Help with LED overflow scrubber design

[therman]

Hi all,

I'm very new here (joined today). I've decided that an LED lit ATS built into my overflow is my next reef priority. As you can see from the full tank shot, I'm no stranger to LEDs:
[attachment=1:3m1vqgmt]frameless full tank-sm.jpg[/attachment:3m1vqgmt]
This is the third major Cree fixture I've built, and consists of 156 XR-Es with 20 degree optics running about 36" above the 36"x37"x18" display. Overkill I know, it runs between 450-600mA depending on the string and color and only draws around 250W. I prefer underpowering and having flexibility to turn it up to wishing I had more light :)

I've decided to try to incorporate three turf scrubber screens into my 6"x12" overflow box in the back of the tank, and drop the water level in the overflow box down as low as I can. Due to the design water already flows in perfect sheets down the inside of the box (~900gph).
[attachment=0:3m1vqgmt]overflow.jpg[/attachment:3m1vqgmt]
I will line the overflow box with 1/8" black ABS sheet, and paint the inside white with krylon fusion to improve reflectivity in the overflow. Then I'll hang roughed up plastic canvas sheets down each of the three sides, and light with LEDs through the back wall of the tank with a flat array slipped between the wall and the tank.

From all my LED tinkering I should have 4 warm white XP-Gs and 8 or 9 cool white XR-Es sitting around to run off a spare Meanwell ELN-60-48.

So my main question is, do you think these 12-13 LEDs will be sufficient to grow turf, or do I need to buy some reds (deep?) too? They are almost as efficient as they come, and the warm white XP-Gs should have a pretty substantial red component. I imagine I'll have a total around 200-250 sq in (6"+12"+6" W x hopefully a good 12"H above the water level in the overflow, minus some area blocked by the plumbing). This should be just about the right flow/area/etc. for the roughly 100g display volume from what I have read in SantaMonica's guide. Just concerned about the lighting since it seems few are using LEDs for turf scrubbers.

Adding red LEDs would complicate things, as they would limit my drive current to 700mA if I put them on the same driver.

Would love to hear everyone's thoughts on the design, and the lighting. FWIW 4 cool white XP-Gs have done an awesome job of growing turf in my fuge, but not sure if turf prefers a different spectrum.

Thanks!
Tim

[SantaMonica]

Nice idea.


I will line the overflow box with 1/8" black ABS sheet

Good idea, if you want to keep the scrubber light out of the tank. Although once the screen is growing, it will block all the light. It may not hurt, therefor, to just let the light in, and keep it simple.


paint the inside white with krylon fusion to improve reflectivity in the overflow

Not really needed underneath the screen, because it will be covered completely with growth, and will see no light.


From all my LED tinkering I should have 4 warm white XP-Gs and 8 or 9 cool white XR-Es sitting around to run off a spare Meanwell ELN-60-48.

What will be the overall LED wattage? Experience has shown that it needs to be about the same wattage as the recommendations for fluorescent lights for scrubbers.


So my main question is, do you think these 12-13 LEDs will be sufficient to grow turf

Remember, it's green hair algae that you want to grow; not dense turf. And growth will mainly be a function of wattage and focus of the white LED's.


or do I need to buy some reds (deep?) too?

It's highly recommended to have red's. They are the primary growth color for what we are doing. If you look at LED horticulture lamps, they are mostly red too.


I imagine I'll have a total around 200-250 sq in (6"+12"+6" W x hopefully a good 12"H above the water level in the overflow

Keep in mind that the side panels, since they are at 45 degrees to the oncoming light, will not be operating at the same efficiency as the back panel; they will be operating at about 70 percent instead. Also, the LED's have to shine through that thick back panel; that will reduce the light too.


4 cool white XP-Gs have done an awesome job of growing turf in my fuge

Yes but it's a different story when you want it to be your primary filter. When you harvest a scrubber each week, the nutrients in the tank drop to very low levels. I'm guessing you did not harvest your fuge; in that case, nutrients did not drop much. When nutrients drop low, growth requirements change because the algae has to make up in one area (light) what it is losing in another area (nutrients to eat). That is why the highest powered scrubbers need very strong light to get good green-hair growth as the nutrients get low. Display tanks have the advantage of large area, so scrubbers make up for it with strong light.


seems few are using LEDs for turf scrubbers.

There is a reason. And this is where the experimental part comes in. In my opinion, there has never been a "successful" tank run on an LED scrubber as it's only filter. Successful means operating for at least 3 months with regular feeding, no water changes, and having the N and P tests come out "zero" (and of course, no nuisance algae in the display).

There has been screen growth, and nutrient removal, with LED scrubbers, but they are so hard to get operating correctly that very few have gotten anywhere with them. And even this was with months of tinkering and rebuilding. The colors required have pretty much been found, but one big problem is beam convergence. As you may know, the closer a light is, the more light-power (number of photons) is delivered to the screen. However, with LED's, if you get too close the individual LED beams don't converge, and they start producing individual focus spots instead. With red/blue combo's, this stops the growth since algae needs red+blue to grow. White does have some of both, but in a different percentage.

So the overall recommendation I can give is to approach this with "experiment" in mind, instead of "results". Make your build very adjustable and/or re-buildable. And read all the LED threads first:

viewtopic.php?f=9&t=822 (http://www.algaescrubber.net/forums/viewtopic.php?f=9&t=822)

[therman]

[quote]paint the inside white with krylon fusion to improve reflectivity in the overflow

Not really needed underneath the screen, because it will be covered completely with growth, and will see no light.

I'd think it would be very helpful right after cleaning, when you're seeing the most growth and most transparency. Black would absolutely suck up the light.


From all my LED tinkering I should have 4 warm white XP-Gs and 8 or 9 cool white XR-Es sitting around to run off a spare Meanwell ELN-60-48.

What will be the overall LED wattage? Experience has shown that it needs to be about the same wattage as the recommendations for fluorescent lights for scrubbers.

Wattage is a very poor measure of light output, particular with LEDs, and most especially when comparing two different lighting technologies. For example, Luxeon III and Cree XP-G could both be called "3 watt", yet at 1000mA, one puts out 80 lumens and the other 139 lumens. And the Chinese knockoffs that people are using are of even more ambiguous quality.

Also, due to the design of the bulbs, virtually all of those lumens from an LED are directed at your substrate, whereas a large proportion of fluorescent lumens are lost to restrike and reflector inefficiency. When you are referring to watts/gal, what fluorescent design are you referring to? T12, T8, T5, u-lamps, and spring bulbs will all have different effective lumens/watt.


or do I need to buy some reds (deep?) too?

It's highly recommended to have red's. They are the primary growth color for what we are doing. If you look at LED horticulture lamps, they are mostly red too.

The more I look at spectral plots, the more I think that red LEDs are not the way to go. They are only hitting a certain wavelength (well...2 if you count the deep reds) which is marginal to the chlorophyll peaks. Plus these are not vascular plants and may not have chlorophylls which peak like the widely available a and b graphs do and very likely have pigments specialized to cope particularly well with the blue spectrum if they find themselves in water of any depth.


Keep in mind that the side panels, since they are at 45 degrees to the oncoming light, will not be operating at the same efficiency as the back panel; they will be operating at about 70 percent instead. Also, the LED's have to shine through that thick back panel; that will reduce the light too.

of course.


4 cool white XP-Gs have done an awesome job of growing turf in my fuge

Yes but it's a different story when you want it to be your primary filter. When you harvest a scrubber each week, the nutrients in the tank drop to very low levels. I'm guessing you did not harvest your fuge; in that case, nutrients did not drop much.

I meant to say "chaeto" instead of turf. And I do harvest big handfuls of it occasionally


So the overall recommendation I can give is to approach this with "experiment" in mind, instead of "results". Make your build very adjustable and/or re-buildable. [/quote:2ga6d2p8]

I'd consider any turf scrubber experimental just like you'd consider any LED setup experimental. There is every bit and more the cadre of doubters in regards to LEDs for reef tanks. Both are highly tailored to the individual system just as the needs of any system vary. So it does take some experimenting to make it work just the way you want it. But if LEDs can grow corals and vascular plants at ridiculous rates (which I've done myself), I'm sure they can grow algae :)

Thanks for all the tips, and keep the advice coming.
-Tim

[rygh]

Very interesting strategy.
Given the space restrictions, my guess is that LEDs are pretty much your only option.
BTW: SM is correct in that while a lot of LED systems are "pretty good", it is unclear how many if any
are "total success". I have 0 N/P, but still have algae in my main tank.

I have looked at the warm XP-G warm whites with a rather cheap spectrometer.
The red end is weak. They are warmer because of a lot more yellow and orange, not deep red.

I would say that you badly need red. That is probably the #1 mistake of people doing LED.
I have direct experimental proof that Algae does not grow nearly as well with just blue.
See pics near end of modern led thread.
Absolutely positively add red.

Do not be fooled by the great lux XPG numbers from Cree. That is weighted by what the human eye sees,
so a really efficient deep red can look rather bad on a datasheet, yet be far better.

LedEngin Deep Reds are not restricted to 700 mA. They run to 1.5A.

Given your extra LEDs and driver, my suggestion:
4 x Warm white
2 x Cool white. (add more to just below 48V)
8 x LedEngin Deep Red
As a substitute for deep reds, you could use 2 strings 8 standard cheap ebay reds in parallel, so 650 mA each.
Likely good enough, and saves a lot of money.

Other Concerns:
1) Getting enough light to the sides.
If the LEDs are facing forward, not a lot will go to the sides.
You actually may want to make a small square acrylic tube that goes INSIDE the overflow, and put the LEDs
in that, with some facing forward, and some to the side.
2) Single sided.
Regardless of LED, single sided has had efficiency issues.
3) Cleaning.
I imagine cleaning the screens is fine.
But the rest of the overflow will need cleaning. Algae gets all over the place.
Something to think about.

[inkidu]

Adding red LEDs would complicate things, as they would limit my drive current to 700mA if I put them on the same driver.


I am by no means an expert but if you want my two cents I would stay away from mixing led types even if the forward voltage is similar.

I figure it this way one of the main benefits to led is their longevity so let similar/same ones live together harmoniously.

You can you get away with it sure but why risk it with cheap small drivers like this meanwell lplc-18-700. $10

http://www.meanwell.com/search/lplc-18/default.htm

Another bit of advice

Rygh turn me on to this site although I have yet to buy anything from them (The whole drilling and tapping is boring and tedious to me)

http://www.rapidled.com/servlet/the-Hea ... Categories (http://www.rapidled.com/servlet/the-Heat-Sinks/s/10/Categories)

In particular this one (holds 18 led which would give you some room to add 5 deep reds if you so desire)

http://www.rapidled.com/servlet/the-84/ ... %22/Detail (http://www.rapidled.com/servlet/the-84/Drilled-and-Tapped-5.9%22/Detail)

You could also add the Meanwell lpc-35 $16 and the shipping would still be $6 (in my area at least) if you got the heat sink also.

http://www.rapidled.com/servlet/the-3/d ... ers/Detail (http://www.rapidled.com/servlet/the-3/driver-700ma-48v-drivers/Detail)

I could keep going on but I got a couple things I got to do. Try to keep the build flexible so you will be able to tweak it.(dimming, moving etc..)

Good luck and I hope this helps.

p.s. I have been sitting on 8 x-pg 3000 K warm whites that I have yet to hook up so keep us posted I am curious how well they will work.

I haven't found the need so much I have a successful led scrubber.

Although if I feed more than 10 times a day some times I get some algae in the tank. lol