LED Reef lighting help, consumer input needed!

[ChargerOnDavins]

I work for a company who does LED lighting for vehicles.. and I got to thinking about my other hobby, saltwater aquatics.

I am trying to talk Steve (the owner and good friend of mine) in to jumping in to LED Reef lighting. This guy is an absolute lighting genius and he is receptive.

Instead of just slapping a product together, he wants to know what you (the consumer) would want. Any information would be appreciated. At this point, we don't sell a product. If we did, we would become a supporting vendor. The professional footprint of this website is what attracted us.

What he wants to know is this:

What do you (the potential consumer) want in an LED fixture?

Do you want a retrofit kit? A hang-over fixture?

What color temps are necessary? 12000k? 14000k? 20000k? (I know a bit about HID lighting which we also do)

What is a good price to make LED's worth it?

LED lighting would save on your electrical bill, is a cheaper technology, doesn't require light replacements, you won't need a cooler because of heat, and have FAR more potential. Much more durable, LED's are the way to go. I will be testing a prototype by replacing my T5 with a 48" LED fixture on one of my reef tanks.

Consumer input is necessary to potentially release a stellar product. Fire away, feel free to private message me. We don't make any reef lighting at the moment, we are contemplating jumping in. What do you folks think? I can show you some pictures of the world's first LED halos that we invented!

-Jason

[wld1783]

Just got a pair of LEDs for my BMW 1200 RS.

They seem to be a cool white from twisted throttle. Only use them day time or night off road.

Tagging along.

Bill

[ChargerOnDavins]

Originally Posted by wld1783

Just got a pair of LEDs for my BMW 1200 RS.

They seem to be a cool white from twisted throttle. Only use them day time or night off road.

Tagging along.

Bill

LEDs are a great product!

[g8gxp]

LED's are the absolute hot market right now, and getting in now is ideal especially since very few companies have a solid plan in place. However, this is also a drawback IMO because it's so new (similar to computers) the technology (meaning the lights themselves) will improve extremely quickly, efficiency will continue to become better, packaging and manufacturing processes will become better, etc. Which means you will have to move very quickly and be constantly ahead of alot of competition.

As a hobbyist, of course i would say do it - the more options for me the end user, the better. Start by going through the ton of DIY threads - they have a lot of information in regards to what type of applications have been working and which have not. You can find technical articles all over the place, advanced aquarist has a lot of good stuff about color spectrum, test sheets and data charts, effective life, PAR and PUR measurements, etc.

[zygote2k]

I'd like PAR38 style with the addition of a manual and/or electronic dimmer capability.

[username in use]

I want a fixture with an easy to navigate control panel that allows me to independently dim not just white and blue, but small sections of the whites and blues. It also needs to give me the ability to slowly ramp up the light and slowly ramp it down, none of this on/off nonsense. I also want different options for not only length of the fixture, but width. a 12"/18"/24"/30" width's would be great, and Im surprised it hasn't been done yet with so many people doing deeper tanks these days. I also want to option to choose what optics are on it i.e. 40/60/80 degree so that I can hang it at the height that I want, not the height that the manufacturer wants.

[jtma508]

Personally, I think as far fixture design goes I think the AquaIllumination fixture is the best. Modular, easy to setup, can vary module spacing based on tank landscape. Elegant and simple.

Optics, as I'm sure you know, are a big issue. Optimize penetration and you get light columns. Optimize spread and you sacrifice penetration. I've often wondered if you could impove light spread if the stars were mounted at slight angles relative to the water surface. Not on the same angle or with the same orientation but in a manner meant to slightly disperse the beams.

I think AI would have a better light if they changed to royal blue LED's. That would shift their 'actinic' from about 470nm to 450nm. Ideally we'd like to have some coverage in the 420nm vicinity to get that real actinic look but no one makes high power LED's in that range (maybe add some UV's?). The 6500k whites are where the really high power LEDs live but it might be worthwhile to add another led color to better smooth out the color spectrum (some greens maybe).

As was mentioned, better programmability. Multiple ramp schemes for sure. That's my $.02.

[HCO]

Originally Posted by username in use

I want a fixture with an easy to navigate control panel that allows me to independently dim not just white and blue, but small sections of the whites and blues. It also needs to give me the ability to slowly ramp up the light and slowly ramp it down, none of this on/off nonsense. I also want different options for not only length of the fixture, but width. a 12"/18"/24"/30" width's would be great, and Im surprised it hasn't been done yet with so many people doing deeper tanks these days. I also want to option to choose what optics are on it i.e. 40/60/80 degree so that I can hang it at the height that I want, not the height that the manufacturer wants.

+ 1 and the ability of programming settings like sunrise, thunderstorms, rainy days, etc.

[coolfishy101]

I think the largest problem right now is the cost of LED's. If eliminating the programming would drop the price down dramatically, I would buy LED over halides any day because I can't program my halides and LEDs are way more energy efficient. Incorporating an "opptional" program attatchment would still keep those customers too.

As far as what type of LED, this changes all the time. Right now I think the cree XP-G R5 cool whites and cree XR-E royal blues is what I would go with. Mixing in other colors is a great idea as long as you could blend them in together properly. I would definately seperate the on/off switches for the cool whites and blues though. I would shoot for a 12k - 14k color as this seems to be the most desirable color range right now.

Another thing I would do is pack in more LEDs than the current manufacturers to avoid the "spot light" effect, and to ensure they are powerful enough to grow sps. If the fixture is 175-200 watts and is replacing a 250 watt halide, I would still go for it because I wouldn't have to worry so much about bulb replacent.

Hopefully this helps, good luck!

[jcolletteiii]

I think that with LED lighting, all successful designs will have some kind of modularity. This is the reason I chose to go with AI for my new 135 build. I can start with x number of modules, play with spacing between units and above teh water, and order more if I have to. I think AI has the design right now. However, there are a few ways that the AI design can be improved. If I were in the engineering dept at AI, here is what I would be doing for the next gen of the module:

Fan: absolutely nothing, the current gen AI fan is dead silent and only comes on when called for.

LED lamps: I would definitely find a way to get more blues in there. The current mix makes good color, but you have to crank the blues up to 100% while having the whites at 50 or 60%. In the current gen units, there are 8 groups of 3 LEDs each, one blue, 2 white. Ideally, I'd like to see 8 groups of 4 LED's with 2 rb and 2 white LEDs, and keep the XPGs, those things are monsters! While were at it, lets use royal blues instead of the blues that AI currently uses - these blues do not give as much actinic 'pop' as the royal blues. It might be good to make the LED's user-replaceable as well. Some kind of clip-connector for wiring, and simple stainless steel machine screw/plastic washer combo to mount the stars to the heatsink. This would not only be good for the consumer - who would not have to ship a unit back to teh manufacturer for a replacement LED, but also very good for the manufacturer in cutting costs associated with warranty work - the consumer can self-repair, just send them the star and a couple screws and washers! (...and good instructions in pdf format, of course!)

Housing: I wasn't a big fan of the AI look at first, but it's growing on me. In any case it will be fine for my in-cabinet installation where a 'slick' finished look isn't required. However, I currently have one of these modules hanging above my 12g nano - lots of beautiful light, but the light itself is not beautiful to look at. So, a sexy, low profile, radius edged housing for the 'underside' of the modules would be great, finished in high-gloss black or white enamel. The top would still have to be open for the heatsinks to be exposed to maximum air and for the cooling fans to operate. A hanging mount could be designed right into the modules as well. But some kind of rail system also makes sense for users like myself that may need to adjust spacing between units.

Control: Unlike many, I actually like the AI controller. I do get mad at the push button sometimes when I have to go back through a menu when it scrolls to the next choice as I push the button, but it's not that big a deal. A separate button for confirming menu choices might be better...

Modules: Perhaps offering 2 sizes of modules would be good. The standard 2 rows of 4 LED groupings is good, but maybe also a 2 rows of 5 LED groupings for tanks that are deeper front to back.

[ChargerOnDavins]

Right now we are thinking about "chainable" LED's. Meaning for those who would like to retrofit, it would be easily done... mount and be done. Programming we are looking in to but it seems like the masses really are more worried about price. They'd like all of the above mentioned toys but don't want to pay for it.. does that make sense?

Does anyone feel that LED's with the same functionality as T5HO and Metal Halide would be something they are interested in? A lot of the more "trick" set-ups we may need some help testing and designing. Those who help could potentially try this unit out at cost. Any help would be appreciated.

Thank you all for your responses. This is definitely easier said than done. I too think it would be neat to be able to set sunset and storm and all of that but they don't seem to move very many of those. This is very expensive to start, we are trying to look for a way to cut cost so we can offer the best price, yet get the job done to the consumer's satisfaction.

So far I have gathered that possibly two options available: An advanced model with more program-ability, and a generic model that competes with the same functionality of a T5HO or MH set-up.

Please, the more input you send the better. I'd love to jump in to this boat and provide you all with some quality LED lighting. Steve (the owner) of my company (flashtechusa) has been working with LED technology and is on top of his game. In this economy though, for a small company like us we need to make sure we don't hit any snags.

Thank you all for your responses!

[jtma508]

The programmability isn't where the cost comes from. It's from the LEDs/drivers. To do this right you need to be able to get alot of LED's into a relatively small space at relatively low cost. You're basically trapped by the driver-to-LEDs ratio and that will ultimately define how tricky you can make the controllability.

I'm not sure what you mean by T5HO and MH 'functionality'. A small percentage of T5 and MH users have dimming. Do you mean just on/off as opposed to ramping sunrise/sunset kinda functionality? Again, from everything I've read that's not where your cost is. Apart from the housing, you need to develop an array of LEDs that can compare favorably to T5 and MH in terms of spectrum, output (PAR at depth), cost, efficiency, heat output and subjective 'look' of the light. As I mentioned, you are constrained by the number of LED's per driver and by the number of drivers per power supply. If you use several different LEDs with distinct power requirements you are further constrained in how you combine LED's to their driver.

Alot of the development engineering has been done for you. Look at AI's light. That gives you a decent benchmark of what you need in terms of LED density to match the penetration. You can't drive fewer LED's and be competitive because you'll lose penetration and spread. The first generation of LED fixtures proved that approach won't fly. Where you could improve things is in the output spectrum. That's where both T5 and MH has LEDs beat. Better (lower w.l.) actinic and better 'lift' in the 550nm area.

[ChargerOnDavins]

Originally Posted by jtma508

The programmability isn't where the cost comes from. It's from the LEDs/drivers. To do this right you need to be able to get alot of LED's into a relatively small space at relatively low cost. You're basically trapped by the driver-to-LEDs ratio and that will ultimately define how tricky you can make the controllability.

I'm not sure what you mean by T5HO and MH 'functionality'. A small percentage of T5 and MH users have dimming. Do you mean just on/off as opposed to ramping sunrise/sunset kinda functionality? Again, from everything I've read that's not where your cost is. Apart from the housing, you need to develop an array of LEDs that can compare favorably to T5 and MH in terms of spectrum, output (PAR at depth), cost, efficiency, heat output and subjective 'look' of the light. As I mentioned, you are constrained by the number of LED's per driver and by the number of drivers per power supply. If you use several different LEDs with distinct power requirements you are further constrained in how you combine LED's to their driver.

Alot of the development engineering has been done for you. Look at AI's light. That gives you a decent benchmark of what you need in terms of LED density to match the penetration. You can't drive fewer LED's and be competitive because you'll lose penetration and spread. The first generation of LED fixtures proved that approach won't fly. Where you could improve things is in the output spectrum. That's where both T5 and MH has LEDs beat. Better (lower w.l.) actinic and better 'lift' in the 550nm area.

The functionality I am describing is the 4 modes: Off, All on, Moon light, and Actinic.

The LED spacing issue isn't really an issue for our company, check these out







These halo lights have VERY closely spaced LED's and you can actually drive with nothing but them on and see very well. (Not that it's recommended for legal reasons)

Thank you all for your input!

[benrokh]

First of all good luck with the initiative.
You were right to suggest that cost is a limiting factor and in truth a cost effective answer to the well established MH and T5 fixtures is something that would be quite difficult to achieve.

It will all come down to the simple fact that achieving high light intensity over large areas would require quite a few high power LED's, the cost of which makes even DIY projects quite expensive.

While many noted the success of AI fixtures their greatest limitation is spread. While using LED sistems as pendants is often easier than blanketing the tank with light, the latter often proves more usefull in the long run. Another thing that is often forgotten is that the overlap from a "blanket source" light such as T5 is the reason why they achieve such high PAR numbers while the tubes themselves produce less light per watt than a MH.

Chinese fixtures like maxspect would prove quite a challenge in terms of competition and the better quality AI fixtures and their counterparts would prove hard to manufacture at a lesser price.

Perhaps it is best then to consider a cost effective answer to T5 instead. However as I mentioned earlier the problem with such applications is the number of LED's that would be required to replace even a single tube.

If you could find a cheap enough high power LED you could potentially compete with other manufacturers. Even then you should carefully consider the cost for R&D and the time it would take to move into production.

[billsreef]

Sorry, but this isn't the place to create (and build interest) in a future product.