RB/NW vs. full spectrum

[uglybuckling]

I've been thinking about / designing a side-by-side to test a royal blue and neutral white LED array vs. a full spectrum LED array. There are a number of concerns that need to be addressed here, since the idea would be to minimize interference from other variables.

 

First of all, I'm thinking two identical tanks connected to a common sump. One might ask, "why not just do a single tank and divide it in half with a divider?" Well, the way I see it, the overflows are going to be kinda rough on a system like that; the water might flow from the sump to one side of the tank and then to the other side (through the divider); if the two overflows are not 100% even in terms of height. This would be suboptimal because badness or goodness could be delivered to / rinsed into one or the other side of the tank in an unequal fashion. Obviously an impermeable divider would work too, but the issue there is that the tank will then no longer be usable for anything else (giant silicone seam down the middle; my skill at applying silicone is about that of a five year old), plus if I'm using a cheap tank (10 gallons) each side will have a fairly small volume (5 gallons).

 

So, two ten gallon tanks connected to a 20 gallon common sump, with a forked return pump that gives equal flow in both tanks. Shared water (common sump) so no differences there. Since I own one 10g tank and one 20g tank already (which are not in use), total cost so far is $10 for another 10g tank. Plus two bulkheads (another $10), and some PVC (probably already have it sitting around). I would need a return pump but given that the one I'm using on my main display tank is almost 10 years old, I'm about due for a replacement anyway. So, no biggie on the plumbing.

 

Toss in powerheads for flow; of course the same brand and model, pointed in the same direction. I have a couple of Maxi-Jet 900s that I'm going to use in an upcoming build, but I don't mind this experiment getting in the way of that build.

 

The next issue is the scape (or lack thereof) in the tanks. Now, I'm all about keeping it even, so bare-bottom would probably be ideal (since you can't buy two pieces of live rock that are the same). However, that does not exactly simulate well the practical usage of these lights. It also leaves little space for biological filtration (the fuge/sump?).

 

Stocking would probably be light because of the above lack of filtration; I would like to have at least one fish in each tank though--again, probably as even as I can, maybe clownfish since I can get some captive-bred little ones for $5 each and I will be able to easily just plop them in the big reef afterward. As far as corals go (the real focus of this experiment), ideal would be a nice mix of softies, LPS, and SPS. In a system this size, though, and particularly since I'll be setting it up on a fairly temporary basis, I'm hesitant to try to grow SPS. The issue is, you can arguably grow softies and LPS with Wal-Mart lights, so the focus here should be SPS--but I'm just not really set up to do the SPS test right. Maybe birdsnests or something.

 

As for the lights themselves, there's also the problem of how to balance them as much as possible (since they are the experimental variable, it's okay for them to differ, but they should differ ONLY in meaningful ways, ways we want to test). My thinking is that their output power in lumens should be equal, even though this may mean a larger number of emitters on one or the other array. Balancing for PAR seems like gaming the system a bit, since PAR is very spectrum-dependent and the whole point here is that the spectra are different, and balancing for cost or for number of emitters seems overly simplistic.

 

Also, should I use cool whites instead of neutral whites on the blue/white array since (1) that seems to be common in practice still and (2) it will maximize differences in outcome variables between the two arrays, or should I instead assume that people actually pay attention to the advice given on this and other forums and use neutral whites in the blue/white array because that's what we'd LIKE manufacturers and DIYers to do?

 

With that in mind, I'm looking for input. Input on plumbing. Input on stocking. Input on any other aspect of this system that is worth thinking about in the planning stages. How to minimize interference from other variables. How many royal blues and how many neutral whites; how many of each LED to use in the full-spectrum world. I can do layouts and I know what will grow corals, but when you start getting into total luminous output of an array, I'm about clueless.

 

Finally, would it be easier to just buy or borrow a couple of PAR38 bulbs, one LEDTRiC and one blue/white one? If so, how do I balance them?

[Lingwendil]

Very interesting idea, if it wqs me i would do a full spec, a standard neutral/royal blue, and a cw/rb.I think putting a scratched up 33 long, or similar tank with black acrylic divider between sections. Individual return pumps sounds good to me, but a split return would work as well. Are you doing a dimmer for sunrise/sunset

[SDartifex]

I like the idea, and I feel like you got your plumbing and balancing of other factors down. I'd just like to know why you don't feel your equipped for SPS?

 

As for the lighting I think if you got two dimmable PAR fixtures you'd be set, but I'm not sure how many commercially available dimmable PAR "full spectrum" bulbs you'd find. If you end up going this route it is possible (and fairly easy) to reflow solder a couple dimmable Feit PAR38's with most chips used in our applications. Then you just need to find someone to borrow a par meter from.

[Veng]

I would say put your LR in the sump.

 

Same number of NW & RB on both fixtures, just add the full spectrum to the second since that's what people seem to actually do.

 

NW over CW because CW sucks.

 

Bare bottom is fine. Maybe an up flow algee scrubber in the sump so you don't have to worry about over feeding and algee too much.

 

Clownfish on both sides sounds good.

 

For coral stocking, cheap zoas, mushrooms, birdsnest, GSP, and a frogspawn on each side. All are high growth rate corals so you won't have to wait 2 years for results. Also a nice mix of soft, LPS, and SPS. Most of the above you can buy something and frag it into two equal parts for even more control.

[MeepNand]

Sounds good! Hope you can get back to us in a few months

[jedimasterben]

For biofiltration, Seachem Matrix and deNitrate, preferably in a reactor, will work much better than live rock for a biofilter.

[kgoldy]

Isn't your wifey coming up to my neck of the woods soon? I could donate some pink-n-gold paly's that grow like the damn plague for me.

[m28cruiser]

Sounds interesting. I m looking forward to the results.

 

+1 for NW instead of CW

+1 for mix reef

 

If you/your wife are stopping by NJ I can frag some my corals for the experiment.

[BulkRate]

Looks like a very useful experiment, given the sheer volume of lighting spectrum threads on this and many other reef forums. I'm not aware of any that involve an empirical "all other things being equal" tank build that also normalizes water chemistry between the compared setups, which is odd considering just how much money is spent on reef tank lighting in general.

 

You'd think this had been authoritatively answered already by one of the big names in LED lighting.

 

Why not do your test with an identical number of emitters on both rigs? Most DIYers tend to buy more LEDs than needed for their build (until you or Jedimasterben set them straight), and most fixture manufacturers tend use the same number of them to control costs & build complexity when offering custom emitter choices on a fixture. Determining exactly which give the most bang for the buck would be a nice side conclusion to go along with your "better for growth/coloration" testing.

 

Can't wait to follow along on this!

[Fnard]

Sounds like a good idea, but you may be making it a little too difficult. You mention having a single tank with an impermeable divider, but why would that be necessary? Forget all the plumbing and use a single 20g and no sump with a divider say of egg crate or none at all. IMO, by using a single sump/fuge for 2 tanks, that is the same as a common tank. Also, you can then use LR, possible shelf rock flat on bottom, in the tank and all corals would be subjected to the same environment. The only thing you would need to do, is make sure the lights are shielded well so that neither side has spill over. Interested in the results either way you go with this.

[uglybuckling]

Very interesting idea, if it wqs me i would do a full spec, a standard neutral/royal blue, and a cw/rb.I think putting a scratched up 33 long, or similar tank with black acrylic divider between sections. Individual return pumps sounds good to me, but a split return would work as well. Are you doing a dimmer for sunrise/sunset

 

I'll probably wind up going with NW/RB and full spectrum. Including CW/RB effectively multiplies the cost of the experiment by 3/2 PLUS when I get done I have to throw the CW emitters in the trash or sell them to some sap (which I'd have moral issues with). Since I'm not made of money, a plan involving emitters that I can re-use sounds optimal. I was planning a single forked (split) return pump. No dimmer; AA and a few other sources have quite convincingly demonstrated that sunrise/sunset do nothing for coral growth or coloration. It's just for us silly humans.

 

As for dividing a single "test" tank rather than using two, I really want to minimize any flow from one test group to the other (since arguments could then be made about water chemistry not being truly equal; stuff about allelopathy and the like). An impermeable divider would work fine to minimize that but then I have to build two overflows. Plus two 10g tanks cost a total of $20. Around here, 33g long tanks might as well be made of gold.

 

 

I like the idea, and I feel like you got your plumbing and balancing of other factors down. I'd just like to know why you don't feel your equipped for SPS?

 

As for the lighting I think if you got two dimmable PAR fixtures you'd be set, but I'm not sure how many commercially available dimmable PAR "full spectrum" bulbs you'd find. If you end up going this route it is possible (and fairly easy) to reflow solder a couple dimmable Feit PAR38's with most chips used in our applications. Then you just need to find someone to borrow a par meter from.

 

Yeah, dimmable would be an issue with PAR38s; and if they're not dimmable I'd have trouble balancing the amounts of light. I'm leaning toward a custom build; every time I think about doing anything with PAR38s I get annoyed at the limits of what you can do with them.

 

As for SPS...I dunno, I suppose I could throw some easy stuff in there. My main reefs are z/p and ric heavy; I'm not exactly into what most people would consider "challenging" corals. I just like them to look good. Preferably as good as possible, hence this experiment.

 

 

I would say put your LR in the sump.

 

Same number of NW & RB on both fixtures, just add the full spectrum to the second since that's what people seem to actually do.

 

NW over CW because CW sucks.

 

Bare bottom is fine. Maybe an up flow algee scrubber in the sump so you don't have to worry about over feeding and algee too much.

 

Clownfish on both sides sounds good.

 

For coral stocking, cheap zoas, mushrooms, birdsnest, GSP, and a frogspawn on each side. All are high growth rate corals so you won't have to wait 2 years for results. Also a nice mix of soft, LPS, and SPS. Most of the above you can buy something and frag it into two equal parts for even more control.

 

Yep. I think most of these details are the way I'm leaning. However, adding on the full spectrum emitters above the simple RB/NW rig introduces an extra variable--intensity of light. Some folks will then try to say "well, the ones with full spectrum light only grew better / showed better color because they had more light available; not because of that light's spectrum."

 

This is why I wanted to balance for actual intensity (I suppose using the rated outputs on the emitters; downside being conversion between mW and lumens for those emitters outside the spectral window where lumens are easy to measure).

 

Sounds good! Hope you can get back to us in a few months

 

Might be more than a few--cost and time constraints + the fact that the system actually needs time to get itself set up properly; testing in a non-established tank will just confound results with bad water quality. Admittedly given proper control over all variables, water quality that is equally crappy across the board, but nonetheless--I'd like to demonstrate the kind of ridiculous growth rates I'm getting in my main reef.

 

Isn't your wifey coming up to my neck of the woods soon? I could donate some pink-n-gold paly's that grow like the damn plague for me.

 

Yes she is. This tank may not be ready by that time but I still want some of your pink and gold weeds. She's obsessed with nudibranchs as well. And Saturday is my 30th birthday. I'm just listing facts here. =)

 

Sounds interesting. I m looking forward to the results.

 

+1 for NW instead of CW

+1 for mix reef

 

If you/your wife are stopping by NJ I can frag some my corals for the experiment.

 

I think her company has an office in something that is pronounced like 'wee hawk in' (I have no idea how to spell it) but that's not where she's going this trip.

 

Sounds like a good idea, but you may be making it a little too difficult. You mention having a single tank with an impermeable divider, but why would that be necessary? Forget all the plumbing and use a single 20g and no sump with a divider say of egg crate or none at all. IMO, by using a single sump/fuge for 2 tanks, that is the same as a common tank. Also, you can then use LR, possible shelf rock flat on bottom, in the tank and all corals would be subjected to the same environment. The only thing you would need to do, is make sure the lights are shielded well so that neither side has spill over. Interested in the results either way you go with this.

 

The idea of using a single sump/fuge for two tanks is indeed (as you suggested) to make it equivalent to a common tank in terms of water parameters (in other words, to equalize water parameters between the two groups). Unfortunately LEDs without optics have 120 degree light cones and any surface motion will make light scatter pretty wild. As such, some kind of isolation is needed not only around the lights themselves but also extending downward all the way to the bottom of the tank, in order to prevent light spillover from the RB/NW side to the full spectrum side and vice versa. Egg crate won't get that done; the barrier needs to be at least light-impermeable. With regard to making it water impermeable, it seems overly complex to try to divide a tank; hence using two separate ten-gallon tanks for the test groups plus the common 20g sump.

 

 

 

Thanks all for the responses.

 

I'll sit down after while and give myself a headache trying to figure out numbers of emitters and layouts.

[Paleoreef103]

As someone who designs and executes experiments for a living, it seems sound enough for what we're doing. It seems like you have your controls well isolated, but I think you can do some more isolation of your experimental variable by adding dimming to that side of the tank. Ideally, you'll want roughly equal amount of light for both sides and that's best done with a PAR meter. Set your NW/RB side at full and then dim your full spectrum side until you have equal amounts of light. The full spectrum side should probably have about five to ten percent less PAR than the NW/RB only side to help compensate for the PAR meter's bias against low wavelength light. Also, make sure you have equal placement of colonies and use the same genetic stock on both sides. I'm excited to see your results!

[Jooce01]

As someone who designs and executes experiments for a living, it seems sound enough for what we're doing. It seems like you have your controls well isolated, but I think you can do some more isolation of your experimental variable by adding dimming to that side of the tank. Ideally, you'll want roughly equal amount of light for both sides and that's best done with a PAR meter. Set your NW/RB side at full and then dim your full spectrum side until you have equal amounts of light. The full spectrum side should probably have about five to ten percent less PAR than the NW/RB only side to help compensate for the PAR meter's bias against low wavelength light. Also, make sure you have equal placement of colonies and use the same genetic stock on both sides. I'm excited to see your results!

 

 

This.

 

Plus I would do no livestock in the tank. There are no two individuals with the same personality. You could have one clownfish that likes to rub up against your coral, stressing it and reducing growth. If you're trying to make this as even as possible things like livestock are never truely equal amongst individuals, and even if none of your fish did stress the corals, its just another reason for your little experiment to be torn apart by doubters.

[uglybuckling]

Thanks for the review. I too design experiments, although these days I'm more clinical than research. The point about fish is a really good one--and ostensibly the same would go for motile inverts. Nonetheless, I'm going to need SOME kind of clean-up crew. I agree though, clownfish probably are not the best idea. Snails tend to have less personality than fish and are therefore, I think, closer to "controlled" as far as being a variable.

 

I still think balancing for PAR is gaming the system since the whole point here is spectrum and PAR is very heavily influenced by spectrum. A good bit of the reason for using the LED wavelengths that we do is that they theoretically provide more PAR than lights that don't use those wavelengths (like RB/NW lights).

 

Hence I agree with your point that the full-spectrum side should have a different PAR rating than the RB/NW side; that being said (1) I don't own a PAR meter and I don't know anybody local from whom to borrow one, and (2) I'm not entirely convinced which side should have lower PAR--the inclusion of red and cyan will boost PAR readings significantly on the "full spectrum" side above what the light's brightness would suggest, but as you pointed out, true violets are not "seen" well by PAR meters and so these will have an opposing effect on the "full spectrum" side.

 

I still think mW/lumens are the answer.

[Paleoreef103]

Thanks for the review. I too design experiments, although these days I'm more clinical than research. The point about fish is a really good one--and ostensibly the same would go for motile inverts. Nonetheless, I'm going to need SOME kind of clean-up crew. I agree though, clownfish probably are not the best idea. Snails tend to have less personality than fish and are therefore, I think, closer to "controlled" as far as being a variable.

 

I still think balancing for PAR is gaming the system since the whole point here is spectrum and PAR is very heavily influenced by spectrum. A good bit of the reason for using the LED wavelengths that we do is that they theoretically provide more PAR than lights that don't use those wavelengths (like RB/NW lights).

 

Hence I agree with your point that the full-spectrum side should have a different PAR rating than the RB/NW side; that being said (1) I don't own a PAR meter and I don't know anybody local from whom to borrow one, and (2) I'm not entirely convinced which side should have lower PAR--the inclusion of red and cyan will boost PAR readings significantly on the "full spectrum" side above what the light's brightness would suggest, but as you pointed out, true violets are not "seen" well by PAR meters and so these will have an opposing effect on the "full spectrum" side.

 

I still think mW/lumens are the answer.

Groovy. Maybe you should check out a small wrasse for the fish in the tanks. Parasite removal is always nice, but I'm going to assume you don't want to end up with two yellow coris wrasses. Still, something to think about.

[BulkRate]

<must fight...physics background...>

 

Determine if you want your experiment to be theoretical or applied. Either can advance the state of the hobby's knowledge. But more people appear to be asking "what" now "how" questions regarding LEDs in reefing these days.

 

Also since by your own admission you're not made of money and time's also a limited thing...focus on definitively answering the basic question you pose in one experiment. Just looking over your contributions in the Lighting section of NR alone it looks like you've got a damn good idea of how to put together an LED array (as do several other notable members). But there's no definitive answer on the whole "given X emitters of Y spectrum, what works best". Not just what looks the best...you could put 10 people in a room with a tank and get 14 answers, each one being "right" to the one venturing it. But to know whether 11-12 emitters set up as a 2:3 NW:RB versus a FS setup yields the best growth/coloration over some term of time would not only be relevant to the nano-sized nano tank owners but also scale up to other common capacities.

 

+1 on choosing veeeery passive critters unlikely to interact with the corals other than by contributing ,erm leavings to ensure there's sufficient dissolved nutrients in the tank to allow growth to happen. Clowns may not be the best choice as they may not stick to being a control reliably.

[Jooce01]

<must fight...physics background...>

 

Determine if you want your experiment to be theoretical or applied. Either can advance the state of the hobby's knowledge. But more people appear to be asking "what" now "how" questions regarding LEDs in reefing these days.

 

Also since by your own admission you're not made of money and time's also a limited thing...focus on definitively answering the basic question you pose in one experiment. Just looking over your contributions in the Lighting section of NR alone it looks like you've got a damn good idea of how to put together an LED array (as do several other notable members). But there's no definitive answer on the whole "given X emitters of Y spectrum, what works best". Not just what looks the best...you could put 10 people in a room with a tank and get 14 answers, each one being "right" to the one venturing it. But to know whether 11-12 emitters set up as a 2:3 NW:RB versus a FS setup yields the best growth/coloration over some term of time would not only be relevant to the nano-sized nano tank owners but also scale up to other common capacities.

 

+1 on choosing veeeery passive critters unlikely to interact with the corals other than by contributing ,erm leavings to ensure there's sufficient dissolved nutrients in the tank to allow growth to happen. Clowns may not be the best choice as they may not stick to being a control reliably.

 

+1 to everything.

 

I know you're in the planning phase, and I'm sure you would before you started, but you should name your actual hypothesis. Try to follow the scientific method, you've made your observation, next is form a hypothesis, do your experiment and (you said you have experience performing experiments, so I know you've got this one) document evvvvvveryyyythingggg. Make it as reproducible as possible, so that anyone who tries to claim your results are inaccurate has no excuse for not attempting to reproduce on their own.

 

There are a lot of fan boys for every type of lighting in this hobby, and you'll have people lined up to offer scrutiny when you actually try to show that full spectrum lighting provides better growth/color than regular blue/white.

[BulkRate]

Also as an aside... be prepared emotionally to accept either outcome. An actual experiment serves to produce a quantifiable result, not just validate a personal position. Refusal to be proven EITHER right OR wrong is the #1 way to invalidate your findings.

 

I suspect the full-spectrum rig built with properly chosen, quality emitters across the desired wavelengths will prove the most effective in fostering both growth and esthetics... but that's based on a lot of reading and not a lot of doing as of yet on my part. It could just as easily turn out that the findings are more qualified... that RB:NW arrays foster faster growth of the stocked corals but with a pronounced fading of the pigmentation that makes particular specimens desirable. Or that the whole full-spectrum approach is based on hooey and it's simply that the larger arrays happen to have enough blue and white emitters to do an OK job as the base lighting and its all just to our eyes that the corals develop better color.

 

Unless you're prepared to set up a fair trial, observe & record impartially and share your findings without bias, turn back now and embrace fanboidom. Nothing inherently wrong with embracing a personal bias on an online/public forum, so long as you're not representing the outcome as "science".

[jedimasterben]

For fish - mollies.

 

 

 

/thread

[uglybuckling]

Most of the reason to do this is because I have run across fan boys for both sides and I don't know who to believe. Believe it or not, despite being able to form occasionally coherent recommendations about how to make a full-spectrum array, I actually don't have much in the way of personal attachment to the full-spectrum thinking. I strongly suspect that parts of it may be completely bunk--specifically, since heavy stimulation of zooxanthellae by simply increasing output on the light causes coral bleaching, I see no reason why heavy stimulation of zooxanthellae by finding a spectrum that hits all the activity peaks for zooxanthellae's photosynthetic pigments would not have a similar effect.

 

I currently have two lights that I've built over (my own) functional tanks (plus a few others over friends'): of mine, one is RB/WW/CB/TV (no ocws; also the dose of TV is essentially holistic, making this pretty close to a typical blue-n-white), and the other is what would pass for "full spectrum" (specifically, two of figure 8 on my full-spectrum layouts thread). The two grow corals pretty equally. Aesthetically I prefer the color rendering and fluorescence on the full-spectrum rig but as you pointed out, that's simply personal taste and is (1) not quantifiable and (2) entirely subjective.

 

As far as hypotheses, the null hypothesis here would be that there will be no significant difference in growth or coloration. For statistical purposes that would be the one to stick with; however, there's no easy way to quantify the findings here, so statistics may be kinda tough. Nonetheless, I feel that "no significant difference" would make an adequate hypothesis. It is in keeping with the current state of research regarding coral growth under these two types of lights, which essentially is "no evidence in favor of OR against full-spectrum lighting." (the problem is lack of testing, not a demonstrated lack of difference, but nonetheless if you discount anecdotal stuff there really is no evidence either way).

[Veng]

I'm not sure how much the OCW in my current fixture is bringing to the table TBH in terms of coloration, I know the TV adds an insane amount of flourecene (which is not exactly the same as coloration).

[GHill762]

you should then pit the "winner" of this experiment against MH, and that "winner" against t5ho..

[BulkRate]

Not meaning to preach at 'ya. Really. It just gets my goat when people weigh in on a subject and then proceed to ignore any facts and information save that which exclusively supports their viewpoint. So largely I spend a lot of time these days sans goat.

 

I appreciate all the time and effort you've put into threads like these as well as your youtube clips. It's very informative and more interesting than I would have given the subject of lighting credit for.

 

I happen to have a very odd perception of color... so getting my own tank lit has been an exercise in finding a balance between what the corals need, and what I find appealing. Neither's gotten exactly what they want up to this point, but the BA True Specrum PAR38 seems to be the closest compromise to date. A little bit on the smurf-blue side, but I'm giving myself a chance to come around, given how vibrant all the existing coral in the tank has become over the past few days.

 

Wife is amused and surprisingly tolerant.

[jedimasterben]

you should then pit the "winner" of this experiment against MH, and that "winner" against t5ho..

blaster already did that. hands down his 'evil cluster' spanked a 14k phoenix, i believe was the bulb it was against. better color and more pop.

 

i would liken my build's 'look' to a 20K Radium, but for pop and color, it handily beats one.

[uglybuckling]

you should then pit the "winner" of this experiment against MH, and that "winner" against t5ho..

 

Why does T5ho get a bye to the third round???

 

=)

 

EDIT: also, what Jedi said. LEDs will ALWAYS beat halides for looks because they are adjustable. For growth might be a different question, but (I think?) that experiment was run by blaster (I haven't seen the thread). My blue/white rig is based off his, and there is no comparison between it and my old phoenix halide.

 

T5ho is a different question, because of the distributed light it gives off. Sadly again I am not made of money and I don't own a T5ho rig. If I buy one for my plant tank maybe I can invest the $80 or so for bulbs in order to run the experiment, but I'm not sinking $300 for a fixture PLUS bulbs; I'm a resident, not a private practice attending. =)