DIY LED Lighting Concept

[chevyrulzs2010]

Okay, first let me start off by saying this is just a concept. I do not claim to be an expert in lighting, leds, or electronics. I do not plan on building this in the near future, but i would like to use this thread as a central location for ideas for led lighting. I truely believe that LEDs are going to be the way to go in the future. I am just taking information I have gathered from other sources and bringing it here.
The Concept
In theory the three colors you need to make any color out there is Red, Green, and Blue. This is also the color of the pixels on your computer screen or Television set. If you made a LED hood light, you could obtain any temp. color you wanted. This would be the best method of LED lighting so you dont have to make a decision about what color leds to buy to get the right effect. This also allow you to adjust for better or worse plant growth in your tank by including or blocking the spectrums that the plant life needs to grow.
With the three color leds giving you the ability to adjust the color of your lighting you could also program (with some knowledge of computer programing) an internal clock to change the color of the lights as the day moves on, morning colors and dimmer light, high noon colors, and evening red colors that dim. Your blue leds could also be programed that only a few stay on after dark to provide your luner light.
The Questions
1. How much area does a single LED cover.
2. Would you need a solid pannel of LEDs or could you just use 6 or 7 RGB modules.
3. What is the PAR rating on LED lights, do they provide the light needed for life to thrive or are they just lights that simply add color.
4. Would Red Green and Blue truely give you the ability to reach any color. (I saw a project with White Green and Blue set up in strips of 5 WGBGW)
This concept is not yet finished, i am still researching the LEDs and adding to my ideas. Please feel free to post any information you know, or any ideas you have. Please dont post "this will cost to much" or anything along the lines of high priced. I havent even looked at prices yet. This is just a concept, if you dont want to pay the money, dont build it.

 

[swimmer4uus]

I'll try to make this short, and hopefully comprehendable. I haven't had my coffee yet.
Conceptually you are correct. I think when you take this, and try to apply it, you run into some flaws. Take what I'm starting to build for example. I have a 48" by 18" retanglular area that I need to cover(90 gallon tank). Ideally, I would want something in the range of 300 to 500 LEDs. Lets use 300 to make the math simple. So, that's 100 LEDs of each color. Now...in an even application, (all things else being constant), one would probably run 1/3 of each color (and I think your color combos are off). One would also probably be running blues 100% capacity, and the others varying...most tanks are like this. So out of 300 LEDs, you're only probably using 60~80% of their actual capacity, meaning less par, less lumen...etc. I believe a better mix of coloring is actually 50% blue, 50% white. Blue LEDs actually produce more lumens, than other colors of same build.
The possibility for dimmig is very promising. Too bad that stupid company has the patent under wraps, and is prohibiting other companies from making process in this area. Most your control is going to be DIY, and I'm just not up to par in that area yet. Although I want to work with some Electical Engineers at school to see if we can make a driver for this. Maybe have a senor do it as their senor project.
As for your questions....
1. LED coverage varies greatly with the build of the LED, and the depth at which you want to cover. Also, optics can be used to increase the penetration of the LED to lower levels, but you decrease the angular spread of the LED this way. It's really a factor of the type of LED used, and the viewing angle.
2. I'm pretty sure you could do either. I can't comment on the RGBs. My array is coming out to be 288 3W LEDs(these bad boys add up cost VERY fast), arranged the length of the tank. First row of banks only though, as I have the back half of the tank to design differently.
3. Par rating on LEDs is difficult. It depends greatly at the efficiency that the LED is being driven to, and the depth of the tank. Really the only way to tell is to measure it in the tank, after installation.
4. This is probably answered in my begining paragraphs.
I believe the biggest area for customization, and progress for LEDs, is in the circuit design of the drivers. To just get the LEDs up and running, one can DIY your own driver for a couple bucks (1 driver per 6 3W LEDs, I'm using 48 drivers). Really cuts down on cost. Also, one could DIY a more complex driver, set it up with a computer control board, and have automatic dimming, and a truely realistic setup of actual atmospheric conditions of gradual brightness increase and decrease. Or, you could just buy a premade. But that's where the system will either excel, or not.
My 2 cents....

 

[scsinet]

I don't know why you would want to produce "any color." IMO white and blue LEDs are probably all you need to re-create most natural lighting scenarios.
The ability to dim the different colors is definitely possible - in fact this is what PFO did with the Solaris - to acheive different effects. By independantly varying the white and blue sets of LEDs, you should be able to achieve a broad color spectrum.
Conceivably, you could probably even dim individual columns of lamps. If you were savvy enough, you could tie these into a serial bus driven by a PIC, Atmel, etc microcontroller, and drive them all from a computer.... once there, you could do all sorts of things... dusk to dawn, lunar cycles, cloudy days... even clouds passing over or lightning could be simulated. There are really several pieces here... each dimmable channel would need a PWM driver circuit to run the LEDs and mange their current consumption, then an interface to a data bus that a Microcontroller could communicate with, then a serial interface to a PC. It's all doable, but you need quite a background in electronics to accomplish this, as well as how to program microcontrollers, which usually requires experience in assembly language.

 

[chevyrulzs2010]

Ok on a beginning note I'm glad someone was able to understand my babbling on about nothing I typed that this morning after work and was tired from a late night. Tomorrow I will revise it and make it more understandable and will continue to add information as I learn
Now I realize that this whole thing takes a lot of knowledge to make successful but I believe anyone with a lot of time and dedication and a strong desire to do this would be able to pull it off.
Thanks to thos who replied your arguments provided just what I needed to make sure my first post was clear and also provided me with information. I need to study.
About the comment about why you would need to make any color possible... My thinking is not that I want odd colors for my aquarium but that with my knowledge of lighting at work (work at petsmart) different bulbs for different uses can be rather pink or purple but I don't see it possible to achieve the color of a power glow 2 bulb with just white and blue or even green in the mix. This also applies to the rising sun and setting sun colors. You would need red to achieve the correct color. I have read somewhere that red light also helps with coral growth (unsure on source or if true). My question would be how well could you get the light to mix from the RGB leds.
I will work on my concept some more and correct any problems I see. I hope more people on the forums will join in this. I know that as a whole we have the knowledge to make this work.
Next update: layout and design

 

[geoj]

Photosynthetically active radiation, often abbreviated PAR, designates the spectral range (wave band) of solar radiation from 400 to 700 nanometers that photosynthetic organisms are able to use in the process of photosynthesis. This spectral region corresponds more or less with the range of light visible to the human eye.
Since the impression of white is obtained by three summations of light intensity across the visible spectrum, the number of combinations of light wavelengths that produce the sensation of white is practically infinite. There are a number of different white light sources such as the midday Sun, incandescent lamps, fluorescent lamps and white LEDs; due to the phenomenon of metamerism. The impression of white light can also be created by mixing appropriate intensities of the primary colors of light, red, green and blue (RGB), a process called additive mixing, as seen in many display technologies. Such sources can appear similar whilst having quite different spectra.
From Wikipedia, the free encyclopedia

 

[geoj]

White light
There are two primary ways of producing high intensity white-light using LEDs. One is to use individual LEDs that emit three primary colors[44] – red, green, and blue, and then mix all the colors to produce white light. The other is to use a phosphor material to convert monochromatic light from a blue or UV LED to broad-spectrum white light, much in the same way a fluorescent light bulb works.
Attachment 238041
Combined spectral curves for blue, yellow-green, and high brightness red solid-state semiconductor LEDs. FWHM spectral bandwidth is approximately 24–27 nm for all three colors.
Attachment 238049
Spectrum of a “white” LED clearly showing blue light which is directly emitted by the GaN-based LED (peak at about 465 nm) and the more broadband Stokes-shifted light emitted by the Ce3+:YAG phosphor which emits at roughly 500–700 nm.

 

[king_neptune]

Originally Posted by SCSInet
http:///forum/post/3220238
I don't know why you would want to produce "any color." IMO white and blue LEDs are probably all you need to re-create most natural lighting scenarios.
Ive seen posts on other sites a while back on this topic.
Evidently throwing in hints of green and red LED's can really cause an explosion in reflected light off the corals. Kinda how everything lights up neon under Actinic's.
Something like 36white/12Blue/6green.
Im not being exact...but you get the idea. Tossing in a splash of a third color evidently pays off Aesthetically in major ways.

 

[scsinet]

Interesting...

 

[tim514]

Im not much of an electrical engineer so I dont know much about LED's and such. But what type of LED's should one use? PN? Manufacturer? Type? Ect.
Where would you purchase such LED's for this application? Radioshack? Homedepot? Online?

 

[king_neptune]

Cree and lumex are good manufacturers or LED's.
3w is a popular choice. But there is so much more to take into consideration. Buck packs, drivers ect. And you wont really be finding what you need at radio shack. fortunatly you can order everything online.
Google:
evilcc66 led guide
that will get you pointed int he right direction. this guy is basically the LED guru for anything Reef keeping.

 

[jstdv8]

why can't a person just buy this setup? Is there no company that puts it all together, dimable, effective amounts of LED's for your gallons and water depth? throw in a controler to take the place of the computer so you get true daylight hours.
I havn't seen it, why is that?

 

[wangotango]

Originally Posted by Jstdv8
http:///forum/post/3242263
why can't a person just buy this setup? Is there no company that puts it all together, dimable, effective amounts of LED's for your gallons and water depth? throw in a controler to take the place of the computer so you get true daylight hours.
I havn't seen it, why is that?
The only good fully assembled LED fixture is the Aquailluminations unit, but you're looking at some serious coin if you wanted one. The main reason why there aren't more LED fixtures available in the US is because of a patent lawsuit against PFO (which is why PFO doesn't exist anymore). It is not easy for companies to do much in the way of controlling the LEDs like outlined in the Orbitech patent. AI is one of the few that can and does.
The Chinese have tons of LED fixtures which are dirt cheap in comparison to the Aquailluminations, but they are crappy knock-offs.
At this point the only real economical way to go LED is to build it yourself, either with the CREE xr-e or xp-g LEDs and optics, or the LED PAR38 lamps.

 

[jstdv8]

ok next question,
I looked up the aquaalumination system. Seems pretty cool, exactly what I was talking about. so for a 48" tank you get a system with 4 12" banks with 24 lights each, thats 96 lights, but this thread is talking about using 300 bulbs. is the AI unit bulbs that much better? I don't see where it even mentions what bulbs come with the unit, it just claims to be as powerful as a system with a 250w MH. so if a guy wanted to go and blow 2400 on that setup how can you be sure exactly what you're gettting?
does the other guy need 300 lights becaause they are weak? I'm not really in the market at this time, I just find the whole total daylight control thing really interesting and think that everyone will have these on thier tanks as soon as that lawsuit thing gets handled and the prices start coming down.

 

[wangotango]

For the tank the OP mentioned, 300 LEDs is excessive and just not worth the money going into it. There is no "rule" for how many LEDs you need over a given tank. A lot of numbering has to do with the optic you use (for DIY units anyway). The average seems to be one LED (3w CREE XR-E or XP-G) per every 10-20 square inches of tank surface. The optic you use depends on how deep the tank is.
The AI unit uses the CREE XR-E and drives them at 1000mA I believe. Their units have the LEDs in clusters and use a tighter optic, so there is more of a spotlighting effect depending on how high you hang it.
I don't think AI released PAR tests, but there are other LED projects that have PAR tests at given depths, and a good setup is comparable to the typical metal halide fixture. If you're using tight optics and small spacing there is no reason why you can't get 400w halide PAR numbers.
Hope that helps. There is a lot of information on other sites. Plus knowing Sanjay Joshi helps

 

[jstdv8]

TW,
So when you say tank surface area you mean all glass surfaces? orthe surface of the water? and then do you also include the bottom as a surface area?
The water surface alone would need 10 lights by your rule for my 90 gallon.
So I take my 48" I'm calling it 50 for easy math x 25" im calling it 30 so thats 150 x 2 for the back is 3000, sides are 30x 20 so 600 x 2 = 1200 for a total of 4200 not including the bottom or the water surface. 4200 sqin
devided by 15 which would be the median range of 10-20= 280 lights.
So doing that math seems like alot and just the water surface seems like not nearly enough.

 

[wangotango]

Maybe footprint would be a better choice of word. You're not passing light through the vertical surfaces so I don't know why you would even consider those. For my 55gal (48x12) I was planning on 48 LEDs on about 2.5-3" OC spacing. But honestly, spacing and optics has more to do with the total number of LEDs you use than anything than that guideline.
Google search for some of the other DIY LED builds (there's a lot of them out there).

 

[jstdv8]

I'm still lost. I'll go check out the other threads.

 

[jstdv8]

ok, I checked out soundwaves extremely large thread on another site. he as some 4000+ posts. i read the first 1000 and the last 100.
He seemed to get pretty good par and good spread by not using any optics and puting the bulbs fairly close to the water. and that was a year ago.
had 48 bulbs on his 75 gallon.
most of it gets pretty technical about how many LED's you can run off of a 24v power supply and buckpuck ect ect.
I'm guessing sooner or later they are gonna come up with something that plugs into the wall, has decent spread and is dimmable all in one
Pretty cool looking at some peoples builds. works of art really.