Open letter to the LED industry

[TropTrea]

Someoe refered me to this tread because of the spectrum plots on some of the T-5 bulbs here. A little over a year ago I was able to get some spectrum plots on various LED's and T-5 bulbs as I was searching for a better lighting system. Orior to running these plots I had been in the process of converting from all T-5 Lighting to all LED's. I was getting mixed results with some corals reacting very favorable to the LED's and others much less so especialy in growth.

After the study on LED's I did I saw some holes in the LED spectrums that were not cecessarly in the main photosynthesis regions cut did corelate to many of the reported florescense excitation areas.

With the short wave lenght LED's I tested the 420nm leds produced only about 1/10th as much light at 405 and 435 nm as they did at 410. The 455 nm LED's again were about 1/10 at 440nm and 470 nm. This left a gap between roughly 430 and 445 that seemed to low for me. I also found simular gaps in the 480nm to 510 nm range.

As I also ran plots on the Blue Plus bulbs yes they had definate peaks but they also seemed to fill in the gaps that I was missing with just the LED's. Since then I started converting to a combination of Using Blue Plus T-5 bulbs with LED's. The results on my corals are showing a extrem positive results all the way around at this point.

Any comments on my thoughts here are welcome. And I'll continue to tag along on this tag as there are many interesting points here being brought up.

 

[TropTrea]

Pasific Sun RD. The plots I ran on the ATI Purple Plus bulbs are extreml;y simular to the plots that you are showing. The only difference being posible calibration differences on the frequencies as my peaks are roughly 10nm different from your on the low short end and 15 nm different on the longer wavelenghts.

An interesting thought is the three bulbs that your ploted are all extremly simular in my mind. Yes slight differences in magnitude of the peaks but they could all fall within manfacturers specs for the same bulb. Could these three bulbs be made at the same factory.

The reason I ask this is years ago I worked for a lighting company that produced florescent tubes. Very often we would produce tubes with other brand names that they would farm out to us under there specs. Sometimes some of these specialty bulbs would be identical to others with the only difference being the stamping on the tube.

 

[hypnoj]

great research! Thank you for sharing.

 

[clay12340]

This is an interesting read. However, it seems like it might be easier in a lot of cases to just have whichever lights aren't responsible for pigmentation, but are needed to look good to human eyes and might contribute to browning of corals on a separate light/switch. You could just turn it on when you want to look at the tank or set it on a motion switch or something to come on when you enter the room etc. It seems easier than trying to make a perfect compromise all the time.

As long as the relative intensity change isn't huge I wouldn't expect adverse reactions to the changing lights. At the same time I could be completely and totally wrong about that. My SPS experience is extremely limited, and even that wasn't very productive.

 

[penfold2]

This is an interesting read. However, it seems like it might be easier in a lot of cases to just have whichever lights aren't responsible for pigmentation, but are needed to look good to human eyes and might contribute to browning of corals on a separate light/switch. You could just turn it on when you want to look at the tank or set it on a motion switch or something to come on when you enter the room etc. It seems easier than trying to make a perfect compromise all the time.

I don't think this thread is about compromise. It's about creating a light source that maximizes our perception of coral fluorescence. If some people prefer a more natural daylight appearance, they should pursue that, but I don't think there would be any point to switching back and forth when viewing and not viewing. Coral will grow well and maintain the preferred coloration in either scenario.

 

[TropTrea]

The issue is if you look at the florescent excitation chart you need light from roughly 400nm to 190nm to get 95% of the pigments. This is not very far away from full spectrum lighting with the exception of the red light. Now we also know that in nature most of the red light is fiulterered out, and there are some reports that red light is the biggest cause of bleaching and browning in some corals.

So balancing florescense and Growth I would think we want a fairly even light spectrum from 400nm to 590nm with some additional peaks at precise photosynthesis wavelenghts exspecialy in the blue range

 

[unitedbga]

Someoe refered me to this tread because of the spectrum plots on some of the T-5 bulbs here. A little over a year ago I was able to get some spectrum plots on various LED's and T-5 bulbs as I was searching for a better lighting system. Orior to running these plots I had been in the process of converting from all T-5 Lighting to all LED's. I was getting mixed results with some corals reacting very favorable to the LED's and others much less so especialy in growth.

After the study on LED's I did I saw some holes in the LED spectrums that were not cecessarly in the main photosynthesis regions cut did corelate to many of the reported florescense excitation areas.

With the short wave lenght LED's I tested the 420nm leds produced only about 1/10th as much light at 405 and 435 nm as they did at 410. The 455 nm LED's again were about 1/10 at 440nm and 470 nm. This left a gap between roughly 430 and 445 that seemed to low for me. I also found simular gaps in the 480nm to 510 nm range.

As I also ran plots on the Blue Plus bulbs yes they had definate peaks but they also seemed to fill in the gaps that I was missing with just the LED's. Since then I started converting to a combination of Using Blue Plus T-5 bulbs with LED's. The results on my corals are showing a extrem positive results all the way around at this point.

Any comments on my thoughts here are welcome. And I'll continue to tag along on this tag as there are many interesting points here being brought up.

I am finding the same thing with my lights. I am now supplimenting my two D120 LED fixtures with a 150w Metal Halide sandwiched between them. I am using it to fill in the spectrum that the LED's miss. My acans are turning for the better.

I've changed out a lot of the LED's in my Evergrow D120's.

Current configuration is

390 - 1
400 - 2
410 - 4
420 - 8
430 - 4
440 - 4
450 - 16
460 - 8
480 - 4
495 - 2
630 - 1
660 - 1

 

[Pacific Sun RD]

problem is when we forgot that leds are monochromatic type of lighting.
They produce narrow band and built proper spectrum without using many different types of leds is not possible..
When we started our research two years ago(before SMT(S) series was released ) we firstly measured all available "best" lighting setups for growth/coloration(t5, t5+mh etc).
That allowed us built led panel which have very wide spectrum without any gaps.
If you will compare spectrum of most popular(actually) t5 tubes which give best pigments on sps dominated tank(Aquablue/coral plus bulbs and other manufacturers models + blue plus type tube) to newest technology panels - you will see thats actually its possible to mimic needed spectrum by them without problem.
Coral plus t5 tube (and similar type) spectrum:

Aquablue ans similar tubes:

Blue plus and similar:

SMT matrix:

Only one gap which can be observed is near 550nm - that light is not so important in pigment built, because in this area there is not so many GFP pigments:

Most important area is between 490-510nm - there we can find many pigments and many manufacturers still dont remember about that important fact, typical multi chip fixtures have still gaps in that area ...
If led panels dont use white leds - spectrum built by leds(placed very,very close together) give perfect color blending and not comparable spectrum control.
Other charts to compare:

 

[zachts]

One large failing of LEDs still (and only from a human visual perception point of view, ie. perceived brightness) Is that there are no LEDs available that can output high power light with a narrow peak between 540nm and 550nm. (the large green spike in all fluorescent tubes)

It is possible to achieve in LED lighting by use of a phosphor converted green LED (blue led with a narrow band green phosphor only) but no one is using them to date.

This is a major difference in the spectrum and part of why many people find their viewing experience is enhanced by addition of some T5 lighting.

That very narrow high power violet spike on the T5 graphs also cannot be replicated by LEDs at this time. It is providing energy to the corals for growth by targeting their photosynthetic peaks while staying outside the range that us humans can easily see the light. Thus making the light look less blue and therefore brighter to us while still giving the coral the energy it needs.

 

[Pacific Sun RD]

That part of spectrum is not so important for corals - its only for "human eyes" - and its a major problem. Why?
Because many people set led lamp power based on "human eyes perception" - and they ar setting it too high(than it should be).
This 550nm are is "lumen window" for human eyes - most brigthnest lightwaves - and "properly" configured led lamp seems to be too dark - and every Customer increase power to get 'similar" results like under T5(for example).. Its a common mistake..
Too much power = troubles..
Remember that this green peak in T5 is not possible to remove - its depend from gasses used to produce t5 tube - and in each tube it "have to be".
its not possible to built blue looking t5 tube without green peak..

 

[TropTrea]

One large failing of LEDs still (and only from a human visual perception point of view, ie. perceived brightness) Is that there are no LEDs available that can output high power light with a narrow peak between 540nm and 550nm. (the large green spike in all fluorescent tubes)

It is possible to achieve in LED lighting by use of a phosphor converted green LED (blue led with a narrow band green phosphor only) but no one is using them to date.

This is a major difference in the spectrum and part of why many people find their viewing experience is enhanced by addition of some T5 lighting.

That very narrow high power violet spike on the T5 graphs also cannot be replicated by LEDs at this time. It is providing energy to the corals for growth by targeting their photosynthetic peaks while staying outside the range that us humans can easily see the light. Thus making the light look less blue and therefore brighter to us while still giving the coral the energy it needs.

I think that t40-550 spike spike cannot be duplicated to day not because they cannot vreate a LED in that spectrum but because there is little demand for it. If you look a medical LED's you will find a lot more selections than what most LEd resellers handle. But investigate proce and you would not want to touch some of tose LED. The driving matter on price is with a very limited demand the cost of set up can exceed the cost of low run production. So if there was a large demand for those LED's they would be available at reasonable costs.

As far as the vioet LEd are concerned I think the same princile applies. Most medical uses for leds indicate the narrower the wave lenght band is the better. It could be very possible to create a LED with a band width between 310 and say 470 nm but what is the neamand for such a beast. If the market studies show the demand is so small that it is not economical why put an effort to makes such a beast.

 

[zachts]

I think you both misunderstand my statemnt to a degree. I'm not saying that 540nm light is there for the corals or for pigment development. It is only included in T5 bulbs for the sake of the humans viewing the corals under those lights.

We want to have a tank that is brightly illuminated and easy to view our corals and fish, While at the same time maximizing coral pigmentation. T5 includes this wavelength only because it makes the light look bright to humans and the end result is what hobbyists are after. T5 lights specifically use a phosphor that creates this spectrum, they could not have it and the light would look purple, or not include the red phosphor as well and then you have an actinic blue bulb.

In order to achieve a bright light under LED the same principle needs to be applied using narrow red and a narrow green band as the spikes to create the brightness for the human eye that we desire for our tanks.

Red LEDs are narrow enough for the red spike. However no high power Green LEDs with a narrow spike are available at the moment because as you've stated it is not "in demand" so the chip manufacturers are not producing them (philips is rumored to have one but they are keeping it to themselves and using it in their "HUE" lights)

A company like Pacific Sun could easily implement their own chip assembly using a remote phosphor applied to over a blue LED. The technology is very inexpensive and a prototype could be made for a modest R&D cost. This is the last Key component I believe is still missing from LED lighting for Reef tank use.

For Example One such phosphor readily available from a US supplier:
https://secure40.securewebsession.com/2nc99ci87.site.aplus.net/data_sheets/HTG540.pdf

For a modest $5,000 and a minumum order they will even create for you an LED phosphor sheet that can be applied to any LED using the principles of remote phosphor but replicating very closely (using the right combination of Blue LEDs) the spectral output of the Coral Plus Bulb (or whatever spectral plot you give them). If you want such a bulb to achieve a CRI of near 100 they can do it.

 

[fewponds]

What is the green spike in nm?

 

[TropTrea]

While the green spike may be inportant to duplicate what we see compared to T-5's I'm more concerned about the Blue end of the spectrum for the corals benifit. What I would love to see is something like a 440nm and 485 nm chip to work in the series of brue blue (470), Royal Blue (455)

Actualy the technology exists for a true atinic/blue chip where the spectrum from 400 to 490 could be covered with a peak around 445 but the 1/2 power points being around 425 and 475. But would there be enough demand for someone to produce and sell them at a reasonable price. Yes for our Reefs we would buy them but would anyone else?

 

[zachts]

What is the green spike in nm?

Around 545nm +/- a few depending on the bulb your comparing.

While the green spike may be inportant to duplicate what we see compared to T-5's I'm more concerned about the Blue end of the spectrum for the corals benifit. What I would love to see is something like a 440nm and 485 nm chip to work in the series of brue blue (470), Royal Blue (455)

Actualy the technology exists for a true atinic/blue chip where the spectrum from 400 to 490 could be covered with a peak around 445 but the 1/2 power points being around 425 and 475. But would there be enough demand for someone to produce and sell them at a reasonable price. Yes for our Reefs we would buy them but would anyone else?

Most all of those wavelengths are available when you look at all the bins, We cant get them as DIY hobbyists due to resellers not stocking all the possible wavelength bins. Really the half widths of what we can get will overlap sufficiently to cover the blue end very well from 400nm up to 500nm.

Just look at the Pacific Sun Hyperion graph that they have posted numerous times, it has very good coverage over that range using many different chips. there is a slight dip around 490 it appears but not enough to worry about.

 

[fewponds]

A little bit different topic... Common judgment is white LEDs cause sps bleaching, but do we know which of the white's attributes (spectrum, intensity,...?) is responsible for the bleaching?

 

[TropTrea]

A little bit different topic... Common judgment is white LEDs cause sps bleaching, but do we know which of the white's attributes (spectrum, intensity,...?) is responsible for the bleaching?

There have been several articles ob bleaching from light. They usually all come back to an excessive amount of red light especially in the 680nm range.. But the ratio is something that is very controllable with your LED selection and ratio.

Warm whites produce almost twice as much Red light as the neutral Whites
Neutral Whites produce a little more than twice the red of cool whites
And Cool whites produce just a little red light.

On a bulb fixture with 24 LED's you have about the same amount of red if you use 3 warm whites, 5 Neutral Whites, or 10 cool whites, with the rest of the LED's being in the /UV family. Because of the green and blue ratio the 3 warm whites will be just about as bright as the 10 cool whites to most people eye. I personally like to use the neutral whites but would only run 4 of them in a 24 LED fixture. I do not think you will get enough red with the Cool Whites unless you start adding red LED's. hen you start adding red LED's you only want to add a little as your looking at three different LED colors to handle the entire red spectrum.

The trick is they need a little red light but to much will bleach or even kill them.

 

[penfold2]

There have been several articles ob bleaching from light. They usually all come back to an excessive amount of red light especially in the 680nm range.

Do you have links to any of the articles? The only info I could find is in regard to bleached corals having reduced fluorescence at 680nm.

 

[TropTrea]

Do you have links to any of the articles? The only info I could find is in regard to bleached corals having reduced fluorescence at 680nm.

I would have to search the web for them. But yes 680nm is considered an interesting red frequency. It has been proven to bleach some coral when in excess. It is also an off switch for some corals to shut them down to protect them from excessive light. Then to add to the spice Corals containing Carotene and or Chlorophyll D can utilize that wave length. So the secret is fining the right ratio.

You should also note that phycocyanin is excited by light at 620nm and this is the primary protein in cyano algae and seldom found in other corals.

 

[conix67]

There is no pigment which would require OVER 600 nm radiation (absorption).

Hello PSRD,

Thanks for the informative posts and sources. I enjoy reading your post a lot.

I have a question, in the chart above, many species listed sit on maximum absorption spectrum of around 580nm, how do you achieve this in the LED lighting system? Green LEDs?