DIY LEDs - The write-up

[skippyreef]

Here is the one I was looking at.

http://mouser.com/ProductDetail/Led...GAEpiMZZMsgllGlynFdfrWRskIe%2bjpERY9/Lb1W7ys=

I was thinking with 60 degree optics I might be able to add these to some of the arrays.

 

[der_wille_zur_macht]

IMHO if you're adding a small number of any color of LED, you probably don't want optics. If there are only a few of a given LED spaced out over the tank, optics will make that color really "splotchy."

FWIW I wouldn't get wrapped up in spacing. Determined the total number of LEDs you need to get the intensity you want, then just space evenly over a reasonably-sized heatsink.

 

[jerwin]

So as I read for the T5 lights the 420nm is used for coral growth and the 460nm is for visual. If we are putting LEDs that are in the 465 and above and nothing in the 420nm are we actually not helping our corals by doing this? I know I am just an newbie at this but, I know we all love our tanks and we want to do what is best for them.

 

[der_wille_zur_macht]

IMHO it's a VERY steep and slippery slope that quickly leads to misinformation when you start to try to guess which wavelengths of light will produce certain results (better growth, better coloration, etc.) There is very little actual scientifically-generated data out there on the subject, and that data covers an extremely narrow range of circumstances that aren't likely to apply to a typical hobbyist's tank.

The one thing we do know is that the "standard" mix of royal blues and cool whites is capable of producing good colors and good growth, so IMHO the best approach is to not over-analyze things or try to extrapolate based on results from other types of lighting. We're certainly not missing out on growth or coloration by not having 420nm LEDs. If you want to experiment with other combinations of colors or UV LEDs, then by all means go ahead and report back your findings - but do it as an academic venture, not because there's any obvious deficiency with the standard approach.

 

[skippyreef]

Well said

I get what you're saying about the splotchy efffect when using just a few of a certain color temp mingled with a more consistent pattern.

 

[jerwin]

Thanks der_wille_zur_macht. I want to do the LED project but I want to first do it to help the corals. I Just want to have all my "ducks in a row" and not missing anything before I start the project.

 

[der_wille_zur_macht]

One thing to keep in mind is that if you get results you don't like, it's *very* easy to change things around - swap out a few LEDs for different colors, adjust drive currents, etc.

 

[skippyreef]

The versatility of this type of lighting will take some getting used to. I mean to be able to dim the whites, swap a few colors here and there, accent with high power spots, use angled aluminum to aim under braces......it is like a whole new approach.

 

[jerwin]

I have not seen any one use some kind of strip that the LEDs can plug into. I know we are doing them in serial. But it would sure be nice if we could have a power side and a ground side. and just plug the LEDs into that. That way if you do want to change a single LED out for a different color or need to replace one. You would not have to unsolder and resolder. You would just put the ends on and mount and then attach the ends of the wire to a power and ground block.

Even yet would be better if we had something that the starts would snap into that could be mounted and just change the stars out.

 

[ReefEnabler]

do a search for threads by spacedcowboy, he is attempting to build a modular rig that snaps and holds LEDs together just like you describe.

He's done some actual prototyping also.

 

[madmike]

Does anyone see a problem with this setup.

I am going to run a 56" U channel aluminum strip along the back of my 5 foot ATI T5 fixture. I want to run 8 RB and 4 white XPG with 4" spacing in a BWBBWBBWBBWB pattern. All i'm going for in this setup is for shimmer and the ability to "tweak" the actinic color of the current setup. The ATI is hung about 6-8" from the water surface, will this be a problem with spotlighting at 4" spacing if I dont use optics? I have the meanwell D driver so I can dim as needed to desired result. If this test run goes ok I would like to build another for the other fixture (tank is 12 feet long) and run the blues off one meanwell and the whites and a few blues off the other so I can really tweak colors

let me know if you see any flaws in my logic. My main concern is the 4 inch spacing at only 6-8 inches off the water surface causing some weird spotlighting...

 

[SpacedCowboy]

I have not seen any one use some kind of strip that the LEDs can plug into. I know we are doing them in serial. But it would sure be nice if we could have a power side and a ground side. and just plug the LEDs into that. That way if you do want to change a single LED out for a different color or need to replace one. You would not have to unsolder and resolder. You would just put the ends on and mount and then attach the ends of the wire to a power and ground block.

Even yet would be better if we had something that the starts would snap into that could be mounted and just change the stars out.

Look here for the announcement, and here for some more progress. You can see the top and bottom of the model design, and the corresponding top and bottom of the printed parts (photos taken without me trimming off the excess "raft" plastic - they can look a lot better )

I'm hopeful that this weekend, I'll have enough time to "print" out a few more clips and I'll see what it's like to have a series-of-6 LEDs mounted on the bar, fed by my CAT4101 driver.

The plan has morphed (slightly) from the original - I intend to use clear plastic not opaque white, and probably switch to a module that holds the LED in-place on both sides, rather than one that locks into a single side (thus requiring N+1 modules to hold N LED's)

So, it ought to look like shiny Al bars, with clear-plastic mouldings holding the LEDs in-place, and shiny copper strips going from LED to LED. I think the combination of the metals will look rather nice

I've just ordered a cheap CNC mill to help with some of the larger parts (my wife has christened the garage "CNC-central" - there's the CNC PCB milling machine, the CNC 3d printer, and now there'll be the more heavy-duty CNC mill).

Hopefully I'll have an update after the weekend. We'll see

Simon

 

[der_wille_zur_macht]

let me know if you see any flaws in my logic. My main concern is the 4 inch spacing at only 6-8 inches off the water surface causing some weird spotlighting...

Putting the channel on the back of the fixture might create a "weird" appearance in terms of front/back differences in lighting, depending on your rock stack - corals on the front of a pile of rock won't be illuminated. If your setup is a typical arrangement where the "back" of the tank is against a wall and most viewing is done through the front panel, I'd consider putting the U-channel on the front of your current fixture.

my wife has christened the garage "CNC-central" - there's the CNC PCB milling machine, the CNC 3d printer, and now there'll be the more heavy-duty CNC mill

LUCKY! :lol:

 

[The Grim Reefer]

So as I read for the T5 lights the 420nm is used for coral growth and the 460nm is for visual. If we are putting LEDs that are in the 465 and above and nothing in the 420nm are we actually not helping our corals by doing this? I know I am just an newbie at this but, I know we all love our tanks and we want to do what is best for them.

I don't know who keeps putting that out there but based on personal experience and a whole lot of reading it's bunk.

 

[The Grim Reefer]

For what it is worth in dealing with T5's I found running blue lamps that are supposed to be in the 450nm range actinic lamps don't really make a difference with all the lights on. Looking at the Royal Blue LED spectral output I suspect they will be the same. I may order up some of these in 403nm to prove the theory

http://www.fosterandsmithaquatics.com/product/prod_display.cfm?c=3578+3733+10704+21598&pcatid=21598

If nothing else they would make a nice Dusk/dawn set

 

[der_wille_zur_macht]

For what it is worth in dealing with T5's I found running blue lamps that are supposed to be in the 450nm range actinic lamps don't really make a difference with all the lights on.

From personal experience I can tell you that the RB LEDs do make a significant visual difference! Not sure how to "prove" it in terms of spectrum and numbers, but you can definitely see the light they produce and differences when you're dimming.

 

[jerwin]

well I think the way to really tell this would be to take one tank and divide it in half with something dark then put a couple of the same frags on each side. Then put a 1 (white light and Royal blue) on one side and put in the 1 (UV (400-420nm) and 1 white) light on the other side. Take pictures at the beginning and then wait a month and compare the growth. Might think about also doing 1 (white, Royal blue and UV). This should be done in the same tank. That way the water is the same quality. Figure this would be an easy test and specially with the guys having LEDs laying around.

What do you think of trying this test?

 

[The Grim Reefer]

From personal experience I can tell you that the RB LEDs do make a significant visual difference! Not sure how to "prove" it in terms of spectrum and numbers, but you can definitely see the light they produce and differences when you're dimming.

I mean prove that running Royal Blues will give you all the fluorescence. With T5's the Blue Plus lamps give a lot of glow. I notice a very slight difference in some green colors kicking in actinics but it's hardly noticeable. My only question would be are the LED's more focused in the specified range so they don't drift as far into the UV range as the T5's. Judging by all the pictures I've seen I think there is plenty of UV range light there already just running RB's.

I'm thinking a row of those 403's running with a string of white lunar lights would look wicked for night time viewing.

 

[The Grim Reefer]

well I think the way to really tell this would be to take one tank and divide it in half with something dark then put a couple of the same frags on each side. Then put a 1 (white light and Royal blue) on one side and put in the 1 (UV (400-420nm) and 1 white) light on the other side. Take pictures at the beginning and then wait a month and compare the growth. Might think about also doing 1 (white, Royal blue and UV). This should be done in the same tank. That way the water is the same quality. Figure this would be an easy test and specially with the guys having LEDs laying around.

What do you think of trying this test?

I know of someone who tried to raise corals under a "Black light" as an experiment. I forget exactly what all he placed in the tank but most didn't do very well but a couple things did seem to live under the light just fine.

The Actinic experiment would be interesting, I know actinic lamps have lousy PAR BUT, PAR meters don't measure blue light very well so it might be a lot better than we think.

 

[der_wille_zur_macht]

well I think the way to really tell this would be to take one tank and divide it in half with something dark then put a couple of the same frags on each side. Then put a 1 (white light and Royal blue) on one side and put in the 1 (UV (400-420nm) and 1 white) light on the other side. Take pictures at the beginning and then wait a month and compare the growth. Might think about also doing 1 (white, Royal blue and UV). This should be done in the same tank. That way the water is the same quality. Figure this would be an easy test and specially with the guys having LEDs laying around.

What do you think of trying this test?

That's basically what the few "tests" that have already been done did, but IMHO it's tough to extrapolate growth and color data from a single coral or even a small set of frags to reef tanks in general - it leads to making statements like "X light is good" when it might not be the best for everyone's situation.

It WOULD be a cool experiment; I'm not trying to shoot you down - just trying to say that I think you'd need to temper your results carefully.

I mean prove that running Royal Blues will give you all the fluorescence. With T5's the Blue Plus lamps give a lot of glow. I notice a very slight difference in some green colors kicking in actinics but it's hardly noticeable. My only question would be are the LED's more focused in the specified range so they don't drift as far into the UV range as the T5's. Judging by all the pictures I've seen I think there is plenty of UV range light there already just running RB's.

I'm thinking a row of those 403's running with a string of white lunar lights would look wicked for night time viewing.

OK, gotcha. It's tough to compare spectral results that aren't measured under the same conditions, but it does strike me that the LEDs are probably producing a narrower wavelength of color than a T5 lamp.