DIY LEDs. How inexpensive?

So if I'm understanding everything in this thread, the basic components needed are:
-dimmable drivers
-Heat Sinks
- LED's @ 2:1 ratio of RB to white
 
pretty much that a voltmeter, something to dim the ballast I highly suggest the typhon. soldering iron, solder, flux, wire, and thermal paste, and fans
 
- LED's @ 2:1 ratio of RB to white

This part is up for debate depending on intended use (supplemental vs. main tank lighting), personal preference, model and bin of LED involved, drive current, height, optics, and so on.

Supplemental builds are way more forgiving since the only real criteria is "it looks good." Building a strip of a few RB LEDs on channel aluminum is a great way to test the DIY waters with minimal investment and without risking an existing coral population on new lighting.

And since Cully also asked "how hard is it" I'll extend this offer to him or anyone else in the Syracuse area. If you're worried about the difficulty of the actual assembly process, bring me a 6 pack of something worth drinking or a frag or two and I'll help you put it together.
 
And since Cully also asked "how hard is it" I'll extend this offer to him or anyone else in the Syracuse area. If you're worried about the difficulty of the actual assembly process, bring me a 6 pack of something worth drinking or a frag or two and I'll help you put it together.

does this mean i need to ship you a 6 pack?
 
So if I'm understanding everything in this thread, the basic components needed are:
-dimmable drivers
-Heat Sinks
- LED's @ 2:1 ratio of RB to white

Those are the most expensive parts, you'll need quite a bit more for the entire project but its a lot of little stuff. Some things are obvious though: wire, soldering iron, pots, thermal grease, washers/screws/bolts/etc for hanging, a hanging kit, 10V adapter (for dimmable), 12V adapter for a fan + the fan and converter, plexie glass (not necessary but its a good idea). Some stuff might change depending on your exact build though.

Also depending on exactly how many your doing 2:1 might not be ideal. The dimmable drivers from RapidLED will do 8-14 LEDs each, so if you were doing a 24 LED build for example, you could do 14 Blues and 10 Whites for the same price, but to do 16 Blues and 8 Whites it would require 3 drivers, bumping up the price a little more. Just something to keep in mind.
 
Rapidled is OK but there are one or two vendors cheaper, and honestly I am not a fan of the solderless approach.

A typical supplemental build might be done with 12 RB XPE D316 and a driver to run them at 700-1000mA. The LEDs are maybe $3.50 each, so $40 for 12. Driver might be $30. Say $20 for heatsink, hardware, wire, other stuff. So, $90. On a tank as big as yours you might want two of these units for nice strong suplementation. So you're looking at $180 total, plus or minus.

Can you tell us the names of some other good vendors? or is that not allowed on RC? Thanks---Rick
 
There are a bunch of good vendors, some hobby oriented and others not. Prices tend to fluctuate from time to time so it's probably not fair to put it down in stone that a given vendor is the cheapest (though there are definitely trends and some vendors that are consistently cheaper than others).

If you're curious, shoot me a PM and I'll let you know who I was referring to.
 
Dont want to hijack this thread, but thought I'd ask this question to the Guru's here.. I read somewhere about the color combinations of LED.
Assuming emulating a 8x T5HO, with something like 4x ATI AquaBlue Plus (Actinic blue), 3x ATI AquaBlue Special and 1x KZ Fuji Purple, using the spectrum I've found online, I should be using the following color bulbs

400-420nm for purple range - both ATI bulbs have a tint of 420nm.
445nm Royal blue - that's the peak wave length shown in the blue spectrum for both ATI blue bulbs.
460nm Royal Blue - that should fill the missing gap between 450nm and 480nm blue.
480nm Blue - that's the near-cyan color spectrum.
No 503nm cyan, because according to the ATI bulb spectrums, there's very little cyan color in their bulbs, and the 480nm Cree XP-E blue already have a strong tint of cyan.
10000-12000K white - that will cover most of the green to red spectrum
16000K white - these will be turned on only 5 hours a day to simulate the brightest period between 11am-4pm.
620nm red to add a tint of purple to the overall appearance, to emulate the KZ Fuji purple color.

This is the sprectrum his lights are designed, I believe.

3w 12000K C. White Semi LED SGD12 - 12 LED's
15W 16000K Semi Led - 4 LED's
3w 445nm Royal Blue Semi LED SGD12 - 10 LED's
3w 620nm Red Semi LED SGD12 - 2 LED's
3w 405nm Violet SGD40 - 4 LED's
3W 460nm Royal Blue XP-G Cree - 4 LED's
3W 480nm Blue Cree XP-Q - 4 LED's

His FTS
DSC_2931.jpg


Can this sprectrum be technically achievable via LED's? Also does anyone know where one can find these LED's as I cant find them anywhere online ? Else what other commonly found LED color combinations can be used to achieve this sprectrum?


Thank you,
 
I dunno if I agree 100% with the explanations given in that post but it's hard to say when it's taken out of context - and I'm not a total expert on T5 spectra anyways.

The thing about monochromatic LEDs is, they have an EXTREMELY narrow spectral width. You get ONE wavelength of light, that's pretty much it. No other form of reef lighting is like this - even "actinic" T5 or VHO lamps have a wider spectrum typically. IMHO it's dangerous to try to match the look of a specific existing light source by mixing lots of monochromatic LEDs. Even if you get something that looks the same to our eyes, it almost certainly will not have the same spectral distribution as another form of lighting, which means the livestock might not react the same.

A specific example of this. Most people who have done the typical cool white/royal blue combination have decided that their builds aren't "warm" enough. Some people solve this by adding a small number of red or "deep red" LEDs (typically 630nm or 660nm). Other people solve this by swapping out some cool whites for warm whites. The visual appearance of these two methods can be very similar, but the spectral results are TOTALLY different. With the warm white approach, you get a nice big broad bump in intensity across the entire "warm" spectrum. With the red approach, you get a really narrow spike at the specific wavelength. It's hard to say in general which one is better, but IME the warm white approach is "safer" - consider for one thing that when people build LED lighting for algae turf scrubbers (which are designed to grow hair algae) they use monochromatic red LEDs, because hair algae responds EXTREMELY well to those specific wavelengths!

I guess my point would be, if you are deviating from the "norm" of LED builds, you're doing it on an experimental nature, and at this point in time you're kind of on your own.
 
let me point out a specific comment based on the quotes you made - again though I don't mean to be critical since it's hard to know the entire context of where you got that info.

620nm red to add a tint of purple to the overall appearance, to emulate the KZ Fuji purple color.

In our human brains, when you mix red and blue light, you get purple. But in the spectral sense, there's a BIG difference between having a spike of actual purple light, vs having a spike of blue light mixed with a spike of red light. Again, I can't conclusively say if one or the other is "better" for a reef tank, only that they are "different" and it'll likely be the case that photosynthetic livestock will respond differently.
 
Wouldn't corals (and all things with chlorophyll) benefit from the 660nm wavelength? Or do alga tend to have a higher percentage of chlorophyll A than corals?

Sorry Cully, we're steering your thread in a whole 'nother direction.
 
The panoramas are nice and you could grab 2 of the 12" for about $200. Corals need to acclimate to them. Without dimability, they pose a risk to my corals. The reef brites pose a similar problem. It wound be nice to have them dialed back, and dialed way back for moon lights.

Just to throw it out there, ecoxotic now makes a RBG panorama pro LED module that is dimmable and also has the ability to put out many different color variations. It's a bit more expensive than their other 12" model because you have to buy the controller for it, but it gives you a ton more flexibility with color spectrum and intensity. I was going to go with this for my nano tank, but I couldn't financially justify this model when I already had one of the older models with just blue LEDs. I decided to purchase the panorama pro model that mixes 12k white LEDs and 445nm LEDs. When I bought it, I got the reflector for free, and it's definitely worth it. The two light fixtures together give me a nice color spectrum over my small tank. I just have to figure out how to build a mounting unit for the lights now.

Another option might be the ecoxotic stunner LED strips. They've come a long way. Jesse just got one for his 40 gallon long tank, and it is bright and is perfect for a moonlight. The LEDs are quite strong. He got the 48" 24 watt Stunner LED strip with 445nm Blue LEDs. He got the reflectors as well, which definitely make a difference. Dr. Fosters and Smith had a sale in January on their LED lighting, so it only cost $140 for the whole set up for his tank. Even without a sale, IMHO, they're definitely worth it. They work well on deeper tanks too (we tried it out on my 55 gallon first before we put it on his tank, and I bet it would still be all you'd need for a moonlight over a 75 gallon). If you look at ecoxotic's website, they offer a LED inline dimmer as well for the stunner strips. I don't have any experience with the dimmer, so I don't know how well it would work, etc.

Hope some of this helps. Whatever you decide to do, keep posting up pics of your amazing tank! :thumbsup:
 
Wouldn't corals (and all things with chlorophyll) benefit from the 660nm wavelength? Or do alga tend to have a higher percentage of chlorophyll A than corals?

I don't think it's safe to make sweeping statements like that, given the role of various other pigments in corals and the fact that corals are often not growth-limited by light. If you spend time trolling LED threads for results reported by people who add red LEDs, you frequently get answers on both sides - some people claim better coral growth, other people claim their tank was overtaken by pest algae. Looking at results from people who add warmer wide-spectrum light, I can't recall anyone claiming it caused a bloom in pest algae, a few people have talked about better coral growth, and almost everyone claims it makes the tank more visually appealing.

I hope it doesn't sound like my last few comments in this thread should be taken as gospel, I'm just as clueless as the next guy for the most part - I'm not trying to provide answers, but rather point out potential pitfalls that might not be obvious.
 
You definitely know more about LEDs than the rest of us here, and your opinion is highly valued. Every LED thread that I've read, you have commented in, so your thoughts and ideas should be well taken. I'm sure we all appreciate it.
 
Thanks for the vote of confidence. I'm looking forward to the point in time where the components become so cheap that we can just buy a few hundred of each and actually do some real solid back to back experimentation.
 
Back
Top