I sell a lot of frags to a lfs in town and when I mentioned that I was thinking of experimenting with LED's, he went in back and came back with a box of stuff and asked me to evaluate.
Anyway, I haven't gotten the name of the manufacturer yet, and apparently as a way to get around possible patent infringements, this system comes disassembled and could be used for anything, I just happen to be testing it to see if it'll work for a marine aquarium EDIT: Name of company is http://www.coralskyLED.com

So it's set up in modules, 6 LED's to a module, from the lfs owner 3W Cree's with some kind of optics attache, not sure what. Some modules seem to be Royal Blues and others are supposed to be "10000K White". Hopefully I can get more info on that later, with PAR readings and other stuff.
Anyway, 4 modules are connected in an array to a set of 1/2" aluminum rails, all driven by a meanwell LPC-35-1400 power supply. Not sure if it's parallel or series.
So here are some photos of individual modules front and back, and of an array.









First impressions:
Kinda raw looking, very much a retro kind of system. I'll have to figure out how to mount them to a hood or something . Any suggestions about how high to hang them above the water will probably be respective of what kind of optics are used, or trial and error.

With six LED's of the same color on the same module, it'll be tough not to have a kind of patchy lighting effect in the water, unless I ditch the rail idea and set them up as spots. Have to think about that one more.
Any and all ideas and comments on how to evaluate them and arrange them are more than welcome.
Right now I have two arrays mounted, one royal blues and one 10000K white and 8 more modules, half of blue and half white. I might mix and match or even angle them so that two colors hit the same spot. Again, seeking any suggestions anyone might have
Thanks!
Roger

I guess I should have done some of this research before starting the thread, but I promised Evil the info on it quickly so didn't do it. Anyway, here's a link to the power supply. It's a 35W unit, so obviously the LED's aren't 3W'ers. 6X4X3=72W out of a 35W supply. Not sure what happens in that case, the LED's are underrun if they are 3W or are they just 1.5W or less?

Watching.

Still not sure I buy into the LED craze.

Ok. As it stands right now, that driver has a max output of 24v @ 1400mA. This means that it can only run up to 6 LEDs in series, so each module is run in parallel. In english, each LED is only running at 350mA (1.225W if you want to look at it that way). What you could do if you wanted to just use the blue modules is take out the whites, and now everything will be running at 700mA and will have better output. That driver isn't dimmable, so intensity will have to be adjusted by height.


Come on Fosi. Everyone's doing it. Don't you want to be cool too?

I don't blame you Isaac. I've always been a metal halide man. They've produced good growth and with VHO actinic supplementation are pleasing to my eye.
I've also spent a bunch of money on T5's with the hope that my power and bulb replacement bills would go down. Unfortunately I haven't seen the growth that others are touting. If the LED's can produce good growth I'm in, but no reason for a "poor college student" to bother doing the testing when others are willing. When it's time, I know you aren't so set in your ways that you will resist the change if it's been demonstrated.

Mixing the colors on a module will make the color blend better and give a more pleasing apearance, yes. The modules are set up with single colors more out of convinience I think. Many of the places that sellthe populated MCPCBs will only do solid color setups.

Let me back up a bit so I can clarify what you stated Evil.
To define, a module is a set of 6 LED's, all of one color.
An array is 4 modules, for a total of 24 LED's, driven by one power supply.
As it's set up now, there are two arrays, one blue and one white.
Are you saying that I need to reduce the number of modules in an array or just don't mess with the white modules at all?
Oh, and a cursory look at the wiring harness showed that as you said, the modules are hooked up in parallel to the power supply.

I like LEDs on principle because I like efficiency. Not just the idea of "cost effectiveness" but actual electrical, thermal, etc. efficiency.

However, I'm still not on the bandwagon of LEDs as marine tank lighting fixtures b/c I don't trust them. LEDs are great for other types of spectrum-generic lighting but for growing photosynthetic livestock? I've seen some of the same claims as you have but I'm still going to wait for more people to jump in and give them a long-term try.

If the lamps last for as long as people predict and they grow healthy, good-looking livestock, I'm in. Until then, I'll be a MH man I think. I love my Sunpod and I'm not ready to dump another $300 on another fixture.

Maybe in 2-3 years when it's time to move enough data will be out there regarding fixture/lamp longevity and utility for me to justify an LED fixture over my next tank.

If you want to increase the output, you will have to remove modules. For every module you remove, you divide the total current by one less module. 4 modules, 350mA. 3 modules, 467mA. 2 modules, 700mA.

The modules will be wired in parallel. All the positive (red) wires will be tied together, and all the negative (black) wires will be tied together.

I warned you I'd be asking a lot of questions.
so what about "diminishing returns"? In your opinion, since the fact that you'd have more power over a smaller area, and given that some of the setups I've read involved dimming down the led's because they were too intense, what setup would you think would be best? And without using a lux meter do you think that just go with the setup as is will produce any satisfactory results? I know it'd just be a guess and sometimes I've been lucky, if a light is really bad you can figure it out pretty quick, but if it's borderline it can take quite a while to realize, time wasted.
Something I've been wanting for a long while is a light meter. Looks like I'll be ordering one. Hopefully also I'll get some more information about these setups.

I'll be testing it for you pretty soon in the Pico I'm setting up now.
You can thank me later.

Well, 350mA per LED isn't going to do a lot for you by itself on your tank. If you wanted to use the setup as is, and it's certainly usable, I would order another one of those drivers. Put the whites on one and the blues on the other. This will bump the current up to 700mA and will make the setup far more usable.

Don't waste your time with a light/lux meter. It's not going to tell you anything really useful. What you need is a PAR/quantum meter. Not cheap though.

Owww. You're sure right about the prices of those quantum meters. $3-400.
Okay, so if it's as simple as adding another driver (or I'm assuming, using a single more powerful driver), I can do that. There are two more unassembled arrays in the box, so it'd be a simple matter to take one of the drivers and hook it up to two of the modules, so 2 modules per driver. But is that assuming that the led's are Cree 3W? Which would also mean that they have been running off 1.225W? I didn't realize they could run at a lower wattage than "rated". If it turns out that they are not 3W bulbs, would hooking up two modules per driver blow them out or shorten their life?
I did get some response from the lfs owner. He said he'd get the info "by Friday". Hopefully that will shed some more light on the subject.
On another LED front, I just got confirmation that the 15W Evil lamp was shipped today.
15W=5 LED's vs. 35W=24 LED's. I can definitely see a problem there.

Well I for one hope it is not just a craze. I have had many PC and MH setups in the past. Long ago I eve had one of those 175w MH setups with the huge fat bulb (forget what they called them). At the time it was the only type of MH setup readily available.

I am about to setup a small nano so I ordered yesterday a eco-lamps KR91-14. They now come with the programmable timer like the KR92's. I would have gone with you evilC but wanted a sleeker look and not really into spot lights for this setup.

Their PAR spec for this is about 310 @ 8" and 270 @ 12" (distance to sand I am planning) so if they were telling the truth in their PAR specs all should be good for a small nano even down on the sand.

Anyway had to sell a kidney to pay for the setup so I hope it works

Sort of. Lumen values *are* a good reference for PAR *if* you have a fixed reference point. Let's say, you have a particular fixture that puts out N-lumens of cool-white. If you have another fixture or array that puts out double the lux of cool-white, it's a pretty safe assumption you have about 2x the PAR.

While LEDs certainly vary in terms of efficiency, it's a general trend that within the cool-white category they all have pretty much the same spectral curve because they are all using the same technology. A cool-white LED that throws 200lumens is going to have twice the PAR as different cool-white at 80lumens. Since Evil has the PAR meter perhaps he can confirm or deny that. It's this nature of LEDs that make them superb for plant growth because unlike MH or T5, you have at least some industry conformity towards spectral emission.

However, you can't calculate much from how much power the fixture uses because if we use Cree as a reference their LED efficiency varies tremendously from flux bin to flux bin. They could be 240lumen at 3-watts, or 110.....who knows. You really have to fire up the fixture and measure it as a whole.

I like the idea of ditching stars and re-flowing them onto a heatsink because it simply saves money. It also allows you to get the emitters closer together and eliminate color mixing problems, although it's not used here. I just wish I had the gear to do this.

Also, do the fixtures here have optics? If they don't, I don't get why you couldn't just use monster die arrays and reduce the cost lower.

They do have optics, but don't start me lying about what kind they are. Information on these setups has not been forthcoming.
I think you are right about the lumen meter also. As long as I use it as a comparison between different setups I've used or have I think it could work okay. Just need to keep it home and not make comparisons about other people's numbers.
Wish I had something now though. I'll be honest. While I know the eye is a pretty coarse comparator of intensity, those damn royal blues are INTENSE! The fluorescence they give to a large percentage of the corals in my tanks is downright freaky. I need to set them up for Halloween because the colors are more vivid than a black light poster. If I was smokin' herb I'd be "Freakin' Out, Man".

Thought I'd renew this thread, as I've got an array of 2 white and 2 blue modules over my 10 gallon tank contest entry.
If you are interested in any more information on these lights, the contact is http://www.coralskyLED.com
The stuff I got is designed to be installed by professional installers at lfs, but the company may have underestimated the "power of the DIY'er"!

I wish I had seen this thread a bit earlier, so I had a chance to minimize the speculation and maximize the information regarding our systems. All the while you were hoping to get more information from your LFS, I was leaving multiple, unanswered voicemails and emails, trying to find out what he was doing with the systems we left with him.

Yes, these systems are intended to be sold and installed through our dealer network. Unfortunately, your LFS will not be a part of our network. Without knocking the DIY community (I am a DIY-er myself, after all...), we are really trying to avoid folks from trying to figure out our systems by trial and error (or shocks, in Yardboy's case). There is nothing very complicated about the system though, as Evilc has already confirmed.

I do want to address a couple of things that were mentioned in the thread. First of all, our systems consist of 24 Cree XR-E LEDs on 4 separate aluminum core PCBs (we call them "engines"). The small lens you see on each LED is actually part of the device (called the "primary optic"). We decided not to use secondary optics in this system in order to improve color blending and reduce cost. Using a secondary optic will increase the intensity of the emitted light at the expense of beam angle. If you have ever seen an LED system that produces a "flashlight effect", where everything within the light beam is very bright, and everything just outside the beam is very dark, then you were looking at a system with secondary optics.

Our systems can be supplied in 3 standard colors: a 10,000K "white", an 18,000K "reef" and a 455nm "blue". The white engines are actually a combination of white and royal blue LEDs in a 5:1 white:blue ratio. The reef engines have a 2:1 blue:white ratio. The blue engines are just royal blue. (Custom colors are possible, but you must work through our dealer network, and we will help design each system to the requested color temperature and light intensity.)

Our systems do not come mounted in anything. It is up to the installer to mount the engines in a frame or enclosure. The linear arrangement that Yardboy photographed was something simple that I built from scrap aluminum channel in order to demonstrate our system to your LFS without having to juggle 4 engines and a driver.

We selected 35W Meanwell drivers for a couple of reasons. First and foremost, since we designed a system to be installed in a variety of locations, including enclosed environments, we were very much concerned about heat. Cree LEDs must be kept below 125 C at the solder junction. If they get hotter, their lifespan will be adversely affected, or they may fail. Increasing the drive current over 350 mA greatly increases the cooling requirements (it's not a linear relationship...driving an LED at 700 mA takes more than double the heatsink capacity than an LED driven at 350 mA).

Our systems do not require any active cooling as designed. We did not want someone to lose their system just because a $12 fan fails.

Finally, there is a reason that Cree reports the light output of their units at a nominal drive current of 350 mA. This is the "sweet spot" for Cree LEDs, which means that the LED is operating at its most efficient state, in terms of power consumption and heat generated. At 700 mA, you do not get twice the light output from the LED than when it is driven at 350 mA. In fact, it takes a full 1000 mA to double the light output from 350 mA.

In short, while you could certainly expect more light out of these LEDs by overdriving them, if you are not extremely careful with the heatsink design and provide active cooling if needed, you may end up shortening their useful life. Additionally, you will end up lowering your lumens/watt, essentially reducing their ultimate efficiency.

Yardboy, I really like your 10g setup, and I hope you are satisfied with our lighting system. I would be happy to assist you with any technical information you may need.

If anyone has more questions about our systems, you can contact us via this thread, or my email below, or feel free to post on our
facebook page.

Sincerely,

Jeff Littlejohn
www.coralSkyLED.com
jlittlejohn@coralSkyLED.com

Thanks for the information Jeff. Clarified a lot of stuff for me, at least.
I will be posting the pics and description of how I laid out the light fixture for my 10 gallon shortly. It was really quite easy. Your engines really take most of the DIY out of it, leaving only a little innovaiton in making up the fixture itself.
And as for the shock, it was a connecting nut between the engine and power supply that came off, so it was only a little tingle, not as bad as when I stuck my hands in another of my tanks only to discover that a powerhead was shorted. YOW! Now that one really tingled!

Any chance you can do a comparison between your ghetto tank and led tank?
eg, take 2 frags and put one in each, then compare over period of time?

Also update be nice

I could do that, but unfortunately I didn't at the beginning, and the 10 gallon tank I have set up with the LED's is a contest tank that I can't add any more corals to. The conditions in those two tanks would really have to be set up with as many variables the same in both except for lighting. My test now is to find out if, with these LED lights, corals will grow. A later test could include what you are asking for.
Coralsky offers three different setups, the circular engines containing different combinations of white and blue LED's.
Actinics - all 6 Blue LED's
18000K - 2-white, 4-blue
10000K - 5-white, 1-blue
I started the contest tank with two all blue engines, and two 10000K. The white LED's very much overpower the blues in visual perception, though you can tell the difference between the different combinations. On the other hand, using the different engines aligned over the tank didn't cause any problems with spotting, or seperated areas of white and blue.
I like blue tanks, but all blue LED's caused certain corals to fluoresce heavily, while making others appear nearly invisible, plus the tank looked too dark.
Jeff offered to swap out the 10000K engines for 18000K ones.
This has resulted in a, to me, very much more pleasant looking tank. A few corals do look more washed out with this combination, but overall I like the effect.
Here is how I've got the engines arranged in a simple fixture, from the top:


and from the LED side. I had to stop the camera waaaaay down to get this shot, I hope I didn't damage the sensor as there were lines shooting from each LED on the screen when I took it!

Here is another tank, showing how I mounted the LED's in a different way. I just took an existing canopy, that originally had a PC fixture mounted in it, and using aluminum angle and channel mounted the engines to it.



Please don't laugh. This tank has been set up for 4 years, a 10 gallon with corner overflow, species tank for a yasha hashe goby/tiger shrimp pair. that is connected to a 400 gallon system. The lighting was a 70W MH with Astralux 14K bulb, but it hadn't been changed in several years. There was also an infestation of majanos, aiptasia, and red cottony algae. After I put the LED fixture on it, I began to clean it up and it's looking better, but still pretty poorly. After I get it squared away, I'll add a frag of some sps and also put a similar frag in another tank on the same system, under 175W metal halides. That way they will both be seeing the same water, anyway.

any update on this little project. photos of growth please please pretty please.

Everythings growing, except my collection of photos!
Look here

Do you anticipate any problems with the Coral/LED patent that I've heard about?

I thought I'd bring this thread back from the dead since I'm convinced that LED's are here to stay. I was in an LFS in Gainesville Fl a few weeks ago and the owner, who had a very nice shop told me that LED's were "not quite there yet" and when I told him that I'd been keeping sps under some, he expressed some doubt.
While the CoralSky's are not the most powerful LED's on the market, the results speak for themselves, compare to my previous post from Jan. Sadly, the large purple monti I nearly lost when the tank got too hot in the summer (remember this tank is set up on the sump of my 150 display and frag tanks so there are a lot of MH's heating the water in the other tanks) but the growth of the green slimer, Caulastrea and brain corals has been quite good, the slimer developed much thicker branches than the same corals under metal halides, though it has even less flow on it. Doubters, give it up.

High five dude.

In defense of the LFS guy, I hear the same thing "LEDs aren't there yet" from every store I've been in. This brain damage is based on three factors; the need for LFS stores to sell legacy lighting gear to stay in business and the fact commercial LED units are all over the place in terms of performance. While on a frag hunt this past fall I visted several living rooms running big tanks with obscene halide wattages and not so marvelous color. The owner(s) were whining about the electric bill and putting off replacing bulbs, but were skeptical of LEDs. Hence, I've made the offer to all the reefers in my area to build any size LED light at cost simply to get them off their halides, and I'll guarantee the same if not better PAR.

Reason 3 is more controversial, but one thing that really irks me is 30-40 years ago your average HS kid would have little trouble slapping together an LED light -vs- that same kid now. That's...because we actually made things back then from an industrial / manufacturing perspective and school science classes reflected this. I've noted that techs over the age of 50 seem to have have far fewer issues initially grasping LED circuits than those under 30. {shrug}

The guy bad mouthing LEDs in the LFS store simply doesn't get it.

Interesting observations. What can I say, I'm 54 and feel I've maintained my edge by being willing to accept change. It really surprises me too, as your observations about younger generations are mirrored by my own experience. Jeez, I have a 30 something friend with a PC setup, that he bought new! And you know, this whole attitude is mirrored in the greater society, where managers are willing to tolerate (and pass on) high O&M costs while avoiding Capital costs that will make the O&M costs lower.
I applaud you for taking matters into your own hands. I am promoting LED's through my local club, but don't have the time or money to puruse the ideals that you uphold! I to try a little harder.
What can I say though. I'm definitely not a trend setter, not willing to pay more to get less. I bide my time and wait for others more eager to have the latest and greatest to bring the price down! With LED's, I think one of the things that might end up being more of a help than a hindrance was the patenting of the use of LED's over tanks. Forced us all to build them ourselves to get the best product available.
I like to note dates though. Today, I bought my last spiral CF bulbs for my refugium. New LED PAR 38's are out for $25 and the expensive ones that were selling for $50-60 are on deep discount to get rid of them. For now the new ones are all warm white, but soon there'll be daylight ones too, then it'll be goodby spiral CF's, and good riddance.

That tank is absolutely beautiful. Thanks for the update Yardboy!!!

I absolutely love to hear this, because the kid from the LFS who is practically jerking off his marine biology professor, is going to have to suck it when I show him the growth on the blue tort that he claims I won't be able to grow under my LED lighting!

You have no idea how much trouble you are going to get me into with a local forum with that above statement. I linked to this post to demonstrate some nice coral growth, and, well.....long story lol......

redfishsc, you can't help what other people say, and while I know what genj is tryng to say, well...
At least he's excited about growing corals under LED's!

One thing I have to remember though, is that reefing requires paying attention to three things,

Hehehe, sorry red. But I've been put to the test by this guy. Don't get me wrong, as I like all the people at the LFS. They are comfortable with doing things a certain way, and I don't blame them. They do have a business to run. I'm hoping to be the blasterman for the area and help them see that it does work and that the tech is ready to adopt, even if it is a hybrid manner using LED and t5.

Oh, and I'm just as anal about water quality and motion... it's just that the LED aspect of the hobby really gets me going.

Since my first marine tank, I've been a "metal halide man". Jeez, I even had a 175W over a 10 gallon tank, and still keep them over my big tanks, and all this time I've read Sanjay Joshi's articles in Advanced Aquarist on which bulb/ballast combos put out the most PAR. He's had the definitive info on metal halide systems as far as I'm concerned.
Well, finally he decided to compare MH. T5's, and LED's. the results may surprise you, though the "LED-heads" expected it! Read it and weep, all you non-believers!

The original power supply for the lights is a LPC-35-1400 and it wa suggested by Evil that I run two of these on the setup to get more intensity. While the manufacturer talked about the sweet spot and heating issues, the warranty is over and the second set of this setup I had I've used over half of my 40L Gulf of Mexico tank.
I'm setting up a DIY over the other half, using a RapidLED kit and want even intensity so I'm wanting to order another meanwell. Rather than the LPC-35-1400, would either of these work?
They'd be used on 12-Cree XR-E LEDs.

Mean Well LPC-35-700W
Mean Well LPC-60-1050

Yes, they should replace the 35-1400 just fine.

You need 2 of them obviously, they will handle 12 XRE's each. How many of the 35-1400's did you have? It seems like they would only handle 6 or 7 LEDs each unless you ran two parallel strings of 6 (meaning, same voltage output, but the drive current would be half... and you risk a cascade LED failure if one or more blew on you).

I didn't notice the 1400mA drive current, that's pretty nuts for an XRE. For XPE or XPG it fine, but I blew an XPG when I was running them at 1440mA.

If you want dimming, get the ELN60-48D for a few dollars more (worth it... by far... if you ask me).


Your corals are used to some pretty wicked intensity if you were running them at 1400mA so you probably don't want the 700mA version.


I would be interested in your used driver if you are looking to sell them, send me a PM .

I see you edited your post, pardon if I'm a bit confused, I guess I need a "driver model number" lesson.
Right now I'm running 4 sets of 6 XR-E LED's in parallel on the LPC-35-1400, which resulted in each led running at <1.5W. That's what the 10 gallon tank in the previous post is running on, and doing fine.
I've got another of those setups I want to put over a 40L breeder so I want to raise the intensity level.
Evil suggested I run just 2 sets of 6 in parallel on a single driver, so I 'd need to buy another one, but I can't find it. Rapidled has the LPC-35-700 so what would the 2 sets of 6 in parallel be running on that?