55 gallon DIY LED display build (from groupbuyled.com)

[joelespinoza]

My heatsink for my 55 gallon DIY LED project came in today, its probably going to be a bit before the rest of my stuff comes in but I figured I would start the thread. Here is what I ordered from groupbuyled.com (who by the way are AMAZINGLY helpful):


2 x Arctic Alumina Thermal Adhesive 5g
◦ 1 x OEM Digital Multimeter DT-830B
◦ 3 x Inventronics 40w driver - 700mA
◦ 7 x CREE XT-E Royal Blue XP-G Neutral White 3 UP
(Color: 2x XTE Royal 1x XPG Neutral, Quantity in Package: 2 Pack)(This product is available for pre-order only)
◦ 7 x 3UP - XT-E XP-G Optic- 2 pack - 40 Degrees
◦ 3 x 10K Ohm 1/2 watt Linear Taper Potentiometer Easy Solder
◦ 1 x Solid Wire 24awg (300 volt) - 25 feet - White
◦ 1 x Stranded Wire 24awg (300 volt) - 25 feet - Black
◦ 2 x Solder Tube 10g (63/37)

$404.20 shipped.

10.000" Wide x 46" Long Heatsink

Heatsinkusa.com:

10.000" x 46" Fin Height: 1" Base Height: .300" Weight ~23lbs (because overkill is underrated):

$170.10 shipped.
http://i18.photobucket.com/albums/b124/JoelEspinoza/Heatsink.jpg

[kerry]

Nice heat sink!!!

[joelespinoza]

It took a while to get everything in, some stuff was backordered and I changed the order a bit, I added violet/red/turquoise LEDs for extra color, but these could be considered optional. Here was what I got total (I subtracted prices for stuff that was optional):

6 x CREE XT-E Royal Blue XP-G Neutral White 3 UP - $162.00
(Color: 2x XTE Royal 1x XPG Neutral, Quantity in Package: 2 Pack)

6 x 3UP - XT-E XP-G Optic- 2 pack - 40 Degrees - $31.50

2 x Arctic Alumina Thermal Adhesive 5g - $13.98 (only needed 1 -$6.99)

1 x OEM Digital Multimeter DT-830B - $8.88 (could have left this off -$8.88)

5 x Inventronics 40w driver - 700mA - $185.00 (if you leave off the violet/red/turquoise LEDs only need 3 - $74.00)

1 x True Violet LED - 420nm - $27.00 (optional - $27.00)
(Quantity in Package: 6 Pack, Optic Degrees: 60°)

3 x Deep Red - 660nm - $11.94 (optional - $11.94)
(Optic Angle: 60°, Quantity in Package: 2 Pack)

3 x Turquoise - 495nm - $15.00 (optional - $15.00)
(Optic Angle: 60°, Quantity in Package: 2 Pack)

4 x 10K Ohm 1/2 watt Linear Taper Potentiometer Easy Solder - $12.00 (If you leave off the violet/red/turquoise LEDs you only need 1 for white LEDs IMO -$9.00)

1 x Solid Wire 24awg (300 volt) - 25 feet - White - $5.75

1 x Stranded Wire 24awg (300 volt) - 25 feet - Black- $5.75

2 x Solder Tube 10g (63/37) - $8.50 (One is plenty -$4.25)


Subtotal:
$487.30

Insurance for shipping:
$14.13

Grand Total:
$501.43 (what I paid)
$344.37 (what it would cost to just have Royal blue and Neutral white LEDs)

[joelespinoza]

The stuff from groupbuyled.com:
http://i18.photobucket.com/albums/b124/JoelEspinoza/0406121731.jpg

The LEDs glued on to the heatsink (groups of 3 are violet/red/turquoise):
http://i18.photobucket.com/albums/b124/JoelEspinoza/0406121846a.jpg

All the LEDs tinned and the first string of Royal Blues wired and tested (all I had time for tonight before it got dark):
http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0222Medium.jpg

More to come tomorrow.

[joelespinoza]

LEDs Wired:
http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0237Medium.jpg

Blues and Violets on:
http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0245Medium.jpg

Neutral Whites, Greens, Reds on:
http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0244Medium.jpg

All on:
http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0242Medium.jpg

[joelespinoza]

Not quite done yet, but getting close.

Old lights (with some daylight):
http://i18.photobucket.com/albums/b124/JoelEspinoza/OldlightsMedium.jpg

New Lights white on full (No daylight):
http://i18.photobucket.com/albums/b124/JoelEspinoza/NewlightswhiteonfullMedium.jpg

New Lights white on lowest (No daylight):
http://i18.photobucket.com/albums/b124/JoelEspinoza/NewlightswhiteonlowMedium.jpg

I am going to get 2 more 3up LEDs and optics for the middle, and I am also going to get 6 more Royal blues. I can easily add all those with existing drivers.

[Ace25]

Looking good!!! That is one BEAST of a heatsink! You are never going to have to worry about heat with that big boy, and certainly won't be needing any fans with it. :D

Did you just setup the tank? Why is it so cloudy?

[joelespinoza]

Looking good!!! That is one BEAST of a heatsink! You are never going to have to worry about heat with that big boy, and certainly won't be needing any fans with it. :D

Did you just setup the tank? Why is it so cloudy?

Yea, the heatsink weighs almost 25 lbs by itself, so its passivie cooling ability is pretty great. I also have a big ceiling fan thats always on in that room and the house is always kept around 72 degrees so I dont see it ever being a problem. After being on 12 hours the heatsink right next to the LEDs is not even body temp and just a couple inches away frm the LEDs the heatsink is room temp, it doesnt feel warm at all.

Its always cloudy. I feed blended oysters (and other frozen food mixed in) and it makes the tank very cloudy. I wrote about it in the post SM has about $5 oyster feast, but there really isnt a resolution. Basically I just dont have enough filter feeders to feed what I am feeding. I should go get some prawn roe and feed my fish that, but I have been to lazy to go get it. I am going to go buy all of a friends livestock today, so hopefully he has some corals that will like it.

[Ace25]

Not sure if you found this out by now, just thought I would mention it for others though. Hot glue does not hold down the wires on a heatsink for most people. ;) I know, it was the first thing I tried as well thinking it was the ideal solution. Problem ends up being the heating and cooling of the heatsink makes the hot glue just pop off the heatsink in about a weeks time. Maybe you will get lucky since your heatsink is so large you won't have the heating/cooling issue I am speaking of, but for most of us "normal" people who don't buy the biggest heatsink possible using hot glue doesn't work well (which buying a giant heatsink is good thing and I think it looks nice on your tank, very slim without fans, just most people aren't willing to spend that kind of $ on a hunk of metal, LOL).

[joelespinoza]

It probably depends what kind of hotmelt glue you use, and how hot the heatsink gets. That being said I did not use the high temp nice hotmelt glue, its just the cheap craft stuff from Wallyworld. So yea, hopefully it will stay on, if not its not a huge deal. Most of the wires are tight enough that they will be ok by themselves if the hotmelt glue pops off, The few that I left lots of slack in are going to additional LEDs (already ordered), and there will be a bit of reorganization once they come in. The first string I wired I didnt think about skipping every other LED and bringing the chain back, so I have 1 4 foot wire that will probably just get glued to the plastic cover when it is made, unless I get a wild hair up my ass to resolder all those LEDs, which I might.

On the heatsink front, my thinking is that massive heatsink cost $170 shipped. When you factor in the cost of fans ($13 each for 120mm fan+silencer, so at least $26), The cost of the powersupply for fans ($10), the fan controller ($10) the cost of electricity for a decade(4 watts/fan, 12 hours/day for a decade at $0.12 per kW/hr ~ $42, and the cost of a smaller heatsink ($88.69 for a 5.886" x 46" Heatsink), I could have paid about $177. So over the course of the fixtures life I would have paid at least $7 dollars more (assuming 2 fans was enough and the fans and powersupply never had to be replaced in 10 years) for the inconvience of increased cleaning schedule (from fans pushing in dust), and the increased noise level of any and all fans. To me buying a larger heatsink seems like a bargain.

[sklywag]

I like your thinking Joel. And that you are thinking.

[joelespinoza]

Thanks, I try.

[joelespinoza]

Last night I bought a friends whole livestock from him, he had a small ~20 gallon but had a bunch of stuff in there. I acclimated it last night and then just carefully set it in my display tank. My display is so foggy I cant really see where I am putting stuff right now.

This morning when just my Royal Blue and violet lights came on it looked AMAZING, everything was glowing like crazy, looked a bit like a scene from alice in wonderland under black lighting. I tried to get some pictures of it, however my camera was just not capable to capturing the images. These 2 are as close as I got and they dont do it justice:

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0218Large.jpg

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0230Large.jpg

[Ace25]

I hear you.. it sucks that there isn't a camera out there that seems to be able to capture tanks lit with LEDs like we see them in person. It is a night and day difference between what you see and what the picture shows.

Have you stopped feeding your liquid food? IMO a tank should not be cloudy for more than 30 minutes after feeding liquid food, if it is your putting in way to much unless you have a completely NPS (non photosynthetic) tank. I am not a fan of them, but in your case, I would be using filter socks to clean up the water and then remove the sock when it is clear.

[kerry]

I agree on the cloudy-ness.

[joelespinoza]

Have you stopped feeding your liquid food? IMO a tank should not be cloudy for more than 30 minutes after feeding liquid food, if it is your putting in way to much unless you have a completely NPS (non photosynthetic) tank. I am not a fan of them, but in your case, I would be using filter socks to clean up the water and then remove the sock when it is clear.

Its been about a week since I stopped feeding blended oysters, I setup a aquaclear 70 HOB on it this morning.

[kerry]

I sure do want to see pics when it clears up!!

[joelespinoza]

The last of my LEDs came in today, and my cover, while rough, is almost done. So I hope to finish the light soon.

Sadly the way my tank is setup there is no place for a filter sock. My overflow is my algae filter, there is no seperate pipe. I think that my tank cloudyness is a result of my bare bottom sump. I have lots of cleaners down there, but there is just a buildup of small particles sitting on the bottom of the tank, those end up getting sucked up my pump everytime I bump it. I think I need to put my sump pump in a tray of some kind then put a small layer of sand in the sump, that should allow more bacteria to live down there and to break up the tiny particles.

[Ace25]

I use a Koralia 3 in each chamber in the sump.. nothing can settle in my sumps, heck, the water barely stays in the chamber with a K3 pump in a 8"x8" box. You could also just put a pump in your tank with a hose going up and into the sock and have it filter in the tank itself. That is IF you have filter socks laying around, if you don't, then it isn't worth the headache of even thinking of going that route, but I would get at least some filter floss in the system somewhere.

[joelespinoza]

I use a Koralia 3 in each chamber in the sump.. nothing can settle in my sumps, heck, the water barely stays in the chamber with a K3 pump in a 8"x8" box. You could also just put a pump in your tank with a hose going up and into the sock and have it filter in the tank itself. That is IF you have filter socks laying around, if you don't, then it isn't worth the headache of even thinking of going that route, but I would get at least some filter floss in the system somewhere.

Sadly I only have the kinda wide pore aquaclear sponge filters. The only other "filters" I have are these, and they wouldnt do well in salt im afraid:

Freshwater filter 1 (christmas moss):

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0217Large-1.jpg

Freshwater filter 2 (Jacques the vampire shrimp):

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0223Large.jpg

Hopefully the aquaclear 70 will clear it up a bit til I think of a more permanent solution.

[joelespinoza]

I still have to clean up the wiring from the fixture to the drivers, and finish sealing the cover to the heatsink, but other than that its all done.

Final LED count:

14 XP-G Neutral White (40 degree optics, 700mA max)

36 XT-E Royal Blue (40 degree optics, 700mA always)

6 3W True Violet (60 degree optics, ~600mA)

6 3W Turquoise (60 degree optics ~600mA)

6 3W Deep Red (60 degree optics~600mA)

All LEDs on full:

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0232Large.jpg

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0230Large-1.jpg

Neutral whites on lowest setting, everything else on high:

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0226Large.jpg

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0225Large.jpg

[Ace25]

Looks great! The first 2 pictures with them ALL ON, just going off the picture (which I know is bad the begin with) it "appears" brighter than what even a 250w XM 10K bulb puts out. If it appears brighter, then it must be REALLY bright. Figure a 250w 10k MH will put out around 800 PAR at the surface of the water if the light it at least 12" above the water, taking a wild guess going off the picture, I would bet your getting 1800 PAR at the surface with your LED light. Just an educated guess. I know the picture was just for the comparison shots, and you are more than likely never going to run them all at 100%, just be very careful with the intensity.

I would bet that if you ran all your LEDs at 25% right now, you would still be putting in more light than your previous lighting. You are going from a Volkswagen to a Ferrari in terms of lighting upgrade. Corals can take the Ferrari lighting.. eventually.. but you have to slowly acclimate them to higher lighting. If you go more than 10% above your previous lighting output in one shot it shocks most SPS corals (not sure if you have any) and it usually means near instant death (overnight), talking from personal experience. I was slowly ramping up my lights to acclimate my corals, started at 40% and every 5 days moved up 5% but one day I went from 60-75% in one shot and had some nice white skeletons in my tank by the next morning.

[joelespinoza]

Well I do have a few SPSS that I got last night, however they came from a very shallow tank (14" tops) with a 4 bulb T5HO fixture, so they are used to much higher lighting than what used to have. They are also currently very low in my tank.

I plan to leave 28 of the royals blues on high (they are not adjustable) and have everything else turned down to low. I will slowly adjust it up from there.

[Ace25]

4 bulb T5 Fixtures usually puts out around 400 PAR under the bulbs at the surface (type of reflectors and bulbs play a big part in T5 output, but the range is still usually 300-500 PAR depending on variables), about 1/2 the output of a 250w 10k bulb, or about equal to a 250w 20k Radium bulb, just to give you an idea of what the corals used to be under.

Also, just an FYI, it is the royal blues that cause the most damage to corals. Reason being is royal blue is very dim to the human eye so we crank them up to get color we like, and the glow of the corals, but the problem is the spectrum is ideal for photosynthesis (455nm), and this is the light that does the most damage to the corals because when switching to LEDs, most people end up putting 2-5x MORE 455nm spectrum than what any previous type of lighting could provide at that spectrum. Cool and Neutral white LEDs actually put out a lot of royal blue spectrum as well, but the rest of the spectrum they put out washes out the blue, but it is still in there. White LEDs all by themselves will provide the same, if not more, of the 455nm lighting than even most daylight T5 bulbs or MH bulbs. You add royal blue LEDs on top of the whites and you really start blasting the tank with that spectrum, but as we all know, royal blue LEDs are really a must have for our viewing pleasure so it isn't like you can do without them.

[kerry]

I run a mix of 7 12,000 and 19,000K LED's combined to 1 455nm LED and my corals seem to react very well to this. I have 32 total over a 40G breeder type/size tank.

[joelespinoza]

Still trying to clear the water, I bought some fiberfill from wallmart and filled up the aquaclear with it last night, and today I bought some Seachem clarity (a flocculating agent) and added twice the recommended amount, the directions say up to 3 times the recomended amount is ok for faster/stronger clarifying. It is slowly clearing up, hopefully the clarity will help some.

I think the cloudyness at this point is due to micro algae, it probably was aggrivated by cleaning both my algae screens at one time.

[joelespinoza]

Yea taking some pictures of it with the white flash made it clear what the problem is, its not very obvious without strong white lighting though.

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0232Large-1.jpg

[joelespinoza]

I know LEDs are kind of a mystery to many people, this was quite the experience doing all this. It took days of research and tons of emails to Milad at groupbuyled.com before I even started, and then I changed my plan like 15 times before I finished.... So I figured I would list the final costs of this light and what all you would need to make one if anyone is interested in doing something similar.

The following is what would be needed to make a light exactly like mine, and also one without the extra colored LEDs.

Heatsink: 10.000" x 46" Fin Height: 1" Base Height: .300" Weight ~23lbs - $170.10 shipped (heatsinkusa.com)

Subtotal: $170.10


Cover:

OPTIX 24 in. x 48 in. x .093 Acrylic Sheet - $23.98 (local homedepot)
GE Silicone II 2.8 oz. Window and Door Caulk Clear - $3.89 (local homedepot)
Rattle can of black spraypaint ~$5.00 (local homedepot)

Subtotal: $32.87 (plus tax if applicable)


LEDs, I will list the cost of what I got, and the price of just the basic 28 Royal blues and 14 Neutral whites:

7 x CREE XT-E Royal Blue XP-G Neutral White 3 UP - $189.00
(Color: 2x XTE Royal 1x XPG Neutral, Quantity in Package: 2 Pack)

7 x 3UP - XT-E XP-G Optic- 2 pack - 40 Degrees - $36.75

1 x Arctic Alumina Thermal Adhesive 5g - $6.99

5 x Inventronics 40w driver - 700mA - $185.00 (if you leave off the violet/red/turquoise/6 individual Royal Blue LEDs only need 3 - $74.00)

1 x True Violet LED - 420nm - $27.00 (optional - $27.00)
(Quantity in Package: 6 Pack, Optic Degrees: 60°)

3 x Deep Red - 660nm - $11.94 (optional - $11.94)
(Optic Angle: 60°, Quantity in Package: 2 Pack)

3 x Turquoise - 495nm - $15.00 (optional - $15.00)
(Optic Angle: 60°, Quantity in Package: 2 Pack)

1 x CREE XT-E Royal Blue - $19.38 (optional - $19.38)
(Quantity in Package: 6 Pack )

1 x XP/XT Optic - 6 Pack (40 Degree) (optional - $9.54)
(Quantity in Package: 6 Pack )

5 x 10K Ohm 1/2 watt Linear Taper Potentiometer Easy Solder - $15.00 (only need 3 for for just white/royal blue LEDs -$6.00)

2 x Solid Wire 24awg (300 volt) - 25 feet each- White - $11.5

2 x Stranded Wire 24awg (300 volt) - 25 feet each- Black- $11.5

2 x Solder Tube 10g (63/37) - $8.50


Subtotal:
$547.10 (The cost to get exactly what I have)
$384.24 (The cost to just have 28 Royal Blue and 14 Neutral White LEDs)


Grand Total:
~$750.07 For a fixture exactly like mine.
~$587.21 For a basic fixture similar to mine, but with just 28 Royal Blues and 14 Neutral Whites.

[Ace25]

You really can't beat that price. You get the added bonus that if anything goes bad you know you can quickly and easily fix any problems yourself with your light and also got to pick exactly the colors you wanted.

I know exactly what you mean about all the research.. I spent a good 12 months researching LEDs before making a purchase, then after the purchase I changed the plans several times after I got to see it running. So for me, it took about 2 years to really get a firm grasp on how LEDs work on a reef tank, 1 year of researching and thinking I had it down, then building, then another year to understand what I missed in my thinking during the first year of research.

For me, an added upgrade to your LED light would be to use PWM drivers and use an arduino controller (like the Typhoon) to be able to control each channel of lighting and have wireless capabilities so you can control the light with a smartphone, computer, or iPod touch. I am a computer geek though, so I understand that is another giant step in complexity. I did the same as you originally, I used pots to control my LED light intensity when I first made it, then 6 months later replaced the drivers and added a reef angel controller with 8 PWM channels so I can do all the extra cool stuff I wanted to do. Getting the light working and looking good is first priority though, adding extra features can always come at any time later.

[joelespinoza]

Yea, some of the controllers seem pretty neat, but I doubt I will be spending any more money on lighting for a while. This REALLY strained the college student budget as it was. The next project for me is probably going to involve upgrading the algae scrubber somehow.

I am not sure yet if I will just add 2 LED panels behind my existing screens to have them lit on both sides (currently lit on only 1 side with 26 watt curly flourecents) or if I want to build a while new enclosure for the scrubber. I am having to fight my natural desire to go way overkill on the scrubber, because I would like to be able to keep softies and some decorative macro algae. Its a mental battle =)

[kotlec]

Arduino controler is not a big investment like one you just did. I made mine for about $30-40. Drivers are another case. But there you can go DIY route as well.

[joelespinoza]

Arduino controler is not a big investment like one you just did. I made mine for about $30-40. Drivers are another case. But there you can go DIY route as well.

I am not that up on electronics, it looks like my drivers would work with a Arduino controller as long as I had a low level filter installed on the outputs. However i am not sure I have enough free time to learn everything it would require to build and program a Arduino controller. Do you know anyone who has whole kits and or programs them for LED lighting applications? I am dealing with a 0-10v analog signal. I am decent with soldering but terrible with even somewhat complex electronics.

[kotlec]

There is thread on RC for complete nobs. Not that hard at the end. Will be easy for you , I am sure. Just check it : www.reefcentral.com/forums/showthread.php?t=1987110&page=54

If your drivers are dimable then all you need else is this :
http://www.ebay.com/itm/Duemilanove-2009-ATmega328p-Board-Free-USB-Cable-Arduino-Compatible-/280851303722?pt=LH_DefaultDomain_0&hash=item41640acd2a

$17 shipped :)

And something like this :

http://www.ebay.com/itm/LCD-Keypad-Shield-Module-F-Arduino-Duemilanove-LCD1602-/290698028015?pt=LH_DefaultDomain_0&hash=item43aef3e3ef

Oh yes you also need RTC. Like this
http://www.ebay.com/itm/New-Arduino-I2C-RTC-DS1307-AT24C32-Real-Time-Clock-Module-For-AVR-ARM-PIC-/270849749804?pt=LH_DefaultDomain_0&hash=item3f0fe7332c#ht_2811wt_1249

$7 shipped

[joelespinoza]

There is thread on RC for complete nobs. Not that hard at the end. Will be easy for you , I am sure. Just check it : www.reefcentral.com/forums/showthread.php?t=1987110&page=54

Yea, I had already read through about 30 pages of that...... Maybe someday someone will condense 55 pages of discussion of complex theory (for me and anyone else with near no experience assembling electronics) about electrical circuts and coding into something managable for someone without a degree in electrical engineering..... If thats the easy version for noobs there is no way I am even in the class of noob.

I understand the hardware required for the most part, and could certainly assemble it if given exact diagrams with hardware that closely matched what I have, but as far as making it work..... It is kind of like saying it isnt hard to learn to use a ratchet, and pretty much everyone knows that a car needs an engine, a transmission, a final drive, axles, brakes, wheels, and tires. So that means anyone should be able to get on ebay, order everything they need, and assemble a working car they can drive to work everyday, right?

Basic knowledge and a working understanding are very different animals. If you do not understand how something works, or even seen a complete working model, most people will need step by step instructions on how to assemble it.

[kotlec]

There is general wiring diagram there. Just need some luck digging :) And if you read 30 pages, then you has to find step by step instruction. If you read more you only find that controler has become more complex and hi tech. But you can just stop at light controling. That what cachupoy is agitating to do.
I am sure you can easily do it . You underestimating your abilities. If you can keep reef aquarium , then asembling arduino controller is much easier. When I first heard about that, my reaction was exactly as yours. But now I am sitting in front of my working controller. And know what ? - I have already tweaked sketch to my liking. People on arduino forum are very helpful too. So never say never :)

[joelespinoza]

Honestly I dont think I am going to worry about that for now. The lights can come on and off with timers and it works fine. Maybe I will consider a sunrise/sunset controller for several projects down the road.


My next 3 projects:

1: Make some damn money because now I am DEAD broke. =)

2: Beefing up the scrubber somehow, probably additional LED panels on the backside of existing screens to make them lit on both sides, although possibly building a new enclosure of somekind when the SM releases his new plan. My current microalgae bloom is a sure sign I need to make it more competitive.

3: Getting another 50+ lbs of dry rock to the sump to get it started as live rock.

[joelespinoza]

You have to slowly acclimate them to higher lighting. If you go more than 10% above your previous lighting output in one shot it shocks most SPS corals (not sure if you have any) and it usually means near instant death (overnight), talking from personal experience. I was slowly ramping up my lights to acclimate my corals, started at 40% and every 5 days moved up 5% but one day I went from 60-75% in one shot and had some nice white skeletons in my tank by the next morning.

I have tried starting off as low as possible and running my normal time, and while I did not kill anything (yet) I definately gave some stuff a sunburn:

My hammer coral bleached up the sides, and has exposed white now, but the tops still look good, I moved it to the very bottom so hopefully it will recover.

My purple fan gorgonian retracted all its polyps for 2 days, I moved it to a shady spot, then it started putting them back out, now it looks fine except it had a FULL body peel.

My green star polyps looked great for the first few days, now all 4 seperate colonies have retreated and there is not a polyp to be seen.

The SPSS I put in the tank has still never shown a single polyp, its base growth on the rock looks fine, but I assume by this point it is probably dead.

I decided today to approach it from a new angle, I am going to leave the 6 royal blues and 6 violets on all day (15 hours) at about 75%, I am going to set the rest of the LEDS (28 royal blues @100% everything else at 30-40%) on a 4 hour per day schedule until everything seem to be out and about, then I will start adding time, probably half and hour every other day or every 3rd day or something along those lines.

How does that sound?

[joelespinoza]

The tank is not totally clear yet, but is getting much better, so I thought I would post some pics. Also I measured today, like LEDs are almost exactly 5" from the water (from the optics to the waterline)


Video of the color seperation over rough water, and the color consistance over smoother water:
http://i18.photobucket.com/albums/b124/JoelEspinoza/th_PICT0265_mpeg4_001.jpg (http://s18.photobucket.com/albums/b124/JoelEspinoza/?action=view&current=PICT0265_mpeg4_001.mp4)

All LEDs on full:

http://i18.photobucket.com/albums/b124/JoelEspinoza/Allonfull.jpg

Mandrin dragonnet is watching you:
http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0249Medium.jpg

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0253Medium.jpg

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0248Medium.jpg

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0246Medium.jpg

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0243Medium.jpg

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0239Medium.jpg

Sad Hammer with a sunburn =(
http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0238Medium.jpg

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0237Medium-2.jpg

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0236Medium-2.jpg

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0235Medium.jpg

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0232Medium.jpg

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0229Medium.jpg

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0228Medium.jpg

http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0225Medium.jpg

[moncapitane]

Like the layout. Do you have a paper sample or something that shows what lights you put where and how you wired it? The light is way better than the light before the LEDs. Would love to hear from you.

[joelespinoza]

My light is not bad, everyone who visits me thinks it looks amazing and usually stares at the tank for quite a while, however knowing what I know now, I could make a better light WAY cheaper.

I would use the same heatsink, that thing rocks. But I built that with 34 Cree Royal Blues, 14 Cree neutral whites, 6 turquoise, 6 deep red, 6 true violet LEDs and 700mA drivers with 45 and 60 degree optics.

If I did it again I would use Luxeon 2700k 95 CRI warm whites with Luxeon Royal Blues from steves LEDs and drive them at 1000mA with steves drivers. Luxeon 2700k whites look incredible and they are all I use in displays now. With them the added colors are simply not needed.

With how much more efficient Luxeons are I could easily get away with 28 LEDs total, although I might do a 35 LED setup and go with no optics.

$170 10" x 46" heatsink shipped

$104.65 35 luxeons at $2.99 each

$77.86 Stevesled drivers for a 35 LED setup

$15 for good thermal adhesive

$5 for wire, although honestly ethernet wire works like a charm and I have tons of it, so thats being generous.

So while my original fixture cost over $750 by the time I was done, I could now make a better fixture for ~$375, one that would have more light, better coverage, better colors and better drivers (plug and play basically with arudino controllers) for half the price.

It would obviously be even less if I went with the 28 LED build, which would probably be about the same amount of light I have now.

[Floyd R Turbo]

I just saw your link to steve's LEDs in your other thread. I didn't realize they were now carrying 660s and true violets. Steve's is starting to look better and better. I like their blurbs about the Philips LEDs vs Crees. Good timing I'm pondering convincing a customer to switch to LEDs instead of replacing 8 T5HO lamps every year....

[joelespinoza]

I just saw your link to steve's LEDs in your other thread. I didn't realize they were now carrying 660s and true violets. Steve's is starting to look better and better. I like their blurbs about the Philips LEDs vs Crees. Good timing I'm pondering convincing a customer to switch to LEDs instead of replacing 8 T5HO lamps every year....

Yea, Steve is great. He has the pricing of chineese no name brands, but with premium products and great customer service. I have never seen anyone respond so fast, to emails or for shipping.

My advice on LED builds is now start low and optics are usually optional on tanks up to 20" deep. Figure with Luxeons at 1000mA you should aim for about 1 LED/per 20-30 square inches of tank surface area. 20 for SPS and like 30 for for softies. You can always easily add more later, but its hard to get back corals that got fried.

Here is a post I have on reefcentral about the warm whites, it amazes more more people dont use them, they are really fantastic:

http://www.reefcentral.com/forums/showthread.php?t=2194214

I am still amazed by how great that freshwater tank looks, its ok in pictures, but in person its just beautiful. That build is a tad ghetto because the person I built it for wanted to stay as cheap as possible, but even going insanely overboard on LEDs it still only cost like $85 for the whole retrofit. I mean you cant even buy a T5 setup for a freshwater tank that cheap. Although she might end up paying for it when she has to buy all her fish sunscreen and tiny sunglasses... =)

This was the build thread for the freshwater: http://www.reefcentral.com/forums/showthread.php?t=2192354

[jnad]

Hello!

Is the led unit you made from Steves led still running good?

I have ordered an ADA tank with dimensions 75x45x45cm, it is going to be set up as an "Holland" type of planted fresh water tank.

I am considering bying leds from Steves for display lightning of this ADA tank, any suggestions of led setup?

Jnad

[joelespinoza]

Hello!

Is the led unit you made from Steves led still running good?

I have ordered an ADA tank with dimensions 75x45x45cm, it is going to be set up as an "Holland" type of planted fresh water tank.

I am considering bying leds from Steves for display lightning of this ADA tank, any suggestions of led setup?

Jnad

Sorry i havent checked on this thread in forever.... VERY busy with school, but yea the stevesleds are doing great

[crashmushroom]

Joel i think you may have convinced me i have a diy cw & rb cree xte build. Since changing from my t5 a year ago i was so dissapointed wigh colour loss i put my old t5 unit back up and my expensive led build is gathering dust under the stairs. Im going to swap out my cw with 2700k from steves and confirm what you are saying. I am correct in saying xte RB and 2700k from steves is all i need. If so im glad i didnt go ahead and waste more more on cyan red green and violet leds. U may have saved me a fortune. One more request can i see pics of your 2700k build please?
Thank you

[joelespinoza]

Steves 2700k warm whites definately bring out color better, and I think they are far superior to the cool or even neutral whites most people use, however I think that a combo T5/LED unit still looks a bit better. Personally I think something like 4 warm whites : 2 royal blues : 1 ~420nm + 1/4 T5 lighting is pretty spectacular. That sounds complex but let me give you an example:

So if I was making another fixture from scratch, for a 90 gallon tank (48" x 18" = 864), I would figure roughly one of steves LEDs per 30 square inches, if driven at about 1000mA, which is about 28 LEDs total. So probably 16 2700k Warm White Luxeons at 1000mA, 8 Royal Blue Luxeons at 1000mA, and 4 ~420nm violets at 700mA. Then I would add a single 48" T5HO bulb in the middle, probably a 10k one, although maybe a 50/50 if I wanted it less white.

[Floyd R Turbo]

That is a really low figure joel, when I was going to do a 2 RB : 1 CW array, I was told that the rule of thumb was based on surface area and you divide by 16 for low light/softies, 12 for LPS to easy SPS, and by 10 for heavy SPS.

For my 120g (24x48 = 1152) that was 72 for low light, 96 for mid-range, and 120 for high light.

For a 90 at low light, that's 54 LEDs, minimum, and you're going to have a hard time growing anything past LPS at mid to low positions in the tank even with that few LEDs I would think.

[crashmushroom]

Oh good i was hoping to get away from t5 hopefully the cri95 leds will do me they arrive monday. Do u have any pics of 2700k mixed with RB?

[joelespinoza]

Oh good i was hoping to get away from t5 hopefully the cri95 leds will do me they arrive monday. Do u have any pics of 2700k mixed with RB?

Sadly any pictures I send you wont mean anything, any pictures I take simply wont represent reality.


That is a really low figure joel, when I was going to do a 2 RB : 1 CW array, I was told that the rule of thumb was based on surface area and you divide by 16 for low light/softies, 12 for LPS to easy SPS, and by 10 for heavy SPS.

For my 120g (24x48 = 1152) that was 72 for low light, 96 for mid-range, and 120 for high light.

For a 90 at low light, that's 54 LEDs, minimum, and you're going to have a hard time growing anything past LPS at mid to low positions in the tank even with that few LEDs I would think.


The "Classic" rule of thumb is 1 3w Cree XP-E driven at about 2 watts (~700mA max) per 15-20 square inches of tank surface area, up to about 24" deep, with 1/15 square inches are very high light, and 1/20 square inches as medium high. Looking up specs sheets from different companies feels like comparing apples to oranges, but it you stare at them long enough you realize that the old generation XP-E Crees running at 700mA have significantly less then half the output of more modern Luxeons running at 1000mA.

Since this is reef lighting, its quite likely that whoever told you 10-16 was applying the "more is better" theory that has been pretty applicable to all previous types of reef lighting, and is so hard to get away from with LED lighting. In reality its INCREDIBLY easy to burn the crap out of corals with LEDs, and pretty difficult to starve them to death.

In the end, I am no expert. All I have is my softies and my LPS, I dont bother with SPS at this point. However I can tell you that my fixture has plenty of spread for a 90 gallon tank, and with my 28 XTE Royal blues and 14 mixed whites that if I turn the fixture up more then 70% (~490mA) everything starts shrivling up and doesnt do well at all, however everything is perfectly happy indefinately at 10-15%.

Cree XTE at 1500mA and Luxeons at 1000mA are pretty comparable in output, so if I wanted a 70% cap on my lighting and to make it out of Luxeons I would be looking at nearly 1/3 the LEDs, so 42/3 = 14 give or take. This is an estimation, but its probably in the ballpark.

Keeping all that in mind, the fixture I estimated was twice that number of LEDs and a single T5HO tube (for visual brightness). Honestly, I think its overkill.

PS I meant 4 Luxeon Royal Blues @ 1000mA max : 2 Luxeon Warm Whites @1000mA max : 1 ~420nm violet @ 700mA max. I dont know why I reversed the ratios of royal blues and whites, sorry.

[Ace25]

I run a 50/50 mix of RB/CW 24 LEDs on a standard 4' 60G, light is 18" above the surface and I still can grow acropora on the sandbed. My other LED light is 48 LEDs (mix of all types of LEDs, cyans, reds, all types of whites, blue, royal blue, and actinic) over a 75G (30x24x24) and it is way to many LEDs, could have easily got by with 36.

Video I just took with my new iPod Touch.

http://www.youtube.com/watch?v=bYaRJK7co7M

[joelespinoza]

As I said, these are not ANYTHING like reality. You can tell because those crazy ass green things are actually metalic green star polyps, not irradiated alien blobs as you would think from the pictures. Also, sorry about how filthy my tank is, not only is this the end of finals but it also hasnt had a working algae filter (or any filter besides the sump) for over 3 months.

Also dont mind the bizarre LED ratios, its just kinda how it turned out, not how it should be.

This is with my ghetto Galaxy S3 smartphone camera:

This is with 28 Cree Royal Blues at about 70%, 14 Cree Neutral whites at 10%, 10 Luxeon 2700k Warm Whites at 10%:

http://i18.photobucket.com/albums/b124/JoelEspinoza/20121214_201642Medium_zpsbf0ad902.jpg

This is about 70% on the Royal blues and 10% on the neutral whites and about 50% on the 10 warm whites. This is about how I normally run it, just a tad brighter overall:

http://i18.photobucket.com/albums/b124/JoelEspinoza/20121214_201819Medium_zpse216c780.jpg

This is 70% on royal blues and 100% on both whites. I would never actually run it like this, in person it is glaringly yellowish white and doesnt appear to have even a trace of blue or purple to it. I included it here just because:
http://i18.photobucket.com/albums/b124/JoelEspinoza/20121214_201858Medium_zps6230d33c.jpg

[alum]

does all corals happy with your new light ???
how abaut they grow ?

[crashmushroom]

I have to agree with you joel pictures are a waste of time when it comes to colour spectrum it needs to be seen in person, but thank you anyway. luxeon 2700k leds arrive monday so by monday night maybe tuesday i will know. I currently run 20 XTE RB, & 20 XPG CW looks fine apart from certain colours look washed out so i hope the CRI of the luxeons may change that. Will report back my opinion. I was also thinking i may have to order some NW for the mix but ill see first what this looks like. Do you think UV 420nm are needed?

[Ace25]

Do you think UV 420nm are needed?

UV will harm corals/fish... 420nm on the other hand will help immensely as it is the primary spectrum for coral photosynthesis. There is no such thing as "UV 420nm" because UV starts at 399nm and goes down from there. 420nm spectrum is known at 'actinic' or 'violet', not ultra violet which is beyond what the human eye can see.

[crashmushroom]

Yeah i knew that ace, but its what our friendly led suppliers call the 420nm spectrum i wonder why?

[joelespinoza]

Yeah i knew that ace, but its what our friendly led suppliers call the 420nm spectrum i wonder why?

Calling it Ultra Violet is either a marketing or ignorance thing. What we want is actually "True Violet"

Try these, I havent had a chance to try them personally but I have heard good things:

http://shop.stevesleds.com/3-Watt-True-Violet-LED-TrueViolet.htm

As far as growth goes, its hard to say. This is my first saltwater aquarium and my first real light. Before this I just had a ghetto ass T8 shoplight. Also with the extra nutrients and algae growth I have right now I am pretty sure my corals are not going to be doing their best.

Also I think I need to start dosing Iron, I am about to start searching about how to go about it. Any suggestions?

[alum]

Calling it Ultra Violet is either a marketing or ignorance thing. What we want is actually "True Violet"

Try these, I havent had a chance to try them personally but I have heard good things:

http://shop.stevesleds.com/3-Watt-True-Violet-LED-TrueViolet.htm

As far as growth goes, its hard to say. This is my first saltwater aquarium and my first real light. Before this I just had a ghetto ass T8 shoplight. Also with the extra nutrients and algae growth I have right now I am pretty sure my corals are not going to be doing their best.

Also I think I need to start dosing Iron, I am about to start searching about how to go about it. Any suggestions?

Hello Joe I believe there is so many kind of UV that we found in the net but not of all for best for corals, your url might be true of not unfortunately there is no review on there, I am also building for new light too.

[Floyd R Turbo]

I do believe I stand corrected on my figures. Thanks Joel and Ace for the insight.

[joelespinoza]

Hello Joe I believe there is so many kind of UV that we found in the net but not of all for best for corals, your url might be true of not unfortunately there is no review on there, I am also building for new light too.

You want to aim for 420-430nm and avoid much of the curve dipping below 400 into the UV range. 420nm is the main receptor for chlorophyll a, anything less then 400nm starts to cause UV related tissue damage, and if it doesnt have much of the curve at ~425nm-435 its not going to be very visible to the human eye.

I have heard that steves true violets are amazing looking and visibly bright, and they work out well for the other requirements. I plan to get some eventually, but right now I am a broke college student.

[joelespinoza]

I do believe I stand corrected on my figures. Thanks Joel and Ace for the insight.

Hopefully it helps, I know how much misinformation is out there about LEDs. As Ace pointed out when I got my setup, its just SOO easy to burn corals with them, and while you do want some range of adjustment, if you are anything like me you will try cranking them up at some point just to see how everything does.... And at least with our advice, if you do find our experience to be different then yours, you just have to add more LEDs later. The alternative is blowing too much cash and killing your corals up front. I spent almost $800 on my fixture, knowing what I do now I could build a better one for $350-400


You want to aim for 420-430nm and avoid much of the curve dipping below 400 into the UV range. 420nm is the main receptor for chlorophyll a, anything less then 400nm starts to cause UV related tissue damage, and if it doesnt have much of the curve at ~425nm-435 its not going to be very visible to the human eye.

I have heard that steves true violets are amazing looking and visibly bright, and they work out well for the other requirements. I plan to get some eventually, but right now I am a broke college student.

[Ace25]

One other thing I don't mention enough, but I think plays one of the biggest roles in LED lighting is SHIMMER. This is an old article written before LEDs were used in our hobby, but it is my belief that LEDs do mimic nature in respects to shimmer/prism effect where as other types of lighting do not according to the article. I have experimented on this quite a bit, with my 24 LED light and no surface agitation to cause shimmer the acropora on the sand bed start losing color and polyp extension after a few days, add the 'shimmer' back and it regains color and polyp extension in a day. The tank has the same flow, I just move the power heads and returns 1/2" downwards so it doesn't cause ripples, and I have an overflow box so I don't think it has anything to do with organic buildup on the surface or O2 related variables, I think it is the prism effect that is making the LED light that much more efficient and my PAR testing also shows me this.

I just tested again to verify and in the center of my tank, 1/2 way down, I get a steady 200 PAR with no surface agitation, when I point the powerhead at the surface I get a range of 170-330 PAR in the same spot, so I am getting over 30% more light at the peaks, couple this with the fact a PAR meter reads 40% lower in the Royal Blue spectrum and the light contains much more of that spectrum than any other type of light, you can see how the 330 PAR peaks in reality can be closer to 600 PAR, which with LED lighting that much PAR can be deadly. There is so many variables to think of.

http://www.advancedaquarist.com/2006/5/aafeature

At least one previous report of photosynthetic activity under conditions of artificial glitter lines seems to have indicated an enhancement in the rate of photosynthesis (Walsh and Legendre, 1983). Although an explanation of this apparent contraction is beyond the scope of this article, it should be noted that natural glitter in shallow waters differs considerably in frequency, intensity and prismatic quality from that of artificial glitter.

Figure 4: Natural glitter lines dance across the surface of a Porites coral in a tide pool in Kahalu'u (Kona), Hawai'i. Note how these glitter lines differ from artificially-generated glitter lines in Figure 2.

Even when various parameters were manipulated in this experiment (such as water depth, surface agitation and distance between the lamp and water surface), it was not possible to replicate glitter lines as seen in shallow tide pools here in Kona, Hawai'i. Natural glitter lines can produce light pulses about three times that of surface intensity (which is immediately followed by 'de-focused' light resulting in pulses of light intensity much lower than average). See Figure 5 for measurements made with a Spectrum Technologies (Plainfield, Illinois, USA) Watch Dog™ data logger and two PAR sensors (one above water, one below). In no case in this experiment did artificial glitter lines produce the high intensities seen with natural glitter lines (data not shown).

Figure 5: Light intensities both above and below water in a Hawai'ian tide pool. Various factors (such as wind and tidal water motion, among others) affect the focusing effect and hence light intensity of the glitter lines. Underwater light intensity at about 7 cm depth can be briefly almost 3X that of surface light intensity due to glitter lines. By the same token, light intensity can be dramatically lower due to de-focusing of light by wave action.

The lack of differences in photosynthetic yield under artificial glitter in the present experiment is fairly easy to explain. Since yield at each light intensity was initially measured with 'calm' (i.e., no artificial agitation, hence 'no glitter') conditions and moments later 'with agitation', it was easy to determine the high and low photosynthetically active radiation amplitudes. The average PAR value of glitter lines measured over short periods (a minute or two) always very closely equaled the PAR measured when a condition of 'no glitter' prevailed. In other words, the average intensity of the light field produced by glitter lines over a few seconds' time cannot possibly exceed the average light intensity in a 'no glitter' condition under the same set of circumstances. One may wonder why photosynthetic yield, instead of electron transport rate (ETR), was used to compare effects. The answer is quite simple – computation of ETR requires a known PAR value be used in the equation, and this was not possible in the conditions of this experiment. The PAM 210 does not include a PAR meter (more expensive models do have an internal PAR meter) and it was not possible to synchronize instantaneous recordings of Yield and PAR (a measured with my Li-Cor quantum meter) in a highly variable light yield such as that seen 'with' glitter lines. In addition, one should not confuse the drop in yield in Figure 3 as Photoinhibition. It is natural for photosynthetic yield to drop with increasing light intensity.

The real question is whether the high light amplitudes produced by artificial glitter lines would encourage a 'ramping up' of photosynthesis able to maintain a higher rate of photosynthesis during the brief (and inevitable) lower light intensities sure to follow the bright 'flicker'. This report suggests it does not. The story may be different with 'real' glitter lines in nature, or in aquaria capable of generating larger waves with high frequency (which could approximate the waves seen in the observed tide pools). For now, it seems that glitter lines are more 'show' than 'go' (photosynthetically speaking) in most aquaria situations.

[Floyd R Turbo]

This is a well known phenomenon, there have been studies that investigated 'flashing' and photosynthetic efficiency gong back to the 1930s, when they discovered that if you flash light at a 1kHz rate you get the same production as constant light. Later studies revealed that the dark period can be 10x as long as the light period and still result in the same production. Anyone who has been to a pool in the summer on a sunny day has seen these "glitter lines". These are essentially performing the "flashing" effect. I have discussed this on other thread, but essentially think of it like the light causing part A of the process to fill a 'bucket' and the dark period allowing a secondary process (strangely enough referred to as the "dark cycle") to 'empty' the bucket. If there is constant light, the bucket can only 'spill over'.

With respect to corals, there have been specific studies that have investigated this effect, many of them, and the above referenced article I'm sure was pointing to one of those studies. These 'glitter lines' are the result of point-source light (sun) and in our tanks this is one reason why people love MH so much (point source) and why LEDs are so attractive.

With the speed of technological advancement of LEDs, it's likely that the average user is going to 'get burned' by following a 'guideline' that is only 6 months old, but they are using newer style LEDs. That's the trouble with technology advancing so fast!

http://reefbuilders.com/2012/12/11/xml2-crees-highest-performing-led/

on a side note, w/r to scrubber and LED lighting, we are likely already getting this flashing/glitter effect both in waterfall and upflow scrubbers due to the motion of the water and bubbles, respectively.

[Ace25]

The big difference here though is 'the studies have shown in the past that shimmer in nature is not equal to shimmer from a MH light'. Shimmer from a MH is just cosmetic, it doesn't provide anything useful according to the studies and my own testing with a PAR meter. This is where I think LEDs differ because it is a much tighter point of light than a MH and LOTS of them. In the beginning days of LEDs people always complained about the 'disco effect' when they spaced their different colored LEDs too far apart from each other and they didn't blend well (I experienced this myself and quickly fixed it so my sandbed didn't look like a 3D movie). Compared to any other type of light, I still get a slight disco effect, but it no longer looks like a rainbow/prism on the sand bed after getting the correct spacing and optics for my setup.

I just don't think people realize this aspect because it isn't really talked about, but with LEDs, surface agitation is a big factor to take into consideration in regards to amount of light and placement of corals. If you measure a new LED light 12" above in open air and then put it over a tank with agitation the PAR readings will be quite different, and shockingly to most, it will actually be higher 12" under water during peaks vs 12" in open air (more so the tighter the optics you use and closer to the surface you place the light). This can't be said for MH because MH is a much larger point of light and radiates out 360 degrees and needs reflectors to focus the light down.

[Floyd R Turbo]

Interesting. I was planning on building an array of fixtures that had a ratio of 5 RBs : 1 CW : 1 NW : 1 WW 2700K : 1 WW 4000K : 1 Green : 1 true violet : 1 Cool Blue : 1 Cyan-Turquoise. Going to have 6 sets of each (30 RBs and 6 of everything else), and I have 6 small Maker's heat sinks so I am going to create "pods". Each color will be on a separate channel (lots of wires!!). I have a bunch of those DIY LED driver boards from TheFishMan65 on RC, each can drive 8 strings of 6 LEDs and can be configured for up to 8 channel PWM dimming (but will probably only do 4 or 6 at the most) and I have an Arduino to control it all. The idea was to be able to test different bandwidths for color rendering. I was planning to group the LEDs very tightly together to prevent exactly what you're talking about with the disco effect.

[crashmushroom]

Just cost myself 30 euro i think wired up 10 ww leds but not working and my cw strip does. I tested them with an led test on a multimeter and they light up so im not sure the driver definately works as the solderless cw rig works. This is my first diy solder atempt i normally buy solderless. :-(

[crashmushroom]

Great all is working had them wired wrong and badly floyd gave me some soldering advice and its up and running. My clownfish look mental with those luxeon 2700k leds im well impressed thanks for the recommendation joel washed out look is gone :-)

[joelespinoza]

Crash im glad you liked it, the 2700k warm whites really seem to help the color depth and gradients dont they?

Ace, have you seen the vitamin c and light post i have up yet? If so any thoughts? I am also curious how much the benificial shimmer effect is lost on my tank with no optics and the LEDS being less then 5" off the waterline.

Floyd, that setup would be good for a testing or demonstration piece, but before actually building a full sized fixture, I would suggest trying out the 4 royal blue : 2 2700K- 4000k warm whites (maybe one of each) : 1 true violet. Warmer whites bring out colors better but the light itself looks visibly dimmer, cooler whites make things looked washed out but the light looks visibly brighter. Considering that you service local businesses you would probably pick your whites ratio based not only on "optimal" but also take into account ambient light and desired atmosphere. IE 2700k warm whites would look great in a dimly lit resturant, but for a brightly lit dentists office with lots of windows you would probably want mostly 4000k-6500k whites with maybe even a few cool whites tossed in.

Also you should take a look at steves drivers, I have heard great things and they are made to be about as plug and play as it gets with aurdinio. They were designed for that purpose.

[Floyd R Turbo]

The ones made by thefishman65 are CAT4101 drivers just like Steve's (in fact, it looks to me like Steve copied them) and are 8 channels instead of 4 and cost $50 each

[Floyd R Turbo]

There is a looooong DIY driver thread on RC that shows the development of the CAT 4101 based driver. I'll have to take a few pics of the boards I got and post them. PWM dimmable and all

[joelespinoza]

There is a looooong DIY driver thread on RC that shows the development of the CAT 4101 based driver. I'll have to take a few pics of the boards I got and post them. PWM dimmable and all

Awesome, I will have to look into that.... I bought invotronics drivers for my setup, and while they work fine I would eventually like to step up 1000mA PWM drivers. I think that with my current heatsink+leds and 7-10 true violets (probably just driven on one of my 700mA invotronics drivers) should be more then enough for a 48x24 120 gallon tank.

[crashmushroom]

Yes i am very happy joel my colours are back like my t5 the washed effect really irritated me.

I am looking at steves drivers as i want to swap out my meanwell pwm drivers.