chucelli

This statement is not accurate, as one of the main reasons for not using fans would be to cut down on manufacturing costs. Lumen Maintenance This spec sheet shows test results and projected lumen maintenance of Cree X-Lamps under conditions of varying ambient temperatures and its' effects on junction temperature. Employing active cooling (fans) will substantially increase range of "safe" operation. Not using fans for units with densely packed LEDs doesn't mean they will burn out all the sudden. It just means that they will not maintain their rated output as long as an equivalent unit with a fan, especially in extremely hot environments. -Robert

Not a good choice even given space. They have very low long term survival rates and eventually starve. Ask paulb, I think he got lucky with one... Forgot to add, Green Chromis, although don't technically school, look very nice in large numbers and I found contrary to what most have experienced, long term survival rates can be 100% if given multiple feedings. Basically, if you treat them like the much more expensive Anthias, you should not lose any to aggression. IME, the reason they start dwindling one by one is due to the fact that the more dominant ones harass the weaker ones and prevent them from feeding. Feeding methods need to be modified (ie. multiple feedings per day, smaller food particles with strong current to spread food to all territories) so that even the weakest ones can gorge themselves with food before the day is out. -R

They are offered here http://www.aquariumspecialty.com along with some other LED choices. -Robert

I don't mean to be a buzz-kill, but I would like to clarify some misconceptions many people have regarding LEDs. Hopefully this information will help hobbyists make more informed decisions in an arena fast filling with misleading ads or hype. First off, wattage doesn't mean anything when talking LEDs. There are many flavors of LEDs manufactured by a handful of companies. 90% of which will degrade when pushed to limits needed to provide adequate intensity for our reef tanks. Therefore, detailed specs are usually not published. Instead, they focus on wattage or other universally understood advantages of LEDs like energy savings, cooler running, and the overused and claimed 50,000 hour lifetime. That 50,000 hour rating depends largely on one factor; heat management. This is something which manufactures don't always design adequately because they need to save money. Also, with higher emitter densities, cooling needs increase. Take this "Photon Cannon" for example. This 50w EdiStar is made up of 49 ~1 watt emitters pushed together into a relatively tiny area. The 50,000 hour rating is dependent on keeping the junction temps under 65 degrees C (roughly 150 degrees F). It would take quite a beast of a cooling system to dissipate the amount of heat which will be generated from this small area. Regarding light output: The Edison spec sheet says this 50 watter is rated at 3200 lumens. This amounts to 64 lumens per watt. To put things into perspective, the bins of Cree emitters most DIY'ers and AquaIllumination uses, rate a minimum of 100 lumens per watt. The newer XPG emitters run between 140-160 lumens per watt. This may not mean anything to the Average Joe on paper, but that is twice the amount of light output for the same wattage! I don't have Cree stock or anything, (missed the boat on that one) but there is a reason why the AI units are expensive. This new market is quickly emerging, but unless hobbyists become more informed, manufactures will just keep creating hype and take advantage of fads and trends with little to backup. -Robert

To minimize the amount of already existing confusion and misinformation, let's just make sure we're comparing apples with apples. The PAR38s have emitters spaced ≈1.5" apart, while El Camaron's fixture design uses a ≈3-4" spacing between emitters with a mixture of 60 and 80˚ optics. They are not driven at the same current, nor use LEDs of the same Bin. If one wants to see differences in output of the various optics, my original LED thread contains a comprehensive table of PAR readings with varying optics conducted in an environment where all relevant variables remain constant. Also, "Taqpol" from RC recently conducted some very thorough PAR readings of his DIY build, taking coverage into account. It is accompanied by 3D graphs which plot intensity vs coverage using different optics. Since intensity (PAR readings) is directly proportional to coverage with optic changes, it is important to factor total number of LEDs and coverage provided when attempting to compare PAR readings. -Robert

My Stealth was luckily in my water change tub when I discovered it was stuck ON and leaking current into the water. I found out when I got a pleasant shock when I stuck my hand in the tub. The RC post linked above contains postings from multiple users who have had failures with their stealth heaters, including explosion. I'm not sure how they can get away with releasing such a product. Seems like a lawsuit waiting to happen. -R

Those numbers are very consistent with Inline designs due to the parallel arrangement and spacing of the emitters. This design favors even coverage over intensity. Additional arrays could be added to create overlaps, which would raise PAR numbers. However, I don't think it necessary judging from the numbers you are getting when mount height is factored in. If you were to lower the fixture to match previous T5 mount height, the PAR numbers are sure to exceed current numbers as well as the previous T5 setup. Having said that, I don't think you should be changing anything if your corals are growing and displaying pleasing colors. If you want to up the intensity, you could lower the unit an inch at a time, keeping in mind that intensity changes with height more so than with other forms of lighting (T5, MH). -Robert

Here's my submission: Robert My 75 display Nikon D40x -Unsharpmask -reduced blue Teal stag on top of Yongei

Couldn't agree more. There's a reason why he's been there for 30 years. -Robert

The problem is, I and probably others, are not convinced that your skimmerless systems are capable of sustaining SPS corals for the long haul. There are not that many SPS corals (let alone mature colonies) in your tank... unless there is another tank you have not posted pictures of... -R

I would seriously re-think, or at least bring to the end-user's attention, that he/she is paying $24 per led when using those PAR38 LEDs vs 23.60 per LED for an Aqua Illumination unit. The price difference may seem negligible, but consider the additional engineering (better cooling, modular design, published specs, etc...) in the AI unit and the difference is not so small. -Robert

Nate is going to go buck wild once he get's his new 300g display situated.

old reefer, a dozen of the whites with tight optics will be the way to go. However, the shimmer would definitely be slightly washed out. Hard to say since I've never seen that application in person... The shimmer could be further enhanced by limiting the amount of T5 whites so there is more contrast between the two light sources. -R

Jim, jut FYI, for 300 3w Cree emitters, you are 3 Meanwells short if we are talking about the 48v Meanwells. Anything more than 12 emitters per 48v Meanwell is pushing it slightly. To your question, this inline design does not need active cooling due to the relatively low density of LEDs in the given area. -R

audible, tanks pictured in this thread are not my tanks. I only designed and built lighting units for them. You can go here to post comments relating to Renato's tank: -R

EDIT: Channel length for Renato's 70g cube is 2' not 3'. I have too many numbers running around in my head. -R

Do not pull at it no matter what you do. You risk tearing the bysal foot. If you can't reach the filaments with scissors or razor, it's best to chip off the piece of rock it's attached to. -R

Thanks guys. Ideal (and conservative) emitter spacing I have come up with through testing is 3 emitters per foot of this type of channel, when LEDs are run at 700mA. This could be pushed, but I would not be comfortable putting them next to hot halides with densities exceeding 3 per foot. Jim, the cube is 70 gallons in volume I believe. Channel length for this particular build is 3'. Here is a link to his dedicated tank thread: http://www.wamas.org/forums/topic/33546-el-camarons-70-gallon-cube-upgrade/page__st__50 David, endcaps are an option. However, maximum heat dissipation is better achieved leaving ends "open". I want to add that, although there is more room in skimping on cooling, I opted for a more conservative approach, due to the fact that I want to ensure my units will maintain lumen output for the spec'd 11 years (when run 12 hours per day), since any decrease in output won't be detected by the user until a 50% output decrease has happened. Without a robust cooling design, it is hard to calculate proper replacement intervals. -R

Forgot to add, for comparison reference, The unit at TFW is mounted next to a 400w SE metal halide lamp. The one at TL is mounted next to a 250w HQI metal halide lamp. -R

mcmaster also has solid channels: http://www.mcmaster.com/#3230t63/=60klur -R

Hey Renato, adding some of my pics.... using same WB as my own tank shots. Your water is definitely clearer, since there is a lot more blue showing in your tank pics compared to mine... No post processing except unsharpmask. Closer shots. Post processing: Reduced blue levels to 0%. Unsharpmask. The two pictures below are closest to in-person. What do you think Renato? -Robert

Finally got a chance to document a recent inline style custom build. This has been in the works for a while, developed in tandem with the pendent style, but didn't get a chance to do any proper documentation until now. Emitters used: Cree XR-E Q5 bin CW and 450nm RB. Emitter number: 12 per array, total of 3 arrays = 36 LEDs for this build. Drive current: 700mA LED color ratio: 1:1 blue/white Optics used 80˚ center unit, 60˚ side units. Inline Array12. testing. Delivery! Originally designed as three separately adjustable units, it was decided since there were corals growing on the glass, that one single unit would work better. So the trio were fused together. FTS. Wires still need to be organized and hidden. Some closer shots to show color. Anyone who is interested in seeing these inline systems in person, talk to Nate or Renato. The original pendent style units can be seen at Tropical Lagoon in Silver Spring as well as Tropical Fishworld in Gaithersburg. -Robert

Av8BlueWater, sorry for the late reply. I just noticed this thread today after Jaesun posted... My RC thread is located here. It contains links to some of the most informative info regarding LED lighting. You can also check out my article at ReefAddicts here. I think it was posted earlier in this thread, but I am using Cree XR-E emitters from the Q4 bin. I recently switched to using Q5 bin emitters for new builds because suppliers are no longer carrying Q4s. -R

I don't think I ever argued against your point regarding bleaching undersides, or darker caves and shadows. When I was running metal halide, some of the larger colony undersides started receding also when I removed my sandbed (due to removal of the reflective bottom surface). You are correct in that LED output tends to be more directional, which is why system specific variables need to be taken into account when drawing designs for specific setups. The same recession will be experienced in any system if you simply tilt some of your larger colonies at different angles different than what they have grown accustomed to. This effect is not LED specific, but may be more exaggerated. This is another example of fault with application and not technology. This particular problem can easily be resolved with smaller multiple fixtures providing light from more than one angle. Smaller colonies and frags will never experience the problems you mentioned since they will naturally grow to maximize surface area exposure under directional LED lighting. Finally, I also want to add that my apprehension with commercial units is not so much with where it is manufactured, but rather the components used, and the fact that detailed specs are not usually offered to the end consumer. In addition, the lack of customizable options results in an ineffective "one size fits all" model. -R

LED technology for use in captive reef systems is still a relatively new concept, as such, applications of this technology still needs to mature. Note, I am not saying the actual technology needs to mature. Because LEDs output light in a very different fashion when compared with traditional metal halide or T5 lighting, design and applications need to be customized to take care of differences and utilize advantages. The problem I see currently with LEDs is the countless variables involved when attempting to apply this technology. Many simply do not understand, or still are thinking in traditional terms, like wattage, reflectors, color temp, etc... In terms of what's best when talking component specifications, a DIY fixture (assuming you have the know-how, ability, and tools) can always be more advanced than a commercial unit, simply because the resources available to us hobbyists have grown tremendously, and you have total control over the quality and relevance of each and every component. I don't know about the most current commercial units, but I have read on various forums and heard from a few LFS owners that the Solaris units had many issues, including but not limited to different parts used over several units, LEDs actually burning out, etc.. Of the many DIY'ers, There has not been one single report of an LED failure, which is more consistent with the technology. I believe it has to do with the quality components people tend to spec out for themselves when building their own unit. Commercial companies have less reason to do so and more reason to cut corners to save production costs. Several important things to watch for (which are not always published by the manufacture) when shopping for any LED system, are: 1. LED type and Bin. There are mainly two trusted LED manufactures, when it comes to reef lighting, as defined by majority of LED builds. Philips and Cree. Cree currently seems the winner, as they currently have claim to the most efficient production LED on the market (lumens per watt). They have also been extremely reliable. Within any brand LED, the emitters are sent through QA and separated into bins. These bins determine the overall quality of the LED within that company's line of LEDs. Differences between the lowest performing bin and the highest can vary by more than 50%. As such, price differences between the different bins also vary greatly. This does not mean very much when it comes to commercial lighting or cutting edge flashlights, but it will matter when we are talking about lighting for a reef tank. 2. Drive current. The amount of light a particular LED can output is directly related to how much current is running though it. The higher the max spec'd current, the more current that particular LED can handle. However, with higher currents come higher temps. Keep in mind, this is not the kind of heat that would add anything to the ambient air, but it is this relatively small amount of heat that can shorten lumen maintenance (the amount of time the LED keeps specified light output). It takes more than a 50% decrease in light output, before the human eye will register a change, so keeping the LEDs cool is key to lumen maintenance. 3. Optics in design. choice of optics can mean the difference between 50w metal halide or a 400w metal halide, all in one LED fixture. It is perhaps one of the most important and overlooked aspects of LED lighting. LEDs do not use reflectors, as it will do little good, since the light emitter wafer in these LEDs is a tiny flat square 1-2mm in diameter. There is absolutely no light beyond the 180˚ half circle, which is why optics are used to gather that light into something less than 180˚. The tighter the angle, the more intense the light. Optic angles play a direct role in intensity vs coverage. 4. PAR output at desired color temps. Many of the complaints regarding earlier commercial systems were related to inconsistent PAR output at a given color. This is due to the fact that these manufactures allowed end users to change overall light color through the use of dimming. The problem with this approach is that once the user has settled on desired color, the total output can be very different then the unit's factory specs. Dimming is not an efficient approach to color manipulation, since it involves running certain banks of LEDs at less than ideal efficiency rates. Modification of LED ratios is a better approach to achieve color temp (example 1:1 blue/white vs 2:3 blue/white). I think LED technology for captive reef use is here and ready, it is the application of this technology which still needs to be explored and refined. -Robert