Minimalistic multichip DIY LED build

would this be a good buy? Im thinking two blues and one white. thats 60 watts and i've been told by others with biocubes that that is plenty. i just want to know if you think these would do well. is it a good idea to buy these leds with their intended drivers?
http://www.ebay.com/itm/1pc-20W-WHITE-HIGH-POWER-LED-1pc-AC-85-265V-LED-DRIVER-20-Watt-/400306040077?pt=LH_DefaultDomain_0&hash=item5d34196d0d#ht_2925wt_1127
http://www.ebay.com/itm/1pc-20W-Royal-Blue-High-Power-LED-Lamp-AC-85-265V-Driver-20Watt-/400306040020?pt=LH_DefaultDomain_0&hash=item5d34196cd4#ht_2925wt_1127

Considering this is a 6,000K white LED the 2 to 1 ratio should work good for you. My only concern is getting a good balance of the color with only 3 LED's It you set them up in a triangle the side with the two blues might make that side of the tank look slightly bluer than the other sides. Personaly I'd put the bluer side to the front of the tank and would not use lens to give the colors a better chance to blend together.
 
In my experience, 60 watts without lenses is too little for a tank type biocub (around 100 l) if you want "SPS light" down to the bottom. The blue light is not just for aesthetics but it has a biological purpose, too. Giving more power in the areas that the corals need for their photosynthesis. Do you use, in addition, a white source at 6500 K, you need more power in the blue area. If you work with lenses so you can for example place a series of RB (3 - 4 pcs) with 90 degree lenses in the rear and in front of them two white pieces with 60 degree lenses. Personally I would prefer to use white with 10-16000 K. Dimming is also good - then you can adjust the light as you want. Do you use, 10-16000 K so you do not need the same distribution of white and blue. 1:1 works well.

Sincerely Lasse
 
I did ask the question in the lighting forum and got some replies, but I think you guys here might be able to help much more.

This is an existing 75x20x12 frag tank (water level is actually about 10"). That's 190x50x30cm for Europeans with water level at 25cm. Does anyone have an estimate on how many hybrid LEDs would be needed to light this for SPS corals?

We are thinking about using these LEDs: http://www.ebay.com/itm/EPISTAR-20W-Super-Actinic-Blue-Hybrid-Led-Panel-Aquarium-/220990452654

Thanks a lot!
 
With no lens, a 100w hybrid *measures* 410 PAR at 1" below water, and 137 PAR at 18", with an apogee MQ-200. Note I said measures, because supposedly the apogee is not perfect at measuring the blue.

Probably better off with a row of 20's or 30's.
 
This guy claims his 20" heat-sink supports 50W LED with no fan.

Do you think that is true ?
The profile of the aluminum in the link you provided might be able to handle 50 watts, but that's if the power is applied across the surface in a more or less uniform matter, at least in a length wise configuration. Putting a single LED on it, it's damn near 20 inches long! will more likely than not be inadequate for the application you want to do. Granted it may be fine, I'm just saying if he took the rating of the linear heatsink and applied that to a point source, it's wrong.
 
I did ask the question in the lighting forum and got some replies, but I think you guys here might be able to help much more.

This is an existing 75x20x12 frag tank (water level is actually about 10"). That's 190x50x30cm for Europeans with water level at 25cm. Does anyone have an estimate on how many hybrid LEDs would be needed to light this for SPS corals?

We are thinking about using these LEDs: http://www.ebay.com/itm/EPISTAR-20W-Super-Actinic-Blue-Hybrid-Led-Panel-Aquarium-/220990452654

Thanks a lot!

The aquarium is shallow and that's most likely enough with 20 watt chip to get the "SPS" light on the bottom. Likely, it works with 10-watt chip's also but then it is probably necessary to use lenses. If you use 20 watts, I would personally try without lenses first. The problem with LEDs is often that it is too bright light rather than too weak so I would also prefer to use dimmable drivers. How many and how they should be placed is not that easy to answer. That I would prefer would be to arrange them in two rows along the aquarium. Approximately 20 cm between the two rows. The first line of about 15 cm from the front window and the other 15 cm from the rear window. Weave LED chips by the first line starting approximately 30 cm from the short side and in the back row start to 45 cm from the same short side. Then leave about 30 cm between each chip. This means 7 chip in the front row and 6 in the rear. There are many watts if you choose 20 watts but most likely needed to get an even distribution. I think it's important to test - my distances are just guesses. You could also mix 20 and 10 watt chip.

Personally I would opt out of the hybrid solution and instead use the white 16-20000 K in the first line and RB with a wavelength of 445 nm in the back row. But this is just my personal opinion.

Sincerely Lasse
 
Lasse, how many watts would be good for a bubble tip anemone?
an how about for sps in a frag tank of the dimensions : 10"L x 20"W x 12"H
 
@ JoshuAcOOk: I do not really know because I never had a bubble tip in my tank. Anemones are notorious for crawling around and choose places that they enjoy themselves. Perhaps you can control that behavior by having different light levels at different locations so they may choose for themselves. At 12" depth, however, both 10 watts and 20 watts is probably enough, especially if you use lenses on the chip´s, and I believe they give "SPS" light also down to the bottom of the tank with lenses.

If I never tried something myself I always write that I think so and so. There is no guarantee that it works - it makes each one to try themselves, but maybe I can help with a probable way out from the problem.

Sincerely Lasse
 
Just wanted to say this is a good thread. I am reading with interests and currently looking to replace my t5's with LEDs, and someday, replace my 250MH with these multichips. This is definitely where the hobby is headed. A friend of mine has a few of the Radium LED fixtures and he gets the color, but not the growth. I think this is the bridge for that, but however I still feel these MH replacements are in its infancy. I can certainly be wrong but reading through this and other threads leaves me with that impression. Good luck to all that embark.
 
Holy cow, this is a great thread, but I'm not even half way through, so I'm gonna have to mark it and finish later. Thx for all the great info
 
I'm curious as to whether anyone here (Original poster?) can recommend a piece of hardware that would allow me to control [preferably through a GUI based software application] both the time (on/off) AND sequential dimming of several channels?

the goal would be to program an array of multichips to simulate spectral and intensity changes throughout the day (i.e. 0600-0900hrs: royal blue channel sequencing from X current [wattage? amperage? i suck at electrical, sorry] to Y current [simulate dawn into morning]; 0800-1200hrs: 20000k [maybe also a 10000K] channel sequencing from X current to Y current [simulate morning-daytime transition]; 1200-1600hrs: 20000K channel sequencing from Y current back down to X current [simulate afternoon transition to evening]; 1600-2000hrs: Royal blue channel sequencing from Y current back down to X current [simulate evening into night]) as well as an occasional cloudcover simulation. I know this sounds complicated, but between 7 billion people, I'm sure someone has figured out how to do it easily. I KNOW the LED's have the capacity to be controlled as such.

I've more or less wrapped my head around the minimalist construction aspects (largely thanks to this thread) but if i'm going to use such capable lighting elements, i'd love to use them to their full potential of natural condition simulation.

Thanks again for a great thread.
 
Colin,

I was planning to use an Apex to do what you are talking about. Apex has tables that may help you do what you are envisioning. Check out the Neptune forum in the Sponsor section, there is a programming guide that shows you some possibilities.
 
Last edited:
$$$$$ Out of my budget. also, i'm not a high-tech reefer. i like old school, balanced ecosystem based paradigm with as little man-made equipment in the way as possible. I don't plan on purchasing a controller (although i have looked into it, believe me) even for the basic functions they offer, let alone in order to purchase additional modules.
I looked into the apex and in order to do what i want, i would be required to also purchase the aquabus lunar simulator and variable speed/dimming module on top of the base unit, which i'm not entirely certain are even compatible with lighting systems other than the proprietary ones (aqua illumination and maybe another) mentioned on their site. Those three pieces of equipment would cost me $680 + tax msrp. I haven't spent half of that on my entire system to date. $680!!!

that's part of the reason i want to do DIY. Building something myself, knowing how it works and fully controlling the functionality is more valuable to me than just 'making it work'.

I feel like a dolt - i read a post of someone's on some forum mentioning a *cheap* programmable module (arduino-based, if i'm not mistaken) which allows for functionality like i desire. Of course, i can't remember it...
 
I saw a few posts back that you're starting to play with the multichips for lighting refugiums. How effective do you think one of these: http://www.ebay.com/itm/16W-Greenhouse-Fast-Grow-Led-Panel-/220844846100?pt=LH_DefaultDomain_0&hash=item336b60ec14 would be at lighting a 12"x18"x14" refugium, compared to my current 23w CFL bulb? Would I be better off using a standard 6500k mutlichip like this: http://www.ebay.com/itm/Epistar-20W-6500K-White-High-Power-LED-Panel-/220787520321?pt=LH_DefaultDomain_0&hash=item3367f63341?
 
Back
Top