LED lighting: Cluster vs. Panel

[TimmyD16]

So the title says it all. If designing a new LED light fixture should I lean towards a cluster arrangement to get more of a spotlight effect or should I arrange the LEDs in a panel to get more even coverage throughout the tank (ideally).

I would like to focus on more scientific reasons for this such as coral and system health, and less on aesthetics like "this one looks better".

Looking forward to the discussion!

 

[Cheapreef]

Spread out panel will be far better for SPS, once they grow in the cluster will cause a lot of shadowing. That's if your going with SPS.

 

[TimmyD16]

Any concern over un-even color spectrum? With LED colors where there are only a few emitters for the whole fixture, wouldn't these not fully blend with each other potentially creating hot and cold spots?

 

[Ron Reefman]

Any concern over un-even color spectrum? With LED colors where there are only a few emitters for the whole fixture, wouldn't these not fully blend with each other potentially creating hot and cold spots?

Minimal issue IMHO. In fact I shut off the red and green leds except for the 45 hour midday lighting. And even then I only run them at 10%. The other spectrum, violet, royal blue and blue have enough leds to mix with the white ones so there really aren't any 'hot spots'.

I went a step further and added 2 led strip fixtures to my main led fixture (Photon V2). I have all blue at the back of the tank and a mix of white and blue at the front of the tank. I have not seen any real difference with how the corals react at the front vs the back of the tank.

Clusters of leds can been fine as well, assuming you have a number of them spread over the tank, like 3 or 4 Radions or a Reef Breeders Aquasanrise which has multiple pucks.

 

[ca1ore]

If one were clearly better than the other, the industry would have completely migrated that way - though my unscientific observation is that more and more seem to be going the clutter route. I've used both and have no real preference for one over the other.

 

[pisanoal]

If one were clearly better than the other, the industry would have completely migrated that way - though my unscientific observation is that more and more seem to be going the clutter route. I've used both and have no real preference for one over the other.

I wonder if this is for aesthetics, as in better blending/less disco, and better shimmer? That seems to be a big focus of led manufacturers.

The main arguments against leds especially for sps being uneven coverage, my opinion is that the bulk of the industry is going towards aesthetics over functionality. Might not be true, but that is where my headspace is right now with it.

You see it more and more with other products too. Curb appeal wins over functionality for a lot of people these days...

 

[jda]

If you are designing a totally new LED fixture, then get a whole bunch of them and shine them up into a reflector from a small cluster in the center. That actually has not been done. That should get rid of shadows and color mixing. It will be likely be more power, though.

 

[chaudhry]

From physics point of view of the light beam, if you cluster them together, they start off mixing better this taking away any disco effect but adding more shimmer as they are clisely packed(kessil).
Compared to when you have them spread out. Every bulb emits their own beam and they get a chance to blend with other colors only half way, before hitting the tank water. So chance of disco effect is more. If you are spacing them apart on apanel then make sure not to cluster too many greens and reds together, but spread them out from each other, so that they can blend in better to avoid any hotspots or color focuses.
Personally i thing lights places closely and covered with wider angle lens gives the best blend. Thats what kessil does!


Sent from my iPhone using Tapatalk Pro

 

[jayball]

A panel of clusters, similar to the GHL Mitras. Each cluster is densely packed to provide a blended spectrum while the 6 clusters are spread apart and pushed to the sides/corner on the panel, not in one line.

 

[oreo57]

Can't think of any scientific reason for a cluster...

All are aesthetics or design issues..

Best would be a cluster on a moving arc across the tank simulating a normal sunrise/sunset geometry..
Sun's rays on the ocean are basically perpendicular to the surface at noonish (emulated more by panel than cluster) and the angle changes by rising and falling (emulated more by moving a well collimated cluster)

the "scientific" basis would be catering to the natural geometry of the world..
so the rest is more "cultural and aesthetics"

kind of depends on your definition of "scientific"..

 

[oreo57]

From physics point of view of the light beam, if you cluster them together, they start off mixing better this taking away any disco effect

disco effect is aesthetics.. There is no scientific reason to care..

not meaning to pick on you per se.. just a random thought..

Secondly you can solve all blending problems by "proper" phosphor contents on a rb pump (or better yet a violet one) diode..
That is a manuf design choice.

As to shading and killing branches. that will be detrimental to a coral form not function or possibly even growth

An aesthetic issue..

 

[jayball]

disco effect is aesthetics.. There is no scientific reason to care..

not meaning to pick on you per se.. just a random thought..

Secondly you can solve all blending problems by "proper" phosphor contents on a rb pump (or better yet a violet one) diode..
That is a manuf design choice.

As to shading and killing branches. that will be detrimental to a coral form not function or possibly even growth

An aesthetic issue..

"Disco" is a visible effect of uneven spectrum. In the wild a colony is receiving the same spectrum all over. While I can not say that it harms a coral it certainly is not natural. I hate to reference the BRStv videos but they have certainly shown that even with clusters like the hydra (that exhibit disco on the sand) there can be large variance from point to point in the tank. This is made worse with the panel design and made better by using tightly clustered LED's in a single lense or reflector(see kessil or GHL mitras) or diffusion like the phillips coralcare.

For your second I would love someone to come out with a RGB phosphor on a violet pump in a 14 - 16 K color temp. That would be great but it looks like current offerings in areas like high end photography and art lighting is only low powered. I would guess we are still having problems pumping the higher energy violet through phosphor coatings at high power or there is no other commercial use to develop it.

The third I am going to say I disagree. One of the biggest arguments against LED compared to MH/T-5 is how hard it is to keep the base and lower branches alive. Dying coral falls firmly in the "coral health" argument from the OP.

 

[Bronx19]

Panel is far better than cluster. I wont buy any of these single/two cluster units, just no coverage.

 

[Cheapreef]

Panel is far better than cluster. I wont buy any of these single/two cluster units, just no coverage.

Don't you run Kessil?

 

[jda]

For your second I would love someone to come out with a RGB phosphor on a violet pump in a 14 - 16 K color temp.

This is possible and it has been done, or at least close to this. They are more wasteful than other types of light - there are no electrical savings since they cannot cut spectrum to save wattage. If there was more use, then the development would probably get it to where they are the same power as incandescent or fluorescent. If these were possible, then there would be no color controllability, just dimming.

 

[Bronx19]

Don't you run Kessil?

Na, T5. Considering a Orphek for a new build now.

 

[tmc1313]

Why not both? This is my preferred method of lighting right now. Many clusters, but in panel form. No optics for a much more even spread and no disco, blending is fantastic.

 

[LQT]

Why not both? This is my preferred method of lighting right now. Many clusters, but in panel form. No optics for a much more even spread and no disco, blending is fantastic.

Nice. That looks like one of Dave's Nanobox creations. Is that yours? If so, did you use the 2 available end T5 bulb slots for T5 supplentation?

 

[tmc1313]

Nice. That looks like one of Dave’s Nanobox creations. Is that yours? If so, did you use the 2 available end T5 bulb slots for T5 supplentation?

You know your lighting!! Yes, it's one of Dave's creations he did for me. I'm using two ATI actinic T5's in the slots. I didn't want to take away from the Led's in any way. With that many clusters, it was basically designed to be LED dominant. I almost went with all led. I still might just put two led strips in place of the T5's.

 

[outy]

I went with the new ocean revives instead of mars aqua solely because they had there panels spread out with higher wattage going to each led.

I get no real disco effect and shimmer similar to my MH lighting before.

hate the shadowing but I may raise lights higher to adjust that, since most do not run at 100%, I can raise lights and get better spread and not loose any par once the light is turned up afterwards.