize things for high or low LAI, or high or low k. On my last post, I just wanted to let clear than most about penetration is ruled out for them, and not for light source caracteristics.

I was pretending to enter on next post in which situations are better lit for narrow beam angles and which not, and when using crowed LED panels or distributing LEDs along all the grow area is better, or when using side lighting is the way to go. Understanding that the answer depend on plant's caracteristics is the key here.

All them depends of the LAI and k of the grow area, and optimization can be worked out both by changing the lighting setup, train plants or both.

I didnt make any asumption about LAI yet, not about k. Given values of them may be considered to produce a "continuos canopy". But my reasoning was general, for practival values found on our grow areas, that can vary a lot for some setups to others.

I dint postulate yet any way to optimize things or recomended any technique. Each grower must decide for himself what is more confortable and the balance work/yield whatever the style of growing he decide to use.

All I said is the way of optimizing penetration must consider the plant's pattern, whatever it is, to desing the lighting that adapts better to it. And that penetration, as its not determined by lighting caracteristics but on a minor degree, may be improved by using growing techniques.

So in my estimation, these are the situations which can benefit by the higher beam concentration to enable a taller growing plant. In my humble opininion technology should not dictate the style of cultivation, by the nature of the strain and the style of the grow accommodated by the choice of supporting technology.

What do you think?

Absolutely agree, I think we must decide what technology to use to serve better out purposes, and not the opposite.

But as we have different technological solutions, we must understand first in which cases ones work better than others, so we can decide with knowledge.

The only other point I wish to comment on is from the previous post:

.

LED light is not actually directional, it only appears to be as an artifact of the was it is emitted. Light is emanated from the face of the device so it is a quasi-collimated light source. The light is not coherent like a laser, nor is it actually collimated, so it does spread out with distance. My empirical measurements with a meter have shown that the concentration or strength of the beam falls nicely with the inverse square law.

Discussion?

I call any ligting that emits on a given direction as directional. But we shouldnt lose on semantics. The fact that a LED not spread his light emission along a sphere, but on best case, along a semisphere means the light density drop is below the square of distance. On worst case, it drops at half the square of distance. The narrower the beam angle, the stepper drop in light density with distance.

Of couse any actual light source, still the more pure directional ones, spread with distance, so light density drops (still a laser beam will do, it you let enough distance it will become evident). And in general, as in the beam angle light spreads on all directions, an square factor is always present, but corrected by a coeficient, which depends on the beam angle. It varies for 1/3 to lower than 1/10 for very narrow and colimated beam angles. Measure light density with distance for aspheric lens of 2º and check how close of the square it is.

Whats happen is LEDs, still narrow beam angles ones, emits lots of light outside the "beam angle". There is 2 ways manufacturers defines beam angle:

The most used is half max intensity, meaning "beam angle" is that angle at which intensity (photometric unit cd, radiometric W/sr) is half of the max (intensity on the optical axis). So it lets lots of light emission outside the beam angle.

Other manufacturers use as the beam angle that at which 90% of the total light is emitted. Usually, for same LED, beam angle using this definition is larger.

So photons emitted by a LED are spreaded along a wide area, whaever the rated beam angle. Most of them fall into the defined "beam angle", but still many falls outside it (with secondary lens, usually there is a well defined beam angle and little light falls out of it). So the correction factor respect the square tend to be low, around 1/3. But 1/3 of the square is not the square.

Anyway, we dont use a single LED. We use many. As light density with distance for a single LED drops, because footprint increases with it, when you put many, light being spreaded for each enter on the area lit by others, and that is what keep light density near constant: same photons spreaded on same area. So the rate at which a single LED emission drops light density with distance is almost irrelevant on a multiLED setup. Due that I prefered to skip the question, as it only serves to mislead us.

Of course there are differences between narrow and wide beam angles when you consider plants blocking light. A narrow beam with higher intensity entering the canopy across a hole of a given size is able to carry more photons than a wide angle one with less intensity. But on the other hand, it suffer more for shadowing, as a leaf into the main beam angle blocks most of the light. I want to make a detailed analysis about this on next post, and show the advantages and disadvantages of each.

Again, I don't dispel what you say, merely bringing out some finer points for practical application. As a retired engineer specializing in X-ray and imaging design and implementation for a very large German engineering firm I have had a fair amount of both theoretical and practical applications. I very much enjoy this type of interchange with you as I seldom have an opportunity to converse with someone with the same background and education.

For sure that we can understand mutually. At some point I decided to give up dispelling wrong myths about lighting when I realized most people dont understand my reasoning. I enjoy the oportunity of exchange opinions with people that follow me and can point out where Im wrong. Learning is a process of findind the weak points on our explanations.

None the less, we must bring these things down to a language and context where "everyman" can understand and use the fruits of our discussion.

Once again, thanks for your timely and valuable input, please continue to do so, my friend.

H.G.

Yes, that is what I want to do. Especially on the context of pragmatical aplication of LEDs. But some theoretical background is required, and dispelling wrong myths is a must if we want to be able of work on the right direction.

Once explained some things, we can enter on explaining what are the advantages/disadvantages of each solution, using as plain languaje as possible.

But if I wouldnt wrote my two last post, misundertanding of that will continously stop the progress of the thread by questioning of misleaded people.