DIY LEDs - The write-up

[der_wille_zur_macht]

The driver in the first kit is preset to 700mA and not dimmable or adjustable.

The driver in the second kit is adjustable via internal setscrew over a wide current range, AND has an external dimming circuit.

In both cases you're running under the capacity of the driver in terms of LED count.

 

[adova]

Anyone ever use one of those "cold solder" soldering irons that used to be advertised on TV? I have one as a gift from a couple of years back that I never used.

If not recommended (I would not be surprised) - what is a good iron?

Shawn

 

[dnahas]

I agree but am not willing to test connected to the cree's, I don't want to let the smoke out. Tell me another way to test it and I'll give it a try.

 

[der_wille_zur_macht]

Anyone ever use one of those "cold solder" soldering irons that used to be advertised on TV? I have one as a gift from a couple of years back that I never used.

If not recommended (I would not be surprised) - what is a good iron?

Shawn

Not recommended. Any iron with good power (40-50w+) and good temperature control would work fine. I use this one:

http://www.amazon.com/Aoyue-937-Dig...ref=sr_1_1?ie=UTF8&s=hi&qid=1268928273&sr=8-1

I agree but am not willing to test connected to the cree's, I don't want to let the smoke out. Tell me another way to test it and I'll give it a try.

You could check the meter's accuracy by measuring something known - i.e. measure current in a simple circuit with a voltage source and a resistor, then measure voltage across the resistor. Once you know the meter is reading accurately, if you really wanted to test the Meanwell "beyond spec" then you could hook up LEDs in parallel, such that the total projected current was less than each LED's limit when split among the number of parallel lines you set up. OR, you could use a big stinkin' power resistor. Either way, IMHO, it would be purely for curiosity's sake, I don't see any reason to go above spec on the DIM circuit on purpose.

 

[ti-christ]

Has anybody measured the light intensity of their LED fixture when the fixture is at different distance from the water (like 6", 1', 2' or 3')?

Because some people tells us to the put the LED fixture away from the tank to avoid salt deposit on the LED which is a great idea. But, if I remember correctly my physic, light intensity (lux) decrease by the square root of the distance! Maybe within a 1 meter, it does not matter, I do not remember the details !?!

Of course, there are a lot of good point to move away the light fixture from the water but I am afraid that the light intensity will decrease significantly.

 

[der_wille_zur_macht]

A small number of people have measured intensity at different distances, but you'll have to dig because the results are buried. User Santoki was probably the most thorough.

But, if I remember correctly my physic, light intensity (lux) decrease by the square root of the distance! Maybe within a 1 meter, it does not matter, I do not remember the details !?!

It depends on the "type" of light source and where you're moving in relation to it. The lessons we were all taught in Physics classes were abstracted for unrealistic "ideal" situations. The relationship you're remembering essentially has a slope component that's generally not considered because of the order of the equation. But, the distance changes we're talking about are on a scale that the slope CAN play a factor. The slope is essentially related to how laminar the light is. FWIW, the light leaving an LED is very laminar, which means the slope of the that "intensity vs. distance" curve is not as steep (i.e. we don't lose intensity as quickly) as with other types of light sources.

If you have a hard time grasping this, consider the light leaving a regular incandescent lamp (not laminar at all) vs. the light leaving a laser pointer (very laminar.) Over a 10 foot change in distance, you'll get a marked drop in intensity from the incandescent lamp, but the intensity measured from the laser pointer will barely change.

This is another area where LEDs are smashing some of the "truths" we're used to in the hobby. Even with very good reflectors, traditional reef lighting rigs never get very laminar light, which means that your assumption is true - the slope of instensity vs distance is such that changes we'd make in our light rigs (i.e. raising an MH to 24" instead of 12") will make a big difference in intensity. However, slap a 9 degree optic on an LED, and changing from 12" to 24" won't make nearly as much of a difference, in terms of percentage of the original intensity.

 

[The Grim Reefer]

OK so maybe some of you LEDolics can tell me what you think this layout will look like. I am thinking 50/50 Blue and white is too blue using XRE's and too white using XPG's. I want to try using a 50% mix of XRE Q5, 25% XPG and 25% XRE warm white. What do you think?

Oh, the Royal Blues are XRE's Not XPE' as the chart shows.

I've ordered the blues and XRE's just want to make sure before I order the rest of the whites.

 

[sammy113]

Whats the better option (talking about heatsinks) for doing LED strips of Royal Blues just to supplement T5? I want 12" modules.
Using aluminum "L" or angular channel
1"x1" or 1"x1/2" squared aluminum tube
getting any profile heatsink from heatsinkusa (lets say 8.4" x 12" and have it cut at 1" wide) final pieces will be 1"x12"

OR just getting 8.4 profile x 1" which is cheaper $2.25 but the fins run the short side

And last, 1/8" thick aluminum sheet (again 1"x12")

Thanks!

 

[Skeptic_07]

sammy: What drivers are you running at what current and how many leds?

grim: thats a pretty cool idea. I'm not sure how that would look because I've never seen a warm white cree in action.

 

[der_wille_zur_macht]

I want to try using a 50% mix of XRE Q5, 25% XPG and 25% XRE warm white. What do you think?

Looking at your diagram, it looks like your "total ratio" would be:

50% XR-E RB
25% cool white XP-E
12% warm white XR-E
12% cool white XP-G

It'll depend on the color bins you get for the various white LEDs but that strikes me as a very warm mix overall. I've never tried it but I bet it would look like an average 10kk MH lamp with no actinic supplementation.

If I were in your shoes, I'd either up the RBs by 10%, or switch some (or all) of the warm whites to neutral whites, or make sure my cool whites were a very blue color bin.

All within the realm of personal preference of course. . .

 

[koensayr]

Because some people tells us to the put the LED fixture away from the tank to avoid salt deposit on the LED which is a great idea. But, if I remember correctly my physic, light intensity (lux) decrease by the square root of the distance! Maybe within a 1 meter, it does not matter, I do not remember the details !?!

Let me try to explain this a bit more clearly. The inverse square law comes from the spreading of light over distance (it's actually the square of the distance, double your distance, intensity becomes 1/4). I'm guessing that is probably what you were remembering from physics class. If you imagine the sun, light comes out from the sun in all directions. At any distance R from the sun, the light is spread out over a sphere with radius R. Now, the surface area of a sphere is a function of R^2. So, if you double your distance (R), you get 4 times the surface area to spread the same amount of light over, which means your intensity now is 1/4.

However, with LEDs, the light is generally coming out at a much smaller cone instead of a sphere, so the area over which your light is spread over increases much slower with distance. If you make it such that pretty much all your light from the LED is making it into the tank instead of hitting outside of the tank, there is basically no loss. The only other thing there would be absorption and scattering by air and particles in the air, which is negligible for any distances we will be looking at for this use.

 

[kcress]

much smaller cone instead of a sphere, so the area over which your light is spread over increases much slower with distance.

You sure about that? The light coming from a small point is still going to be subject to 1/R^2. Even worse than if the sphere of the source was larger. As the source sphere gets larger,(rather than smaller), more of the rays are parallel and that's what reduces the spreading.

 

[TheFishMan65]

I think he is trying to say that the tighter the lens the less the radius changes over distance. It is still the inverse square law. If angle A is half of angle B. Then at the same distance B will have 4 times the amount of light as A in the given circle of the lens.

 

[koensayr]

Yes, it is still inverse square. What I was trying to say is that, if your solid angle is small enough, and your entire surface of your solid angle falls inside your tank, then the distance from the tank doesn't matter because even though the light has spread out, all of the light is still making it into your tank. So you are not wasting any light due to the inverse square law. I guess I wasn't quite as clear as I wanted in the first response

 

[stugray]

I'll pipe up here a little:

The inverse square law applies to a POINT SOURCE in free space.

All the stars in the sky are point sources that radiate light in all directions (essentially) equally.

A device that directs a beam does NOT follow the inverse square law.

As stated above, a "perfect" laser beam with no divergence would lose zero intensity with distance.

Most of our LEDs have a "divergence pattern" or "cone" that ~90% ( or the 50% FWHM thing ) of the light is emitted in. Therefore the inverse square law does not apply. They are not "point source emitters".


So an LED with a very narrow optic will lose almost zero intensity at distance as long as the optics are "decent".

Stu

 

[dnahas]

How are people attaching the little 10414 Carclo optics with the 4 pegs? Did I not order something? yea they fit into the holes on the PCB but that's about it, they will not stay if you just tap on the canopy.

 

[ReefEnabler]

Dab of adhesive, like epoxy or polyeurethane, superglue also works but some datasheets say no superglue.... I still used it

 

[akindbro4u]

Dab of adhesive, like epoxy or polyeurethane, superglue also works but some datasheets say no superglue.... I still used it

the reason they say no superglue, is because if you ever touched your led, and didn't clean the oil off the fumes will adhere to that and harden. Trust me it's no fun to realize you made that mistake.

 

[TheFishMan65]

stu,

You have me confused. You say that LEDs they LEDs are not point souce. They are they just don't spread in every direction only a limed area. You also agree that the have a FWHM angle. That means they do spread, Lets make this simple say the FWHM is 90 degree. At three feed they will cast a circle of light 3 feet in diagmeter. At six feed the radius will be 6 feet. Here is your first chance to disagree. Now lets say the LED put out 100 lumen. In the first case all that light falls in the 9pi square foot circle. Yes? In the second case it falls in a 36oi square foot circle. Yes? So lume per square foot has decreased by a factor of 4.

 

[stugray]

TheFishMan65,

"You have me confused. You say that LEDs they LEDs are not point souce."

No they are not. Lets assume we had an LED that had a FWHM of 120 degrees ( XP-G ).

Then you could assume that ALL of the light is radiated out in one HALF of a full sphere then that would be 2PI steradians.

A point source puts out all of its light over 4PI steradians.

So for a point source the inverse square law works, but for an emitter that emits over 2PI steradians, it should be = (Inverse Square LAW) / 2 for an emitter that emits all light over 1/2 sphere (hemisphere).

It goes down as the emitter FWHM reaches zero such that an emitter that is collimated has zero loss at distance (theoretical perfect laser).

Stu