LED Aesthetics: What do you really think of your color?

[blasterman]

That's all fine and dandy, but to be 'blunt' again where are the spectral graphs of those 10k -16k emitters?

 

When you add royals to a typical cool white emitter you get 10k-20k CCT. If this is the goal, why are we screwing around with all these oddball chinese emitters? Is there some kind of award paid for who can light a reef tank using the biggest variety of LEDs?

 

Again, you are not covering different spectrums using different CCT white emitters. You are just overlapping different spikes.

[Mr. Microscope]

Again, you are not covering different spectrums using different CCT white emitters. You are just overlapping different spikes.

 

True, but this thread is about aesthetics. We're talking about what looks best. I'd say that tank looks great! Though, I agree that the setup is a bit on the overly complicated end. Hopefully we can figure this out with four or less colors. I've never heard of 16KK LEDs. Kinda neat. I can't wait to see what the best ratio ends up being on those TVs.

[graham_more]

Been following this thread for a while has been really interesting.

 

In June I went for an array of 6CW(xre), 8 RB(xpe), 1 UV(led engin) combo over my frag tank, and just decided on upgrading my 24 Gal cube lighting to part led.

The UV Led hasn't made any real difference(no notable flourescence/colouration).

 

After reading this thread I decided to mix up the emitters a bit for the 24Gal.

Wanted to replace a 24W PC moonlight for a actintic/fiji purple.

I went with a 7 Led strip RB,CB,RB,RED,RB,CB,RB @700ma(all xpe). Tried to get a violet look without using a true violet.

Found the 1 red to 6 Blue is still way overpowering, so made a home made diffuser from white tubing.

would estimate a 1:10-12+ for red would be closer to a purple hue.

Looks far better than just the Rb on the frag tank , much more colour depth.

Also going to be adding some extra whites to help the sps growth.

Have tried to take some pictures but i cant seem to get the white balance to be anywhere near right :[.

 

Thanks for all the info :]

[Hans Dorn]

Again, you are not covering different spectrums using different CCT white emitters. You are just overlapping different spikes.

That's not quite true. The warmer emitters have their peak phoshor output at higher wavelengths and thusly enhance different parts of the spectrum.

 

It's not like they only vary in the amount of yellow/red emitted by the phosphor.

 

[skunker]

From Acrotrdco on RC

http://www.reefcentral.com/forums/showthre...17#post17836317

He also posted all that is this thread: http://www.nano-reef.com/forums/index.php?...40&start=40

 

Been following this thread for a while has been really interesting.

 

In June I went for an array of 6CW(xre), 8 RB(xpe), 1 UV(led engin) combo over my frag tank, and just decided on upgrading my 24 Gal cube lighting to part led.

The UV Led hasn't made any real difference(no notable flourescence/colouration).

 

After reading this thread I decided to mix up the emitters a bit for the 24Gal.

Wanted to replace a 24W PC moonlight for a actintic/fiji purple.

I went with a 7 Led strip RB,CB,RB,RED,RB,CB,RB @700ma(all xpe). Tried to get a violet look without using a true violet.

Found the 1 red to 6 Blue is still way overpowering, so made a home made diffuser from white tubing.

would estimate a 1:10-12+ for red would be closer to a purple hue.

Looks far better than just the Rb on the frag tank , much more colour depth.

Also going to be adding some extra whites to help the sps growth.

Have tried to take some pictures but i cant seem to get the white balance to be anywhere near right :[.

 

Thanks for all the info :]

There is a difference between purple and violet. The spectra of the Fiji Purple bulb is in the post by OneTrickPony about arcotrdco's build. By combining blue and red you made purple, not violet. You get the color depth like you say, but not the florescence that comes from true violet.

[ReggieF]

I've read most of this thread and I think I can add some helpful information to this topic that should shed some light (hehe) on your color quests:

 

Links:

- http://hyperphysics.phy-astr.gsu.edu/hbase...on/rodcone.html

- http://en.wikipedia.org/wiki/Color_vision

- http://en.wikipedia.org/wiki/Color

 

1) The human eye can't see blue.

 

"Current understanding is that the 6 to 7 million cones can be divided into "red" cones (64%), "green" cones (32%), and "blue" cones (2%) based on measured response curves."

 

Atleast not as much as Red and Green. This is a major reason why people are complaining about their tank being "washed out". 10K White is already brightest in blue and green and red come afterwards. Warm whites on the other hand are brighter in the red spectrum, and they would probably be a good addition to any tank with strong Royal blue LEDs to add balance and color visible to the human eye. I remember hearing that the human eye can detect 16 different shades of Blue at once, 120 shades of Red and 160 shades of green, so keep that in mind.

 

2) Cyan IS missing! Cyan is at 485 nm and notice the huge drop off at that frequency in Hans Dorn's image. Cyan is also a mixture of Green and Blue light, so a 4-8K white + Royal Blue should create some nice cyan. The below color spectrum was taken from an LED setup that's supposedly 14K Cree LEDs. Notice the large amount of Cyan in sunlight, and about 20% of that in the MH. Then notice there's about 4% in Fluorescent lighting and basically none at all in this 14K LED lighting.

 

3) MH lighting has many sharp peaks where LED is a much smoother color curve which is probably better for coral. This makes the MH light seem a lot shinier to the human eye because it highlights certain colors with sharp peaks, almost in the same way that jewelry looks appealing with sharp peaks of color. This is actually a much more complex point than I can type about, but the idea is essentially that these color spikes will make things look nice to the human eye, but that doesn't mean they're good for photosynthetic organisms, in fact I'm pretty sure that it's the opposite.

 

[doctaq]

crazy idea for the cool blue..biconcave lens, right like the opposite of optics we have dealt with currently

[Mr. Microscope]

crazy idea for the cool blue..biconcave lens, right like the opposite of optics we have dealt with currently

+1. Great idea. For smaller setups closer to the waterline, it'd help blending a lot and the LEDs wouldn't need to be super close to each other. Though I imagine PAR would drop off very fast.

 

However, without lenses the LEDs pump out light at 120 degrees already. Maybe it could be pushed to 140, but perhaps not much more than that. Also, with biconcave lenses we're working with a beam whose collumation has been crossed over. This can cause some astigmatism and abberations towards the edges. Perhaps not worth the added coverage depending on the severity of artifacts.

[jfarabaugh]

Getting back into things and have been toying around with the LED layout.

12 NW

12 RB

6 CW

6 B

6 violet from nanotuners when they become available.

 

This is what I have come up with so far. Feel free to comment. Once I finalize and build I will make sure to post pics.

 

[doctaq]

at the edges of the 120 deg the output is nearly half of what it is in the middle, i guess it could be worth getting a single lens and playing with my lux meter to see how it affects the spread

[joelsaxton]

Getting back into things and have been toying around with the LED layout.

12 NW

12 RB

6 CW

6 B

6 violet from nanotuners when they become available.

 

This is what I have come up with so far. Feel free to comment. Once I finalize and build I will make sure to post pics.

 

How are you going to drive these LED's? I am wondering if the NW's will overpower the other colors.

[jfarabaugh]

How are you going to drive these LED's? I am wondering if the NW's will overpower the other colors.

 

How I will drive them

http://www.nano-reef.com/forums/index.php?...=245630&hl=

 

The whites and blues are on seperate channels so I can control the intensity of each color to get get the proper mix.

[Mr. Microscope]

42 LEDs? How big is your tank?

[jfarabaugh]

42 LEDs? How big is your tank?

 

BC29

 

Keeping in mind that nothing will be running any higher than 500mA, no optics, and will be dialed down via my controller.

 

It is a lot of fire power but I will not have to worry about color blending. I am about to pull up visio again and mess with it so it may change.

 

I am very open to suggestions

[jfarabaugh]

New design and I like it better as it should help avoid strips on any high rocks in the tank

 

[Genj]

That's a lot of LEDs. I was at the same number for the Solana, but I backed off because I'm pretty sure it would be overkill.

 

I should have some pictures of the setup, minus the TV, next week or early the following.

[Machupicchu]

Taking blue out of the picture, lets get a quick vote on who thinks Cool White should be eliminated. And if not why, and what ratios of CW and NW should be used.

[fishfanatic88]

so the warm whites are too far down the spectrum right? they are too hard to blend with blue?

[blasterman]

I don't think CW's should be 'eliminated' per say because;

 

(1) There are a lot of people using cool-whites and RB's that are happy.

 

(2) CW emitters encompass a broad range of color temps, ranging from 10,000k to 5600k. Cool whites warmer than that exist but are rather rare in the lighting industry in general. Somebody running WG Crees or cool-white Bridgelux (5600k) is going going to have an entirely different aethestic than somebody running Chinese CW's, DX brand Cree's (often WC bins or higher) or those ghastly 7000-10000k emitters that somebody is advertising as '100% natural color'. Don't get me started on the later because if those are 'natural color' because the CCT is approx the mid day sun then a twinkie is health food. Editorial / off

 

At worst I'd say go no higher than WG bin CW's, and if at all possible try both NWs and CWs small scale to see what you prefer. Somebody running a lot of SPS -vs- LPS and softies might not mind a CW centric light while those of us with acans, zoos etc., have found CWs to be pretty 'blah' and in dire need of an over-haul. As I keep telling people, CW emitters are designed to bright to the human eye and work efficiently in street lights. They weren't designed for reefing (like any other white LED).

 

In general though it seems RBs and NWs deliver a much broader pallette, and at a ratio of around 1.5-2 : 1. It's tougher to narrow it down more than that because of the big variance in efficiency of NW emitters. A NW XPG can be twice as bright as a mediocre flux XRE or Rebel.

[Machupicchu]

So for those of us who are trying to keep it as simple as possible, myself for example( im trying to keep it down to two drivers), What would be a very rough ratio of NW to CW be? lets just say i keep all types of corals and want the best overall look. From what iv read it looks like a slightly higher ratio of NWs to CWs would be best, perhaps 1.5:1?

 

And as for blues it would be a much higher RB ratio, something like 3:1 or 4:1.

 

All blues run at same mA and all whites at same mA.

 

Also, i think someone had an idea earlier that you could sand down the lenses in the optics on certain LEDs if they were to powerful but you were not able to dim them separately. Im sure theres alot of things you could do.

 

Im speaking from a poor reefers point of view

[reef-luva]

I am using one Panorama Module 8000K/453nm Actinic Blue and two stunner strips of the same spectrum.

I don't mind the fact that I can't control them separate and/or dim them. I have a 12" cold cathode lamp (blue) for the moon and it is what comes on first in the morning. Colors is great, effect is great and every one is happy....btw, nice set up DPR.

[fishfanatic88]

So for those of us who are trying to keep it as simple as possible, myself for example( im trying to keep it down to two drivers), What would be a very rough ratio of NW to CW be? lets just say i keep all types of corals and want the best overall look. From what iv read it looks like a slightly higher ratio of NWs to CWs would be best, perhaps 1.5:1?

 

And as for blues it would be a much higher RB ratio, something like 3:1 or 4:1.

 

All blues run at same mA and all whites at same mA.

 

Also, i think someone had an idea earlier that you could sand down the lenses in the optics on certain LEDs if they were to powerful but you were not able to dim them separately. Im sure theres alot of things you could do.

 

Im speaking from a poor reefers point of view

 

i think that is what i will be doing as well about 2-1 nw to cw and a 5-1 rb to cb

all whites running the same ma and all blues running the same as well

[Genj]

This thread needs more pictures! Hopefully I can provide a few by the end of next week.

[ReggieF]

Here we go!! This is the PERFECT graph to understand the science behind this LED "washed out color" issue. Look at the green graph for a goal essential for the human eye. Notice how blue cool white is and how red warm white is! The missing color in standard LED setups is actually red + green as cool white and RB are all near the blue light spectrum.

As I've said in other posts, to those of you upset with your LED lighting, throw in a standard warm-white CFL bulb and take a look at the difference!

[JSVAND5]

I have a system coming for my NC 24. It will be 3 rb strips and 2 cw. Would adding a cheap PC fixture to my hood improve the color output at noticeable level? I was just looking around on ebay and thinking maybe 2 of these (http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=320563924751) with 2 18w bulbs would be a cheap way to maybe fix the color issue. I don't have my hood yet, so I don't even know if it would fit in there, but if I were able to fit it would it help? Any ideas for a better solution? I think replacing individual bulbs in my LED unit with NW bulbs may be a little beyond my skill level and I would assume it would void any warranty on the system which I don't want to do.