DIY LEDs - The write-up

Status
Not open for further replies.
Hey everyone,

I'm trying to encourage people who have played with LEDs to post in this thread:

http://reefcentral.com/forums/showthread.php?t=1718642

I PM'd some of the people who have build threads, but I know there are a lot more people out there, so please stop by that thread and share your results if you've run DIY LEDs over a fish tank. :)

Cool, I'm writing mines up now and this will be very helpful to everyone thinking or currently building an LED rig!
:idea:
 
I'm getting ideas for my next array....keep the websites coming folks, especially multiple LED's on printed circuit boards with one set of (+)/(-) soldered leads, perfect!
 
Didn't think to check cutter, since they're in oz. Looks like $6.67 USD for the R5 bin on a square or round PCB. Not bad. I'm assuming they ship to the US, given that they show prices in USD (in addition to AUD, EUR, and GBP).
 
215 lumens @ 4W from the datasheet. That's around 53 lumens per watt, a little more than half of what a good Rebel or XR-E will do. In other words, they're not very efficient. And I can't imagine how you'd control or dim them.

In fact, I can't imagine how they even work, since an LED is inherently a DC device. I'm guessing there's more than just an LED on those chips.
 
That CutterXPG20STR at ~$7 ea mounted on a star is awesome!.

The data sheet shows that with the R5 Bin, you will get ~250 lumens at 700mA.
The XR-E gets about 170 lumens at 700mA.

However ( this is curious ) they drop slightly LESS voltage than the XR-E at 700mA.

I figured that we could only put half as many LEDs on a driver compared to the XR-Es but that is not true.

You can put just as many on a driver for about the same price as the XR-E and get almost 2X the light output.

I wish I hadnt just bought 80 XR-Es @ 6.66 ea. :-(

Stu
 
No you have the R3 bin, but there's very little difference other than ~0.60 ea, for the higher output.

Try this link

http://www.cutter.com.au/products.php?cat=Cree+XPG


then click on Select XPG bin, click R5

Then click XPG Light Engines & select the "CutterXPG20STR " Option & they charge you about $1 to mount it to a star.

Stu

So, XPG vs XRE, does this have an impact on how folks are creating the array of LEDs? Or would you have just as many, just more intensity?

I guess I was wondering if I could use few LEDs if I had the XPGs?

thanks!
 
So, XPG vs XRE, does this have an impact on how folks are creating the array of LEDs? Or would you have just as many, just more intensity?

I guess I was wondering if I could use few LEDs if I had the XPGs?

thanks!

I think it would depend on your goals. You could do the same LED count and get more intensity, or fewer and get the same intensity. Or, you could do the same LED count and run them at a lower current, which would give you similar intensity, similar light spread, but less power draw, i.e. more efficient running. This would pan out both in terms of your monthly bill, and in terms of lifetime of the LEDs, since they'd last longer at a lower current.

If you do fewer XP-G's for the same intensity, you'd be spacing them out more, and hence more likely to get columns of light, if you used optics. So probably best to stick with much wider optics if you went that route. You'd face a similar challenge if you went with multi-emitter stars or chips, like the Cree MC-E or the 3-up Rebel. IMHO there's almost a point of diminishing return where we will have to really re-think our designs if LED development continues on the same path. It'll be more like putting a dozen of 70w MH lamps over your tank, and less like 100 little tiny LEDs.
 
wow... I read about 200 post b4 I realized this went on for 1000s of posts... I've been trying to supplement my bio cube's lighting for months... just simply boost it a little. I have a 75g sps tank but I just want to boost my bio... I've been trying to find some simple small t5 actinic bulbs but no luck. I figured I'd just drop in maybe 2 and then take the actinic pc thats stock and put in a second 1000k bulb so I'd have twice the light... but no such luck... but lately I've been hearing about leds... so I just wanna know if anyone at this point in the research thinks there is a way to add light to the existing fixture or not, and would it be better to add a Pc 1000k bulb and do actinic leds or leave the actinic pc and add white leds. the most I was looking to add were 3 or 4 leds... can anyone pm me... I've been reading and reading and am still quite confused.
 
wow... I read about 200 post b4 I realized this went on for 1000s of posts... I've been trying to supplement my bio cube's lighting for months... just simply boost it a little. I have a 75g sps tank but I just want to boost my bio... I've been trying to find some simple small t5 actinic bulbs but no luck. I figured I'd just drop in maybe 2 and then take the actinic pc thats stock and put in a second 1000k bulb so I'd have twice the light... but no such luck... but lately I've been hearing about leds... so I just wanna know if anyone at this point in the research thinks there is a way to add light to the existing fixture or not, and would it be better to add a Pc 1000k bulb and do actinic leds or leave the actinic pc and add white leds. the most I was looking to add were 3 or 4 leds... can anyone pm me... I've been reading and reading and am still quite confused.

I'm a little confused, are you looking to add light to your biocube or 75 gallon SPS tank?
 
For the biocube, just get a dimmable 700mA buckpuck, a 24v power supply with at least 1A capacity (1.5 or more would be better, IMHO), a small heatsink - even a CPU heatsink - and a few LEDs.

The biggest challenge will be finding a good quality heatsink that'll fit in that hood, I'll bet. Look at some of the nano-specific forums out there, lots of people have done similar things.

And, IMHO, your best course of action would be to ditch the PCs totally and go with LEDs. If you just wanted softies and easy corals, 6 LEDs would probably be fine, and you can run 6 off a single buckpuck, so it won't be *that* much more expensive to implement. It'll pay for itself in a year - no lamps to replace, and cheaper on your power bill.
 
Status
Not open for further replies.
Back
Top