5mm LED primary lighting for my nano reef

[fishlibrary]

1. First of all, planning how to drive the LED array, how many voltage & forward current. Prepare the tools & materials for this project.
Tank size : 24"L x 11"W x 14"H DSB single tank.
Tools : 1 x Cutter, 1 x wire wrapper.
Materials : 400 x 50,000mcd white LED, 400 x 13,000mcd blue LED & 100 x UV LED, 4 x pre-drilled PCB board, copper wire, power supply(not include inside the photo)


2. Planning the white/blue combination and build a prototype first.
Yellow one and silver one is white and blue LED respectively. Arranging different combination of white & blue LED before wiring it up.


3. After confirm the white/blue combination, I can start to finishing the whole PCB.
No solder iron is require. Just wire up the LED by copper wire with wire wrapper as your designed white/blue combination.


4. Finished of single PCB, each PCB consist of 100x50,000mcd white LED, 100x13,000mcd blue LED & 25xUV LED. That's all!



5. At this state, four PCB finished and placed over a acrylic sheet on the top of my RT.



6. Check the DIY light effect on photo below and wait for the test result.


7. Please get any comment on this project for further my improvement.
Thanks!

Previous discussion here :
http://www.reefcentral.com/forums/sh...hreadid=975445

[BeanAnimal]

You did a nice job on the layout. What are your current and voltage specs? It appears that you have 4x4 series parallel arrays for each section.

I am eager to see the PAR readings when you get them. I suspect it will be rather low but testing may show differently.

My biggest concern is the wire wrap connections. They will fail in the humid salty environment. Because they have aleady been exposed, it will be VERY HARD to get them clean enough to take solder.

It may be better to get a few PCBs made and solder everything. You COULD use a good flux and solder right after you wire wrap. This would save you the trouble of having PCBs made.

Nice work.

[fishlibrary]

Thanks BeanAnimal,

Each PCB contains 45 parallel x 5 series arrays that supply from a 15VDC notebook PC ballast, each LED claim 3VDC.
Total consumption of the 4 PCB is listed below and measured by FLUKE 79 Series II Multimeter:
DC: 15V 1.1A
AC: 223V 0.11A (~24W)

Un-soldering is a easier way to replace defective LED.

I will made a housing to store the PCB inside, and also to prevent it from corrode.
You can see from the photo, I keep both the leg of LED for cooling like a heat sink.

Anybody can tell me how to find a easier and cheaper way to test PAR as well as the light spectrum?

[BeanAnimal]

The apogee quantum meter is likely the cheapest PAR meter you will find. Measruing spectrum will be more expensive. You may wish to contact Sanjay or JBNY for more information.

[fishlibrary]

Originally posted by BeanAnimal
The apogee quantum meter is likely the cheapest PAR meter you will find. Measruing spectrum will be more expensive. You may wish to contact Sanjay or JBNY for more information.

Thanks, but I think using US$300 to have a apogee quantum meter just for a single test is not reasonable for me.

[toastman]

Very nice work, I love the shimmer of the LEDs more like the shimmer of MH, then T 5s I would like to see how it fairs long term

[six.line]

That is EXCELLENT wire wrap work!

[Tigger240]

you could look at it as an investment - you get good numbers, im sure you could sell quite a few of them... or someone may let you borrow theres... or you could send the device off to a trusted someone to let them test it.. if i had a meter id pay shipping too again, contact sanjay. nice work.

[Ellery]

Awesome project - can't wait to hear about your results

[MJT82]

So 900 LED's? Wow.. Was that expensive?

[hahnmeister]

I have those LED's, but not as many. Tomorrow I get my Apogee PAR meter, so I can take a reading of what about 60 would make... you can figure out the rest if you like.

[hahnmeister]

Okay, I have some figures for you. Ambient light in the room where I took the readings showed up as 1 or 0 on the PAR meter... so Im not worried too much about control.

On a 5mm LED array of white or blue LED's like you are using (the whites and blues had almost identical outputs), the PAR at the LEDs was about 800. Not bad... this us just under what T5s can do (1000ish) at the bottom edge of the reflector (about 1/2" from the bulb). At 7-8" from the LEDs the PAR was around 200. At 12", this dropped off to 100. Now, I only had 60 LEDs of each color hooked up for this, so an array as large as yours would be a very nice lighting choice for a nano. Most higher light corals I know of only need a PAR of 400 to do well. The other thing to keep in mind is how directed the light is. If I moved the sensor more than 10 degrees from the edge of the LED 'spotlight'... the PAR levels dropped to less than 1/4 almost instantly. LEDs may not make as much output as a halide or T5s, but they sure do get the little bit of light exactly where they want. Im impressed considering they are only 5mm LEDs.

Now I dont feel so bad about only using 180 5mm LEDs to light my 4 gallon desktop nano.

[golf4lyfe]

If you wouldnt mind sharing roughly how much did this cost? also where did you purchase your leds from

[hahnmeister]

I bought my LEDs pre-wired inside spotlight (medium base) units... PAR 20 floods. Id have to dig around to find the link, but they were about $25 each. The cool thing is that every 60 LEDs only use 4 watts.... so my 4g nano is only using 12 watts of lighting, 8 watts of pump (MJ1200 provides flow and runs a skimmer), and a 50watt heater.... and its fine for what I keep in there... those rhino/frilly green mushrooms, purplr mushrooms, GSP, and a couple caps up at the top. The tank houses my delicate smaller critters... porcelain crabs, sexy shrimp, boxer crabs, a small frilly-arrow crab, a clown goby, and a neon goby. My only hope was that I would be able to sneak a green BTA anemone in there, and I think Ill be able to.

[aomont]

Can you post some pics from your light setting hahnmeister ?
Anderson.

[dga]

Originally posted by hahnmeister
I bought my LEDs pre-wired inside spotlight (medium base) units... PAR 20 floods. Id have to dig around to find the link, but they were about $25 each. The cool thing is that every 60 LEDs only use 4 watts.... so my 4g nano is only using 12 watts of lighting, 8 watts of pump (MJ1200 provides flow and runs a skimmer), and a 50watt heater.... and its fine for what I keep in there... those rhino/frilly green mushrooms, purplr mushrooms, GSP, and a couple caps up at the top. The tank houses my delicate smaller critters... porcelain crabs, sexy shrimp, boxer crabs, a small frilly-arrow crab, a clown goby, and a neon goby. My only hope was that I would be able to sneak a green BTA anemone in there, and I think Ill be able to.

so you disassembled the par 20's for the led's @ 25 dollars each? how many led's were in each flood? sounds like you may have wanted to shop ariund.

[hahnmeister]

no, I have left them stock... why would I rip em apart? I bought them because no wiring required.

[dga]

got your post mixed up with another one. i also like the prewired ones. i use mr16 and festoon led bulbs in my custom track lighting under my canopy.

[fishlibrary]

hahnmeister,
Good take!
Very thanks to your information & measurement.
Getting from my experiment, the density of LEDs also control the light intensity. Higher the LED density would have higher light intensity, but I don't know does PAR directly proportional to LED density!!

Also a good news to me that the distance from my LED light to sand bed only have 11"(1" air and 10" water) and most SPS allocated at the upper half level.
Good PAR can keep SPS alive, but I think nobody want to keep a brown SPS! Right? In what combination of blue, white LED can color-up all the SPS will be the coming challenge to me.

[hahnmeister]

yeah, you have a good mix of colors... similar to the solaris really. They just use 420nm instead of UV-A LEDs (I assume you are using UV-A LEDs, not B or C, right?). I would add 420s and maybe some 3000K (indoor bulb color) to fill in the red/yellow spectrums... something that the LED's you are using lack all together.

[fishlibrary]

Yes! You are right! I choose 400-410nm UV LED to improve invisible to blue portion.
Why use UV LED? I find most of the shop promote their SPS with VHO actinic 03 bulb and the SPS color is excellent colorful, so ................... UV LED was selected inside my combination.
I have an idea to buy an affordable spectrometer stated in the below link. Just for your information!
http://www.advancedaquarist.com/2006/7/aafeature

[hahnmeister]

I already have a computer based ocean optics spectroradiometer. Um, just so you know, actinics arent designed to peak at under 400nm... they are designed to peak at about 420-430nm, and UV-A is 315-400. UV-A bulbs are not the same as actinics.

[golf4lyfe]

fishlibrary, where did you purhcase your led's from?

[badpacket]

Another option I'm hoping someone tries before bonus time comes and I play guniea pig...

http://www.luxeonstar.com/sub_catego...k_str=189::191

[Tachyon]

if you get the luxeon stars they are already attach to heatsinks so you can do an array like u did on this one. was just thinking about doing it. but i took like 1 electrical engineering class... so yea...programmer here.