Upgrading to leds

The friendliest place on the web for anyone with an interest in aquariums or fish keeping!
If you have answers, please help by responding to the unanswered posts.

CoryB

Aquarium Advice Activist
Joined
Apr 7, 2011
Messages
135
Hi all I'm looking to upgrade my 29 biocube to LEDs. I'm am going to build it myself. Below are the LEDs I have selected.

7 XT-E royal blue
3 XP-E blue
7 XP-G cool white
3 XP-G neutral white

All are Cree LEDs the whites are 5w and the colors are 3w. Colors and whites will be on separate drivers for blending.

I plan on using aluminum u Channel as a heat sink and mount the drivers outside of the hood to keep the heat out. I am not going to use any reflectors or secondary lenses.

Any light experts see anything wrong with my plans. Is this enough light? Too much light?

I want to be able to keep anything and everything (corals/nems) and not have to worry about my lighting.

Thanks for reading/replying.
 
As long as you run it through a dimmer that sounds awesome to me! You should post your progress, so many people are wanting to do this but don't know how ;)
 
Yes, the plan is to run both drivers independently and have them on dimmers. It should be a fun mod.

I'll post pics as a progress.
 
But back to my original question. With that setup would I be able to keep any coral or nem I wanted?

If not please explain so I can spec out the LEDs differently.
 
I am not a lighting expert but I do believe your lights will be more than enough to grow anything. Perhaps even too much which is why I asked about the dimmer. From what I have learned the general rule of thumb is that each LED diode is 3w each so sounds like your good. I am sure there will be someone with more knowledge out there to confirm :)
 
Cool just wanted to make sure. I have done a lot of research but have not found a definitive answer. Always good to ask people with more knowledge than me.

Now to go pop the hood and get dimensions.
 
I am not sure if that U channel with be sufficient. You might wanna check out the cost of an actual heat sink instead.
 
I thought about that. Had a thermal engineer run some numbers and it would seem sufficient. Will add fans to be sure. Once it's all together I'm gonna monitor with a thermal imager for 24hrs and see, I know it's overkill but whatever.

Sometimes it pays to work in a manufacturing environment.
 
I thought about that. Had a thermal engineer run some numbers and it would seem sufficient. Will add fans to be sure. Once it's all together I'm gonna monitor with a thermal imager for 24hrs and see, I know it's overkill but whatever.

Sometimes it pays to work in a manufacturing environment.

Haha! I love it. I just bought LEDs from rapid led and will be putting it together this weekend. I am a process engineer and work with class 3 electronics all day. I am sure I am going to scrutinize my workmanship on the solder and wire routing. I am also working on my masters in mechanic engineering and decided it would be fun to program my own controller, version 1 is almost done. I'm glad to know I'm not the only nerd out there. I am thinking about running a thermo couple to read my temps so I only run the fan as much as is needed but that won't be in the original design. I haven't fun any thermal analysis because I really didn't like that class.
 
I was thinking about a thermo couple myself but decided that's just outrageous. Don't want another controller. I was trying I do this cheaper than a pre made retro kit. I'll wire up the stock fans to turn on when the lights do.

I started programming one of our cnc machines to make a proper heat sink but then I realized I might be insane and stopped.

I will monitor the temp for a week and decide then.

Now I just have to wait for shipping the suspense is killing me.
 
Yeah, waiting is the worst. I realized that even though a thermocouple would be kinda fun to implement it is definitely over kill and as far as the performance of the led goes the best option would be to run the fans on high when the LEDs are on. You should let us know how the thermal analysis goes once you get them running.
 
Well the package arrived and been assembling in spare time.



image-1601893154.jpg

Quick mock up for testing. Was going to hid the wires behind the heatsink but decided against it seeing that they will be hidden when the hood is closed.




image-2744910162.jpg

Power on at 10%

These things are bright. I don't see a need to ever go over 40%. Keep in mind I am also under driving them at 750ma. Could bring that up but don't see a need to.
 
So, no optics then? I would not have any if I was running your set up. I would be afraid you would see the different colors in beams.
 
How come you left 2 LEDs out? Are they for night lighting?

Of course, now you will have to provide the details on final cost so some of us others who can dabble in electronics can consider the switch. Please.
 
So, no optics then? I would not have any if I was running your set up. I would be afraid you would see the different colors in beams.

I decided against optics seeing the lights sit directly on the water Column. Didn't think they would be necessary.

Don't know how to quote twice on a phone but...

The two LEDs that are left out are moonlights. I need to wire them up and check their intensity, I fear that's way to much light for the cube after seeing what the other LEDs are capable of.

As far as cost I used a lot of material I had laying around but the electronics were $180. Again judging by intensity if I could do it again I would purchase fewer LEDs which would result in a smaller drivers and bring the overall cost down considerably.

Assembly is a cake walk assuming you have some soldering exp.

On a side not the dimming system is not so great you can't dim lower than what I call 10% it just doesn't have have that much resolution.
 
Mounted to the tank. Currently at the lowest setting and still brighter than the stock PC.

image-2538431760.jpg
 
Back
Top Bottom