Odd scrubber... lighting questions

[jedimasterben]

Hey guys. I made my new sump to have a combination of a skimmer, algae scrubber, and refugium (may or may not stay lit, depending on whether or not macroalgaes survive lol).

What do you think of this:

http://i5.photobucket.com/albums/y189/jedimasterben/Aminals/Fishies/Shallow%20reef%20system/renderings/sump1.png

Water comes in the filter sock area in the rear left (probably won't run one, but in case I want to), to skimmer, and then flows up and over that angled glass in the front (that is where the scrubber screen will go), then the L shaped bit is the fuge, then flows into the return section.

Ok, so for that scrubber, the glass area is about 9"x10", and I plan to have a screen be that size. Flow rate will be roughly 700GPH over the screen, as it is powered by my return pump. I plan on doing something close to the Blu Coral method, so ultra-high-protein food at least once a week, and once the scrubber gets going, I will do a constant drip of the food. As it is now, my fish eat better than I do, with around three feedings daily.

I don't know what the filtration capacity of this scrubber will be - it is definitely not considered a UAS, but it is not particularly a waterfall scrubber either, so I'm not sure of the total filtration capacity. It will be lit with LEDs, all Luxeon Rebel and Rebel ES. I'm thinking somewhere around 16/4 deep red and royal blue at 700ma (350ma for the RB, they will be in parallel). So that's roughly 12,160mW of deep red light and around 2,400mW of royal blue light.

Any ideas as to the capacity? I don't think it will be able to take 100% of the filtration, but I'd like to have an idea of what I'm working with. :)

Thanks!

Ben

[SantaMonica]

Welcome.

It's a horizontal river. 90 square inches.

[jedimasterben]

So it is a single cube screen, and needs 16x deep red LEDs like I had thought?

[SantaMonica]

Seems about 2 cubes, and would need 36 watts = 12 stars.

[jedimasterben]

Oh, I figured that the numbers for a horizontal scrubber would be equivalent to the 1 sided screen. Cool.

And I've been wondering - is the wattage based on Chinese '3w' LEDs or higher-output ones like the Rebels?

[SantaMonica]

They are all counted the same, 3w

[jedimasterben]

But shouldn't we go by output and not by count? Because those cheap knockoffs on ebay have around half the output of a Rebel.

[SantaMonica]

Too difficult.

[jedimasterben]

Not particularly. 'If Rebels (or other high-powered LEDs), use XX, if no-name, use XX'.

At any rate, so if I am using Rebels, I have much more light than is recommended. This will not increase the filtering capacity of the scrubber, correct? Just means I can run it less hours for the same effect?

[SantaMonica]

More light will grow more, if there is enough flow, and attachment.

[jedimasterben]

Well, it'll have plenty of flow (about double the recommended). Hmm, I wonder how high I can take the light then? (as in power level)

[Garf]

I think you'll need to get the screen submerged a little more than it looks on the diagram. 1 to 2 inches depth seems optimum. And like SM implied, your gonna really struggle to over illuminate these things. They do grow under lower lighting but nowhere near their full potential when it's up and running. Gonna be a great sump :)

[jedimasterben]

Hmm. I will have to see what I can do. The sump is already silicone'd together, but I should be able to slice off the horizontal bit and go from there. Hell, I can probably get a new piece of glass cut and make the horizontal bit much longer.

And overillumination? You obviously haven't seen what I'm capable of ;). http://www.nano-reef.com/topic/323517-building-an-evil-cluster-diy-led-on-steroids/

http://i5.photobucket.com/albums/y189/jedimasterben/LED%20Arrays/Evil%20Clusters/20130512-IMG_0118.jpg

Measured at over 1500 PAR at 12" using 74 degree reflectors on the Luxeon M and 50 degree reflector on the BXRA.



So if I can increase the lighting on this beast and make it more efficient, I'll do it. I just would like to know how much light I should not go over, which is really why it would be better to have output numbers instead of just 'XX 3w LEDs'.

[Garf]

Caught a glimpse of what looked like a growlight setup on that link, looked smart. I've got a 50watt UFO on my vertical screen, about 4 or 5 inches away and it could easily handle more than that, if that's any use to you. I'm not an LED techy though, but I know a man who is - Floyd, where are you?

[jedimasterben]

That beauty is one of two or three that will be going over a dedicated clam tank - simply because few corals will be able to handle the amount of light I'm shooting for. Basically, that is about 70w of high-quality (and high efficiency) LEDs, Bridgelux and Philips, that can cover what I will presume to be a 24"x24" area with around 300 PAR on the sandbed. I will have tests soon that will give me real numbers to work with.

As for those grow lights, they usually use cheap 5mm or '1w' class LEDs, which are super low in output and, well, pretty inefficient (though initially cheap and get the job done). I'd imagine most of the chinese knockoff '3w' 660nm LEDs to put out around 300-400mW of light at 700ma. The Rebel deep red I'll be using are binned to have around 780mW at 700ma, so effectively double the amount of light, and should not be classified in the same range as the knockoffs IMHO. I'd love to toss a few of the knockoffs under a spectrometer and get a usable readout (every listing I see measures them in lumens, which means they're either NOT 660nm or they don't actually know how to measure them).

[Garf]

Very true - I only do cheap unfortunately :)

[Floyd R Turbo]

I've been busy working my job unfortunately. Sucking up all my daytime playtime.

I think the photosynthetic saturation happens when you have a new screen with a ton of light. But it can also happen when an existing screen. I have only noticed the latter when running Luxeon RBs at 700ma 2" from the screen.

As far as output goes, yes, there is a difference between Rebels and the knock offs. I've been saying this since I stopped using the SatisLEDs because it's true. Jedi, thanks for supplying a few hard numbers to support that.

For a horizontal scrubber with the light a few inches above it, I would just cram the LEDs in there as tight as you can and let 'er rip. Not many have tried, so just see what happens

[jedimasterben]

So just load it up lol. Where should I start? I don't mind paying for the LEDs, but I don't particularly want to drop $100 when I only needed $60 or so. ;)

[jedimasterben]

I guess before I go out and buy a whole lot of Rebels, i could just use the light I made a while back for my fuge. 28x deep red and 6x royal blue at 700ma. The deep red should be putting out around 500mW of light or so, and the royal blue should be around 600mW each, so around 14,000mW of deep red and 3,600mW of royal blue.

Yes, I know, I already have those LEDs just hanging around, but this build is about efficiency :)

[Floyd R Turbo]

Since LEDs are more configurable and easier to cover larger areas more evenly, we do away with "watts/sq in" rules in a way and instead switch to coverage based calculations.

For a "normal" light LED vertical scrubber, on each side you would want one 3W 660 @700mA for each 8 square inches of material.
For a "high" light LED vertical scrubber, on each side you would want one 3W 660 @700mA for each 4 square inches of material.

For your 9x10, that's 90 sq in / 4 =~ 22 LEDs, but that would be on each side of a vertical waterfall.

The general rule of the past was if single-sided, double the light. Not to rehash recent discussions, but the "high" light is already doubled, so we skip that part.

Now for horizontal, we would (theoretically) double it again for the same given area. This would mean one for every 2 sq in, or 50 LEDs. But, try to cram those puppies that close together? Have fun with your wiring.

So I suggest somewhere in between. Or, just start with 25 and see how that goes. Use a MakersLED heat sink and you can reconfigure it if needed. Use the Steve's 1-part thermal adhesive and avoid the screws/nuts/washers, and then you can move them easily if you need to reconfigure.

[Floyd R Turbo]

I guess before I go out and buy a whole lot of Rebels, i could just use the light I made a while back for my fuge. 28x deep red and 6x royal blue at 700ma.

or that.

[jedimasterben]

or that.

:)


It's a beast. 60 degree optics on them, though, so it'll have to be hung higher off the scrubber. No biggie, though, cause I measured 300 PAR at like 14" with it :)

[Floyd R Turbo]

I think I saw that one in your NR thread, same one Garf mentioned? Take the lenses off and put it close, or like you said ^^

[jedimasterben]

Lenses are built-in.

And no, the one I linked to earlier is a light for a display that FAR outpaces what the fuge light can do. Here's a link to some pics of the whole shebang and the PAR readings at 700ma on all five LEDs. http://www.nano-reef.com/topic/323517-building-an-evil-cluster-diy-led-on-steroids/page-13#entry4328767

[Floyd R Turbo]

I meant the one on post 387, 2nd pic

[jedimasterben]

Oh yep, that's the one I'm talking about (but the lenses are built-in). The first pic is actually what I ran half my old waterfall scrubber on, that thing made some algae. It was just thrown together from leftovers - 4x deep red, 4x royal blue, 4x true violet, and 2x 5000k Rebel ES.

[jedimasterben]

Got it running this past weekend, it's got a light dusting of diatoms at this point.



Soooo, uhhh, is 800+ PAR a bad thing for a new scrubber?

http://i5.photobucket.com/albums/y189/jedimasterben/Aminals/Fishies/Shallow%20reef%20system/IMG_20130630_184127.jpg

[Floyd R Turbo]

haha! At first, maybe. Might want to dim that, but only if you notice that you are getting zero or close to zero growth. I sounds like you are getting something at least and not just a bare white screen, so I would leave it be for a few weeks.

But once you get a strong base of growth, it will generally take whatever you throw at it as long as you provide it enough flow and nutrients

[jedimasterben]

Gotcha. I actually had it running higher, around the 1200 PAR mark, but I realized 'holy mother of god, that's just gonna fry everything!' and raised it by four inches. It is not dimmable, and unfortunately the lenses are fixed, so I'm limited in my options other than raising it, which will cause even more immense light spill. Can't wait til a cop drives by my house and thinks I've got a grow op going lol

And again, this is my 'old light' that has 28x 660nm and 6x 450nm, all generics, not Luxeons. If they were Luxeons and I had that many, I'd probably max out the PAR meter like I did on my 'evil cluster' :)

[jedimasterben]

Also wanted to note - I went to visit the Smithsonian Marine Station in Fort Pierce, FL a few weeks ago. They've got a massive-ass Caribbean reef tank, and a dozen other smaller displays, a mangrove ecosystem, seagrass bed, etc. What I found VERY interesting is that on not one of their tanks do they use foam fractionation at all for filtration - they exclusively use algae scrubbers! Their design was... interesting, to say the least. It is difficult to describe how it worked, but I'll do my best. Maybe I'll make a mockup and a video to describe it better, I just wish I would have gotten pics of one :/

It's a rocking design. Water falls down in the center of a very long, shallow tray that was around 6ft in length, which was on a rotating rod. Water flows into it and down to one side until it falls down and dumps the water out, and then water begins filling the other side. OHHH, like a shishi odoshi (aka deer chaser or knocker) http://www.youtube.com/watch?v=aNbHJDZYr74

I asked the aquarist there about it, and he says that it is the most efficient design they've tested.

[Floyd R Turbo]

Sounds like it's an Adey dump bucket scrubber?

EDIT: no, that is THE Adey dump bucket scrubber

http://www.sms.si.edu/smee/behindthescenes.htm

This is the famous design my Dr Walter Adey from back in the 1980s, included in his book "Dynamic Aquaria"

Dude, you're way behind man!

[Floyd R Turbo]

Also a note, from the link I just posted, under the "chemical filters" subsection


Foam fractionators or protein skimmers are better suited for removing organic compounds produced by animals such as mucus and those specialized compounds produced for chemical defense. All but the Mangrove Model Ecosystem are equipped with protein skimmers.

[jedimasterben]

Sounds like it's an Adey dump bucket scrubber?

EDIT: no, that is THE Adey dump bucket scrubber

http://www.sms.si.edu/smee/behindthescenes.htm

This is the famous design my Dr Walter Adey from back in the 1980s, included in his book "Dynamic Aquaria"

Dude, you're way behind man!

I guess so! Lol


Also a note, from the link I just posted, under the "chemical filters" subsection
When I was there, they told me they used only algae scrubbers and carbon for filtration, maybe that part of the website is old? The guy had been there a while and seemed like he knew what was going on, but he may have been wrong lol

[Floyd R Turbo]

Probably an old website issue. That site looks pretty old school.

[Floyd R Turbo]

Dang if that weren't 2 hours from Ft. Lauderdale I'd visit it during MACNA

[jedimasterben]

Well, this is unfortunately a failure. The diatoms died off and nothing grow on the screen at all, so I took that sump down and replaced it with a standard 29g (really for reasons other than the scrubber not running properly, but you know :P). I'm trying to figure out an easy way to do a waterfall scrubber while also having my skimmer in.

http://i5.photobucket.com/albums/y189/jedimasterben/Aminals/Fishies/Shallow%20reef%20system/2013_08_04__23_31_39.jpg

http://i5.photobucket.com/albums/y189/jedimasterben/Aminals/Fishies/Shallow%20reef%20system/2013_08_04__23_31_59.jpg

The old light is still hanging there because I just haven't taken it down yet :)


And a video of it all to show just how ridiculous the microbubble situation is.
 

http://www.youtube.com/watch?v=vYBSafnJLvE



Not sure if I'm going to use the same LEDs that I had on the old scrubber (4x 660nm, 4x 450nm, 4x 420nm, 2x 5000K white) that I had such great results with, or just use 8x 660nm and 2x 450nm per side (if I can get the waterfall idea going, if not, it'll be a one-sided submerged screen). I wonder if using deep red and a warm white would be a better choice than using straight royal blue - since royals need to be run around half power anyway, why not use a chip that gives less of it and more green/red? Just a thought I have had. I also wonder what role 420-430nm plays. It's obvious that the combo I had was very powerful (around 350 PAR I believe it was), and grew 3D green algae over a large screen (12" x 8") over about a week and a half until I had to clean it off.

[jedimasterben]

I think what I'll do is do my best to afro-engineer a waterfall scrubber. Feed the living bajeezus out of the tank, and light both sides of it with different LEDs. One side being 10x 660nm and 2x 450nm, one side having about the same combo I had - 6x 660nm, 2x 450nm, 2x 430nm, 2x 2700K warm white. All will be Luxeon Rebel/Rebel ES, violet will be Epistar from LEDGroupBuy.

[Devs]

Hi Floyd

Is this with or without a prismatic diffuser ?



I think the photosynthetic saturation happens when you have a new screen with a ton of light. But it can also happen when an existing screen. I have only noticed the latter when running Luxeon RBs at 700ma 2" from the screen.

[Floyd R Turbo]

saturation generally happens without the diffuser on a blank screen. adding the diffuser seems to alleviate this.