6500k T5 Bulbs - Important Spectrums Missing in LEDs?

[jda]

I have seen two multichip tanks in the US. Both were fine. They were not long lasting due to DIY skill, heat issues, there were all the same issues as other point-souce applications, so the owners did not use them for more than a few years. These things get HOT... every bit as hot as a MH bulb when getting your hand near (we did not use a temp gun). They can heat up a hood or room pretty well when you use 10-12 of them over a 300g heavily stocked acropora tank - chiller was still needed in the summer. These guys did try and get the same output, so they did not skim on fixtures or wattage.

I think that both of them would tell anybody that your DIY skills will have a huge impact on your long term success... which makes sense since build quality is paramount.

They both put their mercury stuff back on after they had a few failures since they were not saving any money and it is not free to replace the units. Both had strong thoughts on the performance, but they can log in and post these for themselves if they cared enough to do so.

I used one over a frag tank for a while where I have run many panels to get a feel for them. It stopped working, but it was pretty corroded when they lent it to me. PAR was nearly the same as an equal wattage MH right next to it, but my Apogee 510 only mostly captures the visible range.

 

[oreo54]

It’s a shame if this has been buried as a very quick google search l, and so many multi-chips are now available covering virtually every wavelength from 330nm right up to 1050nm and there completely customizable as well from certain manufacturers... way beyond my diy skills but really interested to learn more

WELL while typing my soapbox below jda posted and is 100% correct (well bit less, being nice..).. See below..
One correction though is modern diodes are a) now more efficient (less heat)
and b)more heat resistant on chip BUT goes to the "quality" chip thing..
He's right that.. well starting w/ visible spectrum some of those old chips were less efficient than MH's (or as he stated equal).
times change.. outputs are pusing 2x the efficiency of MH's and add in the better "delivery geometry" (less loss bouncing in and out of tubes/reflectors of a 360 degree point light source) even moreso...

I do caution people about judging 10 year old diodes against moden equivalents though some of those Chinese COB's may be the same as in the old days..
Things like Bluacro "pucks" LEDGroupbuy or Kessil multichips and a handful of other quality "assemblies" are much better than the "original" multichips.
As to heat.. well there is more than one kind of heat..
........................................................................
Here is the "catch" though.
1) Generally not the best chips on those boards which moves DIY'ers into other directions (low Lumen/watt)
2) Heat can be an issue as well as UV/violet/blue "browning" of the err lenses on board and other heating.
3) Making them look "pretty" is challenging
4) Not as cost effective as one would believe esp. since modern LEd's offer lots of "programming" like WiFi Bluetooth options..though this cost also includes the whole chain from heat sinks to drivers/power supplies.

Actually you would be surprised how easy they are to build..

Kind of comes back to my little "project "above.. Sort of a rebellion from the "kitchen sink" diode approach..
If you note the costs.. it's not cheap but uses sort of state of the art diodes..well one level of them..
Oh and it's really "daylight" centric for the most part w/ 1 "actinic" channel, 1 white..
the orig. Royal blue/10000k-ish LEd's were, for the most part successful IF one understands watts..
Prob. was trying to sell 50W as equiv to 250W MH.. BIG mistake in PR..

Probably plenty of successful tanks (from an owners view, maybe not from some purests vision) still running Royal blue/white diodes out there..

also gets back to the ORIG. thread.. What about the spectrum..
As to wavelengths..
Which are wants.. which are needs.. which make small difference.. which make large differences..
Which encourage growth.. which encourage fluorescent pigments.. which encourage regular pigments..which look the best..
ect..ect
That's almost an impossible challenge considering the number of species and 100's of different environments.

Yes no "UV" in most but "exactly" what does that do?
No IR to speak of.. same issue . "exactly" what does that mean.
One could add "when" does it matter as well........


From my limited point of view there are no "important" wavelengths missing in LED's till one defines EXACTLY what is important..
and how important...
and honestly, from what I've seen (not done) even those are a stretch..

You're dealing w/ a fairly "plastic" organism that is capable of adjusting to its light environment ..to a certain degree.
fw aquatics and terrestrials do the same..
From that point of view, light within reason, is probably the least important factor...

Defining what is "correct" is an individual choice not a mob pick..

to be honest, and I have this same discussion on fw sites, start w/ a vision of what YOU want it to look like, regardless of say how fast it grows..or which corals like it or don't.

Now that also isn't for everyone either some do want "the ultimate" of err something.. Fine but don't forget that costs in a LOT of time and patience.

This may sound like some vote for mediocrity but not really.
Just puts you ahead of the tank not the tank ahead of you.
With that.. LED, MH, t5 combo of any or all make it your choice..They all "work" AFAICT..
Just my 1 cent worth of "philosophy".....

 

[Mortie31]

WELL while typing my soapbox below jda posted and is 100% correct (well bit less, being nice..).. See below..
One correction though is modern diodes are a) now more efficient (less heat)
and b)more heat resistant on chip BUT goes to the "quality" chip thing..
He's right that.. well starting w/ visible spectrum some of those old chips were less efficient than MH's (or as he stated equal).
times change.. outputs are pusing 2x the efficiency of MH's and add in the better "delivery geometry" (less loss bouncing in and out of tubes/reflectors of a 360 degree point light source) even moreso...

I do caution people about judging 10 year old diodes against moden equivalents though some of those Chinese COB's may be the same as in the old days..
Things like Bluacro "pucks" LEDGroupbuy or Kessil multichips and a handful of other quality "assemblies" are much better than the "original" multichips.
As to heat.. well there is more than one kind of heat..
........................................................................
Here is the "catch" though.
1) Generally not the best chips on those boards which moves DIY'ers into other directions (low Lumen/watt)
2) Heat can be an issue as well as UV/violet/blue "browning" of the err lenses on board and other heating.
3) Making them look "pretty" is challenging
4) Not as cost effective as one would believe esp. since modern LEd's offer lots of "programming" like WiFi Bluetooth options..though this cost also includes the whole chain from heat sinks to drivers/power supplies.

Actually you would be surprised how easy they are to build..

Kind of comes back to my little "project "above.. Sort of a rebellion from the "kitchen sink" diode approach..
If you note the costs.. it's not cheap but uses sort of state of the art diodes..well one level of them..
Oh and it's really "daylight" centric for the most part w/ 1 "actinic" channel, 1 white..
the orig. Royal blue/10000k-ish LEd's were, for the most part successful IF one understands watts..
Prob. was trying to sell 50W as equiv to 250W MH.. BIG mistake in PR..

Probably plenty of successful tanks (from an owners view, maybe not from some purests vision) still running Royal blue/white diodes out there..

also gets back to the ORIG. thread.. What about the spectrum..
As to wavelengths..
Which are wants.. which are needs.. which make small difference.. which make large differences..
Which encourage growth.. which encourage fluorescent pigments.. which encourage regular pigments..which look the best..
ect..ect
That's almost an impossible challenge considering the number of species and 100's of different environments.

Yes no "UV" in most but "exactly" what does that do?
No IR to speak of.. same issue . "exactly" what does that mean.
One could add "when" does it matter as well........


From my limited point of view there are no "important" wavelengths missing in LED's till one defines EXACTLY what is important..
and how important...
and honestly, from what I've seen (not done) even those are a stretch..

You're dealing w/ a fairly "plastic" organism that is capable of adjusting to its light environment ..to a certain degree.
fw aquatics and terrestrials do the same..
From that point of view, light within reason, is probably the least important factor...

Defining what is "correct" is an individual choice not a mob pick..

to be honest, and I have this same discussion on fw sites, start w/ a vision of what YOU want it to look like, regardless of say how fast it grows..or which corals like it or don't.

Now that also isn't for everyone either some do want "the ultimate" of err something.. Fine but don't forget that costs in a LOT of time and patience.

This may sound like some vote for mediocrity but not really.
Just puts you ahead of the tank not the tank ahead of you.
With that.. LED, MH, t5 combo of any or all make it your choice..They all "work" AFAICT..
Just my 1 cent worth of "philosophy".....

Thanks for replying and @jda your explanations are experiences are appreciated, ive always gone for Orphek LEDs had 3 models over 9 years and they have delivered what I want..
I’d just never seen these DIY super chips and was very intrigued... Also the very wide choice of spectrums available a quick google and found this site https://www.leds-global.com/customized-50w-multichips-led-for-aquarium-p-301.html
But like you have said until we know what’s needed then is it relevant

 

[jda]

I could not get past the point source... this is why I would choose Atlantik or Photon if I had to move toward LED. Hard to get past this after coming from 20" reflectors that cover massive amounts of area very well.

The dude with the newer chips had a LiCor that captured all the MH spectrum and it put out more PPFD per watt than the MultiChip in the range that we care about (350 to 850), but the 400-700nm Apogee 510 had the Multichip in the lead. The multichips still seemed more geared for human eyes than a coral, but I do not know too much about them. They did have a bit of UV and IR. Other than the UV and IR, this is not too much different than a commercial LED which pops a PAR meter in the visible range, but not much outside of it.

In the end, if you want one of these, you will likely have to build one yourself. Nearly everybody that buys LEDs does it for the apps and sales pitch of decreased power, decreased heat and form factor and multichips do not have much of that (except for maybe form factor if you got really creative). It would be a hard sell commercially since most of their selling points don't work for these. This is just me, so do not get too upset, but I do think that most commercially available LEDs are toys to varying degrees, but these are more of a pro-line thing more in the vein of MH.

From what I understand from my friends, if you are going to mess with this, you need to consider many 50w chips, or just go with much larger ones. Factor in about 1:1 wattage from the T5 or MH that you are leaving. If you can get away with less, then you could have used less T5 of MH as well.

If I ever built one of these, I would probable use a high wattage one, but put in some true UV and IR diodes along side in a manner where they were replaceable when needed... unless I could find a 350-850 (or so) chip.

 

[Lasse]

I’ve never seen this thread before, these multichips are massive.... is this you doing this video on how to build one... great video btw...did you give up on this idea for home aquariums?

No it was not me that produce that video but he use the same methods like me. However - the original Dream Chip had 5 independent channels that could be adjusted separately . One 50 W can be still be bought at the original vendor. It has

• Channel 1: EPISTAR ES-CADBV45P 45*45mil 15000K with red phosphor coated / CRI: 84.91 x 10 chips
• Channel 2: EPISTAR ES-CADBV45P 45*45mil 455nm-457nm Royal Blue x 10 chips
• Channel 3: EPILEDS EP-U4545K-A3 45*45mil 417.5nm-420nm x 5 chips + 427.5nm-430nm Violet(UV) x 5 chips
• Channel 4: EPISTAR ES-CADBV45P 45*45mil 445nm-447nm Royal Blue x 10 chips
• Channel 5: EPISTAR ES-CADBV45P 45*45mil 10000K with red phosphor coated / CRI: 82.38 x 10 chips

This is close to the original Dream Chip´s configuration but the original was 10 rows of 10 LED - max rated at 250 W

Note that they are 45X45 mil which means that they are 50 individual 2.5 W LED and that they manage 700 mA - and it was here we did the big mistake - to run them at max amperage. If we had run them at lower amperage (down to 350 mA - as many commercial brands now do with their fixtures including singular 2,5 - 3 watts LED) - the heat would not damage them. The second mistake I did was to use the glass lens - - too hot inside.

I´m sorry @Rakie if I got off topic but I only want to show that not all people around the world fell in the PAR trap around 2011-2012 - there was other ideas. Why I still not use these LED (and especially the test chip that include red (630 and 660 nm) and also green additives (that I did already around 2014) is another story - I can come back with this in another thread if anyone are interested of this. However - I did not leave the idea because it did not work - it was of other reasons

But this is two video from nov 2014 showing my old tank with three Dream Chip 100 (one with a red channel) with different adjustments of the channels





Sincerely Lasse

 

[TexasReefer82]

A very interesting thread... And if I may jump back to the OP's title question: Do 6500K T5's contain spectral qualities lacking in LED's?

Why do you all think that so many owners of Radion and AI LED modules nearly always run the White LED channel so low, if at all? Perhaps the 6500K T5's don't contain additional spectrum but instead virtually no one is using the 6500K channel on their LED modules.

Just a thought - if the manufacturers of these lights didn't intend for a heavier white spectrum why are there 3x white LED's per puck in the AI's??? If they intended for users to turn the white channel down to 5% why not just use 1x white LED and then include a couple additional colors (like amber, cyan, etc)?

Someone stated previously that the grow-out facilities using Radions operate all channels at 100% and have the fixtures mounted high off the water for good spread. This makes perfect sense - it's probably very close in spectrum to using a 6500K Iwasaki with actinic VHO or T5 bulbs to supplement - which is a proven winner for growing corals.

Maybe using all channels at 100% (or at least all channels at the same setting) is the OBVIOUS way to use the LED fixtures during the main part of the day and then scale dusk and dawn with blue channels only.

 

[oreo54]

A very interesting thread... And if I may jump back to the OP's title question: Do 6500K T5's contain spectral qualities lacking in LED's?

Why do you all think that so many owners of Radion and AI LED modules nearly always run the White LED channel so low, if at all? Perhaps the 6500K T5's don't contain additional spectrum but instead virtually no one is using the 6500K channel on their LED modules.

Just a thought - if the manufacturers of these lights didn't intend for a heavier white spectrum why are there 3x white LED's per puck in the AI's??? If they intended for users to turn the white channel down to 5% why not just use 1x white LED and then include a couple additional colors (like amber, cyan, etc)?

Someone stated previously that the grow-out facilities using Radions operate all channels at 100% and have the fixtures mounted high off the water for good spread. This makes perfect sense - it's probably very close in spectrum to using a 6500K Iwasaki with actinic VHO or T5 bulbs to supplement - which is a proven winner for growing corals.

Maybe using all channels at 100% (or at least all channels at the same setting) is the OBVIOUS way to use the LED fixtures during the main part of the day and then scale dusk and dawn with blue channels only.

Besides people liking "blue" much comes from this:

https://ecotechmarine.com/wp-content/uploads/2016/07/ReefWholesale_CoralLab.pdf

Prior to using Radion® G3 Pros RW (Reef
Wholesale) determined that the best coloration
and growth in their SPS corals were achieved
using T5 bulbs. Specifically *ATI Blue Plus and *ATI
AquaBlue Special in a combination of 80%/20%
and specifically in *ATI Powermodules. In the
same time period given the same parameters and
frags from the same mother colony, this T5 bulb
mix generally outperformed (for most, but not all
corals) their original lighting staple, metal halide.

 

[TexasReefer82]

Thanks... I've read that article before - it's very interesting.

"the best coloration and growth in their SPS corals were achieved using T5 bulbs. Specifically *ATI Blue Plus and *ATI AquaBlue Special in a combination of 80%/20%"

I don't doubt them, but I do wonder what were all the lighting sources they tried before coming to this conclusion?

They also state that this combination wasn't the best for ALL SPS corals.

They state "the best coloration AND growth"... does that mean the best compromise between the two? Or does that truly mean the best of both?

This much is clear: Corals are HIGHLY adaptable! haha

 

[blasterman]

Spectrums for all the light sources mentioned here are available via this thing called google, including a 6500k T5, although the specific spectrum for that T5 will vary depending on the specific phosphor set and desired CRI. The higher the CRI the more energy is expended on a few more spectral spikes that are eyes aren't as sensitive to so efficiency goes down. That's why high CRI fluorescent wasn't a defacto standard in industrial lighting - it wasted energy. Not everybody wants to pay for GE 950 phosphor sets in a 50K square foot warehouse.

If you look at the Spectrum of a 6500k T5 or Halide you see big, harsh spikes and then a noise floor of trivial amounts of energy. A 6500k LED by comparison has much smoother and longer spectral curves with most higher tech LED's having extended red beyond 630nm. Fluorescent does have some energy in the true UV range as it's base emission point but this is inconsequential to corals. Fluorescent tubes also cannot emit near the continuous spectrum than an LED can. The commercial lighting industry settled on this fact a few decades ago. We now have some LED makers adding discrete 470-480nm LEDs and getting a light source approaching plasma sulfur for color demanding applications.

There is otherwise no such thing as a full spectrum reef light other than the half dozen people on the planet that played with plasma sulfur a few years back. When these reef light makers advertise 'Full spectrum' with their LEDs because they toss extra LED colors in there it's not backed up on their spectral charts. Do explain how something can be called "full" when you have 50watts of radiating 450nm blue and half a watt of turquoise?

If you want a practical way to prove this for yourself without investing a lot of money just get a 4 or 5 color DJ par from Amazon for $40, hold it above your reef tank, and adjust the colors to taste. You'll find you're cranking up the blues to max, and just using a tiny bit of the rest of colors and leaving green off.

The real color that gets neglected with current LED reef lighting is 465-470nm blue. This wavelength pops oranges and greens like a boss, cancels out the annoying Barney purple that some LED lights have, and makes your tank look visibly brighter because that wavelength of blue is more sensitive to our eyes than 450nm.

 

[Mortie31]

Besides people liking "blue" much comes from this:

https://ecotechmarine.com/wp-content/uploads/2016/07/ReefWholesale_CoralLab.pdf

Honestly, this article has driven the use of predominantly blue lighting? I’ve not read it before but shows virtually nothing conclusive, and the comparisons are terribly poor.. glad I missed it and rn my LEDs with plenty of white...

 

[oreo54]

If you look at the Spectrum of a 6500k T5 or Halide you see big, harsh spikes and then a noise floor of trivial amounts of energy. A 6500k LED by comparison has much smoother and longer spectral curves with most higher tech LED's having extended red beyond 630nm. Fluorescent does have some energy in the true UV range as it's base emission point but this is inconsequential to corals. Fluorescent tubes also cannot emit near the continuous spectrum than an LED can. The commercial lighting industry settled on this fact a few decades ago. We now have some LED makers adding discrete 470-480nm LEDs and getting a light source approaching plasma sulfur for color demanding applications.

There is otherwise no such thing as a full spectrum reef light other than the half dozen people on the planet that played with plasma sulfur a few years back. When these reef light makers advertise 'Full spectrum' with their LEDs because they toss extra LED colors in there it's not backed up on their spectral charts. Do explain how something can be called "full" when you have 50watts of radiating 450nm blue and half a watt of turquoise?

If you want a practical way to prove this for yourself without investing a lot of money just get a 4 or 5 color DJ par from Amazon for $40, hold it above your reef tank, and adjust the colors to taste. You'll find you're cranking up the blues to max, and just using a tiny bit of the rest of colors and leaving green off.

The real color that gets neglected with current LED reef lighting is 465-470nm blue. This wavelength pops oranges and greens like a boss, cancels out the annoying Barney purple that some LED lights have, and makes your tank look visibly brighter because that wavelength of blue is more sensitive to our eyes than 450nm.

There are outliers to the above including many violet pump emitters w/ RGB phosphor LED's.
They have their own set of problems though..

As to current high CRI whites w/ blue/cyan major player is Luxeon..
"Fresh fish" spectrum.. 95CRI-ish
still need to supplement the blue/cyan range if you want "aqua" tone over all
Advantage to a broad blue phosphor as in tubes..

Their hort. white is very similar..but comes lensed very wide..

CREE has high CRI yet still mostly deficient in blue/cyan. Many ways to get high CRI...

"Fresh Fish" COB..

 

[oreo54]

Honestly, this article has driven the use of predominantly blue lighting? I’ve not read it before but shows virtually nothing conclusive, and the comparisons are terribly poor.. glad I missed it and rn my LEDs with plenty of white...

Well more like "supporting" info..
At least it promoted "some" color.

The Royal Blue /10000k white fixtures sort of started it..
Of course have very little of anything but blue and yellow/green

 

[shred5]

Remember that there is more to coral color than just feeding zoox.

I think this is the most important statement in this thread.
People seem to always focus on chlorophyll.
People tend to forget these are animal and not plants.
They can also alter light or block light.
Parts of the spectrum of light that is bad for SPS can help with coloration because some of that coloration comes from the sunscreens which are meant to block or alter the light they do not like.

 

[oreo54]

I think this is the most important statement in this thread.
People seem to always focus on chlorophyll.
People tend to forget these are animal and not plants.
They can also alter light or block light.
Parts of the spectrum of light that is bad for SPS can help with coloration because some of that coloration comes from the sunscreens which are meant to block or alter the light they do not like.

On that note one also has err.. "skeleton effects"...

Recent studies on the optical properties of coral tissue and skeleton have revealed that vertical light attenuation is predominately in the blue region, thus leaving more red light available for symbionts in deeper tissue layers [21]. Corals also harbor various types of fluorescent and chromophoric host pigments that absorb energy-rich UV, blue and blue-green light and cause red shifted reemission

Point is not only blockage but re-emission at a lower wavelength ..still usable..
like deep water corals providing red light by downshifting the blue/green light..

 

[jda]

Why do you all think that so many owners of Radion and AI LED modules nearly always run the White LED channel so low, if at all? Perhaps the 6500K T5's don't contain additional spectrum but instead virtually no one is using the 6500K channel on their LED modules.

This is only really true in the US and mostly with noobies and people unable to get good color using a whiter spectrum for various reasons - most people tend to get less blue as they stay longer in the hobby (not all, but a lot). In Europe, some vendors and very experienced LED keepers do run a large amount of whiter spectrum for large parts of the day. Most of the all-blue has come from the Manufacturer, Trade Show or LFS tanks that do it to sell coral and hard products.

There were some 6500k/white diodes in older panels that had some harsh quality that burnt coral, but these are likely long gone and I have not seen this in too many new panels - we are talking like 6-8 years ago. I used to really recommend that people keep their whites down low, but this does not seem like much of a problem lately.

A 6500k LED has the same harsh spikes and stuff. Most graphs are smoothed out to help with marketing and sales and sometime to just make things simper. If you put them on a raw spectrometer, 6500k LEDs have their spikes and valleys too... just like some MH and T5s charts have been rounded and smoothed out. Every lighting type has their raw and their rounded charts out there.

 

[oreo54]

A 6500k LED has the same harsh spikes and stuff. Most graphs are smoothed out to help with marketing and sales and sometime to just make things simper. If you put them on a raw spectrometer, 6500k LEDs have their spikes and valleys too... just like some MH and T5s charts have been rounded and smoothed out. Every lighting type has their raw and their rounded charts out there.

I need proof of the LEd stuff. And there are not anywhere near the spikes as in tubes and MH's..

MORE than happy to admit it if I'm wrong.
BTW: I do know LED spectrums are smoothed. It's the degree you are implying I question..

New LEDs arrived! | Details | Hackaday.io

<p>With new 490nm and 470nm LEDs I can finally proceed: </p><p><img data-src="https://cdn.hackaday.io/images/966141439845279607.jpg" class="lazy"></p><p>470nm LED came close to specs:</p><p><img data-src="https://cdn.hackaday.io/images/6420701439846630858.png" class="lazy"></p><p><img...

hackaday.io

Ceramic mh vs led..

 

[Ashley Kekua]

I have notice coral puffmore when I add the 6500k the response immediately happen. I switch back and forth and always see the response so I keep the 6500k

 

[ingchr1]

I've read through the thread and was wondering what the consensus is (if there is one) on white LEDs and coral health?

I have a GHL Mitras LX7 that has 6500K and 7500K white LEDs. Based on all my reading I currently have it setup as shown below.

Is this beneficial, detrimental, can it be improved?

Can that question even be answered? Do I just wait and see?

The power setting gives me 120-130PAR center bottom of the tank throughout the day. I have the light 9" off the water surface, any higher there will be too much light spill into the room.

2 Hours a Day:

6 Hours a Day

 

[Lasse]

I use 2 pcs 120 deep actinic Mitras 2 Bars, 1 pc 120 actinic Mitras 2 bar and 1 pc 110 mitras Ocean Blue Mitras 2 bar. total input effect - around 300 W, I use a seasonal program like this

​

My miday is between 13:00 - 19:00 with noon 16:00. Here is the spectra for that time - the intensity varies between 90 % (13:00 and 19:00) to 100 % (16:00

​

425 nm follow this path

​

450 nm, 455nm, 472 nm, 530 nm and 622 nm follow this path

​

And my whites 4400, 6500 and 7700 K

​

The result - you can see in my build thread

Sincerely Lasse

 

[ingchr1]

@Lasse, thank you.

Your midday spectrum is kinda of similar to mine, with differences due to different lights.

I start with the midday for two hours then ramp to the other spectrum for the six hours.

What spectrum are you running on each side of your midday?

I'm keeping the same power for each spectrum, so my ramps look like this.

425nm, 455nm, 475nm and Blue White

385nm

6500K

7500K

528nm

656nm