Full Spectrum LED tank pictures

[jedimasterben]

I like both, but Steve's newest violets are lower in wavelength - they used to be 422nm peak, now they're 414nm.

[jarob]

Hmm Im thinking two more NW, two HV, and 4 RB added! Thanks!

[Paleoreef103]

^ what he said, but add more violet. A 10g tank is 0.155m², so that is about 7w of 400-440nm light, so that's 6x 430nm and 2x 405nm. Don't have to use that many, but that's the upper ceiling for a 10g tank to make sure you're not adding too much light.

Hey Jedi. Could you go over the calculations you used for this? Just curious.

[jedimasterben]

Hey Jedi. Could you go over the calculations you used for this? Just curious.

Even at 15m depth, corals are still blasted with 52W/m² of 400-440nm light, and 60W/m² of 440-480nm light. To match those output levels would be to match what they get in nature and will provide just about the best growth that you can get. Safer numbers to use would be 45W/m² of 400-440nm, and 40W/m² of 440-480nm. All you have to do is calculate how many square meters your tank covers and then figure out the radiant power of the LEDs you are getting.

 

For instance.

 

My tank measures 48" x 24" x 16". There is about 39.4" in a meter, so my tank measures 1.2192 meters by 0.6096 meters. The surface area of that (length times width) is 0.743m².

 

If I want to provide 45W/m² of 400-440nm light, multiply that number by the total area (in square meters) that you need to cover, so 45 x 0.743 = 33.44W. Then just divide that amount of power by the output of the LEDs you want to use.

 

Let's take LEDGroupBuy's hyper violet - it outputs 0.96W of radiance at full power (note that the LED uses 2.5W of power from the wall - do not confuse this with the output of the LED, which is 0.96W). So 33.44 divided by 0.96 = 34.83 LEDs needed. I would need about 35 LEDs to give my tank 45W/m² of 400-440nm light.

 

Where I got the numbers from in nature: http://www.advancedaquarist.com/2012/10/aafeature

[Paleoreef103]

Even at 15m depth, corals are still blasted with 52W/m² of 400-440nm light, and 60W/m² of 440-480nm light. To match those output levels would be to match what they get in nature and will provide just about the best growth that you can get. Safer numbers to use would be 45W/m² of 400-440nm, and 40W/m² of 440-480nm. All you have to do is calculate how many square meters your tank covers and then figure out the radiant power of the LEDs you are getting.

 

For instance.

 

My tank measures 48" x 24" x 16". There is about 39.4" in a meter, so my tank measures 1.2192 meters by 0.6096 meters. The surface area of that (length times width) is 0.743m².

 

If I want to provide 45W/m² of 400-440nm light, multiply that number by the total area (in square meters) that you need to cover, so 45 x 0.743 = 33.44W. Then just divide that amount of power by the output of the LEDs you want to use.

 

Let's take LEDGroupBuy's hyper violet - it outputs 0.96W of radiance at full power (note that the LED uses 2.5W of power from the wall - do not confuse this with the output of the LED, which is 0.96W). So 33.44 divided by 0.96 = 34.83 LEDs needed. I would need about 35 LEDs to give my tank 45W/m² of 400-440nm light.

 

Where I got the numbers from in nature: http://www.advancedaquarist.com/2012/10/aafeature

Very interesting. Thanks for the link. So. My guess is that you recommended 6-430nm to 2-405s is because the 430s have a broader, lower peak curve while the 405s have a higher peak and more narrow spectrograph?

[jedimasterben]

Very interesting. Thanks for the link. So. My guess is that you recommended 6-430nm to 2-405s is because the 430s have a broader, lower peak curve while the 405s have a higher peak and more narrow spectrograph?

Not really. The photosynthetic peak at 430nm is more important than the 405nm peak (about 40-50% higher), so that's the one that should be focused on.

[xerophyte_nyc]

Where I got the numbers from in nature: http://www.advancedaquarist.com/2012/10/aafeature

 

A couple points I want to make, interpret as you wish.

 

First, we don't really know for sure that matching the sun's output at a certain wavelength and at a certain depth results in the best coral growth. The only claim we can make is that it is similar to natural light, not that it is optimum. The sun midday is like 10000fc but even the most sun loving plants can be grown successfully at 1500fc.

 

Second, watts per volume I don't believe is an accurate representation of light quantity. 60 watts of incandescent light is not the same as 60 watts LED. You know that. Let's try to provide some PAR or PUR values. Is there a way to accurately correlate W/m2 to PAR that is specific to the source of light? I did not look at the article you linked, is that in there?

 

EDIT: Had a chance to read the article on the train this AM, everything is there.

 

And we also need a way to account for optics and distance. Some light spills out and reduces output to the water. And of course lights hung from the ceiling won't provide the same output as measured from the water as lights suspended 10" above the water.

 

I think for our purposes we need to aim for the upper limits of light output, then use a dimmer to start low and work your way up slowly and carefully before coral bleach.

[jedimasterben]

A couple points I want to make, interpret as you wish.

 

First, we don't really know for sure that matching the sun's output at a certain wavelength and at a certain depth results in the best coral growth. The only claim we can make is that it is similar to natural light, not that it is optimum. The sun midday is like 10000fc but even the most sun loving plants can be grown successfully at 1500fc.

It is a function of several things - a full, powerful light spectra is one of four points - the others being flow, food, and nutrients (with a fifth possibly being magic, but that's for another thread ).

 

Second, watts per volume I don't believe is an accurate representation of light quantity. 60 watts of incandescent light is not the same as 60 watts LED. You know that. Let's try to provide some PAR or PUR values. Is there a way to accurately correlate W/m2 to PAR that is specific to the source of light? I did not look at the article you linked, is that in there?

It's not watts as in watts from the wall, it is watts as in radiant output. The output in radiance of a Cree or Luxeon royal blue LED is about 1.6W at 1000ma, which pulls 3.0-3.2W from the wall, so we are around 50% efficiency.

 

Yes, you can convert W/m² and Lumens/m² to PPFD (micromols/microeinsteins per square meter per second (mms) of PAR), but it is really only a guideline and numbers are not going to be particularly accurate.

 

Watts/m² = 0.21*L

Lumens/m² = 140.2*L

Lux = 140.2 *L

Lumens/ft² = 13.03*L

 

Where:

L = PAR

 

 

Keep in mind that with LEDs measuring in watts of radiance (only violet, royal blue, and deep/far/hyper red) it is more accurate than an LED measuring in lumens (cool/standard blue, cyan, white, green, yellow, amber, red), with white LEDs being the least accurate to calculate, as they cover a very broad spectrum in comparison to colored LEDs that only have a 10-20nm spread. Even PAR isn't really all that great to use, as I can get super high PAR readings from a 530nm true green LED, but it won't grow any corals. PUR is much more important, but that's not something I am aware of conversions to or from. All I know is that the more violet, blue, and red, the more PUR as those wavelengths are the most useful to photosynthesis.

 

So we take a royal blue LED that outputs 1.6W of radiance at 445nm to cover a square meter. We then divide it by 0.21 to get 7.62mms of PAR.

 

We can also take a neutral white LED that outputs 300 lumens to cover a square meter. We then divide that by 140.2 to get 2.14mms of PAR. Using Hoppy's calculator here, taking that same 300 lumen LED with no optic installed (so 120 degree), I get approximately 2.65mms of PAR per square foot at 10.4" height. I don't particularly know which is right, without having a PAR meter and being to test all of this, but I know Hoppy has done his testing a while back on older XR-E chips, but as spectrums change, so should PAR - theoretically. The XT-E cool white and neutral white have horrendous color spectrum in comparison to XP-G and XP-E chips, and it seems like that was done to improve luminous output, but that will not necessarily raise PAR in this scenario. That would require testing for me, though.

 

And we also need a way to account for optics and distance. Some light spills out and reduces output to the water. And of course lights hung from the ceiling won't provide the same output as measured from the water as lights suspended 10" above the water.

Hoppy's calculator above is great for that. He has done a lot of work with optics, height, etc.

 

I think for our purposes we need to aim for the upper limits of light output, then use a dimmer to start low and work your way up slowly and carefully before coral bleach.

Agreed. This is definitely the best way to do it! Special care being taken with white, red, and cool/standard blue channels - these three can bleach a coral with little effort because corals aren't always used to them being present and have no defenses against them. I would start very low on those channels in proportion to royal blue and violet channels.

[Paleoreef103]

Not really. The photosynthetic peak at 430nm is more important than the 405nm peak (about 40-50% higher), so that's the one that should be focused on.

Looking at the absorbtion specturm it seems like 430 is actually the most important wavelength you can get. Honestly, the nice thing about low wavelength LEDs is that they appear dim to our eyes which means we can use a lot of them to get the appropriate lighting power without purpling the tank too much. Honestly, a single 465nm Cool blue can APPEAR to overpower 3-4 violets.

This is definitely the best way to do it! Special care being taken with white, red, and cool/standard blue channels - these three can bleach a coral with little effort because corals aren't always used to them being present and have no defenses against them. I would start very low on those channels in proportion to royal blue and violet channels.

It's kind of odd, but I seem to remember more people bleaching their corals by adding violets than when they add the warmer channels. I agree whole heartedly with build high, dim low, slowly raise philosophy, btw.

[RiddleEagle18]

Are there 60 degree optics available for the 3ups and OCW??

[jedimasterben]

Looking at the absorbtion specturm it seems like 430 is actually the most important wavelength you can get. Honestly, the nice thing about low wavelength LEDs is that they appear dim to our eyes which means we can use a lot of them to get the appropriate lighting power without purpling the tank too much. Honestly, a single 465nm Cool blue can APPEAR to overpower 3-4 violets.

It's kind of odd, but I seem to remember more people bleaching their corals by adding violets than when they add the warmer channels. I agree whole heartedly with build high, dim low, slowly raise philosophy, btw.

A single cool blue is brighter than four RB chips. It's so strange how that works, just a few nm up and BAM! All up in your face!

 

A lot of people had that issue (including myself, I lost a loooooot of corals), but I'm not 100% sure that it was the violets for me. I know it was for others, though, so definitely start everything at least 50% or lower, and have the white and red channels lower.

 

Are there 60 degree optics available for the 3ups and OCW??

I think OCW have them available, but not the 3ups.

[blasterman]

Monochrome colors bleach corals faster than white. Been there -done that- seen it happen too many times. The light emitted by colored Leds and halides *does not* exist in nature other than in nuclear reactions. While ocean water filters red an yellow light the result at even great depths isn't anywhere near the monochrome emission of artifical light sources like blue Leds and halides.

 

Plants have a far easier job regulating photosynthesis with longer spectrums (white) than monochrome colors, but the full mechanism for this isn't entirely known.

 

If you look at the energy spectrum of a 20000k halide -vs- a 6000k halide the 20k halide is throwing out orders of magnitude more 450nm light. The 6000k light is throwing most of it's spectrum in the lumen peak curve which is around 530nm (green). Typically because daylight halides are designed for viewing.

 

Now ask any professional coral farmer what they'd rather have for light augmentation; a 20k halide, bunch of royal blue XTEs, or a daylight halide. All will pick the 6000k halide if they want the fastest and most natural growth.

 

I've said this multiple times and will say it again; the reason we're using all these goofy colors on our tanks is to make them look kewl. The coral doesn't give a sh_t. Optics are also counter productive because they force collimated light the produces un-natural growth. Heavy growth SPS tanks lit with T5's have corals that look like basketballs and are round. Tanks lit with opticaly pumped LEDs look like something half dead from Mordor.

[VABumpkin]

I've got this concept that's been on my mind lately and I thought others would like to hear it: A large LED Diode that emits light just like a Metal Halide. It would have a fully blended full spectrum, not distinctly colored wavelengths barely blended together, resulting in better visual appeal, coral growth and color. The light would have a 20W option, similar to a 150W Halide, a 30W option similar to a 250W Halide, and a 45W option to mimic a 400W Halide. It would snap into a fixture just like a Metal Halide or T5, emit little to no heat, and be totally dimmable and controllable like any LED. IMO this utilizes the benefits LEDs and traditional lighting in one small package. Hopefully someday this will be possible and at a reasonable price.

 

EDIT: Also, color options that are similar to 10K, 14K, 20K.

[jedimasterben]

I've got this concept that's been on my mind lately and I thought others would like to hear it: A large LED Diode that emits light just like a Metal Halide. It would have a fully blended full spectrum, not distinctly colored wavelengths barely blended together, resulting in better visual appeal, coral growth and color. The light would have a 20W option, similar to a 150W Halide, a 30W option similar to a 250W Halide, and a 45W option to mimic a 400W Halide. It would snap into a fixture just like a Metal Halide or T5, emit little to no heat, and be totally dimmable and controllable like any LED. IMO this utilizes the benefits LEDs and traditional lighting in one small package. Hopefully someday this will be possible and at a reasonable price.

 

EDIT: Also, color options that are similar to 10K, 14K, 20K.

LEDs don't really work that way, unfortunately.

[VABumpkin]

LEDs don't really work that way, unfortunately.

 

A boy can dream can't he?

[VABumpkin]

Actually after looking at the Kessil A350 I realized led's do work that way. The only problem is the fixtures kessil creates aren't full spectrum, only blue and white. They use several small diodes all on a single board, arranged very closely. If you look at spectrum charts for 14k halides, you will see that most colors are evenly distributed, usually with heaviness in the mid 400's and 500's. I hope we will see this technology in the future.

[Paleoreef103]

Actually after looking at the Kessil A350 I realized led's do work that way. The only problem is the fixtures kessil creates aren't full spectrum, only blue and white. They use several small diodes all on a single board, arranged very closely. If you look at spectrum charts for 14k halides, you will see that most colors are evenly distributed, usually with heaviness in the mid 400's and 500's. I hope we will see this technology in the future.

Not really. Kessil LEDs are multi-chip arrays with a common optic. It's ridiculous difficult to have controllable multi-chip arrays because you have to pack everything into a tiny package. There is a reason that the Kessil 350 only has two color channels (though I THINK they have a few violets on the blue channel. I could be wrong).

[VABumpkin]

Okay, thanks for clarifying. Maybe the manufacturer could put the right amount of each color diode in so that the light as a whole matches the corresponding kelvin temp. And from there it would be controllable with one-channel.

[uxdesigner]

Rapid LED 24 LED DIY kit (inspired by Radion XR30w Pro) in a custom industrial style pendant

• Channel 1 (Blue / 8am - 8pm) - 10 Royal Blue Cree XT-E
• Channel 2 (Color Matrix / 10am - 6pm) - 2 Blue Cree XP-E, 2 Red-Orange Cree XP-E, 2 Cree UV, 2 Green Cree XP-E
• Channel 3 (White / 12pm - 4 pm) - 1 Warm White Cree XP-G, 4 Cool White Cree XM-L, 3 Royal Blue Cree XT-E

[jedimasterben]

I'd remove the green and red chips.

[uxdesigner]

Why remove the green and red? If your going to make a bold statement please explain so that others can learn and you don't come across as negative.

 

That said, I'm getting amazing colors and growth in a mixed reef. I've built many LEDs lights and I've had much success with each one. I recommend the blend I have to anyone and the tank to prove it, reason why I'm sharing.

[jedimasterben]

Green (530nm) is not used in photosynthesis in any significant amount (nearly zero, and mostly is used by unwanted algaes) and all white LEDs contain too much green as it is.

 

Red (630nm) is not used much for photosynthesis in corals, but is heavier in macroalgaes since they contain chlorophyll b. Neutral and warm white LEDs contain plenty of 630nm light, it is 660nm that they are lacking in.

[uxdesigner]

Great insight and thanks for taking the time to explain.

 

I'm luckily not experiencing any excessive algae growth and the color pop is well worth it.

 

I know the retailers offer what sells but if its not harmful and adds visual value I say it's all worth it.

[bendel]

Green (530nm) is not used in photosynthesis in any significant amount (nearly zero, and mostly is used by unwanted algaes) and all white LEDs contain too much green as it is.

 

Red (630nm) is not used much for photosynthesis in corals, but is heavier in macroalgaes since they contain chlorophyll b. Neutral and warm white LEDs contain plenty of 630nm light, it is 660nm that they are lacking in.

 

Saying corals lack 660nm is like saying the lack phosphat. In a natural reef it just exist only in very small amounts. In 10m water depth there is only around 2% of the surface intensity in the 660nm range.

Also the green might not be use by the chlorophyll. But there is a lot more going on, biological systems are never simple. So that might we useful for some processes. See http://www.advancedaquarist.com/2013/2/aafeature for some experiments.

 

Steffen

[jedimasterben]

Saying corals lack 660nm is like saying the lack phosphat. In a natural reef it just exist only in very small amounts. In 10m water depth there is only around 2% of the surface intensity in the 660nm range.

Also the green might not be use by the chlorophyll. But there is a lot more going on, biological systems are never simple. So that might we useful for some processes. See http://www.advancedaquarist.com/2013/2/aafeature for some experiments.

 

Steffen

I never said 660nm should be your primary color, nor that you even need a lot of them. They are supplemental and help with color rendition and the chlorophyll a band.

 

Chlorophyll is the primary producer - while it is important to ensure the accessory pigments have enough light, white LEDs have more than enough spectral coverage to take care of them. Green is wholly unnecessary, and I stand behind that still.