Home Made LED Panels

[Mayor McCheez]

Hi I am currently researching the topic of LED grow lights. I thought I'd share what I've learned so far:

I checked out a few books on the topic of photosynthesis. Chlorophyll a is not the only photosynthetic pigment found in plants (carotenoids, chlorophyll b, etc.) A quick and easy way to check out the TOTAL photosynthesis is to look at a plant's photosynthetic action spectrum. You can run a google image search for this. It will get you a rough idea, but there's more.

Chlorophyll concentrations vary from plant to plant, as well as from strain to strain.

The books I've seen explain how to do a detailed chemical analysis to find the concentrations of each pigment, but doesn't look simple enough to do without a full-blown lab. Though, I have seen devices for sale online which claim to measure chlorophyll a content via light sensors.


In addition, the rate of photosynthesis (essentially the rate of plant growth) can be affected by CO2 levels, temperature, and light irradiance (brightness).

"* At constant temperature, the rate of carbon assimilation varies with irradiance, initially increasing as the irradiance increases. However at higher irradiance this relationship no longer holds and the rate of carbon assimilation reaches a plateau.
* At constant irradiance, the rate of carbon assimilation increases as the temperature is increased over a limited range. This effect is only seen at high irradiance levels. At low irradiance, increasing the temperature has little influence on the rate of carbon assimilation."

-http://en.wikipedia.org/wiki/Photosynthesis#Factors


So, not only do we need to establish general wavelength specs for our lights, but specific experiments conducted on an individual strain will tell us the ideal temperature, CO2, and irradiance.

Here is a good all-purpose plant action spectrum:


Here's my analysis of this spectrum. I'm using this as a general guideline for choosing percentages of each wavelength:

% of my total array - wavelength (nm)
5.7% - 400
6.4% - 414.3
7.0% - 428.6
7.1% - 442.9
6.8% - 457.1
6.1% - 471.4
5.9% - 485.7
5.7% - 500
3.2% - 514.3
2.6% - 528.6
2.1% - 542.9
1.8% - 557.1
1.9% - 571.4
2.1% - 585.7
2.7% - 600
3.1% - 614.3
3.9% - 628.6
5.0% - 642.9
5.6% - 657.1
5.7% - 671.4
6.4% - 685.7
2.1% - 700
0.7% - 714.3

If you guys have a better action spectrum, PLEASE tell me.

Also, does anyone have any info regarding the specific action spectrum for herb?
And maybe you guys know the lumens/candelas/einsteins of photons requirements for the plants? I need more information about how much of each wavelength is required. I have the percentages, but not the specific amounts per plant.

Thanks,
McCheez

[TorturedSoul]

I think there might be... It's like analog vs. digital. Your graph has an analog "wave" and shows a front of light. But trying to imitate that with LEDs is like picking some number of points on that graph's front and illuminating right there - it will never become that analog wave.

Which means I guess that your approach of picking 23(!) different wavelengths instead of two or three would seem to have great merit. The more individual points (wavelengths) used, the better. How did you arrive at that number and those particular wavelengths? Are they just every one that is widely available or...?

The more the better - but how much "better" amounts to "enough," I wonder. I once had a cheap television that had few (digital) volume steps and I often found that while one setting was a little too quiet, the next higher one was a little too loud. Obviously, that television's volume "graph" needed more points. But I have (/had) other A/V equipment that has many more adjustments and much finer steps and have still occasionally found that I have been unable to get something exactly the way I wanted it because of the digital "steps." Other analog equipment allowed me those "infinitely fine steps" so to speak.

Ok, kind of a different animal all together... Except that with the A/V thing, digital (adjustability) looks great on paper or to a computer but not so perfect to me. I wonder if that is because I am a product (and part) of nature and if there is something to that for plants as well.

Meds: 1
Tortured Soul: Also 1 (imagine that lol)

[Mayor McCheez]

I split each 100 nanometers into 7ths... Yea I'm sure these points aren't all available from commercial LEDs, but I could always readjust based on what's out there... I haven't gotten that far yet... I'm still theorizing.

But TorturedSoul you make a good point about the question of digital vs Analogue. All I can say to that is that I've seen results from people who claim to be only using red and blue and the pictures are quite stunning. Also, I think (not sure) the commercial grow lights like the UFO are only a few different wavelengths.

Someone on another forum suggested that you can search google patents for any patent and see the specs. That assumes that the commercial grow light companies have already figured everything out.

Also, I've heard someone say NASA has done LED grow light research? Any info on this?

Are there LEDs with multiple wavelengths, or are they each singular?

What about turning the lights on and off? Would it be better if there was an automated dimmer control, much like the sunrise and sunset? Or does this not matter? I'm looking into using a USB interface in order to computerize a system, and a dimmer doesn't sound impossible.

[Mayor McCheez]

Hey I just found an article about NASA's research project. Greenhouses in space. Very cool: Lettuce and LEDs: Shedding New Light On Space Farming

"The LEDs generate less heat, and while leaves take on a black hue due to the lack of green light to reflect, the plants grow normally and taste the same as those raised in white light. "

[TorturedSoul]

Originally Posted by Mayor McCheez

I split each 100 nanometers into 7ths... Yea I'm sure these points aren't all available from commercial LEDs, but I could always readjust based on what's out there... I haven't gotten that far yet... I'm still theorizing.

Perhaps the best thing - if not the most realistic - would be to divide the spectrum into as many different points/wavelengths as humanly possible. To build the brightest, most powerful fixture that one can with the different wavelength LEDs mixed so that the light output is well-mixed on the plants at whichever distance from the LEDs happens to be the best for their penetrative abilities. Call that the test light. Then set up another one that could be modified and pull only LEDs of a given wavelength & grow identical clones under each. Do that repeatedly until you can assure yourself which wavelengths can be dispensed with (one at a time) with no measurable harm. Then start repeating with combinations to make sure there are no hidden "light relationship dependencies." Then continue to repeat until it is known which spectrum can be permanently removed. Make the setup that you have successfully stripped as your new reference fixture and then start playing around with boosting the different remaining LEDs one wavelength at a time. Then start trying differnt combos.

But who has the thousands of dollars and man-/woman-years to do all of that? A large well-funded organization could use many fixtures at once and shorten such R&D significantly but the cost would increase exponentially.

Originally Posted by Mayor McCheez

But TorturedSoul you make a good point about the question of digital vs Analogue. All I can say to that is that I've seen results from people who claim to be only using red and blue and the pictures are quite stunning. Also, I think (not sure) the commercial grow lights like the UFO are only a few different wavelengths.

Yes but are those results pure start-to-finish LED? Are the harvests as quick as they could be or is some missing component of light keeping the flowering process from being as short as possible? Are the buds as tight as possible? Flavor, potency, affect? Etc.

Originally Posted by Mayor McCheez

What about turning the lights on and off? Would it be better if there was an automated dimmer control, much like the sunrise and sunset? Or does this not matter? I'm looking into using a USB interface in order to computerize a system, and a dimmer doesn't sound impossible.

I don't believe that it is all that beneficial to have dimming capabilities. Gradual sunrise/sunset? Do plants that experience such do better than those that get more abrupt changes due to their location? Does a plant need the occasional cloudy day? Doubtful, I expect. But then I never thought making the change from veg to flower gradually over a period of time was beneficial over just setting the timer to 12/12 and being done with it.

Originally Posted by Mayor McCheez

Hey I just found an article about NASA's research project. Greenhouses in space. Very cool: Lettuce and LEDs: Shedding New Light On Space Farming

"The LEDs generate less heat, and while leaves take on a black hue due to the lack of green light to reflect, the plants grow normally and taste the same as those raised in white light. "

Makes sense that leaves would appear black if they were not being fed wavelengths that they couldn't use (and therefore reflected instead of absorbed). Doesn't show that they are being given all that they can use though. Nor does it show that of the wavelengths that the lettuce do absorb if all of them are beneficial (required).

And of course MJ might well benefit from different wavelengths/ratios than (leaf?) lettuce. It certainly does better with a great deal more of it lol.

Such research could run into the hundreds of millions of dollars in bureaucracy-dollars. And considering the budget constraints that NASA exists under, I don't expect to read about any breakthrough results any time soon. Shame.

Just jabbering....

[Mayor McCheez]

I've been told not to mention our favorite plant because they are scientists and only grow "tomatoes"...

There's a few others, but basically you can reduce the total amount of electricity while still saturating your plants with the maximum amount of light they can handle. This seems amazing.

Also, the thing about digital vs analogue, it looks like there is a certain amount of distortion. The wavelength rating for a given LED is its peak wavelength, but it does have a certain amount of fluctuation (max 50 nm), so perhaps this should be taken into consideration when choosing LEDs for your array.

As I find out more, I'll post more. I'm serious about building this thing, but not until I've got the info to do it right. When I begin experimentation, I'll post that too.

Cheers!

[pan4gold49]

Mayor McCheez all your work is nice. Led's have a very narrow wave lenght and not all frequecny's can be produced. I'm leaning towards white led's as the solution to an led grow light.

Pan4

[Mayor McCheez]

Yeah Pan the pulse thing came from one of your posts I've read them all in this thread!

How's that going by the way? I'm interested to know if you've done any experiments?

I want to get the right LEDs, but like Knna said in his thread it's not just about total intensity, it's also about efficiency. It'd be great if we could get the lowdown on what the best LEDs currently available are, and then we could price them... I'll see if I can draw up some kind of chart based on the necessary wavelengths.

White LEDs sound like a good idea, especially because they not only cover the necessary peaks, but they also fill in all the low points. I'd probably supplement them with certain LEDs of single wavelengths just to fill in the gaps.

My main project at the moment is examining the scientific literature available from online library databases available through my university. I'm studying the experimentally determined spectrums of photosynthesis in vivo. The action spectrums are apparently unrealistic.

Check out Knna's thread if you haven't already he's very knowledgeable.


Peace,
McCheez

[pan4gold49]

Thanks Mayor McCheez
I'm still gathering info I'm close to getting the samples I will need to go further in the experiment. I have found some 110vac whites to try they are 25 watters direct from China.

Pan4

[Mayor McCheez]

WOW if they work, you've got to tell me where to find some. The biggest I've found have been 10 watts I think... Guess I should look around more. Those babies must cost a grip.

[pan4gold49]

Thats why I'm going to get samples.

Pan4

[inferno420]

did u ever get them 25 watters?

[silly D]

Originally Posted by pan4gold49

Racefan I am looking for a new way to grow. This will also work on 12vdc a good thing for solar or wind power. I did a lot of research into the panels and leds. I found that most leds used in those panels 3,000 mcd red and 5,000 mcd blue. Also have to look at the lens of most leds have wider diffuser 30 to 45 degrees . The led I have chosen are 70,000 mcd with a 15 degree diffuser that's 439 lumens at the focal point and 303 lumens at focal point times 2. Since this is no box here I will be free to explore the possibilities.


Pan4

i need help wanna make my own led grow lights for veg stage can you pls help me out should i use just blue or blue white
thanks a lot

[silly D]

Originally Posted by pan4gold49

I guess after looking at ways to grow and seeing threads on using Leds for lighting.
I looked at 168 led panels, supermotojoe setup and Irie prycon lights.
DYI Panels

First was led selection. Since they come in all sizes 3mm to 10mm I chose 10mm.
The 10mm is the biggest bullet size with a big lens without going to 1 watt.
Next I found some leds that had a 70,000mcd over the 5,000mcd led found in most 168 led panels.
Since the leds also come with the resistors needed for a 12vdc supply.
I will be making 12" x 12" panels of 125 Leds. 72 red and 40 blue leds like moto's panels.

I am hoping for 9,000 lumens per panel with a 30 watt power requirement.
Led order has been placed in HK hope to start this weekend

Pan4

pls hook me up do you have a you tube video or something,i'm a newbie thanks