Custom 400 piece, 3w, led panel, help me choose led ratio

[DANNYpms]

First off, great site and lots of helpful information. I just had to register. Thanks everyone for creating a great community.

I have found what I believe to be reputable vendor who is willing to create a custom led panel for me per the following specifications:

Panel contains 400s LEDs
Epistar & bridgelux 3W LED
Each led has an individual lens (like the advance led diamond series lights)
--- Was thinking 60 degree lens (60/90/120 available)

I can customize the color ratio to include whatever color spectrum I like; including UV and IR. This is what I need help with... I can't seem to find definite information regarding which spectrum of light marijuana plants need to thrive. Once the unit is ordered, I will post a review here and let you know how it works.

I asked the representative which LEDs were available and was given the following:

370-420, 450-460, 525-530, 605-615, 630-850

The .pdf they sent says I can customize the panel with:
420nm Violet, 610nm Orange
630nm Red, 730nm IR, White

So far, I have heard that 740nm far red is needed for flowering and big buds. Are there any specific wavelengths/colors that trigger specific results?

Should I be trying to replicate MH and HPS here? I know both MH and HPS waste a lot of light but they seem to be the light of choice for many and that's because they produce results. I want to create the same results using LEDs...

So I have lots of options here, and need help choosing the correct wavelength/color LEDs and their ratio. Remember, 400 LEDS total; 25 LEDS per row. I will be growing marijuana from veg to flower.

 

[Hosebomber]

Wow, there is soooo much to cover here. Without going too deep into my research and giving my IP away, I'll try to help you the best I can.

First, any wavelength range of more than a 20nm range is not a single LED or they are not giving you good specs. The specs from the .pdf seem much more logical.

Second, the far red added into the mix is called the Emerson effect. If you google that you will find a ton of studies that use that study but it is very hard to track down the original. The basics of the Emerson effect is that when you add ~10% Far red (730-740nm) you have a increase in light response from certain photo-receptors in plants that causes plant mass to increase and grow at better rates. The intent was to mimic sunrise and sunset by adding that light for the first 30 min and last 30 min of the light on cycle. Through later test they found that adding the far red light during any and all phases of the light on cycle produced this effect.

Third, the wavelengths listed from the .pdf are not UV (and technically, the 730nm is not IR... it is far red) UV C ends at 400nm, which is where violet begins.

Some companies claim that adding green (515-540nm) LEDs increases growth. From my personal experience and all studies I can find, green has little use and only in "deep penetration" portions of the leaf. This requires a large number of high intensity LEDs for very little return. Some companies like white LEDs and others use the claim that is waste energy just like MH and HPS by crossing spectra that are not used. 610nm is yellow. Some use but I still do not have any information on yellow LEDs mmj growth. 630nm is at the low range of the red spectra. Plant growth performs better with 660nm red LEDs but they are more expensive in most cases.

I'd love to have the exact info they sent you on their products to give you a better run down on their led quality and if their prices are reasonable. Does their build include a housing and drivers or just the leds and lenses.

Hose

Edit: Damn that is going to be a huge LED panal. That will be pulling close to 1000 watts.... I'm not sure there is a producer selling a unit that powerful.

 

[DANNYpms]

Indeed, it will be a very large panel... I am getting it direct from the manufacturer as a production sample. The company uses 1, 2, and 3, watt led in some of their panels. "Reasonable price" sarcasm... include 400 3w LEDs and lenses for each led, housing, fans, drivers, and even a chain to hang the light

Just need to decide on spectrum...

1. Emerson Effect: 730 - 740 nm far red (10% of 400) = 40 FAR RED

 

[Hosebomber]

The exact amount used in the org study was 7% of the value of 660nm red light. So if you have 80% (320 diodes) red light then roughly 20% of that would be 630nm range or 64 diodes, would leave you with 238 660nm Red and 18 730nm Far red diodes.

I'm not saying that the 20% blue 80% red is even close to correct... just using those numbers as an example.

 

[TorturedSoul]

Don't all LEDs used in grow light panels have individual lenses?

that is going to be a huge LED panal. That will be pulling close to 1000 watts.

Depending on how much power those 3-watt LEDs are driven at.

 

[Hosebomber]

Don't all LEDs used in grow light panels have individual lenses?

Yes and no. When the LED is manufactured a epoxy resin covers the diode and forms a lens. In general the lens is in a dome shape and creates a lens that projects a 120 degree lambertian radiation pattern. Again generally speaking, when a vendor/distributor refers to supplying a lens their are speaking of an additional external add-on lens that adjust the pattern to something other than the manufactures angle. This is another area to be careful with as they can greatly reduce the intensity if the lens has poor reflective qualities.

Depending on how much power those 3-watt LEDs are driven at.

I was using an average 2.3W power dissipation over 400 LEDs plus driver and fan pull. Most blue leds ran at ~700mA pull 2.6-2.8W and reds pull around 2.2-2.4W. It's possible that it could pull more or less by as much as 100 watts depending on the cooling features, number of fans, power draw of the fans, quality of the led, power conversion rate of the drivers, and current the leds are ran at.

 

[DANNYpms]

I just got the LED spectrum chart from the manufacturer...

How would I make blue 440 nm and blue 470nm? Those nm ranges are not listed on the below LED spec sheet. One of the manufacturers current products lists both blue 460nm and blue 470nm individually; they list approx 10 blue 460nm and 15 blue 470nm... Is the blue 470 a voltage modification or just overlap?

3w Spectrum Catalog
Color Wavelength(nm) input current(mA) Lumin Value(lm) Input Voltage(v)
Voilet 380 700-750 3.2-3.6 3.0-3.6
Voilet 420 700-750 11-14.4 3.0-3.6
Blue 450 700-750 16-21.6 3.0-3.6
Blue 455 700-750 16-21.6 3.0-3.6
Blue 460 700-750 40-54 3.0-3.6
Green 525 700-750 80 3.0-3.6
Orange 610 700-750 48-72 2.0-2.4
Red 630 700-750 72-90 2.0-2.4
Red 660 700-750 12.8-18 2.0-2.4
IR 730 700-750 1.6-3.2 2.0-2.4
IR 740 700-750 1-3.0 2.0-2.4

White
Color Kelvin input current(mA) Lumin Value(lm) Input Voltage(v)
Warm White 2700-4500 700-750 128-198 3.0-3.6
Day White 5500-6500 700-750 130-198 3.0-3.6
Cool White 7500-8500 700-750 128-180 3.0-3.6
Remarkes, $1/pc for UV and IR leds

WAS GOING TO UPLOAD SPEC SHEET... FORUMS DON'T ALLOW DIRECT UPLOADS? The spec sheet is now my profile picture. lol

BAM; HERE WE GO!

As for lenses, I am referring to extra lenses that they put on each led. I can choose from 60,90,120 degrees. I was considering mixing 60 and 90 degree lenses for a custom effect on specific color/wavelength LEDS. Other than focusing the light, how would a lens affect the light wavelength?

 

[Hosebomber]

Great info. $1 per for any 3watt LED is actually a pretty good price let alone for a UV. The 380nm violet is technically "Near UV" and borderline UV C, not a bad LED and has debatable effects. My personal research shows pluses and minus for it's use but does have some concerns when using a large amount or trying to get a rating for sale in the US. I would have to see the data sheet to give exact info but it looks like the 660 red led is a little on the weak side as is the 450 and 455 blues.

You have to understand that not all companies use the same standards to perform their test and the data sheet information on luminous output and test currents can vary greatly while the actual products are very close in performance.

As for the lenses, make sure you test many different angles, too tight of an angle you get poor overlap of light blend. Too wide of an angle you have less intensity and canopy penetration. A lens should not have any or much effect on the wavelength of an LED. It is possible that there is a flaw in the creation of the lens that may distort it some. The biggest issue is generally the lose of intensity from poor reflective properties on the parabolic.

 

[DANNYpms]

I know a lot of people say white light in an LED is useless and wasteful but consider the following...

White light is composed of all four colors of light right? White LED's produce more lumins than certain LED's. Blue 450 and blue 455 have 16 -21.6 lumins; green 525 has 80 lumins; and day white led has 130 - 180 lumins... I am wondering how much of each light color the various white leds produce and if using white leds to cover certain parts of the spectrum would be more efficient than using certain color LED's... Any thoughts on that?

 

[TorturedSoul]

Yeah. Using lumen output ratings to determine which wavelengths to use is a losing proposition. To put it bluntly (aka "the short answer," lol) a really bright light that doesn't do much for your plants... doesn't do much for your plants.

To explain that a little further: Wikipedia has a pretty accurate page on "lumen" that would be helpful in explaining this. Lumen is a measurement of light output that is heavily weighted for people. Wavelengths that the human eye perceives as being brighter are given a higher rating. That's all well and good when you're picking lights for your living room but as plants do not have human eyes, it is a far less useful statistic in general in terms of plants and, in terms of LED grow lights, you might be shooting yourself in the foot. The reason that a HPS can be an effective grow light and still have a high lumen rating is that, although much of its output is in wavelengths that appear to be much brighter to us than to the plants (so to speak), it's still more or less a broad spectrum in relative terms - and because it's hella bright, lol. It's like a 10gauge shotgun, although if you get shot with one at a moderate distance it's almost guaranteed that 100% of the pellets won't hit you, enough are likely to do so that you'll feel it in the morning (or, depending on your luck, that you'll never feel anything again). Whereas a pistol - or a small group of pistols which are aimed slightly apart - can be really, really powerful, but if that one slug misses you, you're golden. And with LEDs, you're not creating a spectral curve - you are just hitting a few specific points.

With LEDs, you want, I should imagine, wavelengths that are "seen" by the plants regardless of how bright they appear to your eye. I suggest you do a Google search for photosynthetically active radiation (PAR). That'll probably make much more sense than the drivel I typed above (which reads kind of goofy even to me ATM, lol).

 

[Hosebomber]

White light is composed of all four colors of light right?

There are actually seven colors in light spectrum.

White LED's produce more lumins than certain LED's. Blue 450 and blue 455 have 16 -21.6 lumins; green 525 has 80 lumins; and day white led has 130 - 180 lumins... I am wondering how much of each light color the various white leds produce and if using white leds to cover certain parts of the spectrum would be more efficient than using certain color LED's... Any thoughts on that?

As TS stated, PAR is a much better measurement to go by when referring to grow lighting. Lumen is the scale used to tell how bright to the eye it is. The human eye sees greens and yellows the best. Higher or lower and the lumen
numbers drop off with the same relative intensity.

As for how much each of the colors the white led hit... that is why I was asking about the data sheets. The data sheet for the LED will give you the exact spectra and range of an LED with all of the other tech data to give you the information you need to make a good choice. There are a few different threads on LEDs in this forum that I have posted links to data sheets for luxeon rebel leds (both color and white). There are the high end of the standard currently.

 

[DANNYpms]

with LEDs, you're not creating a spectral curve - you are just hitting a few specific points.

That's why I am now considering the value of adding white LEDs to the mix. Since they should be creating light of all colors, they should fill the gaps left by the specific color LEDs. The real questions is, would any white LED produce more usable blue light than lets say "blue 450 nm"?

Also, 480 nm to 515 nm is the second peak for carotenoids. Both MH and HPS cover this peak so it's probably important but I don't see a way to cover it using anything expect white LEDS.

 

[tsmit420]

First off, great site and lots of helpful information. I just had to register. Thanks everyone for creating a great community.

I have found what I believe to be reputable vendor who is willing to create a custom led panel for me per the following specifications:

Panel contains 400s LEDs
Epistar & bridgelux 3W LED
Each led has an individual lens (like the advance led diamond series lights)
--- Was thinking 60 degree lens (60/90/120 available)

I can customize the color ratio to include whatever color spectrum I like; including UV and IR. This is what I need help with... I can't seem to find definite information regarding which spectrum of light marijuana plants need to thrive. Once the unit is ordered, I will post a review here and let you know how it works.

I asked the representative which LEDs were available and was given the following:

370-420, 450-460, 525-530, 605-615, 630-850

The .pdf they sent says I can customize the panel with:
420nm Violet, 610nm Orange
630nm Red, 730nm IR, White

So far, I have heard that 740nm far red is needed for flowering and big buds. Are there any specific wavelengths/colors that trigger specific results?

Should I be trying to replicate MH and HPS here? I know both MH and HPS waste a lot of light but they seem to be the light of choice for many and that's because they produce results. I want to create the same results using LEDs...

So I have lots of options here, and need help choosing the correct wavelength/color LEDs and their ratio. Remember, 400 LEDS total; 25 LEDS per row. I will be growing marijuana from veg to flower.

Ther is lots of information on wavelengths needed all the top companies that have proven to be a superior product have given their wavelength, the hard part is the ratio. there are many charts showign the photosynthetic action spectrum this cant be completely followed though it shows that plants uses blue light more then red which is true for most most plants, but with cannabis red light is more important for flowering, there is not much information on the photosynthetic action of cannabis

 

[tsmit420]

The exact amount used in the org study was 7% of the value of 660nm red light. So if you have 80% (320 diodes) red light then roughly 20% of that would be 630nm range or 64 diodes, would leave you with 238 660nm Red and 18 730nm Far red diodes.

I'm not saying that the 20% blue 80% red is even close to correct... just using those numbers as an example.

actually a 4:1 ratio of red to blue is very ideal for a full cycle light. any wavelengths under 400 Nm is a waste there will be no benefit. only UVB has benefit but u will never find UVB leds greater then half a milliWatt. any led companies claiming to have uvb is a scam. and those that just say UV is just a marketing ploy for uneducated customers

 

[Mars Hydro]

Have you tested 4:1ratio of red to blue?

actually a 4:1 ratio of red to blue is very ideal for a full cycle light. any wavelengths under 400 Nm is a waste there will be no benefit. only UVB has benefit but u will never find UVB leds greater then half a milliWatt. any led companies claiming to have uvb is a scam. and those that just say UV is just a marketing ploy for uneducated customers

 

[Hosebomber]

Yes, I have and no it is not correct. Remember that this post is over a year and a half old. There have been a large number of test since this post, let alone before it.

 

[SuperFunker]

Re 380nm: Initially, research indicated that no benefit was found in the near UV; however, research now suggests that 380nm may expedite early stem and leaf growth. Still, if your LED light does not have this spectrum, your plants can do without. If anyone has found recent research on the near UV, I would love to read it. "Need input!"

Re UVB: Simply put, UVB is dangerous! Nevertheless, UVB can induce greater THC production (3 to 5% increase) at the expense of a reduced yield. Further, I doubt that UVB will ever be available in an LED grow light, because such LED's are expensive, dangerous and most importantly, professionals only dose their plants with UVB for short periods (4 hours per day) and only while flowering.

Re white light: White light is likely the most overlooked LED. As HB and I have discussed, PAR research is still very much an incomplete science. We do not know what critical spectrums are missed by delineating PAR with LED's designed to target specific nm which is why white light is SO important. Adding a few full spectrum LED's to a grow light is an effective way of filling in PAR gaps which is presently the best way of addressing our ignorance.

PLIP

 

[SuperFunker]

HB... what would you suggest as a reasonable ratio of 640 to 660 nm LED's? more 640's or more 660's?

 

[Hosebomber]

I personally use 630 and not 640. My current test has a 3 to 1 and 2 to 1 660 to 630 ratio.

 

[SuperFunker]

Re 380nm: Initially, research indicated that no benefit was found in the near UV; however, research now suggests that 380nm may expedite early stem and leaf growth. Still, if your LED light does not have this spectrum, your plants can do without. If anyone has found recent research on the near UV, I would love to read it. "Need input!"

Re UVB: Simply put, UVB is dangerous! Nevertheless, UVB can induce greater THC production (3 to 5% increase) at the expense of a reduced yield. Further, I doubt that UVB will ever be available in an LED grow light, because such LED's are expensive, dangerous and most importantly, professionals only dose their plants with UVB for short periods (4 hours per day) and only while flowering.

Re white light: White light is likely the most overlooked LED. As HB and I have discussed, PAR research is still very much an incomplete science. We do not know what critical spectrums are missed by delineating PAR with LED's designed to target specific nm which is why white light is SO important. Adding a few full spectrum LED's to a grow light is an effective way of filling in PAR gaps which is presently the best way of addressing our ignorance.

PLIP

Just stumbled onto an LED grow light seller that claims to use UVB LED's in their lights. Really? UVB? I am skeptical. If this light seller is ACTUALLY using UVB LED's in their full spectrum lights, they would need to provide some sort of control mechanism so the UVB's could be turned off until flowering and then controlled to expose plants for only short periods each day.

UVB inflicts stress on your plants which reduces above ground growth during vegetation which is why UVB is reserved for the flowering stage. When flowering, plant response to the stress of UVB is greater production of THC, but a reduced yield.