DIY LEDs - The write-up

Status
Not open for further replies.
<a href=showthread.php?s=&postid=14593516#post14593516 target=_blank>Originally posted</a> by CJerome21
I am interested in hearing more about the distance the LEDs are mounted from the water surface in relation to using or not using optics.
Soundwave what is the distance from your LEDs to the water surface and the same for Lynxvs?

I am planing builds for my two nanos and right now my PC light is about 2.25 inches off the surface.


As you probably know optics will concentrate the output of the light source into a narrow beam so the light intensity is greater farther away but with a narrower beam. Of course this reduces the "spread" of the light to what soundwave calls the "spotlight effect"

There is something else going on when light strikes the surface of the water called "Snells Law" what this means is light will be refracted or bends when it hits the surface of the water so it works sort of like a lens but not as good.

This is only true when the light source is very close to the water. The ideal distance could be calculated I assume.

If you use optics you can raise the light and depending on the width of the beam have the same or greater light intensity with the same light spread. Remember even though you have a narrow beam the light will spread the higher you raise it.

I have tried hanging the light at 6" , 10" with optics and 2" without optics.
 
<a href=showthread.php?s=&postid=14593762#post14593762 target=_blank>Originally posted</a> by ahoyhoy239
lynxvs, look up "impingement jet" or "impinging jet", that is the basic configuration that soundwaves fans have with the heatsinks and it is a very effective method of heat transfer.

I did a brief search and don't quite see the connection but i'm not a mechanical engineer.... can you explain?
 
My fixture is 4 inches above the tank with another inch or so to the water. To lynxvs, the fans push air downward which allows air to flow through and out of the fins.

I've given it some thought an would like an opinion on this. I could add optics to the design but I would have to raise the fixture. So the thought is what if I cut a hole in the canopy for the LEDs and attached the entire fixture to the top of the canopy? I would think this would allow for more spread while using optics and allow the heatsinks to operate more efficiently as they would be pulling outside air instead of recirculating the air inside the canopy.

Any thoughts?
 
So would the consensus be that if the LEDs will be mounted close to the water surface, in the rage of 2 inches that optics should or do not need to be used?
Also with the Nanos they are frequently much shallower than larger tanks, for example one of mine is 8 inches another 15 inches so the depth the light needs to penetrate to reach the bottom is much less maybe not requiring the optic focused beam?
 
<a href=showthread.php?s=&postid=14593975#post14593975 target=_blank>Originally posted</a> by CJerome21
So would the consensus be that if the LEDs will be mounted close to the water surface, in the rage of 2 inches that optics should or do not need to be used?
Also with the Nanos they are frequently much shallower than larger tanks, for example one of mine is 8 inches another 15 inches so the depth the light needs to penetrate to reach the bottom is much less maybe not requiring the optic focused beam?

I would think with a Nano you should be fine without optics..
 
<a href=showthread.php?s=&postid=14595614#post14595614 target=_blank>Originally posted</a> by EBOLII
lynxvs or Soundwave so what does your mind recommend number of LEDS with optics per 24"

The magic number seems to be at least 24 LEDs per foot.. Just looking at the commercial products Solaris was using 30 per foot, Aquaillumination is using 24 per foot. I'm sure they had a bigger development budget then I have so I would shoot for that number.
 
OK the decision has been made. I will be modifying the canopy and fixture this weekend. I will order the optics in a few minutes. The idea here is to cut a hole in the canopy roof about 9x36. I will build an enclosure for the LED fixture tomorrow and attach that to the top of the canopy. This will raise the fixture about 12" off the water surface which should still give me a decent spread with the optics along with making it brighter. Cooler room air will be drawn into the heatsinks instead of warm tank air. I'll post pics.
 
Here's a wiring question:

If an LED has a Vf of 3.6v, and you are using a BuckPuck with 24 VDC input, then this is just barely enough to power 6 of these LEDs in series; the Vf of 6 LEDs would be 21.6v (need 2v margin).

Could you wire two groups of 6 (in series) in parallel?? Basically after each group of 6 the wiring would connect back at the buckpuck essentially making two parallel groups of 6.

Anybody know if this will work? And if so, why don't we use more parallel wiring to power more LEDs with the Buckpucks?

Thanks
 
<a href=showthread.php?s=&postid=14591268#post14591268 target=_blank>Originally posted</a> by Soundwave
OK, let's see what I can answer...

Imzadi: For a 55g, Two rows of whites sounds like it would work just fine. However, if they are spaced too far apart, you will get two rows of white in your tank. If it were me, I would put the whites in the middle and the actinics on the sides.

That sounds logical! Actinics would be omni-direction...ish, where as the LEDs, as mentioned, would be point of source...

Thanx for the thought!
 
one more question thats probably been covered
Is it ok to mix LEDs of different Vf in the same series if they require the same current?

Ie a few cool white 3.85 Vf, 700mA LEDs mixed with some 3.15 Vf, 700mA LEDs.
 
when you ran 10w led's did you run them in series? also what driver did you use to run the 4 of them, and how many did you need?
LEDAquarium2.jpg
 
Soundwave, I am very impressed with your layout and think you have given many the courage to attempt to make one themselves (me included).
I have been following along from the start and have been tinkering around with design layouts for an led light to cover a 72" x 30" x 24"h tank.

To everyone:
My big concern is how many 3W led per fixture(6 total, 30"l x 10.5"w).

I am not sure if my thinking is correct here, but with 25 degree led optics will I get a spread like my attached drawings?

The 2 drawings are not of the same fixture, I am just trying to figure out true light spread.


About 8" spread (each LED) from fixture(6" above tank) to about 18" depth(12" tank depth)...
LedLightSpread.jpg




Light about 6" above tank, LED light spread at bottom of tank about 13"...
ledLightSpread2.jpg



I have pdf files of these which are much clearer but photobucket will not upload them.

Thanks
- Paul
 
I just want to apologize to all of you following along. I changed my mind. Yeah I know...

I decided that I haven't given this fixture a chance to show what it is capable of. Like all people, I just want more. I decided, for now, to cancel the order for the optics and see how this plays out. I'm happy with the current output and I think I'll give this a chance before upgrading or changing anything.

Sorry for getting anyone more excited.
 
<a href=showthread.php?s=&postid=14596969#post14596969 target=_blank>Originally posted</a> by ReefEnabler
Here's a wiring question:

If an LED has a Vf of 3.6v, and you are using a BuckPuck with 24 VDC input, then this is just barely enough to power 6 of these LEDs in series; the Vf of 6 LEDs would be 21.6v (need 2v margin).

Could you wire two groups of 6 (in series) in parallel?? Basically after each group of 6 the wiring would connect back at the buckpuck essentially making two parallel groups of 6.

Anybody know if this will work? And if so, why don't we use more parallel wiring to power more LEDs with the Buckpucks?

Thanks

This always sounds logical..... unfortunately it's kind of hard to explain but if you have two branches of 6 LEDs in parallel and each branch requires say 1 Amp you would have to supply two Amps to the circuit. The Buck Puck is a constant current source so it will divide it's 1 Amp into the two branches so it will only supply 500 mA to each branch. I hope this makes sense.

There is another problem with parallel circuit in that if you could supply two Amps to the circuit and if an LED fails and the circuit is opened on one branch the current has to go somewhere so it flows in the other branch. That means 2 Amps will be flowing though that other branch resulting in a big cloud of smoke.
 
Status
Not open for further replies.
Back
Top