DIY LED Scrubber Lighting Guidelines and Builds

[Floyd R Turbo]

Starting a new thread here. I have been using the nice new search function to search this site for the DIY LED builds in an effort to put together some kind of 'best practice' thread so that a set of standards can be narrowed down. It seems that LED comes with it's own set of rules that differ in many ways from the norm of the past.

DIYing LED is not as hard as you think it is. I haven't built a fixture for myself, but attended a local club meeting where we got a chance to see how to do it in person. Anyone with or without soldering and wiring skills can do this. It seems that it is definitely the way to go. In my opinion, based on observations I've seen on here, CFL and T5HO are over. The benefit to cost ratio of LED over anything else seems to be through the roof, especially considering the new feeding-based screen sizing guidelines.

So, I would like to hear from anyone who has any experience building one. Specifically, I am interested in spreading information on:

1) LEDs: How many (total and per side), what kind (spectrum), brand, wattage, where did you get it and how much did you pay, lenses, distance from screen, etc

2) DRIVER: How many, what kind (dimmable/not), brand, capacity, where did you get it and how much did you pay

3) HEATSINK: Type (extruded aluminum, CPU heatsink, etc), dimensions (L x W, fin length, etc), brand, where did you get it and how much did you pay

4) SYSTEM INFORMATION: Size of your Algae Scrubber screen, enclosure, fans, distance from screen, on/off cycle, flow rate, how much you feed per day, single/double sided, how often you clean, what kind of growth you're getting, etc

5) TROUBLESHOOTING: Please post any problems you experienced along the way, mistakes you make, or things you would do differently if you were going to do it again.

6) PICTURES: post pics of your whole system, build sequence if you have it, and most importantly, pics of your screen growth AND display tank

If you have already started a thread with all this information, please post 1 through 5 here and then provide a link to your current thread. Some of these LED threads are long, and some people (rygh) have half a dozen threads, and some with many, many pictures. So I would like a summary of your system here, and feel free to re-post the pictures here if you want to, otherwise link to your thread.

---------------------------

With that being said, I have figured up a few basic needs, based on threads I've read, information I've gathered when designing my custom-built fixtures, and questions I've asked.

LEDs:

Minimum of one 3W 660nm LED (operating at 2W) for every 15-20 square inches of screen. That is an area between roughly 4" x 4" to 4.5" x 4.5". Perhaps a simpler standard would be to say you need one on each side for each cube of food per day, or 12 square inches.

The mixture of different colors should definitely be limited to 660 (deep red), 630 (red), 455 (royal blue), and 435 (violet blue). 660 and 435 are the Chlorophyll "A" band, and 660/455 are the "B" band. I recently stumbled upon one possible reason you want to focus on the "A" band: this light band is utilized the most by chlorophyll during midday when sunlight is the most intense, and the "B" spectrum is for when the light levels are lower, like during the morning and evening, when the earth's atmosphere reflects or filters out most of the "A" band. So as it turns out, "B" stands for "Backup" haha.

Anyways, it appears based on this that the anecdotal evidence suggesting 660nm Deep Reds are most effective is most likely true. Thus, 660nm Deep Red LEDs are the DIY choice.

With this being said, there is also a NASA study (which I have heard of, but never actually seen) that showed that chlorophyll absorbs red and blue in a ratio of approximately 6.5:1 red:blue. So throwing in a blue helps, but is not necesary. Rygh reported that the addition of a single blue to an existing arrray of red did not so much cause an increase in growth as it increased the strength[/] of the algae; thickness of individual strands increased and algae was more strongly attached to the screen. The problem is that in such a low ratio, if you have a screen that only needs one or two reds, adding a blue would usually require a separate channel that could be dimmed appropriately, and this adds another level of complication. I would appreciate input from the DIY LED experts on how to easily accomplish this.

DRIVER:

Meanwell dimmable driver or a non-dimmable driver and some other method of dimming. The need for this depends of how you build your light fixture, and how close it will be to the screen. Could use some input on this one. To me it seems that it would be a lot easier to dim the LEDs if they are too intense, even with a reduced photoperiod. my point being that it would be nice to be able to easily adjust the system as necessary without having to tear things apart. Trying to think of features that the average person can use and be successful.

Also a DIY driver is not out of the question. Like I state above, I'm really looking for some good options for multi-channel drivers with dimming, and the simpler the better, off the shelf preferred, and of course low cost and easy to DIY.

HEATSINK

I've seen a couple threads with good, cheap heatsinks. I would opt for a large heatsink for passive cooling over a smaller one with a fan. Another consideration for this might be increasing the density of LEDs per unit screen area and running them at a lower current. LEDs get [i]more efficient as you dim them (lower total output, but higher lumen/watt ratio) which means less wasted energy = less heat.

The user TeraHz on RC made a DT fixture using these for heatsinks and needed no active cooling (open air fixture also)

http://www.mouser.com/search/ProductDetail.aspx?qs=wT7LY0lnAe3hVEsFt1WZEQ%3D%3D

and another user on here (can't recall who) posted this link http://icamanufacturing.homestead.com/Heatsink.html

But http://www.heatsinkusa.com/ seems to still have some pretty good prices for what DIYers would use for a scrubber based on the feeding guidelines. Under $11 shipped for a 4" x 4" sink with 3/4" fins for one 3W LED is pretty darn cheap. $17 for two!

---------------------------

Ok, that's what I've got just off the top of my head.

So let's hear some info and turn the Algae Scrubber world on it's head!

[kotlec]

Good heatsink : http://www.ebay.com/itm/1pc-Aluminium-Heatsink-8-3W-Aquarium-Led-Bulb-/150652979161?pt=LH_DefaultDomain_0&hash=item23139e07d9

[andthe_brain]

From all the reading and picking other members brains about them (esp Ace25) the general lighting requirements where 1/5th of the power of cfl at a minimum or upto a third of the wattage of cfl for high wattage filtering. Also Cree LEDs are about the best in the business in pretty much every aspect....but im sure ace will jump in somewhere ;-)

[Floyd R Turbo]

Ok, hopefully there will be a flood of input coming on this. I just blasted the link to this thread on a couple dozen LED threads.

AFAIK, Cree does not make a 660nm LED. I could be wrong, but last time I checked, it was Osram, LedEngin, Bridgeluxe, Epistar, etc that you had to choose from.

Good good point on the CFL comparison. I wonder about T5HO comparison.

The other thing that I am trying to pin down is related to the new double-light guidelines. Since LEDs don't waste bandwidth, a watt does not equal a watt anymore. So 1/5 wattage may be related to some correlation, like only 1/5 of CFL is really effective bandwidth, which is why you only need 1/5 the wattage of LED. Then again, people are getting all-green growth from that 1/5 wattage, so what does that say? Might it be that the absence of non-useable light is causing the absence of non-green growth? Since green growth is the most effective, could one not say that 1/5 of the wattage of normal level CFL is equivalent in filtering capacity to a double-lit CFL screen, meaning that LEDs are actual 10x better?

This is why I would like to gather and pool as much real-world information as possible on this thread.

Thanks ahead of time to all who respond and take the time to post details about their experiences.

[andthe_brain]

I will be following this one with interest. not sure on the cree 660nm :-*

[kerry]

Cree does not make 660nm as of my last research. I am testing two LED scrubbers right now. One is very close @ 1.5" w/o optics and the other is @ about 4" w/optics. Both of my scrubbers are using the china LED's and drivers. It does appear that the wattage is about 1/5 to 1/3 less then gas filled bulbs.

[BobRoss]

excited to see this thread grow! (no pun intended)

[rygh]

Regarding the mixture of blue at low ratios:

1) My advice is actually to use warm white or neutral white instead.
You get a lot of blue that way anyway, plus some other possibly useful spectra.
Plus, they are really cheap.

2) Run reds in series, but whites in parallel.
For example:
R + R + R + R + (W | W | W | W)
If that has a 750 mA driver, each red gets 750 mA, but the whites get 750/4 = 187 mA
So even though you have the same number of LEDs, you actually have a 4:1 R:W ratio.
Note that they have to be the same exact led for that to work.

3) Glue some diamond diffusion plating to a small piece of pipe, and put it over the blue.
Cuts it a bit, and spreads the light around nicely.
https://lh3.googleusercontent.com/-ATcAYLG2g7w/TdHucEtBaWI/AAAAAAAAAUI/ZGqrLRMaA_w/s800/led_red1.jpg

[kotlec]

Too add to blues discusion. I have single blue on 8 led total panel . All 1w chinesse. What is interesting that part of screen that is under blue is not different from those of reds. Not more growth and not less. All yellow :)

[Floyd R Turbo]

Koltec, do you have a link for where you got the 1W reds and blues?

[andthe_brain]

Alot of american ppl use www.rapidLED.com for people over the otherside of the pond www.rapid.co.uk or ebay

[andthe_brain]

Sorry not rapid.co.uk its www.rapidonline.com

[kerry]

Floyd,
This is the guy I like to get my china LED's from. He is in California so there is little wait time (3-4 days) as he has this stuff on hand. So far his products have been 100%. I know its still china stuff but, it is in the states at least so it does not take 3-4 weeks to get. http://www.ebay.com/sch/ledtoplights/m.html?_nkw=&_armrs=1&_from=&_ipg=25&_trksid=p5039 .

[MorganAtlanta]

1) LEDs: How many (total and per side), what kind (spectrum), brand, wattage, where did you get it and how much did you pay, lenses, distance from screen, etc

28 LEDs total
14 per side
10x660nm (from LED-DNA via eBay, 3W China LEDs, $40 for 20 shipped)
4x630nm (from BestShop2008HK via eBay, 3W China LEDs, $20 for 10 shipped)
Spacing is roughly 2.5" between leds
The LEDs are roughly 3" from the screen
No lenses


2) DRIVER: How many, what kind (dimmable/not), brand, capacity, where did you get it and how much did you pay

Two Meanwell ELN30D-48, dimmable drivers, one for each side.
Drive the LEDs at up to 625mA when full on.
Watt Supply, $30 each.


3) HEATSINK: Type (extruded aluminum, CPU heatsink, etc), dimensions (L x W, fin length, etc), brand, where did you get it and how much did you pay

Heatsinks 6" x 20" with 1" fins
Heatsinks USA, $40 each.


4) SYSTEM INFORMATION: Size of your Algae Scrubber screen, enclosure, fans, distance from screen, on/off cycle, flow rate, how much you feed per day, single/double sided, how often you clean, what kind of growth you're getting, etc

Screen 19"x9"

SM-100 style design, but 20"x10" box

No fans needed

The total cost of the LEDs was something like $220, and was quite easy to do, although I built my own LED display lights, so I had some experience, and I have an BS in EE.

Experimenting with intensity and photoperiod. 17 on - 7 off worked at 500mA, but burned the algae at 625mA. Currently at 12 on - 12 off at 625 mA and growth seems good, but I haven't settled on the "best" for my system yet. Growth at 17/7 and 500mA was excellent. I need to stop tinkering.

Feeding:
Spectrum fish pellets 2x per day from an autofeeder, very small amount, maybe 0.5-1.0 ml
Coral Frenzy 2x per day from an autofeeder, about 1 ml per time
Frozen mysis once per day, roughly 2-3 cubes, it depends on how big a piece breaks off the flat-pack. I prefer Hikari.

Clean screens roughly 7-10 days, due to growth up into slot

Flow roughly 800 GPH. I've got a Mag 18 feeding the display, and I run the scrubber off the Herbie-style drains. Two drains, one into each end of the slotted pipe.


5) TROUBLESHOOTING: Please post any problems you experienced along the way, mistakes you make, or things you would do differently if you were going to do it again.

Dimmable drivers are not necessary. A driver that runs fixed at 500mA-600mA would be fine. Forward voltage of the LEDs is only roughly 2.2v, so at 500mA, they are only running at 1.1 Watts each, and only using 15.5 Watts per side, so smaller drivers could have been used. I bought the smallest dimmable drivers I could find. Non-dimmable 18W 500mA drivers would work great.

I'm having problems with growth up into the slot. Even though I have light blocker to shade the last inch of screen, it seems the light almost "channels" up the water to the slot. The growth isn't thick, but enough to cause spraying after a week or so and I have to take the screen out and clean it. I'd rather be able to go a full two weeks.

I used Arctic Alumina thermal adhesive to glue the LED stars to the heatsink. For the wires between the LEDs, I used "thermostat" wire. I think it was 20 or 22 gauge solid wire. I soldered the wires to the pads on the LED stars. This isn't as hard to do as you might expect.

For the wires between the power supplies and the heatsinks, I scavenged wires with male "type M" connectors from old cellphone/electronics/toy power supplies I had laying around, then bought some female "type M" connectors from Radio Shack. I drilled a hole in each heatsink to mount the female type M connector.


6) PICTURES: post pics of your whole system, build sequence if you have it, and most importantly, pics of your screen growth AND display tank

The system is a standard 125 gallon with dual back overflows.

The display is lit with 126 CREE XRE leds-- 42 each of Royal Blue, White, and Blue.

I recently switched to "Herbie" drains with the return coming over the back. This is a big improvement over durso drains because there is no air mixed in the flow going to the scrubber, and the whole system is now much quieter.

Sump is a roughly 30 gallon plastic storage bin.

It's a "fish-mostly mixed reef", although I lost most of my SPS recently due to a big swing in Alk. I'm moving to a caclium reactor soon, off of two-part dosing. It will be interesting to see if the extra CO2 in the system affects algae growth. There could be some "synergy" there.

[Ace25]

1) LEDs: How many (total and per side), what kind (spectrum), brand, wattage, where did you get it and how much did you pay, lenses, distance from screen, etc

14 LEDs total, 7 Per side. 6x 660nm OSRAM + 1 Cree XR-E Royal Blue, run @ 700mA. Bought from www.rapidled.com. No lenses, 2.5" from screen.

2) DRIVER: How many, what kind (dimmable/not), brand, capacity, where did you get it and how much did you pay

Mean Well LPC-35-700 driver, constant current/non dimmable. www.rapidled.com for $16.00

3) HEATSINK: Type (extruded aluminum, CPU heatsink, etc), dimensions (L x W, fin length, etc), brand, where did you get it and how much did you pay

aluminum C channel, 10" long, 2 pieces per side, so 40" of material total. Was actually scrap from a friends DIY build but he bought it at a local metal supply company, think it was like $40 for 25', I just needed a few small pieces though.

4) SYSTEM INFORMATION: Size of your Algae Scrubber screen, enclosure, fans, distance from screen, on/off cycle, flow rate, how much you feed per day, single/double sided, how often you clean, what kind of growth you're getting, etc

Screen Size: Standard full screen (13"x10" I believe). 2 fans over the light box enclosures (which I feel was a bad design and my next ones will not use fans). 12 on / 12 off light schedule. Flow rate around 750GPH. Feed 6 fozen cubes + enough spectrum pellets to equal another 6 cubes. Double sided. Clean 7-10 days. Growth goes from "good average green growth" to "OMG how in the world did it grow so fast!" depending on the week, seems to have an ebb and flow to the growth levels on the screens.

5) TROUBLESHOOTING: Please post any problems you experienced along the way, mistakes you make, or things you would do differently if you were going to do it again.

Since my first DIY LED ATS was more "proof of concept" vs what I would consider a final product I think there are many mistakes with my build, although minor and annoying vs anything that would cause the ATS to work less efficiently. Mainly my ATS is waaay overbuilt for my size tank. My next version is going to use what I learned from this build and make it about 1/10th the size overall. So to answer what I would do differently.. pretty much everything except for what LEDs to use.

6) PICTURES: post pics of your whole system, build sequence if you have it, and most importantly, pics of your screen growth AND display tank

Originally started with Warm White LEDs
http://farm6.staticflickr.com/5228/5638737127_c4e695e31e.jpg http://farm6.staticflickr.com/5143/5644678310_420cc25fcb.jpg

Full Sump picture / Fans on Light Box
http://farm6.staticflickr.com/5029/5651529855_47300c0f58.jpg http://farm6.staticflickr.com/5267/5768317993_3bb0d090f3.jpg

Experimenting with different color LEDs
http://farm4.staticflickr.com/3388/5795464037_f6041312d6.jpg http://farm3.staticflickr.com/2003/5796023176_bf2747bb21.jpg

Current LED configuration
http://farm7.staticflickr.com/6046/6267306909_8569edc301.jpg

Display
http://farm3.staticflickr.com/2451/3914071932_f98b784c10_o.jpg

[kotlec]

I've got all LED from ebay.
http://www.ebay.com/itm/180711673998?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1438.l2649

[Floyd R Turbo]

Thanks for the posts Morgan and Ace! That is exactly what I was looking for.

Rygh, thanks for posting that. There is a cost factor and that might make some sense. I particularly like your suggestion of putting the non-reds in parallel! That's why I started this thread, to get great ideas like that.

[andthe_brain]

Nice build ace! I'd love a sump room but SWMBO (she who must be obeyed) would throttle me lol...Ace did you get my pms?

[rygh]

+1 on the sizing for the food you add, not gallons.
My scrubber is way over-sized.

Another thing to really consider is the dual alternating screen approach.
You have two screens. Left + Right.
You clean only one a week. Thus each really grows for two weeks.
I have been trying that off and on. Seems to be better, plus cuts the amount
of cleaning in half.
Only do that if your screens are not very full of course. (Oversized scrubber)

Note: I am still a big fan of running carbon with an ATS.
Without that, my water tends to get a bit yellow pretty quick.

[Floyd R Turbo]

I have never had that problem at all. I can drain water into a white bucket and it looks like fresh SW

[Floyd R Turbo]

Let's try to keep this on topic. I'm guilty too :)

[Floyd R Turbo]

I updated the Basic Sticky http://algaescrubber.net/forums/showthread.php?1291-Algae-Scrubber-Basics-The-Summary post #26 and revised the LED section. Used a few of these pics! I did just notice that I poo-pooed whites and I could have changed that per Rygh's suggestion but totally forgot about it. Oh well. I'm sure I'll come up with an LED build summary as SM has now said most of the new version will need to be LED.

[SantaMonica]

Remember where the yellow comes from: Dying roots.

[bucks448]

Interesting timing…

This weekend I pulled out my old 55gal sump with a rigged CFL scrubber and replaced it with an acrylic sump modified with a built-in LED scrubber. Already this thread has helped me, more on that in the troubleshooting section. I've had good success with CFL since I started about a year ago, but I've pushed it to the point I need more...


1) LEDs: How many (total and per side), what kind (spectrum), brand, wattage, where did you get it and how much did you pay, lenses, distance from screen, etc

20 LEDs, 10 per side. 9x 660nm 1x Royal Blue running at 2.4v, 650mA, no lenses. All LEDs are Fedy. The Royal Blue was direct from Fedy, leftover from my display light last year, $2.27 in bulk back then. The 660nm were from ledtoplights on ebay, $49 for 20.

2) DRIVER: How many, what kind (dimmable/not), brand, capacity, where did you get it and how much did you pay

Mean Well ELN-30-48 driver, constant current/non dimmable, from onlinecomponents.com, $23.68. All LEDs are wired in series. I'm running it at spec for voltage, and a little above spec for current, but within tolerance. I'll post back if I run into any problems pushing the drive like this. It's been running for two days, too early to tell.

3) HEATSINK: Type (extruded aluminum, CPU heatsink, etc), dimensions (L x W, fin length, etc), brand, where did you get it and how much did you pay

HeatsinkUSA - 4x 4.23" x 12", $16.58 each, two on each side, 5 LEDs on each spaced about 2.5" a part. I wanted to be able to change the shape/spacing (somewhat) which is why I went with 4 smaller heatsinks, rather than 2 larger ones. They're oversized so I can add LEDs and for passive cooling in a partially enclosed box. Also, I know how C-channel aluminum can be from my display light build.

4) SYSTEM INFORMATION: Size of your Algae Scrubber screen, enclosure, fans, distance from screen, on/off cycle, flow rate, how much you feed per day, single/double sided, how often you clean, what kind of growth you're getting, etc

The screen is 13" x 10". The LEDs are about 2.5" from the screen and I'm going with 12on / 12off light schedule. The screen is fed off a full siphon overflow, the return is supplied by a Mag12 which based on a "head loss calculator" is pushing 900-1000gph. On the overflow, about 2/3 of this goes to the scrubber, the rest goes to the other side of the sump through a UV sterilizer. I'm been feeding a 4" x 4" sheet of nori, and the equivalent of 3 cubes of oyster feast, cyclopeeze, mysis, and reef caviar daily. I'll probably stick with feeding the same until the new screen takes off. I did strap pieces of the old screen to it for a jump start. Prior to the change I was clean every 14 days. With the new build, I'll see what the growth dictates, probably every 7 days.

5) TROUBLESHOOTING: Please post any problems you experienced along the way, mistakes you make, or things you would do differently if you were going to do it again.

Some how I got it in my head to put the LEDs as close as possible (1.5" in my case) and run them 6 hours on 6 hours off. Morgan's and Ace's posts helped me straighten it out. They are now 2.5" from the screen and I'm running them 12 on / 12 off. There's no sense experimenting from the start, so I'm going to what seems to be within standards, at least for now.

Not LED related...
I started with the screen being taller, about 4" taller than the top of the scrubber box. The plan was to build a shield to block the light and shield the spray from the 4" at the top to keep algae from growing into the slot. I ended up lowering the pipe to the top of the scrubber box because the screen was to large and cumbersome to manipulate, and I could better protect the LEDs this way. Also, I didn't like the way the flow seemed less and uneven as it got lower when the screen was taller.
I need to find a quick release system that I like to remove the screen from the slot. Until then I'm using zip ties.

It's too early to tell if I'll have any other changes. For now, I'm sure that its 100x better then my previous setup with the 55 gal for a sump.


6) PICTURES: post pics of your whole system, build sequence if you have it, and most importantly, pics of your screen growth AND display tank

I don't have any pics of screen growth yet with LED, but here's a short slide show of where I started a year ago to what I just finished up this weekend. I'll post update as things progress...

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

[kotlec]

Cant see your tube vid. Says private.

[bucks448]

Cant see your tube vid. Says private.

It should be fixed now...late night, long weekend, sorry :)

[kerry]

NO yellow here either, even the FW scrubber water is clear. Even after the switch to LED's

[andthe_brain]

Another website for UK based peeps
http://www.farnell.com/uk/leds/index.php

[vahegan]

Thank you, Floyd, for a very interesting thread.
It was most interesting to find out about the NASA study of 6.5:1 of Red:Blue power, never heard of that before. If you find a link to that, please share.

I back the idea that it may be better using reds with cool whites instead of pure RB. Because the white has a large peak at 440-460nm plus the full spectrum to some extent as a free bonus.

Maybe this concerns algae at a lesser degree than higher plants, but I read on some Russian forum that greenhouse plans were stressed when illuminated by pure spectral light (450nm blue + 660nm red). Cabbage produced dark green and bitter leaves and the effect was similar to stress caused by UV. They've got much better results when some full spectrum white light was added. Sure nobody's gonna eat the algae grown on the scrubber but my point is stress: a stressed plat will show stunted growth vs. unstressed plant under same conditions.

What concerns buying 660nm LEDs, Cree doesn't make them and OSRAMs (http://www.rapidled.com/osram-660nm-3w-red-led/) are probably the best, but at a cost of $5.5. There is a cheaper alternative here: Deep Red - 660nm (http://www.ledgroupbuy.com/deep-red-660nm/). The real price difference is not that much however considering that Osrams can stand up to 1A current (this would account for 2.5-2.6W each, thus yielding 0.45W/$) whereas their Chinese alternative is only good up to 0.7A (and the vendor recommends to running it under 0.5A) which is about 1.25-1.75W, yielding 0.62-0.87W/$ but also note the better efficiency of OSRAMs vs no-name Chinese LEDs.

A better option perhaps is to use the Surexi Led Modules (http://www.theledstore.com.au/product_p/ls0017.htm). A friend of mine is using these on a scrubber and reports very good results. The red:blue ratio for these (the F1 spectra is best for our use) is 4:1 however (and this is the usually recommended horticultural ratio as far as I know). These modules run at up to 6.6W each at the cost of $11.95, which yields 0.55W/$ (of course it is best to use these at 0.5A yielding 5W from each unit).

[kotlec]

May be the best thing about Surexi Modules is that they are made for wet locations. On the other hand being 5wats concentrated in a single spot is not so attractive.

[Floyd R Turbo]

Scolley on RC uses some whites and gets good growth

Post # 207 here: http://www.reefcentral.com/forums/showthread.php?t=2028045

but he uses a mix of NW & CW so there may still be a use for other spectra. But we're talking about maximum efficiency here and I think the end result will probably be that we're splitting hairs when it comes down to it. RB vs. White will probably result in similar growth, but the high Red:Blue ratio needs to remain

[trigman]

Just wondering if anyone knows of a UK based site as easy to use as rapidled.com, i've looked at the 2 UK based sites mentioned on here and they don't seem like they are set up for the likes of me.

[kotlec]

Try ebay.co.uk

[andthe_brain]

Have a look on that farnell link i posted on the other page...they do everything u need

[vahegan]

May be the best thing about Surexi Modules is that they are made for wet locations. On the other hand being 5wats concentrated in a single spot is not so attractive.

It is not quite like that. They are designed with a tiny lens integrated on each crystal, 50 degrees. So that the module produces a large and even rectangular spot, with reds and blues properly mixed up. I consider this is yet another advantage of Surexi modules.

Also, they have a solderless design with special $3 molex connectors available, so that wiring up may be easier for those not friendly with a soldering iron and flux.

The bad thing is that few places carry them. The Australian store being one, and Mouser next, I wasn't able to track them elsewhere.

[tebo]

tonight will try to post my settings after a remodeling


regards

[kotlec]

At 50 degrees you want to place them at a good distance. 3 inch can be too close for 5 watt. Then your scrubber becomes fat.

[joelespinoza]

I read through this whole post but I was a bit confused on 2 things:



LEDs:

Minimum of one 3W 660nm LED (operating at 2W) for every 15-20 square inches of screen. That is an area between roughly 4" x 4" to 4.5" x 4.5". Perhaps a simpler standard would be to say you need one on each side for each cube of food per day, or 12 square inches.

So you mean you need 1 LED per side for about every 15 square inches? Or just on one side for every 15 square inches? So I guess my question is will a single 3 watt LED cover 30 square inches if you have one on each side?

Also is this info still fairly up to date?




The mixture of different colors should definitely be limited to 660 (deep red), 630 (red), 455 (royal blue), and 435 (violet blue). 660 and 435 are the Chlorophyll "A" band, and 660/455 are the "B" band. I recently stumbled upon one possible reason you want to focus on the "A" band: this light band is utilized the most by chlorophyll during midday when sunlight is the most intense, and the "B" spectrum is for when the light levels are lower, like during the morning and evening, when the earth's atmosphere reflects or filters out most of the "A" band. So as it turns out, "B" stands for "Backup" haha.

You mention 630 red as a desirable color and then never mention it again. Is it one of the chlorophyll bands from research you have run accross? Or should it be used in a ratio with the others?

[tebo]

1.- 12 LEDs total, 6 Per side. 6x 660nm OSRAM, run @ 750mA. Bought from www.rapidled.com. No lenses, 2.5" from screen may be less.

2.- Mean Well ELN-60-48D dimmable driver. Bought from www.rapidled.com.

3.- aluminum C channel, 10" long, 2 pieces per side, so 40" of material total. local metal supply company 50$

4.- Screen Size: (10"x 8"). 2 fans over the aluminum. 16 on / 8 off light schedule. but I think for me biological low load I reduce. Flow rate around 600GPH. Feed enough spectrum pellets to equal 1 cubes. Double sided. Clean half every 7 days. good green growth

5.- at first my screen was too big, then reduced it by the advice given to me at this forum, but not put fans depended on the air conditioning, so after I put the fans

First
http://i1043.photobucket.com/albums/b440/tebomarino/DSC01729.jpg
http://i1043.photobucket.com/albums/b440/tebomarino/DSC01731.jpg
http://i1043.photobucket.com/albums/b440/tebomarino/DSC01734.jpg

After

http://i1043.photobucket.com/albums/b440/tebomarino/DSC02300.jpg
http://i1043.photobucket.com/albums/b440/tebomarino/DSC02301.jpg
http://i1043.photobucket.com/albums/b440/tebomarino/DSC02303.jpg

Display

http://i1043.photobucket.com/albums/b440/tebomarino/DSC02127.jpg

[Floyd R Turbo]

So you mean you need 1 LED per side for about every 15 square inches? Or just on one side for every 15 square inches? So I guess my question is will a single 3 watt LED cover 30 square inches if you have one on each side?

I would say it's a minimum of 1 3W LED on each side of a 4x4 to 4.5x4.5 screen. If you re-read the part you quoted from me that's what the last part says.


Also is this info still fairly up to date?

Fairly. I'm currently building one though that has 4 x 3W 660 and 1 x 3W 455 (half power) on each side of a 3x4 screen. It might be overkill.


You mention 630 red as a desirable color and then never mention it again. Is it one of the chlorophyll bands from research you have run accross? Or should it be used in a ratio with the others?

basic chlorophyll charts. "A" is 660 and 435, "B" is 630 and 455. Not many people build 630 arrays that's why it's not mentioned much after that I guess.

[kerry]

Fairly. I'm currently building one though that has 4 x 3W 660 and 1 x 3W 455 (half power) on each side of a 3x4 screen. It might be overkill.



I am not so sure its over kill depending on the light your display makes. I only say this because since I have taken my horizontal on the 10G and made it vertical I am getting green algae growing in my tank now. I updated my display LED light (7 3 watt 19000K) before (by about 2 months) my scrubber and didn't have any algae problem. I would however have slight runaway nitrate with the horizontal (up to 30PPM sometimes) and just correct this with feedings. Now I have taken the 4 LEDs from the horizontal and put two on each side of my new vertical. Since doing this my nitrates are great but I am getting algae growth in the display at a real good rate. I am going to add one extra to one side and see how it goes. My screen size is about 2.75x4.5" When I had all the LEDs on one side of a 2.75x6.75" screen I didn't have any display growth.

[Floyd R Turbo]

Well that makes me feel a little better. Still even if it is, you can regulate the photoperiod. I guess it would also qualify under the double-wattage lighting rule

[kerry]

I cut it back some but its still making head way, I guess I will have to shorten the display period a little more.

[joelespinoza]

The mixture of different colors should definitely be limited to 660 (deep red), 630 (red), 455 (royal blue), and 435 (violet blue). 660 and 435 are the Chlorophyll "A" band, and 660/455 are the "B" band.

Ahhh it makes more sense now, it was just a typo, thanks for clearing that up!

[Floyd R Turbo]

Has anyone used the 660nm reds from satisled?

http://www.satisled.com/3w-high-power-red-led-star-emitter-for-plant-grow-light-660670nm_p433.html

These look like the osram chip but I'm guessing it's a knock off. Still though for < $2 each that's hard to pass up over $5.50 Osram from rapid.

[Ace25]

Haven't used those LEDs yet, but they look much closer to FEDY LEDs vs OSRAM. OSRAM LEDs don't look like any other LED I have used personally, they are an odd looking square instead of round (and why there are no optics for them).

Here are my FEDY LEDs, one thing if you look closely that you will notice, the Red LEDs have 2 seperate LED emitters side by side inside each chip, so they are "dual band" LEDs. One LED (middle row) contains a 630nm AND 640nm LED inside, the other one (top row) contains a 650nm and 660nm LED inside. Bottom row is royal blue and only has 1 emitter inside. They also cost $1.70ea.

OSRAM vs Satisled
http://www.rapidled.com/product_images/e/445/osram_red__22846_zoom.jpg http://www.satisled.com/images/v/201012/12935183520.jpg
vs FEDY LEDs
http://farm8.staticflickr.com/7165/6569941509_13f19bb296_b.jpg

[Floyd R Turbo]

Interesting - how has your experience with the FEDY's been?

[kerry]

Satisled takes FOREVER!! I have them from the last order and they work just fine. They grow algae like the ones I get from http://www.ebay.com/itm/3W-x-10-PCs-High-Power-Red-660nm-LED-Specialist-plant-DIY-item-/110813289474?pt=LH_DefaultDomain_0&hash=item19ccfc9c02 .

[Floyd R Turbo]

This is valuable. One thing I'm noticing is that the Osrams (on the Rapid page) have a max current of 1000mA. The satisLEDs say 750. ledtoplights (ebay link above) say 650. So If I am wanting to use a meanwell driver, I could use the LPC-35-700 non-dimmable one for the osrams, but not sure about the other two. I would think those would need a dimmable driver, which costs more, which offsets about 1/2 the savings ($34 vs $16). So would I be OK driving the satisLED or ledtoplights LEDs with an LPC-35-700?

[Ace25]

Floyd, the way I am doing my next build is using 12 LEDs per side on a 6"x6" screen (I know.. WAAAY overkill) but what I am going to use is a Meanwell 1050 constant current driver. Obviously 1050mA will blow up all my FEDY LEDs if I try and use it normally... soooooooo.. what I am going to do is run 2 parallel strings of 12, so I cut the output in 1/2, so 24 LEDs will each get 525mA, much less than the 700mA max they can handle. My thinking is I will have better coverage, less heat, more efficiency, and LEDs should last much longer. I am also going to use (per side) 420nm (1), royal blue (2), 630/640 reds (4), 650/660 (4) along with a warm white (1). Really mix it all up. :D

[Floyd R Turbo]

Do you have a link for the 420nm LED? That is what should really be paired with the 660s according to the chlorophyll curves...

[kerry]

I run the 12000 and 19000K from them (topled ebay), they are listed at 650ma and I run the meanwell 700ma drive for 16 bucks from rapidled with no problems. My next build will be their reds with a meanwell.

[Ace25]

Do you have a link for the 420nm LED? That is what should really be paired with the 660s according to the chlorophyll curves...
RapidLED sells them now... well.. they were in stock over the weekend when I ordered some, but today it seems they are out of stock. The quotes I got from "china" were about double the cost ($9 per LED) but they had dual emitters in them which is the reason they stated for the high cost. Still, the rapidLED light isn't my first choice, just the easiest for me. The rapidLED one states 410nm-420nm, ideally I would want a 420nm-430nm for photosynthesis, but hoping the rapidLED one is closer to the 420nm area than 410nm like with most LEDs. Royal Blue LEDs state 440nm-455nm and most of them are 453-455nm.
http://www.rapidled.com/violet-uv-3w-led/

[Floyd R Turbo]

Ok I did a little digging and here's what I've learned:

SatisLED uses the Epistar die for their 660. The Epistar dies are very high quality, good output (at least, when they came out, they were top of the pack) so these are OK to use.

LEDtoplights said their LEDs could run on 700mA but would last longer on 650mA (duh). I forgot to ask what brand they are, just asked.

FEDY got back to me on prices but I just asked them also what brand LED they use.

Ace I'm with you on the 420nm LEDs. I've been looking at spectral graphs and it seems that there is a little variance here on exactly where the peaks line up. Here's a few examples

These have been posted before on this site

http://i611.photobucket.com/albums/tt191/FloydRTurbo/Miscellaneous/ChlorophyllSpectrum.png

http://i611.photobucket.com/albums/tt191/FloydRTurbo/Miscellaneous/ChlorophyllSpectrum3.jpg

http://i611.photobucket.com/albums/tt191/FloydRTurbo/Miscellaneous/ChlorophyllSpectrum2.jpg

http://i611.photobucket.com/albums/tt191/FloydRTurbo/Miscellaneous/led_problem.jpg

Here's a few others I found just googling around

http://i611.photobucket.com/albums/tt191/FloydRTurbo/Miscellaneous/phorate.gif

http://i611.photobucket.com/albums/tt191/FloydRTurbo/Miscellaneous/chlab.gif

http://i611.photobucket.com/albums/tt191/FloydRTurbo/Miscellaneous/absorption-spectrum.jpg

http://i611.photobucket.com/albums/tt191/FloydRTurbo/Miscellaneous/8883n22_06.jpg

http://i611.photobucket.com/albums/tt191/FloydRTurbo/Miscellaneous/spectrum.gif

Some of these appear to put the blue A peak around the 420 point, some are hard to tell and probably not very scientifically accurate. However it seems that the 420-430 LEDs are a little hard to come by.

[Floyd R Turbo]

I am also noticing that the trend seems to be that the blue "B" peak and red "A" peak are the larger peaks. Perhaps this is why you see so many stock fixtures coming out that contain 455 and 660 LEDs.

Looking at these curves is also appear that the general consensus on the "B" blue peak is closer to 460nm or 465nm than 455nm.

[Ace25]

Here are a couple more graphs. Some graphs "appear" to show what your saying, the lower spectrum is closer to 460-465nm, but I have never seen it actually written out saying that, only the graphs appear to show that but they are far from precise. The bottom graph has numbers listed, but how accurate that is is anyones guess.

Here is the "photosynthetic response chart" that shows, at least in whatever example they tested this on, that photosynthesis uses quite a bit more spectrum than just the "4 major peaks". Again, I am always on the quest for "the perfect light", but for the majority that want simple and functional, I still say going with only 660nm will work just fine. This first graph is obviously a very "generalized" graph though, they really smoothed it out and I don't feel it shows a true accurate representation because there are small gaps in areas of blue and red that are not used and they smoothed over those. First graph is just more of a general idea graph just to show, at least in whatever plant they tested it on, that it does utilize a much broader spectrum than just the 4 peaks and surounding areas.
http://www.hydroponic-growing-systems.com/images/eye-wavelength-color-chart.jpg

http://www.biologie.uni-hamburg.de/b-online/ge24/03.jpg

[Floyd R Turbo]

More info regarding this...


SatisLED uses the Epistar die for their 660. The Epistar dies are very high quality, good output (at least, when they came out, they were top of the pack) so these are OK to use.

LEDtoplights said their LEDs could run on 700mA but would last longer on 650mA (duh). I forgot to ask what brand they are, just asked.

FEDY got back to me on prices but I just asked them also what brand LED they use.

FEDY got back to me, their 660nm and 440-450nm LEDs are from Germany (did not name manufacturer) and their 450-455nm is USA - Bridgelux

ledtoplights got back to me and theirs are "from a Taiwan company, don't know the brand". They're the more $ of the 3 anyways.

SatisLED has not gotten back to me on their RBs

[Floyd R Turbo]

Update

resonse from SatisLED "Royal blue LED use Epiled dies." EpiLED from what I can tell is in Poland.

Going to place an order with Satis I believe. Might try FEDY at some point too.

[Floyd R Turbo]

Talk about fast. I ordered from SatisLED on Monday and they just arrived today.

[kerry]

DHL is the way to go then.

[fajars2003]

can you post the pic of this Led from SatisLED when it's ON, i want to see the colour, it's red or magenta

[Floyd R Turbo]

It will be exactly the same look at any other fixture made with 660s and 455s

[kerry]

Its a purple glow. I have 4 660nm with 1 445nm and that side is purple-ish color.

[Floyd R Turbo]

Taking pics of these LEDs when they're on does not depict the color very well since LEDs mess with the color balance settings. They all look very similar, this is an e-shine fixture with 45 1W 660s and 5 1W 455s

http://i611.photobucket.com/albums/tt191/FloydRTurbo/2012%20Acrylic%20Projects/Algae%20Scrubber%20-%204%20Cube%20per%20day/DSC00218.jpg

http://i611.photobucket.com/albums/tt191/FloydRTurbo/2012%20Acrylic%20Projects/Algae%20Scrubber%20-%204%20Cube%20per%20day/DSC00219.jpg

http://i611.photobucket.com/albums/tt191/FloydRTurbo/2012%20Acrylic%20Projects/Algae%20Scrubber%20-%204%20Cube%20per%20day/DSC00220.jpg

^^that's my hand

[kerry]

I cant wait for that thing to start growing algae!!!

[Nmcgrawj]

Any updates?

[Floyd R Turbo]

I've got one running on a 4x6 screen with 6 660s and 1 455 on each side, looks like a little overkill, I had to put a piece of diffuser in front of the 455 and that seems to have helped. Still too early to tell. Will post some pics of that after this weekend, doing a few mods and tweaks...want to make sure it's right before posting. My OP said 1 3W for each 3x4 section, I bumped this up to one for every 4 sq in roughly or for each 2x2 section, running on a 700mA non-dimmable driver (meanwell LPC-35-700). The blue pulls much higher voltage drop than the reds so this might be why I needed to diffuse it. Might consider just diffusing the whole array so that it can be closer too.

[kerry]

I get good growth on my 6x6 with 4 reds and 1 blue and I am about 3" away. I have them in a square pattern with the blue in the middle with 120 degree lens's on the bottom two so the light does not flood into the bottom. The LED's themselves are labeled as 120 degrees but the lens makes a difference for some reason, maybe because it gets the focal point out an extra 1/2"-ish or so??

Hey Floyd did you see the pics I posted with the dual chip red?

[Floyd R Turbo]

Yeah what was with that? Are those FEDYs? Did they just send you a could duals by accident?

[DennisC]

I have had my LED's running for 2 weeks now and no green growth yet.

I have a 7x7 screen with 6 x 660 reds in a circle and a blue 455 in the centre. 1 each of this setup on each side of the screen 2" away. Does this sound about right?

Dennis

[Floyd R Turbo]

pics?

[Floyd R Turbo]

One thing I am coming across is that with a brand new screen it seems LEDs take longer to get the screen going. That e-shine scrubber I built was growing hardly anything in the middle near the bright light. My theory is the the fluorescent light had bandwidth in it that the diatom/brown liked and that grew first and slimed up the screen. LEDs don't, they just grow green primarily. Green takes longer to get started in SW. Therefore a bare screen starting on LEDs (which not many have done) takes longer to cure up because of the intensity and bandwidth. My solution was to back the hours way, way down. I'm talking less than 9 hours/day. Though I haven't tried this, it makes sense. Run the fixtures for 2 hours 2 or 3 times a day until you start to get some growth. Let the screen get really gooey and don't clean it. When you start to see green, then switch to a single period and start extending it. LEDs likely don't need any more than 9 hours per day if you have more than one per every 12 sq in. I have 6 on a 4x6 screen and run them 18, but I'm testing the UAS right now and just trying things out. for a downflow scrubber though I would think that you are meeting the double-light requirement and should not be lighting more than 9 hours/day.

[DennisC]

http://i953.photobucket.com/albums/ae17/haden869/Scrubber%20build/IMG_7030.jpg

These are my 2 sets of LED's before fitting.

http://i953.photobucket.com/albums/ae17/haden869/Scrubber%20build/IMG_7031.jpg

Fitted into my ATS (still dry)

http://i953.photobucket.com/albums/ae17/haden869/Scrubber%20build/IMG_7034.jpg

Not a very good pic, but this is how they look when ATS is running.

Dennis

[Floyd R Turbo]

Make sure your flow isn't more than 35 GPH/in until the screen cures. In fact, as long as it's flowing evenly across the screen, you can probably slow it down to 20 GPH/in. Your distance to the screen looks just fine. Nicely built fixtures also.

[DennisC]

Do I need to alter the timing. At present, 18 on 6 off.

Dennis

[Ace25]

I think 12/12 works much better than 18/6 on your/my type of setup.

[kerry]

Yeah what was with that? Are those FEDYs? Did they just send you a could duals by accident?
They are from TopLed, a mistake I guess. I would like to have more if the price is the same, they would be great for my display builds.

[joelespinoza]

This is valuable. One thing I'm noticing is that the Osrams (on the Rapid page) have a max current of 1000mA. The satisLEDs say 750. ledtoplights (ebay link above) say 650. So If I am wanting to use a meanwell driver, I could use the LPC-35-700 non-dimmable one for the osrams, but not sure about the other two. I would think those would need a dimmable driver, which costs more, which offsets about 1/2 the savings ($34 vs $16). So would I be OK driving the satisLED or ledtoplights LEDs with an LPC-35-700?

My scrubber has been barely functional for a long time, my screens are WAY to big, and I do not have NEARLY enough light concentration. Now that I went from a ghetto t8 100 watt fixture to a 140 watt LED fixture it has finally failed to keep algae from growing in the display now, even though I have not even brought my LED up schedule up all the way.

Here is what I have been using:
http://i18.photobucket.com/albums/b124/JoelEspinoza/0430121949.jpg
http://i18.photobucket.com/albums/b124/JoelEspinoza/0430121948.jpg

I just ordered:
12 of the 660 satisleds - http://www.satisled.com/3w-high-power-red-led-star-emitter-for-plant-grow-light-660670nm_p433.html
2 of these heatsinks - http://www.ebay.com/itm/1pc-Aluminium-Heatsink-8-3W-Aquarium-Led-Bulb-/150652979161?pt=LH_DefaultDomain_0&hash=item23139e07d9
1 Meanwell 700mA non adjustable driver off RapidLED - http://www.rapidled.com/mean-well-lpc-35-700-constant-current-driver/
2 Cree XTE Royal Blues off RapidLED

My plan right now is to wire them up and put them on the other side of the flourecent bulbs, later I plan to build a real enclosure (like a real boy) and mount it above the display tank. However right now I dont have the time or money to do so, so we will see how the combo of LEDs on one side of a screen and 2700k Curly Q flourecent bulbs on the other side do. Should be interesting to see.

[SantaMonica]

You could just turn one screen off and concentrate the light on the other.

[joelespinoza]

You could just turn one screen off and concentrate the light on the other.

You mean put both LED panels on one? I suppose I could do that.

Here was what my original setup looked like, but you said the T5HO lighting wouldnt do anything, and without the black plastic/syran wrap covering they were too hard to keep dry, so I removed them:
http://i18.photobucket.com/albums/b124/JoelEspinoza/PICT0001Medium.jpg


I know there is some talk about directing bubbles and clogging airstones etc, has anyone thought about just drilling small holes in a line in the airtubing itself? Or getting some of the clear plastic tubing you can fit an airline over and doing it to that? You might be able to control the flow direction better too.

Also has anyone tried a Upflow and downflow together? I know my waterfall screen seems to grow best at the top, and it looks like SMs UAS grows best at the bottom. The flow directions might fight each other too much, leading to too much turbulance, but I am tempted to try it. Here is a rough idea of what I am thinking for my tank:


http://i18.photobucket.com/albums/b124/JoelEspinoza/ASUAS.png

[kerry]

@ Joel, I Use the Topled's because I can get them fast and I use the meanwell driver you mention as well with no issues so far.

[daninfamous]

Just wanted to say thanks for all this information! Will be trying UAS with LEDS once my new tank is set up, (buying used waiting for him to sell livestock)

[Wopadobop]

do it straight out the gate too. makes a huge difference on cycle time. at least mine usually do. that and your not trying to throw a screen into a high nutrient ( oh s@#t) situation so you can skip the gooey black stuff. I didnt even get diatoms this time around.

still have cyano though (scratches head):confused:

[SantaMonica]

Point a powerhead near the area.

[Timo]

Whats your opinion on using violet 405-415nm instead of blue 455nm leds?

[Floyd R Turbo]

I'll be testing the 660/455 against 660/422.5 (Steve's LEDs "True Violet") at some point in the near future to see if this does indeed make a difference. But the general thought is that 455nm LEDs cover enough of both the A and B spectrum - there is really only a supplemental need for blue, as red is the highly dominant need.

[Timo]

I've just finished my scrubber last week and I'm using violet and red LEDs. It's only been running a week and has started to grow algae but I'm thinking now I should have gone with 455 blue insted, mabye.

[Timo]

This is a view of mine, two red 625-630nm on the outside and one violet 405-415nm in the middle. In the flesh the eye sees the middle Led as very dull almost as if its not working properly but when I took this picture with a camera (iphone) it shows the violet is just as bright as the outer red Leds.

3709

[rleahaines]

I am using a combination of red 660 nm LED's with some blue 455 in the middle. Getting really high growth where the blue LED's are vs. just the red.

long hairy algae where the blue/red combo light is.

Looks like the combination is better than red alone.

[Floyd R Turbo]

Yes Ace or Kerry did the same test and got the same result. The observation he made was that the blue enhanced the growth, as in it made it bulkier or 'sturdier' - harder to scrape. But most people that have compared the two (with and without blue) have noticed the same effect.

[Timo]

I took my screens out today and the side with the violet and red LEDs had much less growth than the side with blue and red in during my 2 week trial so the violets are coming out as soon my new red and blue LEDs arrive from china.

[Floyd R Turbo]

Keep in mind that one blue 455nm LED will have much higher intensity than a 422nm Violet. So you might need 2 violets whereas you only need one blue. Hard to say for sure without doing some experimenting, but that might explain the difference.

[Garf]

I took my screens out today and the side with the violet and red LEDs had much less growth than the side with blue and red in during my 2 week trial so the violets are coming out as soon my new red LEDs arrive from china.

I would have given it a bit more of a chance to be honest. The algae types may just not be acclimated to this light. It may get better, the more cycles you put it through.

[Timo]

I would have given it a bit more of a chance to be honest. The algae types may just not be acclimated to this light. It may get better, the more cycles you put it through.
It was a brand new screen an scrubber. I only used the violets as I had them kicking around (I got the violets to make a light for the main tank but never got round to using them) and I could only salvage one blue from my old scrubber. Planning now to use red 630 and 660 + a blue 455 as that combo worked great on my old scrubber.

[Floyd R Turbo]

The true test would be to use it on a seasoned scrubber for a couple of months. Thanks for adding that bit of info - I'm still planning to bench test a scrubber with 660/455 against 660/422.5

[Timo]

The true test would be to use it on a seasoned scrubber for a couple of months. Thanks for adding that bit of info - I'm still planning to bench test a scrubber with 660/455 against 660/422.5
Also I think my violets are a little to far down the scale at 405-415nm mabye too UV, your test with 422.5nm may have totally different results. Algae did grow it just looks stronger,thicker and greener on the blue screen. BTW my scrubber has 4 separate screens in it.

[Garf]

Also I think my violets are a little to far down the scale at 405-415nm mabye too UV, your test with 422.5nm may have totally different results. Algae did grow it just looks stronger,thicker and greener on the blue screen. BTW my scrubber has 4 separate screens in it.

Why not start a new topic, I'm intrigued, for one.

[xerophyte_nyc]

Keep in mind that one blue 455nm LED will have much higher intensity than a 422nm Violet. So you might need 2 violets whereas you only need one blue. Hard to say for sure without doing some experimenting, but that might explain the difference.
Humans don't see violets that well so they appear less intense than blue, but as far as LEDs, violets typically have a higher forward voltage and need to run on lower amps than blues because of heat issues so it is true that more violet diodes may be needed for a fair comparison if you want to compare equivalent power but even that has all sorts of variables like quality of manufacture.

[Enigma84]

Without having read all of this thread top to bottom, i wonder whether it might be possible to made a hard and fast rule regarding led power to square inches on scrubbers, as 1 x 3w for 3" x 4" seems far too low, and it had resulted in buying various different drivers and more leds to upgrade on a trial and error basis. I cant help thinking it would have saved me a fortune if there had of been more accurate guidance to start with. I will start with what i have come up with, but wonder whether others may suggest power to surface area so we can get a more up to date idea of what is needed. With the upgrades i have made so far, and basing it on nuisance algae growth in tank/phosphate and nitrate level of the water, i would say that just under 1w per square inch on the 660nm 3w leds is closer to wiping the tank clean of all nuisance algae and nutrient levels. From looking at the L2 algae scrubber also, with the leds used versus the 4" x 6" screen, this seems to be a just under 1W per square inch per side. I would like others to share their experience as i see many new starters, with grossly underlit screens. Adding more leds is easy enough but getting bigger heat sinks to cope with heat and different drivers to cope isnt cheap.

[Enigma84]

having said that, i have just found https://public.sheet.zoho.com/public/cdm2013/ascalculator-xls-1 on the sticky page which wasnt there before.
As long as you go closer to the maximum than the minimum, then this should be more accurate.

[Floyd R Turbo]

Without having read all of this thread top to bottom, i wonder whether it might be possible to made a hard and fast rule regarding led power to square inches on scrubbers, as 1 x 3w for 3" x 4" seems far too low, and it had resulted in buying various different drivers and more leds to upgrade on a trial and error basis. I cant help thinking it would have saved me a fortune if there had of been more accurate guidance to start with. I will start with what i have come up with, but wonder whether others may suggest power to surface area so we can get a more up to date idea of what is needed. With the upgrades i have made so far, and basing it on nuisance algae growth in tank/phosphate and nitrate level of the water, i would say that just under 1w per square inch on the 660nm 3w leds is closer to wiping the tank clean of all nuisance algae and nutrient levels. From looking at the L2 algae scrubber also, with the leds used versus the 4" x 6" screen, this seems to be a just under 1W per square inch per side. I would like others to share their experience as i see many new starters, with grossly underlit screens. Adding more leds is easy enough but getting bigger heat sinks to cope with heat and different drivers to cope isnt cheap.

Thanks for the valuable feedback. I developed the LED guideline based on what people had done and had success with. The "minimum" of 1 3W LED on each side of each 4x4 section (16 sq in, or 6w/16 = 0.375 w/sq in total, or 0.1875 w/sq in on each side) always did seems pretty low to me, but there were those that had successful growth with that level of light. How long did you try your setup before deciding to change the lighting?

My Lx series scrubbers generally use 6 3w LEDs on each side, so that's 18W x2 = 36W on a 24 sq in screen, or 1.5W/sq in total (0.75 w/sq in per side) if you don't count the blues into the calculation (and I don't), and that is the coverage you get when you have one LED per side of each 2x2 section. This is actually 4x the "density" of the "minimum".

So perhaps it is worth bumping the minimum up to one 660nm Deep Red on each side for every 8 sq in. So for a 2 cube/day screen, 24 sq in, that would be 3x 3W on each side, 6x3W=18W total / 24 sq in = 0.75 w/sq in total (0.375 w/sq in on each side).

Let's not get too confused with the "total" vs "each side"..."total" is easier to explain and less room for error.

So the revised rule would be:

Minimum: Screen size (LxW) times 0.75 = total wattage of LEDs needed. so 24 sq in x 0.75 = 18W, which is 6x 3W LEDs, 3 on each side.

Maximum (high-intensity): Screen size (LxW) times 1.5 = total wattage of LEDs needed. so 24 sq in x 1.5 = 36W, which is 12x 3W LEDs, 6 on each side. This is how I build mine, and I also include a diffuser, which I am recommending that people leave in place until the screen cures, then removing it if they chooses (the working theory being that after the screen cures, it can take more intensity)

[Enigma84]

The revisions you have are spot on i would say. I ran a 10" x 8" screen with 10 x 3w 660nm per side for about 5 months. Then upgraded to 19 per side which i have now, with some very good growth. My nitrates and phosphates are zero but there is still some brown diatom algae, so i am gonna push it a bit further to finish the last of it off. This will be 6 lots of drivers i will have bought now, 4 of which i will use (two per side). I think that people worry about having too much light, so it is good to see a maximum as well as a minimum, but it should be stressed that the higher end is needed to really get rid of the filth. It might be worth adding that my screens are about 5 inches away from the mesh surface, so i might have to use just a bit more.

[Floyd R Turbo]

5" distance might explain part of it. 3" probably would have produced better results. For reference mine are 2" away with a diffuser at about 1"

[inkidu]

Has it been 4 years? Wow. And still no water changes.
I got an email, and was checking out SM's new surf8X. Great work.
FW
Lost a couple led I see. Happy I checked now.
That is thing about led and algae scrubbers.
It is easy to forget about them.
I tried aeration from below screens at one time, and I did have some green success, but in my horizontal the long air stone got in the way.
I will rotate the three separate mesh/screens i.e. I will pull one at one end, clean, and then replace back at the other end.
I switch to Scoth Brite pads (white) in FW they float well and thought I would just give it a try. So far so good. Year or more, lost track.
I got a lot of catch up post reading to do, just thought I'd post.

[SantaMonica]

Welcome back.

Yes I too will forget about the scrubbers sometimes. By the time I check, the LED lid has 6 inches of growth attached to it.

Glad you horizontal is doing what it needs to do :)

[inkidu]

Thanks SM,
I am noticing now, in my pictures, were I was really neglectful.
The brown algae up the sides above the water line (backed up) in the scrubber.
From what I remember the brown algae (FW) is no good. Probably from not cleaning.
The tank always was running but I only had a few fish swimming around.
Goes to the ability of the scrubbers. Even through complete neglected my only filter was still working.
Any thoughts on whether I should clean the brown algae off or just let it go?
Might correct itself if I stay current with cleaning it and not let the water back up.
Thanks

[SantaMonica]

As long as you harvest it, it's good for filtering. If you don't harvest, it does not filter but does still make pods.

[atoll]

I considered building my own LED unit but after research decided it was hardly worth the effort in truth after I found others are using these off eBay.
http://www.ebay.co.uk/itm/221708885323?_trksid=p2060353.m1438.l2649&var=520555366810&ssPageName=STRK%3AMEBIDX%3AITMy

leaflet box scrubber immediately after installation and prior to installing my LED units.
http://i541.photobucket.com/albums/gg378/lesmelling/Aquareef%20300/20150704_200809_zpsxwncg1ql.jpg


My screen size is 6" wide by 7" long.

One of mine mounted on my DIY supporting rig.
http://i541.photobucket.com/albums/gg378/lesmelling/Aquareef%20300/20150714_174706_zpskmtfargt.jpg

The two mounted on my sump in front and beghind of my leaflet scrubber. Skimmer cup removed to get a better shot of it.
http://i541.photobucket.com/albums/gg378/lesmelling/Aquareef%20300/20150718_110031_zps3xnzjrh2.jpg


A 10w version is just $12 and the 20w $23 They come with the right 650/660nm LED's and a couple of 440/450 blues. They are also splash proof, come with heat sink slim casings, mounting brackets and can be adjusted into different angles.
http://i541.photobucket.com/albums/gg378/lesmelling/Aquareef%20300/_57_zpskycmlvq1.jpg

[SantaMonica]

Those look to be the best on ebay. Does it come with power supply?

[atoll]

Those look to be the best on ebay. Does it come with power supply?
Yes they do it is built in in casing. The only downside is you only get a short length of power cable about 9". I bought 2 tubular waterproof connectors to extend them.
these ones.
http://i541.photobucket.com/albums/gg378/lesmelling/Aquareef%20300/T2eC16VyEFJOgOUHqBSYUFbgSg--60_12%201_zpsfag3oihc.jpg

[b4tn]

I considered building my own LED unit but after research decided it was hardly worth the effort in truth after I found others are using these off eBay.
http://www.ebay.co.uk/itm/221708885323?_trksid=p2060353.m1438.l2649&var=520555366810&ssPageName=STRK%3AMEBIDX%3AITMy

leaflet box scrubber immediately after installation and prior to installing my LED units.
http://i541.photobucket.com/albums/gg378/lesmelling/Aquareef%20300/20150704_200809_zpsxwncg1ql.jpg


My screen size is 6" wide by 7" long.

One of mine mounted on my DIY supporting rig.
http://i541.photobucket.com/albums/gg378/lesmelling/Aquareef%20300/20150714_174706_zpskmtfargt.jpg

The two mounted on my sump in front and beghind of my leaflet scrubber. Skimmer cup removed to get a better shot of it.
http://i541.photobucket.com/albums/gg378/lesmelling/Aquareef%20300/20150718_110031_zps3xnzjrh2.jpg


A 10w version is just $12 and the 20w $23 They come with the right 650/660nm LED's and a couple of 440/450 blues. They are also splash proof, come with heat sink slim casings, mounting brackets and can be adjusted into different angles.
http://i541.photobucket.com/albums/gg378/lesmelling/Aquareef%20300/_57_zpskycmlvq1.jpg

Any update on growth from these lights? I am looking to re design my scrubber to a more encased design with LED's. For the price it would be cheaper to buy these than to build my own.

[atoll]

Any update on growth from these lights? I am looking to re design my scrubber to a more encased design with LED's. For the price it would be cheaper to buy these than to build my own.

Yes its working very well. After a series of different coloured alages growing I now have GHA all be it quite drak green. Here is a pic of the screen just prior to cleaning. taken a few weeks ago
http://i541.photobucket.com/albums/gg378/lesmelling/Algae%20Turf%20Scrubbers/20151108_192718_zpsvqhyi5ln.jpg
This ihttp://i541.photobucket.com/albums/gg378/lesmelling/Algae%20Turf%20Scrubbers/20151201_124029_zpsaaxwa5mc.jpgs a harvest of a small handfull today .