Advanced LED Development Thread.

[OsWiZzLe]

HortScience. 2004 Dec ;39 (7):1617-22 15770792 [Cited: 1]

Green-light supplementation for enhanced lettuce growth under red- and blue-light-emitting diodes.

Hyeon-Hye Kim , Gregory D Goins , Raymond M Wheeler , John C Sager

Plants will be an important component of future long-term space missions. Lighting systems for growing plants will need to be lightweight, reliable, and durable, and light-emitting diodes (LEDs) have these characteristics. Previous studies demonstrated that the combination of red and blue light was an effective light source for several crops. Yet the appearance of plants under red and blue lighting is purplish gray making visual assessment of any problems difficult. The addition of green light would make the plant leave appear green and normal similar to a natural setting under white light and may also offer a psychological benefit to the crew. Green supplemental lighting could also offer benefits, since green light can better penetrate the plant canopy and potentially increase plant growth by increasing photosynthesis from the leaves in the lower canopy. In this study, four light sources were tested: 1) red and blue LEDs (RB), 2) red and blue LEDs with green fluorescent lamps (RGB), 3) green fluorescent lamps (GF), and 4) cool-white fluorescent lamps (CWF), that provided 0%, 24%, 86%, and 51% of the total PPF in the green region of the spectrum, respectively. The addition of 24% green light (500 to 600 nm) to red and blue LEDs (RGB treatment) enhanced plant growth. The RGB treatment plants produced more biomass than the plants grown under the cool-white fluorescent lamps (CWF treatment), a commonly tested light source used as a broad-spectrum control.

 

[meskalin]

Could ccfl's also work?
most blue ones have their spectrum peak around 450nm an the red ones about 620nm and they have arond 60lm/w. they seem to be cheaper than led arrays.

 

[newbgrow]

Using 730nm in flowering instead of 660nm? Phytochrome question

Using 730nm in flowering instead of 660nm? Phytochrome question

Having a lot of interest in LED lighting applications for growing plants, I just finished reading many threads from UK420 and other sources. They certainly answered many questions I've had, but there is one observation I would like to get off my chest.

I believe I found one discrepancy in using 660nm red LED's during flowering phase. (I could be wrong, as I'm no expert.)

First, for you experts out there, could you please validate this following blog as a valid source of information?

Science and Sarcasm - Plants as Clocks

Below is a summary:

...Since the sun produces both red and far-red light, there is constant two-way conversion between Pr and Pfr - an equilibrium develops. During light hours, conversion of one form to the other occurs rapidly - within milliseconds... These reactions both require light energy. So, during darkness neither of these active reactions can occur...
During the night however, Pfr very slowly and passively degrades to Pr in a process called dark reversion. On a summer morning there is typically a high Pr : Pfr ratio. By detecting morning levels of Pr and Pfr, the plant is essentially measuring how much time has passed during the night. When the level of Pfr reaches a critical point, physiological responses such as flowering or root growth might be triggered.

Armed with this information at hand, I came to the following conclusions:
1. Plants can tell night from day by it's Pfr/Pr ratio.
2. Light activates a process that can convert Pr to Pfr (or Pfr to Pr) instantly.
3. Increasing Pfr during night means nights are getting shorter, therefore, triggering vegetation instead of flowering.

Now, if you use purely 660nm wavelength LED's during flowering with limited far red spectrums, light instantly converts this Pr energy into Pfr. Since there is limited Pfr light, the same energy isn't being converted into Pr, destroying the equilibrium. Also, lots of red light during flowerng without far red light would have all the light converted into Pfr energy, creating an excess - and promoting vegetation instead of flowering.

I believe the Emerson effect also coincides with my theory, in that you need both Pfr and Pr for it to work.

Any thoughts? Or am I completely off base?

 

[LED_experiments]

cool white or warm white? i know they both have a bit of far red, but do you think any of them is better than the other?

 

[WTF-over]

I still didnt checked it. Warm whites emits more far red than coolwhites, but anyway is only a little. But currently, warm white phosphors have higher conversion losses, so until they improve a little more, i would use coolwhites.

I buyed some 80lm/w cool white white leds to check how they work together with reds and some blue. Im currently thinking in give the blue light half from blue leds and half with coolwhites, wich emits a very good percentage of blue, but a bit of all wavelenghts, included some far red.

Aditionally, from the visual appaerance, coolwhites are best suited, because after mixing with red leds, the resultant tone would be less pink and a bit more white.

I thought the Whole idea was to target frequencies so as to not waste energy. Why not just purchase far-red emitters and disperse then through-out your array. Using full spectrum emitters defeats the purpose doesn't it? Who really cares about visual effect?
:smile:

 

[gordonliu]

I will have to discuss all of this with shuji.

using a longer frequency for the pulses is a better idea. the plant cannot be considered on a molecular level, because it is a bulk material. the chemical or photochemical properties of a bulk material cannot be determined from the molecular structure

also to make your own leds, you dont just need a lot of money, you need a ****ing shitload of money.

molecular beam epitaxial systems aint cheap, neither are the squids or electron microscopes (or AFMs) that you will need to do physical characterization. simply running a clean room and the cost of all the lN2 and high purity gasses and chemicals would be too much. thats why you have to have access to all of this stuff without paying for it!


also, a good topic to search on that may help with this would be organic solar cells.

poly pyrroles and cyclic polypyrroles like porphyrrins and melanins (poly porphyrins) are all similar molecules to the chromophores found in plants.

pumping them with photons is a waste, because any chemical reaction has a rate. it takes a small amount of time for the soliton in the chromophore created by absorption of a photon with sufficient energy to travel the length of the molecule.

the bottlenecks of the overall rate of the photosynthetic reaction will be in the redox reactions and other proton transfers.

essentially any reaction that is intramolecular will have a very fast rate, and any reaction that is intermolecular will have a slower rate.

look at the rates for the redox reactions and proton transfer and the slowest step will tell you how fast the cycle works. i cant do it now, because i dont have access to journals, but i will again on monday, so i will start looking up articles on physical botany and physical biochemistry.



another idea that can be used to limit the damage to the LEDs is to have only several on at a time, so that you not only pulse the entire array, but within the array you only have one row or column on at a given time. it will complicate wiring like a mother****er and make controlling it hard, but it may save you several hundred dollars down the line.

gonna have to talk to some EEs to have them make some of this shit for me (I aint no EE)

 

[Mr Celsius]

At the second page of this thread, i explained the ways LEDs achieve electric savings.

Spectrum improvement is only one of the ways, and IMHO, the less important.

Most people insist to see chlorofills action spectrum instead of full plants action spectrum, wich is very different, being far more flat. There is a gain from using mostly blue and, specially, red, but unfortunatelly, there is a persistent myth about the "light wasted".

Plants use green light for photosynthesis, too, and most of green light have only a slighty lower quantum yield (=photosysnthesis per absorbed photon) than blue light. Yelow light has higher efficacy than blue light. This isnt a opinion, its a fact many times proved at botany's studies.

25% of improvement in photosynthetic efficacy is the most you can expect over an HPS spectrum (supposing same radiance in photons). I doubt currently any LED grow light can achieve more than 15% of photosynthetic gain just from the spectrum.

The visual appearance has some importance in order to inspect plants health without the need to take plants out the cab. Of course it isnt a requeriment, but if we can improve it without dropping noticiable photosynthetic efficacy, why not do it? I think we can hunt three birds with the same shoot using white leds: provide the little far red required, provide a bit of light of all PAR wavelenghts and improve visual appearance.

Of course, im talking about a small percentage of white leds, just about 10-15%.

Anyway, there is people who disagree with this, and still work only with blue and reds. Nothing wrong in it. I believe its good using differents aproachs to this question, finally from the experience we will be able to dial in LED's spectrum, and growers may choose the concept they prefer.

And always remember: plants use all the light along the PAR range

So knna, if you know so much about LED's and what plants want. Why don't you simply list the proper LED's to use in the right ratios... seems simple enough to me. Then anyone can go out and buy those LED's and make an array themselves.

 

[newbgrow]

Not, you arnt off base at all. But what do didnt take into account is phytochromes have a wide action spectrum range.

All grow lights have a Pfr/Ptot (percentage of Pfr to total phy), between 0.75 (cool MHs) to 0,89. HPS are around 0.85-0.87, while 660nm red leds have a equilibrium on 0.89-0.9. Little difference, because we already are using lights that induces low Pr. All grow lights are considered strongly reddish.

The Emerson effect is independent of this, but think a little light about 700nm is enough for the enhancement effect of red light. 660nm leds still smits some light in this range. The Emerson effect isnt phytochrome mediated, but its related to the electron flow between PS I (photosystem) and PS II.

Just adding some white leds, wich emits a bit on the far red range, improves visual effect and provide more than enough far red.

You seem very knowledgable about all of this; I hope you don't mind to clear some of this up for me. I don't currently grow, so I can't test any of this out at the moment - I'm here purely for the knowledge and the theory, which I find very interesting. If this technology is perfected, it would mean a major revolution in the way plants are grown in my opinion.

I previously thought that at some wavelength there is a cut-off point between what is known as red light (Pr) and far-red light (Pfr), but your post as quoted led me to believe that all wavelengths contain both Pr and Pfr. Is this correct? (For example, if 660nm is 0.9 Pfr/Ptot, then do you mean it has 9:1 Pfr/Pr ratio?)

Secondly, at what wavelength of light (if this is the right terminology) is it balanced at 50/50 Pr/Pfr or Pfr/Ptot?

Thanks in advance for clearing up confusion. If this ever gets bothersome let me know.


It is my understanding that a greater Pr/Pfr ratio governs many things such as plant stem elongation, smaller leaves, less branching, germination, etc. In Short Day Plants, the Pr/Pfr ratio is measured at the end of the night to see if Pfr has degraded beyond a critical level, to induce flowering. Pr is absorbed at peak at around 660nm and Pfr absorbed peak at around 730nm. If so, a 660nm LED should have a higher Pr/Pfr ratio... unless this has something to do with the wide action spectrum range you were talking about.

My observation based on a couple hours of online research shows that more Pr light during the day (signifying summer) requires a longer night for dark reversion to break down the Pfr. Therefore, it should be beneficial to provide a higher Pfr ratio during flowering. I.e. It is the degradation and destruction of Pfr that determines the length of the night, not the actual light cycle per se. So if a grower who uses only blue lights during veg, and then switches to 12/12 with a high Pr to Pfr ratio, that would be slightly contradictory in my humble opinion, because the high Pr would show an increase to Pfr during night, making it seem as if summer were just starting to the plant.

That's not to say that it wouldn't work (since I would be wrong as proven empirically by growers here), but that there could probably be a better procedure. Of course these are just my not-entirely-grounded conclusions...


Edit: Sources

http://www.mobot.org/jwcross/duckweed/phytochrome.htm
http://employees.csbsju.edu/ssaupe/biol327/Lecture/phytochrome.htm
http://www.hcs.ohio-state.edu/hcs300/devel2.htm
and several Google Book excerpts...

 

[Mr Celsius]

But they dont think any asian led manufacturer is capable of copy it and manufacture them far cheaper.

But if things continue this way of ignorance, im going to stop sharing info. I always think of giving most of the info i have, and allow to any grower to do its own led grow light, knowing that if you produce mass quantity, you can offer them the same product cheaper or at similar price than a single person can do it himself.

Many people working so hard to get an aditional 5% of efficacy while they are using linear regulators against switched mode power supplies with more than 10% more efficiency. It do me smile often. I tried to offer reliable and indisputable info to avoid people waste their time in concepts without any foundation. But im far of achieving it. Myths are very persistent, and im very tired of fighting them.

So, in your humble opinion, how should everyone be going about it?

 

[Ace2p]

Hey everyone,
Thank you everybody for sharing all your knowledge, i've been looking into this topic for a couple days now. I'm new to this forum, so to give a quick intro: I'm a CPE student, and I love experimenting with new tech alternatives, i.e. the led grow lights. IMO they are definely not ready for people who's main goal in growing is yield. This technology is very suitable for small and stealth grows. I've previously grown with a 600W HPS and 175W MH both hydro, and then soil. Got excellent results both times. But HID lights are too much hassle: power consumption, need for good ventillation... I originally thought about using LEDs to grow about a year ago and wondered why I haven't heard about anyone doing it, since very little energy wasted in form of heat, and very little maintainance. I ended up revisiting this topic a few days ago.
I found the PWM really interesting and i'm gonna give it a try. I plan on using a basic 555 circuit with a diode parallel to R2 and today I calculated the resistor and capacitor values. Ra = 1,100 ohms, Rb = 107,000 ohms, (Yea, You'll have to hook a some resistors in series to get these values) cap = 2 nF. This will give Ton ~ 1.5 micro seconds and Toff ~ 148.5 micro seconds, achieving a duty cyle of ~1% (thank you SY for the pulse widths) and the frequency of 6670 Hz. I'll hook up the output to a high amp, high volt MOSFET to pulse the arrays. Whats left to do is to calculate how much power to pulse the arrays with, and also how many LEDs I will need (depending on their specs)...
I also plan on pretty much covering the entire grow chamber (for one small test plant, indica) with LEDs, of course in the correct ratios, and as knna suggested, hang some led clusters into the plant.
I'm done with using HID lighting, too costly (energy wise), and WTF is up with the companies selling the LED grow lights at rip-off prices!!!
I'll keep you guys posted on the setup and if I decide to do it...
Let me know what you think of the PWM idea...

 

[Ace2p]

well i've just looked at a datasheet of a 555 timer and im not sure if it can handle a duty cycle like that. It says turn off time is less than 2 microseconds. 1.5 microseconds is pushing it. I'm gonna redo the calculations for lets say 50 microseconds, as sy said anything under 200 us is good.

 

[Ace2p]

So taking into consideration that this is my experiment to see how well a plant will grow under the LED's I considerably reduced the frequency, but kept the same duty cycle of 1%. So I decided to have Ton 50 us and Toff 5 ms, frequency 200 Hz. I will use Ra of 3.3 Kohms, Rb of 330 Kohms, and the capacitor 22 nF. Maybe if I get a good grow, I'll build a more advanced PWM that can handle 1 MHz... So I'm going to build this PWM and start testing some LEDs.
I've also decided to do a PC grow with 2 small indica plants, or maybe lowryder 2 (not a fan of autoflowering). I will do a grow diary and post pics and everything... I'm going to keep everything inside the comp and use it, but im gonna move the power supply to the very front bottom (where the pc speaker usually is), and also take out the DVD ROM and put a faceplate there.
Im place of the real power supply I'm going to convert an older one into an odor filtration unit, anyone know the best way to filter the odor out with little noise?

 

[relief]

Well, from going on my brothers LED grow, it worked great in veg like a CFL, but really lacked during the flower cycle just like a cfl.

It will be the future, but right now the technology just isn't here.

 

[LED_experiments]

it is already, you only have to know how to use it

 

[relief]

In flowering, it's not. In veg, we where both impressed... but flowering was a complete letdown. There is something missing in the spectrum during flowering.

 

[IGrowWithLEDs]

Hey LED_experiments,

What I used before, which were some of the best but are now not that great, was the apollo led's, but most places now that I can find only carry the white. I was out of the loop for about 6 months or so and all the suppliers that I used to goto now dont have anything thats worth it.

I would like to get my hands on some of the cree led's, but I still dont know where to get them. Do you know of any good places to get the top of the line led's now? I really appreciate it.

my only other choice is superflux at this moment. thanx

 

[IGrowWithLEDs]

I'm still working on my setup. I'm planning 400 watts worth for 6 sq feet. but I'll definately do a lot of testing using the pwm and overdriving and try to see if it really does help on energy saving without loss of growth. I've been a little low on $$ so I havent been able to invest recently. But I'm going to be getting some money in so now I finally have a chance to buy some more.

I could do a minor test subject right now because I do have the pwm and overdriving figured all out using a microcontroller and since I killed my other arrays i only have an array with about 7 watts worth of superflux so it's only usefull for about first 2 weeks of the plants life.

Hey knna, where would you buy led's from now?? thanx

 

[r2_]

So i'm gonna buy 3 of these, i'll keep you guys updated.

http://store.thatcrazynight.com/product_info.php/cPath/22/products_id/30

 

[1toke]

Where's Deadly Foez and sy9942 been hiding? I need to be updated on the research being done. this thread is very inspiring..I would like to build some LED arrays for supplemental lighting. but am not experienced with the hardware necessary.

thanks for any info,
1toke

 

[IGrowWithLEDs]

the name deadlyfoez has been banned, unfortunately bc thats the best screen name on earth......but trust me...he still lives....trust me