LED lighting

[davelin315]

I have been looking at the possibilities of LED lighting but do not have enough of a lighting background to understand what to do with them. I know that there are different colored LEDs that will combine to make "white" light and also that the more colors present from the spectrum the more representative it will be of natural sunlight, but what I don't understand is the output (PAR).

 

I have been looking into LEDs with an output of "18,000 mcd" but have no idea what mcd is... I have also read some of the threads on reef central and the threads that are referenced in them from other sites, but can't make heads or tails of them, except that most seem to think that it would take a whole lot of LEDs to light a system.

 

Is anyone familiar enough with them to state the feasibility of them on a reef system?

[Guest Keyoke]

I have been looking at the possibilities of LED lighting but do not have enough of a lighting background to understand what to do with them.  I know that there are different colored LEDs that will combine to make "white" light and also that the more colors present from the spectrum the more representative it will be of natural sunlight, but what I don't understand is the output (PAR).

 

I have been looking into LEDs with an output of "18,000 mcd" but have no idea what mcd is... I have also read some of the threads on reef central and the threads that are referenced in them from other sites, but can't make heads or tails of them, except that most seem to think that it would take a whole lot of LEDs to light a system.

 

Is anyone familiar enough with them to state the feasibility of them on a reef system?

32680[/snapback]

 

I'm not an expert, and I invite any of the folks to correct any of the info I'm about to provide...

 

The problem isn't going to be the quantity of light you can produce - with enough LEDs you can rival the sun in sheer intensity. MCD stands for 'Millicandela' and represents how bright a LED is. MCD = 1/1000th of a Candella, which is different than what we usually mesure the intensity of our lighting with in the Aquaria (and just about everywhere else with) as an expression of Wattage (tho it's really LUMENS that tell us how bright the light is, it just so happens that usually as wattage increases, so does lumens.. more on that in a minute)

 

Usually, we look at three things in lighting (aside from wattage)- 1. Wavelengths produced that are useable for photosynthesis (PAR), 2. Color of the light source as expressed as degrees Kelvin, and 3. the bulb's ability to faithfully reproduce color (CRI).

 

1. PAR has to do with the measurement of a light's output in ranges that are useful for photosynthesis -- in a nut shell, you can think of it as a measurement of the quality of light compared to the animal's ability to synthesise food from it. Higher PAR = More light usable for photosynthesis.

 

2. Degrees of Kelvin can give us a good idea what the color of the light is going to be. If you want to get technical you can look up what the Kelvin temp actually means in relation to a black body, but I don't fully understand that part myself - it's a physics thing.

 

Regardless, the temp of the light will give you a good idea as to the color the light produces. Lower temp lighting will produce a more reddish color, while higher temps produces a more blue color.

 

The reason that color is important, is because seawater (most water for that matter) absorbs all but the blue wavelengths of light. So while there may be 2700K "red" lighting that has a high PAR, in an aquarium setting when shining that light thru water, most of that light would be absorbed by the water itself, and never reach the critters in your tank. That's why the Sky is blue too by the way.. moisture in the air absorbs most of the non-blue wavelengths of light.

 

Which is why we aquarists usually stick with bulbs in the 10 - 20K range, but the horticulturists out there will use a more yellow/red light in the 5500-6500K range. Aquarists need high PAR that's mostly blue wavelengths to pass through the water, while folks that grow marijuana in their closet don't have 2 feet of water to shove their photons through, thus there's no absorption of reds and yellows, and thus the animals they're trying to feed have adapted to use more of the red spectrum than blue for photosynthesis.

 

3. Your also looking at CRI - CRI isn't nearly as important as the PAR and K of your lamp and is more of a personal preference. CRI measures how faithful the colors under the lamp are to natural daylight (i think it's daylight..) - lower CRIs mean that colors might be distorted, blues might be greener, or yellows redder than they are normally.

 

So to be truely effective in our lighting, we want lights that are going to be 10-20K in temp, have a high PAR, and a decent CRI (to taste).

 

Now I don't know what the LEDs your looking at are capable of. You could probably combine a bunch of super-bright-white LEDs, and a handful of blue ones here and there to create an approximation of color that a 10K+Actinic bulb combination would produce - color wise. I don't know how much useable-for-photosynthisys light the LED will produce. It'd be an interesting experiment though.

 

Immagine.. light w/o a ballast.

 

I had a heck of a time finding the lumens produced by a 175W MH lamp at 10K, I did manage to find one with an unknown mfgr that listed lumens at 18000..

 

Lumens by the way, is the measure of the amount of light AT the Lamp. It's different than Candella.. There's no direct conversion, but according to what I was able to dig up on Google, if you divide the # of lumens by 12.57, you can get the approximate # of Candella of the light.

 

So just to see how many LEDs we'd need if they were 18,000 mcd...

 

1800 lumens / 12.57 = ~ 1432 Candella

1432 Candella * 1000 = 1432000 mcd

1432000 mcd / 18000 mcd = ~ 79 LEDs

 

Feasable.. Though another problsm is going to be viewing angle on these. The LED should be rated with a degree of visibility, they aren't omnidirectional sources like standard bulbs are. I think the widest I've seen was a 45 degree viewing angle, which means that the LED is going to project a fairly tight cirlce of light...

 

Ok, so that was the long answer. The short answer? Might be able to get away with it on a nano.. Dunno if you could do it on a larger tank though.

 

now I'm curious... maybe I'll go buy 200 LEDs and see what I can do with 'em.

[davelin315]

I am familiar with kelvin, but don't know the conversion for nm to kelvin. The "color" indications they give are how many nm they are. For instance, there is a lot of information on moon lights, which are at closest 472 nm. Actual moonlight from what I understand is 475 nm. Some of the LEDs are listed by Kelvin, and I have seen up to and around 5500 or more, which is what the basic bulbs are that you could get 10 years ago when I got my first metal halide. Some of them simply list the color temperature by giving x and y coordinates, although when I compare this to a color chart for kelvin, I don't understand how it all fits. For instance, many of them say the color is x=.32, y=.31 or something close to that. I can't for the life of me figure that one out as the color temperatures are on a different scale from what I can see. I think a lot of it is putting one graph together with another. As far as the blue coloration, I think it's actually that blue is not filtered out by the water, it's actually transmitted and reflected. Things that appear a certain color are that way because they reflect the light color that is being shined on them (for instance, that is why a red object appears red, it absorbs all the other colors and reflects red back, hence you can see it with your eye).

 

On to PAR, how do you calculate it? Anyone know?

 

I think that LEDs are cost efficient over the long haul, and even the 15 degree ones could be used because if you stack enough of them, you will have a very precise coverage depending on distance from the surface. Also, I would think that with how water refracts light, the 15 degrees would be diffused over the surface area of a reef. In addition, because they produce little or no heat and use such a small voltage, I would think you could mount them on top of the water, or at least within an inch...

 

Let me know if you'd like to try it, I definitely would as well...

[emissary]

Lumens by the way, is the measure of the amount of light AT the Lamp. It's different than Candella.. There's no direct conversion, but according to what I was able to dig up on Google, if you divide the # of lumens by 12.57, you can get the approximate # of Candella of the light.

 

And candela, by the way, is based on the age old power of a candle. 1cd is about the luminous intensity that a candle puts out. So... just hang a few hundred thousand candles over your tanks!!

 

(brilliant coverage Keyoke)

[Guest Keyoke]

I am familiar with kelvin, but don't know the conversion for nm to kelvin.  The "color" indications they give are how many nm they are.

 

Actually, the Kelvin is the overall color temprature of the bulb, and not nessacarily the wavelengths produced, tho there is some direct corelation if your looking at a spectral graph. I'd encourage you to take a look at "The Color of Light" page provided by Venture Lighting (there's a lot of good stuff in the Tech Center there, tho be aware they're obviously biased towards their own products!) Venture Lighting Tech Center

 

On to PAR, how do you calculate it?  Anyone know?

 

AFAIK, you actually need a meter for this. I know they're available but I have no idea on the cost associated with them. Focus PAR Meter is an example of one, though it's geared towards terrestrial plants - you prolly don't want to stick this under water.

 

I think that LEDs are cost efficient over the long haul, and even the 15 degree ones could be used because if you stack enough of them, you will have a very precise coverage depending on distance from the surface.  Also, I would think that with how water refracts light, the 15 degrees would be diffused over the surface area of a reef.  In addition, because they produce little or no heat and use such a small voltage, I would think you could mount them on top of the water, or at least within an inch...

 

I agree, LEDs would be extremely cost-efficent, provided we could locate ones with the proper wavelengths. The problems with LEDs is that they tend to produce a narrow band of color - that is, the spectrum provided by the bulb is extremely narrow. Blue LEDs for example, are rated at a SINGLE wavelength of light - as you pointed out above, around the 470 nm mark. White ones tho.. well by their nature they would HAVE to be full-spectrum or the light wouldn't be white, right?

 

Huh! I actually just pulled up a spectral analasys chart on a White LED, The link is Here (the superbright site has a lot of good information as well) - According to the chart there, it looks almost as if light output peaks in the 420 - 500nm Range.. with an ideal PAR of 400-700 nm, you might be able to get away with lighting a tank designed for deep water with these (Deep water Acros for example), the animals that utlize more of the wavelengths in the 500-700nm (your shallower animals) would be out of luck though...

 

If LEDs could be found with a broader spectrum output, these would be GREAT lighting sources. Unlike MH, or the various types of FL tubes, LEDs don't fluxuate in temprature/wavelengths produced as they age.. A 10K LED, if there were such a thing, would still be 10K a year after it was switched on, while MH lamps tend to loose their color the longer they burn.

[Guest Keyoke]

Further food for thought.

 

Out of curiousity, I did a search for 'LED Grow Lights', just to see on Google (darn I love google...)

 

And They Exist! They look cool too... the site claims that PAR is 100%..

 

There are both RED and BLUE versions of these, Aquatically speaking, we'd be more interested in the blue than the red, tho it might not be a bad idea to have 1 or two red ones...

 

Now I'm _very_ curious. There's a link to a study done by the U of Minn on LED grow lights, which mentions that they are being studied by NASA for growing plants in space..

 

OK. I think I might just try this.. I've got a 10G tank that's actually partitioned.. I could paint the acryllic black and do 1/2 of the tank with LED, and the other half with conventional lighting..

[davelin315]

Please let me know if you do. In 5th grade we cover light, energy and electricity, plants (and therefore photosynthesis), living systems (includes oceans, ecosystems, and classification) as part of our curriculum as well as experimental design. I can foresee a very cost efficient experiment that would teach the kids how to wire up their own little reefs on their desks using LEDs and batteries and maybe small jars. Using airline tubes and simple designs, they could run experiments with single polyps or zooanthids on their desks and write experimental design diagrams for it. We could then report back our findings of how things did using various types of LEDs.

[Guest Keyoke]

Please let me know if you do.  In 5th grade we cover light, energy and electricity, plants (and therefore photosynthesis), living systems (includes oceans, ecosystems, and classification) as part of our curriculum as well as experimental design.  I can foresee a very cost efficient experiment that would teach the kids how to wire up their own little reefs on their desks using LEDs and batteries and maybe small jars.  Using airline tubes and simple designs, they could run experiments with single polyps or zooanthids on their desks and write experimental design diagrams for it.  We could then report back our findings of how things did using various types of LEDs.

32734[/snapback]

 

Sheesh, where were you when I was in 5th grade? Man that sounds like a blast... I miss science classes, I always did so well in 'em...

 

Dang I wanna build this now. It's full of all kinds of things I like. Reefs, Electronics, challenge, Danger of Electrocution.. I even got a pico-reef sized frag of frogspawn that'll do quite nicely in a small tank - say something the size of a Mason jar. An airline tube and a small pump could be used to provide circulation by displacing the water with bubbles..

 

Hmm. Neat experiment. I dunno about battery life though. For a mason jar, I could see 5 - 10 LEDs, but your looking at a 9V battery at least, maybe two (I'd have to do the math) - unless they were rechargable, or supplemented with a wall-wart..

 

Ok, so I'll build one just to see. It'll be neat if it works.

[craby]

More on Candella:

1 candella is the amount of light 1 candle produces onto 1 sf of surface 1 ft away, and foot candella is the measurement of light produced from the light source at a certain distance. IE: if (1) foot candella or fcd is desired 3' below the light source, then the light source would have to produce 9 candella. Make sense? This is important as you concider the depth of your aquarium & distance the critters are away from the light source.