Open letter to the LED industry

[Big E]

I guess we'll find out soon enough. A lot has been done with configuration of the diodes to do this and some attack it in different ways.......for example the puck style.


Using standard diodes makes this extremely challanging imo. You can see that newer units are working better but they are now using the same or more electricity as T5 setps.

I think eventually muli-chip diodes will be made to do this and have good enough reflectors.

There is a huge thread here with DIYers trying to tackle this whole issue.

http://reefcentral.com/forums/showthread.php?t=2128756&page=153

There's a reason this thread has hundreds of pages and over 350,000+ views.

 

[asid61]

Is it even possible for these tight beamed limited spectrum single units to work together in cohesion to provide optimum results, or will this eventually lead to a single LED that can provide the whole spectrum?

Because reefkeeping is such a niche market, no one led will likely ever be developed that replicates MH lighting.
As a result, you need to combine multiple leds.

I just input some stuff into this calculator:
http://www.1023world.net/diy/spectra/

And came up with this:

It's pretty close to a Radium, but leds actually have too wide a spectrum in order to get those super-narrow peaks a MH bulb has. So the cyan and green peaks blend together.

All Rebel ES, with no whites.
50 RB, 15 Cyan, 15 Green, 15 amber, 1 RedOrange, 2 Red, 2 DeepRed.
And some Semileds for violet:
5 390-400nm, 5 410-420nm.

 

[senne76]

If I calculate my spectrum there I get this:

And I have a 1:1 white/ rb ratio , my rb leds are just stronger than the white ones

 

[asid61]

Using standard diodes makes this extremely challanging imo. You can see that newer units are working better but they are now using the same or more electricity as T5 setps.

Leds don't use more power than T5, unless you opt to go with a very cheap multichip with low efficiency. Most commercial fixtures have high efficiency, with 122 lm/w being the standard for whites. White T5 tubes only get around 90lm/w.

Low efficiency leds, however, can be as bad as 60lm/w.

 

[asid61]

If I calculate my spectrum there I get this:

And I have a 1:1 white/ rb ratio , my rb leds are just stronger than the white ones

How did you get a violet peak then? Or such a strong red peak?

 

[Pacific Sun RD]

He didn't get it, because It's not real led lamp light spectrum - there was putted not proper data(violet peak without violet leds)

 

[senne76]

Oh sorry , I didn't mentioned the other led's , the correct ratio is 4(neutral)w/4rb/2bl/2Tv/1gr/1cy/1deepr.
Does this look like a correct spectrum? If it could get better , its a DIY so I can swap LED's.

 

[Vannpytt]

Again, I need to remind you that participate in this thread:

1. I am not affiliated with the industry in any way.
2. I don't have scientific proof, just observations, questions and hypothesis(statements)
3. I don't do this to hurt anyone, rather try to improve what I feel is missing

And for the statements I've made:

1. "White" junkspectrum LEDs driven at higher than 400mA in a chain of 12 is crap
2. Using lenses on LEDs in tanks that are less than 65cm/mounted 50cm above the tank, is pointless
3. Not using enought LEDs to get decent coverage causes too much shading
4. Having more than 1/4th of the ratio in "white" LEDs are pointless.

I would rather have a fixture with the setup Pacific Sun provided, and honestly, I did not even know of their new SMT lamp without white chips, but I do feel that this really is a big step forward and I'm quite sure more will follow this trend.

Questions I have:

1. Why can you run a 400w MH over a 60cm tank and grow corals through the roof with insane colors, but try the same thing with a few high intensity 1000mA+ "white" LEDs and your corals will brown and eventually die?
2. Corals growth is very much depending on being directly lit by the LEDs. Top down they can look "great", from the side they look mostly boring. This does not happen under traditional lightsources. How can this be countered?
3. Why is no one from industry beside Pacific Sun participating in the thread? (And Vertex, but I think that was a rogue operator, and he didn't bring much information)?

 

[Vannpytt]

Oh sorry , I didn't mentioned the other led's , the correct ratio is 4(neutral)w/4rb/2bl/2Tv/1gr/1cy/1deepr.
Does this look like a correct spectrum? If it could get better , its a DIY so I can swap LED's.

And what do you run your white LEDs on?

 

[asid61]

It's not supposed to be all that accurate.
However, with those leds, replace the red with a warm white. Too much red in your design.

Also, on spectral shifts:
Even if the spectrum shifts, it's so minimal it doesn't make a difference. It's not even listed in most led datasheets.

To Vannpytt:
I think the reason that most people use white leds instead of discretes is because it's easier than looking to replicate white leds. You would need to test different combos of leds, in large numbers, with an extensive control over the led current. And you would need access to a spectrometer. A fixture like that would have to be specialty, as you would need something like 5 of each led to account for tolerances in led spectrum, and thus a lot of space on the heatsink would be wasted. You would also need a big heatsink, but that's not as important.
So testing any of this in real life can only be accomplished with both money and status. That means a regular contributor to the forum with a lot of money and an interest in leds, or a trusted sponsor.

 

[gcarroll]

Do you know which PAR was 10cm under water level?
Im sure that not.

I noticed you didn't say I was wrong. Do you expect me to think that a fixture you designed in 2009 put out more PAR than you SMT?

Did you saw that formosa growth up to water level?
Why it wasnt "burned"? We know that - we checked light condition in that and many other test tanks.. We not only try to make nice looking(for our eyes) lights - but we try to make light the best for corals growth and pigment intensify.

Yes and I think it looks great! I didn't say that LED will burn and kill your corals. I have to ask. Did the lighting on this tank look good to your eyes and was the best for the corals?

Did you saw that formosa growth up to water level?
You dont know also what kind of white leds(maybe more green or red - depending of TINT) we used - and there was Royal Blue(450nm) or Blue used(475nm)?.. We know that - and if you will check our SMT matrix - you will understand why we used so many "not popular" LED chip colors.. Not "popular" - but which give effect which we expected.

We are talking 2009 right. Your choice of LEDs was far more limited than here in 2013. Are you claiming that Vannpytt's assumption that all white LEDs are bad are unfounded? If not what does it matter?

 

[Secondsbest]

While I have no idea why users are experiencing positive or negative results from LED fixtures, dimming has a very small effect on LED wavelength output. The natural available spectrum to wild corals could vary as much, or more than LED dimming due to angle of incidence/ surface scattering throughout the daylight cycle.

http://streetlightingresearch.org/programs/solidstate/pdf/gu-spie6337-17.pdf

 

[jda]

Because reefkeeping is such a niche market, no one led will likely ever be developed that replicates MH lighting.

...say, as much of a niche market as designing aquarium-only Metal Halide or Fluorescent bulbs? Pumps? Heaters? Seriously, what is the reason since every other light source has aquarium-only bulbs and components that are effective and profitable. Is there no way to do it? Is there not yet enough true demand, beyond message board chatter, for somebody to invest?

 

[nickdo]

Do you know which PAR was 10cm under water level?
Im sure that not.
Did you saw that formosa growth up to water level?
Why it wasnt "burned"? We know that - we checked light condition in that and many other test tanks.. We not only try to make nice looking(for our eyes) lights - but we try to make light the best for corals growth and pigment intensify.
You dont know also what kind of white leds(maybe more green or red - depending of TINT) we used - and there was Royal Blue(450nm) or Blue used(475nm)?.. We know that - and if you will check our SMT matrix - you will understand why we used so many "not popular" LED chip colors.. Not "popular" - but which give effect which we expected..
Of course, I will not say you that(it would be unwise to share the results of many years of experience) but really in this thread I said a lot of things, that until now has probably never been discussed in similar threads (certainly by anybody from "aquatic LED industry")
This will be in my opinion the best and unique proof.

Przemek, Thanks about your contribution.
I've no doubt about using LED to keep SPS corals.
Do you have any information/experiments about using LED to keep LPS corals? especially for the brain type corals? Since all "master grade" LPS keeper in my local communities refused to using LED to keep their expensive LPS corals. Some Coral distributors tried to keep their LPS stock using LED(blue only), but he change back to ATI T5 finally(with actinic and blue tubes). He claims that LPS are losing color.

 

[asid61]

...say, as much of a niche market as designing aquarium-only Metal Halide or Fluorescent bulbs? Pumps? Heaters? Seriously, what is the reason since every other light source has aquarium-only bulbs and components that are effective and profitable. Is there no way to do it? Is there not yet enough true demand, beyond message board chatter, for somebody to invest?

No, there's not enough chatter. I think that the reason we have aquarium MH lamps are because of horticulture- a thriving market for lights. Some 10k bulbs are even sold to provide more blue for plant growth. I am totally guessing here, but I think you could just adjust the gasses a bit to make it spit out more blue light.
As for heaters and pumps, both technologies are used in other applications like pools. Only the ascetics are bit different.

Leds are an entirely different matter. You have to mess with the different phosphors in a micro-area to produce different light and spectrums. It's an entirely different thing to do, versus simply putting in gas (which likely is not simple, but is in contrast with phosphors). Not to mention that the led companies would need to do some serious research on what phosphor blend to use- manpower that is hard to spare when you have competitors. The led market changes every year, and they have to keep up. How many reefers would buy such a chip? It would be huge gamble anyway due to the anti-led people.
Someday, we will probably end up with an aquarium-specialty led. But until then, we're stuck with multichips and dense matrixes.

 

[senne76]

I run my white leds on 70% , but my royal blue leds also run on 70%.
But again , I'm asking for an explanation for the succes with the triton lani led which also has a 1:1 ratio

 

[nynick]

Someday, we will probably end up with an aquarium-specialty led. But until then, we're stuck with multichips and dense matrixes.

I don't feel like we are "stuck" with multichips. There will never be a single chip to use for reef tanks so either we use multiple chips or multichips. This is just the nature of the beast and there is nothing we can do about it.

Also, multichips have by far the best looking light avaliable in the hobby today. That intrence sun like shimmer just brings tanks to life, even more than MH. It is small and has atleast the potential to go fanless. It has the potential to be cheaper than any other lighting option. It has every advantage over every other lighting option besides the most important one, spectrum.

 

[Vannpytt]

I run my white leds on 70% , but my royal blue leds also run on 70%.
But again , I'm asking for an explanation for the succes with the triton lani led which also has a 1:1 ratio

It seems you have not paid proper attention. You don't "need" the massive peak in blue that you get from a 1:1 white:blue. You just need less of the "white" than what current fixtures provide.

The Lani LED is easy to explain. They run their LEDs without lenses, same as Pacific Ocean, thus reducing the yellow/red spectrum from penetrating deep in the aquarium, as well as running them on low current. The latter is an assumption based on the fact that they have no active cooling. Lani has not posted a single graph on their website, but I would think that they match my observations quite accurately. Feel free to correct me. This is observations based on their information, but without a spectrum graph, I would be making guesses based on my own observations..

 

[penfold2]

It seems you have not paid proper attention. You don't "need" the massive peak in blue that you get from a 1:1 white:blue. You just need less of the "white" than what current fixtures provide.

Of course you need a massive blue peak. That's the one thing that every led builder agrees on. The builds we're talking about here just substitute carefully chosen discrete diodes for a wide spectrum white diode.

The Lani LED is easy to explain. They run their LEDs without lenses, same as Pacific Ocean, thus reducing the yellow/red spectrum from penetrating deep in the aquarium, as well as running them on low current.

But these factors apply to all of the diodes, so the whites are being run no differently than the blues. Nothing is explained since the ratio still applies.

I think the ideas being described here are very interesting. I got a much better feel for what PS RD was aiming for after he posted the spectrum of his Hyperion-S along with the fluorescent pigment chart. I just wish we could stop with the "white leds are damaging" nonsense. They may contribute to washing out some fluorescent pigments, but there is no good reason to believe they are doing any harm beyond that. Until someone can describe a mechanism for this damage and explain all the contrary evidence, it is just a theory held by very few people.

 

[senne76]

It seems you have not paid proper attention. You don't "need" the massive peak in blue that you get from a 1:1 white:blue. You just need less of the "white" than what current fixtures provide.

The Lani LED is easy to explain. They run their LEDs without lenses, same as Pacific Ocean, thus reducing the yellow/red spectrum from penetrating deep in the aquarium, as well as running them on low current. The latter is an assumption based on the fact that they have no active cooling. Lani has not posted a single graph on their website, but I would think that they match my observations quite accurately. Feel free to correct me. This is observations based on their information, but without a spectrum graph, I would be making guesses based on my own observations..

Most of the lani led users I've seen (including the inventors in Germany) are running all of the leds at the same current. They don't need a spectrum graph because their light just works