Ace's ATS v2.0 - LEDs

[Ace25]

So this is my second scrubber which will be installed tomorrow. I am replacing my old ATS/sump with this new and improved model after learning the do's and don'ts from version 1.

55G tank. 1st chamber 18" long, 1-1/2" overflow pipe feeds the ATS, screen is 12"x10.5". Chamber is painted black all the way around and bottom, and top will have a lid. This is to keep all light contained in the chamber and not grow algae all over the rest of my equipment. 1-1/2" pipe reduces down to a 3/4" with a valve as a safety overflow. I hope the lip that the reducer creates will be enough to give sufficient flow over the screen and allow my to keep the valve open for safety. LEDs are 6 LEDs per side, 50/50 mix of Cree Neutral and Warm whites using a meanwell 750mA constant current driver. 3 LEDs mounted per 10" of C channel aluminum inside a glass box with a lid over the box that will have 2 1.5" fans to keep air circulating in the box. Crossing my fingers this all works as planned.

http://farm6.static.flickr.com/5145/5638737141_a2826da600_b.jpg

http://farm6.static.flickr.com/5228/5638737127_c4e695e31e_b.jpg

http://farm6.static.flickr.com/5181/5639312864_865d8055f6_b.jpg

http://farm6.static.flickr.com/5223/5644678292_b8c3f99d46_b.jpg

http://farm6.static.flickr.com/5066/5644111793_cba21b69b0_b.jpg

[vicbay]

great build. very nice job. keep us posted please

wondering why ya choose to mix neutral and warm white leds instead of all warm whites? and how did you decide on 6 per side?

and do you have a thread or info on the first version?

thanks

[Ace25]

I chose those LEDs because it was just what I had laying around. LOL. I also have some cool whites and royal blues but I didn't think those would be as good to use. I chose 12 LEDs total because that is what one driver does, and it also comes pretty close to matching the PAR output of a 42w CFL at the screen.

My first ATS was more of trial run to learn what to do and not to do.. biggest lesson, don't put equipment in with the ATS, like pumps, live rock, heaters.. they all just get algae growing all over them and clogging the pumps. I tested this using a rubbermaid container as the sump/ATS.

http://farm5.static.flickr.com/4080/4866545539_70ef032c9e_z.jpg

http://farm5.static.flickr.com/4141/4867161586_0fd15e4b1d_z.jpg

[Ace25]

Here it is installed.

http://farm6.static.flickr.com/5150/5652096452_61fbe6f990_z.jpg

http://farm6.static.flickr.com/5029/5651529855_47300c0f58_z.jpg

Video of flow over screen (since this site can't display youtube videos for some reason here is the link)
http://www.youtube.com/watch?v=hnb71BY8Y6w

[maglofster]

Looks very nice! Looking forward to see results!

[mrbncal]

Looks great, could you post a video of how you clean it?

[Ace25]

Just pull 2 velcro tabs and the screen drops right out. Used the same velcro method on my last ATS and the the velcro lasted fine the entire time and never gave me a single problem.

http://www.youtube.com/watch?v=9wGllg25vvQ

[rygh]

Some thoughts:

It seems like a bit risky of a setup using your overflow, with ball valve on the end like that.
I worry algae will clog up slot, and if that valve is not open enough, it will back up, and into main tank.
Yet if valve is not closed enough, you lose too much flow.
You might want to run a quick test with duct tape over the slot.
As a fix (if it is a problem), run the emergency overflow up in the air, maybe 5" or so, then back down.
Basically, gravity becomes the valve.

I would be a bit surprised if you get optimal growth with that mix of LEDs.
Seems both under powered and not enough red.

But I really like the build. Nice design, well built, and yay, LED.

[Ace25]

The valve is open all the way on the overflow, no water comes out right now, all water is going down the slot. If it does start to clog the overflow pipe will come into effect. I ran my old ATS for a year off my overflow in the above picture, never once had a problem with screens clogging.

As far as the LEDs, we will see.. I have plenty of different ones to test with and I do have my own PAR meter and a $3000 spectrometer at my disposal. I do think there is more than enough reds in the warm whites I have, I am more concerned that I don't have enough blues actually.. but I have plenty of time and parts to refine my selection should the need arise. Right now I am getting just over 1500 PAR at the screen with LEDs and have more reds in the spectrum than the old 2700k CFLs I used to run. I was getting 1493 PAR with a 42w CFL at the screen.

http://farm5.static.flickr.com/4081/4765714269_b912d07dc6_z.jpg

[vicbay]

yes blue was my concern as well. thats why i asked if you had reasoning behind your led choices. i think the nuetral white spills a lil bit more into the blue end. but im only looking at the cree graph. and estimating.

mind trying that 3000 piece of equipment to work? lol

also it looks like their is about twice the red to blue(ideal ratio i believe) in the warm whites. so i was gonna try just those on mine.

are you using any optics on your leds? if so what angle? i bought 40 60 and 80 to try. just got them in today though so no chance to play yet

[vicbay]

Btw what model crees are you using? xp, xr-e, etc.

and would you be so kind as to post a front pic of the leds to see how they are spaced out?

have you checked temp on heat sink being enclosed like that? and what heat sink did you use?

your light cycle is 18 on and 6 off right? didnt see it posted.

hope im not pestering ya too much but you beat me to a build like i had planned so i just want to pick your brain and copy what your doing.

thanks for sharing

[Ace25]

No optics on my LEDs, they are already close enough to the screen, any optics would make it way to strong and focused. Right now I get almost perfect spread, just the outer corners are slightly dimmer than the rest of the screen. Optics should only be used when the LEDs are at least 12" away on a tank, even then with 60s you still get some spotlighting/color bars from the optics. I have played with all of them on both XR-Es and XP-Gs to see how optics help LEDs. They do increase PAR A LOT (too much in most circumstances) but the trade off is very focused light and not much blending of colors.

I have been thinking more and more about taking off a neutral white and replacing it with a royal blue, maybe just do it on one side so I can see if there is a difference between growth on each side.

The spectrometer I have at work. I have borrowed it a few times but it is easier for me to bring bulbs I want testing into work so that is what I have been doing with LEDs and CFLs. I brought it home when testing different T5s and MHs in my canopy to try and determine replacement age of bulbs based on spectral shift instead of PAR (intensity) or time. I have a 48 LED light over my 75G and a 24 LED light over my 29G that I made. The 48 uses cool blues, royal blues, cyan, cool white, neutral white, and warm whites, my 29G is just royal blue and cool whites.

The LEDs are 3 per 10" piece of 2"x1" C-Channel aluminum, one dead center, the others are 3" out from center. Right now I am using XR-E LEDs because it was what I had left over from my original 48 LED build, when XP-G neutral/warm whites were not out yet. I do like the size/lense on the XR-Es vs the tiny XP-Gs though, I think it helps with the spread. I think the XP-Gs are around 120degree and XR-Es are 100degree, so slightly more focused, just not as energy efficient. The heat, yes, is a concern. I am making lids for the tops of the boxes right now and using 2 1.5" fans on each side with holes drilled in order to circulate air in the box. Without fans it gets up to 120 degrees on the heatsink, with fans they stay almost room temperature, so fans are required and I use fans on all my builds.

Yes, 18 on / 6 off.. just started though with a new screen, so it will take a little while to fill in. I remember my first one taking about a month to get a good base down and 2 months to really start working good.

Don't worry, your not pestering me at all.. I like to help.

[vicbay]

48 on 75g. sounds nice some builds i see use over twice that. do you have a build thread on it and/or the 29?

and for your heatsink. is it a finned design or a solid plate or angle shape? thickness?

sry. just reread thread. on my phone so couldnt before i posted. c-channel. 1 x 2. i assume its standard channel right? not heavyweight

[Ace25]

Ya, standard aluminum C-Channel, not heavy duty. On my other LED lights I used finned heatsinks. Here is my build thread for my LED light.
http://www.aquariacentral.com/forums/sh ... -going-LED (http://www.aquariacentral.com/forums/showthread.php?239281-Finally-going-LED)

Just to give you an idea of the difference between 48 LEDs and 375w of 10K MH light

This is an XM 10K 250w MH bulb, run on a PFO M80 magnetic ballast, so it overdrives the bulb to 375w, This is the PAR output at 12" from the light.
http://farm3.static.flickr.com/2473/3673337517_de8a27cb28_z.jpg?zz=1

Now compared to 48 LEDs, same distance from light
http://farm6.static.flickr.com/5046/5301367855_d3678199b7.jpg

Conclusion, 48 LEDs is about 2x stronger than a 400w 10K MH bulb.

[vicbay]

thanks. eleven pages. guess i got some reading to do now at work lol

two quick questions

is spectrum same from the mh to led and therefore the reason for wide array of leds in 48 build?

have you grown frags in 29 and 75 from same colonies? and if so. did you notice a difference in growth or color that you could attribute to having only two types of leds on 29 versus the 6 on the 75 or was that just for aesthetics?

ok two and a half lol

[Ace25]

The 48 LED build was my first.. and I was trying to cover the most spectrum possible. Looking at them under a spectrometer, there was a large gap between royal blues and Cool whites between the 470-520nm wavelengths, upper blues and Cyan were missing in the spectrum. Also the entire spectrum under 450nm was non-existent, so I have 4 420nm T5HO Actinics on my 75G to cover the low end of the spectrum. This is one area I really think LEDs need a new breakthough in, blues. The whites are great, put out more than enough PAR and have a great spectrum, now we need a blue LED that isn't monochromatic, more like a blue that covers 420nm-500nm range, that would help out a ton. The rest of the spectrum coming from the LEDs, especially cool whites, are a much better spectrum overall compared to most MH bulbs, which are actually very rigid/lots of peaks under a spectrometer. White LEDs are a much smoother curve than a MH. Only colored leds are pretty much mono-chromatic, so royal blues put put a very narrow spectrum, and now the thinking is it caused HEV, meaning too much photosynthetic light at the chlorophyll B absorption point, leading to over saturation and doing more harm than good to corals. Problem is, you need a lot of blues to balance out the whites if your like me and like the 20k look. If you add more royal blues to counter the whites, you end up putting way too much of that specific wavelength onto your corals, so I chose to go with cool blues and cyans instead of doubling up on royal blues.

Honestly... after all the designing of my 48LED light, I can honestly say that my 24 LED light is actually growing SPS corals better. Not entirely sure why that is but it is a puzzle I plan on figuring out.

[rygh]

Regarding blue versus red needs:
I ran an experiment on my old turf scrubber, where I replaced one of my Red/Blue/Red LED clusters with
one of the Cool white + royal blue from my display tank cluster.
The RB/CW definitely did not grow as well as the one with the R/R/B.
See middle of page 23 on my first LED thread.
viewtopic.php?f=3&t=435&start=220 (http://www.algaescrubber.net/forums/viewtopic.php?f=3&t=435&start=220)
Now that does not say blue is not important, and it might have been a bit overloaded with blue, but it is a useful data point.

---
Second thing : Go look on various sites where they grow algae for biofuel with LEDs.
Almost all of them use Red+Red+Blue.

---
That said, I believe it will grow OK under whites. Simply because you do have decent PAR numbers.
But fun to optimize.

--
PS: I am jealous of access to that spectrum analyzer.

[Ace25]

I appreciate your input rygh and I do agree with you. The 2 photosynthetic peaks are 453nm and 660nm, blue and red, so it makes sense why those 2 colors would be chosen to grow algae with. I just didn't have any reds on hand and since I love to tinker and test all things lighting related (I am a reef geek but an UBER lighting geek) so I figured I would try out what I had to see how well they do. Neutral whites have a good amount of blues in them and warm whites have some reds (although they do not reach up to 660nm, which I can see being a potential problem). So I agree with you 100% that my LED choices are not going to be the most ideal choices for maximum growth, but I wanted a starting point. I don't want to have happen what happened when I put my LED light on my 75G.. nuked the crap out of my SPS corals at first. Even matching PAR for PAR between LEDs and the MH it replaced, the LEDs put out way more 453nm than the MH did and that nuked my corals. PAR is only one piece of a puzzle, knowing the true spectrum of the light used I think is more important than PAR. PAR only equals intensity of the light within the 400-700nm range... so in theory you can have 2000 PAR coming out of all GREEN LEDs, and not be able to grow a thing with them even though you have the PAR output that matches the sun.

One thing about BioAlgae for fuels it is a freshwater algae they use from everything I have read, so it has slightly different requirements in lighting vs marine algae in my opinion. Different topic to get into but that is one area I can talk for days about because I do think BioAlgae is going to save the planet vs things like corn, palm, etc. That one topic gets me more excited than any other topic right now in my life but this is not really the thread for that. I think even my co-workers are sick of hearing about BioAlgae as an oil replacement because I talk about it constantly at work. ;)

The question I have now though is "what is the correct spectrum for maximum growth of marine algae". I do believe the ideal spectrum for marine algae is going to be different than FW algae and really different than terrestrial plants due to the living environment that marine algae comes from. Anything more than a couple inches under saltwater gets a large majority of the red spectrum filtered out (90% at 12") so if algae grows in the ocean at depths deeper than 12", chances are it is not receiving much red light, unlike terrestrial plants. Looking at my tank now, I can say by the daily handfuls of algae I am pulling out lately, that LEDs will grow algae very well, much better than my T5s and MH ever did, unfortunately that wasn't what I wanted to grow, I was trying to grow corals. LOL. I do believe over the eons things like marine algae will adapt to their environment in order to utilize the light spectrum it receives as efficient as possible. So I think marine algae will not utilize the reds as much as a lot of people think it will. It will still use it, I just don't think the standard photosynthesis chart you see that pertains terrestrial plants are an accurate representation of what marine algae or corals will utilize most efficiently.

Bottom line is, my current Neutral/Warm white LEDs put out a better spectrum and the same PAR as a 42w 2700k CFL bulb and I get better coverage so in theory my current LEDs should blow away any CFL ATS out there.. but I do think there is plenty of room for improvement on my end with different LED choices. I just wanted to start "safe" and progress upwards, instead of going the other way like I did on my display LED light, start at the top, nuke the corals, and work backwards to get back to acceptable lighting. Last thing I wanted was to prove Santa Monica right in that LEDs just make burn circles on screens and can't grow algae, which is entirely possible if the LEDs are not setup correctly/lenses used to close to screen. :)

[rygh]

Yes, I know what you mean by wanting to just get it done.
My latest system is a bit sub-optimal as well, because of the LEDs I had around. May improve it later.
I should have my big new diplay tank LEDs done in a week or so, and after that might
have some time for more tinkering.

Yep, agree with you on the whole PAR thing.
BTW: Make sure to consider the LedEngin deep reds. They are one of the very few that really
reach up into the correct upper-red spectrum. Interestingly, they are not a red LED.
They are like whites, in that they use a different base LED, and then phosphors to output.
And as such, seem to have a wider spectrum as well.
(As well as my cheesy diffraction grating spectrometer shows)

You have a VERY interesting point about freshwater versus saltwater algae.
I never thought about that. Need to look into it.
Although I would be a bit surprised if fresh/salt hair algae is very different, as opposed
to coraline, kelp, and things that are more specifically marine.

As far as blowing away T5/MH. Well, hope so.
I actually think the "perfect" LED will simply equal the "perfect" CFL/T5 as far as growth.
Simple because I don't think lighting will be the limiting factor, once done right.
But it will certainly blow it away from a power + long term cost standpoint.
And hey, I would be quite happy if it ends up better.

[rygh]

Funny, I was reading your DT LED build thread, and it looks
like you noticed the LedEngine parts already.
Never mind.

Off topic tidbit: I used both Rebel and Cree royal blues, to help spread
out the blue spike. They are fairly different in spectrum, especially when
bin sorted.

[oildalemonkey]

Hey Ace! Remember me from CCRK? Nice looking unit you have constructed: I hope it works well for you. I was curious how you handle water changes these day? Do you do them at all anymore?

Also, I just wanted to say thanks for getting me onto the ATS track a while back. Both my tanks are now ATS filtered, including one with an SM100-style unit I built myself. I don't see myself ever going back to the traditional filtering methods. :D

[Floyd R Turbo]

I just joined that site with your LED build thread so I could read it last night. Very nice! It also looks similar in LED selection to one I saw on RC, I can't seem to find it though. The build had very few CW/RB and had a mix of cyan, WW, NW, etc. I of course saw that thread after I had ordered 54 RBs and 18 CWs. So now I have to decide what to do!!!

[Ace25]

Hey Monkey.. ya I remember you.. I "moved on" to ccreefs, which you know who won't even allow anyone to speak of that site on his forum, go ahead and try and type it out and see what happens. LOL.. but anyway, good seeing you here and awesome that you made a SM100 scrubber. I am now in the process of making a SM25 type scrubber for my 29G tank. I think I will go with a CFL bulb on that one though like Santa Monica uses because it seems a little difficult to control heat on LEDs on that type of scrubber, but the design will work perfect on my 29G setup.

Floyd, thanks for the compliments.. I guess I should have updated that thread at the end to state that I went back to my original plan with the LEDs .. since it was my first build I made a few mistakes as you probably read, mainly on the voltage for the dimmer. Not knowing about that problem I replaced some neutral/warms with cools to get more PAR, but after I figured out that I was only running 2.5v out of 10v to the LEDs on the dimmer circuit I ended up putting the warms and neutral back on, but ended up with only 50% of the LEDs having 60 degree optics, the other 50% do not have any. I feel that combination of optics on my tank gives me more than enough PAR (1200+, not the 1800+ I was getting with optics on all of them, but still 3x what a Radium and Phoenix 250w bulbs put out). So it is still way overkill but I feel it gives me the perfect looking light for my preference. Perfect blending and good intensity. I currently run the neutral/warm whites and 40%, cool whites at 30%, Royal blues at 50%, and cool blue and cyans at 60%. I do not get any "windex" look from the cool blues like everyone told me I would get, nor do I have any issues with Cyans, although if I put optics on the Cyans it does give me ugly beams of cyan in the water, without optics it just fills in the missing color spectrum and in my opinion looks really nice and makes the green corals pop a little more.

I certainly won't claim to be an expert on LEDs but I have now played with quite a few of them and all kinds of combinations so I feel I can give advice to others if they want it. I also built a 84 LED "actinic" light for a friend that has 3 14" SolarTubes over his 240G to replace his T5HO actinics. He went with "china" LEDs and driver, driver died within 30 days like I thought it would, and china LEDs, while they have almost matching spectrum as Cree Royal blues, they are not built nearly as well and can barely handle 750mA. So I am not a fan of "cheap LEDs" after experimenting with them some. To me it is like buying a $10 MH bulb on Ebay vs a major brand bulb for $50-$100. The $10 bulb only lasts a month or 2 and the spectrum shifts very quickly on them. Just throwing that out there because I know quite a few people now are looking to save a $ on things. I would not recommend trying to save a $ on LEDs though by going with china LEDs. Save some $ by going with ledgroupbuy.com and get real Cree's for LEDs. As far as drivers... mean well's do work great, but they are certainly not the most efficient driver out there. The overhead on the driver is almost as much as the LEDs themselves, ie. 12x 3w LEDs should take 36w of power, with a mean well driver, it takes around 58w total, driver is not efficient at all and you can easily tell that by how hot they get. I have heard of a couple new ones just coming out that are supposedly much better than Mean wells, I am sure some people on RC and Nano-reef have talked about them, I just can't think of the brand off the top of my head right now.

[rygh]

Dang, wish I had heard of that ledgroupbuy before.
But I did get a pretty good deal from cutter, and they do bin sort.

I totally agree with the china-led choice. I bought some cheap warm white ones for my scrubber, and definitely regret it.

I may try Cyan on my main tank as well. I have RB/Blue/CW/NW/Red, but it still seems a bit lacking the the blue-green area.
The standard blue helps, but not quite enough.

[Ace25]

Couple pics

ATS v1 @ Day 4 using 42w 2700k CFLs from above (not ideal mounting, but they were literally 2" from the screen at the top, so very close but not directly facing the screen)
http://farm5.static.flickr.com/4137/4772805306_b0095d928e_z.jpg

ATS v2 @ Day 4 using LEDs. Visually, in person, I would say I have about 2x as much "stuff" on the screen at day 4 with LEDs as I did with my CFL setup, again, I understand design plays a big role in it.
http://farm6.static.flickr.com/5181/5667096856_188d4041f2_b.jpg

[Ace25]

Day 5 picture

http://farm6.static.flickr.com/5023/5671009769_6ab6f649be_b.jpg

[inkidu]

One thing about BioAlgae for fuels it is a freshwater algae they use from everything I have read, so it has slightly different requirements in lighting vs marine algae in my opinion.

I do believe the ideal spectrum for marine algae is going to be different than FW algae and really different than terrestrial plants due to the living environment that marine algae comes from. Anything more than a couple inches under saltwater gets a large majority of the red spectrum filtered out (90% at 12") so if algae grows in the ocean at depths deeper than 12", chances are it is not receiving much red light, unlike terrestrial plants. Looking at my tank now, I can say by the daily handfuls of algae I am pulling out lately, that LEDs will grow algae very well, much better than my T5s and MH ever did, unfortunately that wasn't what I wanted to grow, I was trying to grow corals. LOL. I do believe over the eons things like marine algae will adapt to their environment in order to utilize the light spectrum it receives as efficient as possible. So I think marine algae will not utilize the reds as much as a lot of people think it will. It will still use it, I just don't think the standard photosynthesis chart you see that pertains terrestrial plants are an accurate representation of what marine algae or corals will utilize most efficiently.

Please start a separate thread pertaining to these ideas. (I will not get tired of hearing it)

What obviously is needed is the implementation of ideas.
Thanks and good luck with the build.

I doubt that Rygh would mind if you added to the Modern Led Scrubber Light thread.
I have already taken that thread far off the original thread topic and he started a new thread pertaining to his build.

[SantaMonica]

Not bad LED growth for 5 days if it's a new screen.

[Ace25]

Yup, new screen.

Day 7 picture before cleaning.. didn't grow to much more past day 5, just started growing that brown slime stuff, which I got on my first ATS as well. I know that is completely normal on some setups, obviously mine, and after a couple weeks the brown slime goes away and algae starts to really take off. So I will post weekly updates before I clean the screen to show how it is filling in week to week. There was quite a bit of "stuff" that came off the screen when I cleaned it, much more than my first ATS during the first week, so that is a good sign I think.

http://farm6.static.flickr.com/5141/5681374279_ef2362ecc0_b.jpg

[vicbay]

nice growth. any progress on the fan cooling?

im thinking of trying a long version like the sm100. but with leds ofcourse. was it just the 55g sump that dictated your screen kinda square? or flow concern with a wider screen?

how much were you feeding with your cfl scrubber? are you still feeding the same?

[Ace25]

I ordered 4 of these fans from newegg (http://www.newegg.com/Product/Product.aspx?Item=N82E16835119046&nm_mc=TEMC-RMA-Approvel&cm_mmc=TEMC-RMA-Approvel-_-Content-_-text-_-) over the weekend so they should be here this week. Once I get them in I will install them on the lids and take a picture.

If your making a SM100 clone, that should be really easy to swap out the T5s with LEDs. You can either go with a real heatsink to fit from www.heatsinkusa.com (http://www.heatsinkusa.com) or just a couple pieces of Channel Aluminum, either way it would be very easy to mount a couple 80mm fans on the backside to keep the LEDs cool.

Few things factored into my screen size... but mostly a lot of guessing and crossing my fingers it worked. Before I ran 2 10" screens on my ATS, so I knew it could handle 20" worth before, but this time I T'd in my BRS Dual Reactor and chiller directly to my return line, lowering the overall flow from my overflow... so as a guess I thought 12" would be about right. Another factor was I designed my sump/baffles to lengths I just pulled out of my behind.. 18" for the first chamber, and it just so happened that with the PVC parts all attached, I got exactly a 12" slot.. actually about 11-3/4, I had to cut a couple squares off a top corner to get the screen to fit good.

I am feeding about 4 cubes worth of food a day.. I was up to about 8 cubes at my peak of my first ATS and it was handling it great... then I made a really stupid mistake and tried out bio-pellets, while running an ATS, and let me tell you from first hand experience, absolutely horrible idea and it crashed my tank a second time in a year. I thought before hand it may cause the algae to die off... but never expected it to do it so fast and to make such a nasty mess of my tank and sump.

[vicbay]

what made ya try the biopellets?

[inkidu]

Just a suggestion. (120mm fan might not fit)
I have had success running 12v 120 mm fans at 6v. The bigger the fan the better. (efficiency and noise)
My understanding is as soon as you get any movement of air over a properly designed heat sink you will get dramatic cooling especially if you consider the relative amount of heat you are trying to remove.
The benefits
Less build up of dust,noise, and also the fan is less likely to fail i.e. you are running at half power.
Although some 12v fans don't run at 6v.
Hope this helps.
This site has great info about fans.
http://www.silentpcreview.com/

[Ace25]

what made ya try the biopellets?
Peer pressure.. lol.. I had done Vodka dosing in the past but it got to be too much of a pain doing daily dosing so I quite and went with the ATS... but then Bio-Pellets came out right as I built my first ATS and for months I kept reading mostly positive reviews on them and saw a lot of locals having success so I finally decided to try them out even though the little voice in my head kept telling me it was a bad idea. I just didn't realize how bad an idea it was until I tried. I think I lasted 3 weeks until I removed them and in those 3 weeks it did so much damage that almost 6 months later I am still recovering, although I think I am finally towards the end of my recovery and my tank is starting to look good again. I won't say bio-pellets are bad, too many people have success with them, they are just bad in certain circumstances, ie. tanks that use algae as its main filtration method.

[mrbncal]

Just pull 2 velcro tabs and the screen drops right out. Used the same velcro method on my last ATS and the the velcro lasted fine the entire time and never gave me a single problem.

http://www.youtube.com/watch?v=9wGllg25vvQ

I like that! Well done

[Ace25]

Thanks! The first time I watched SM's video showing him cutting zip ties to remove the screen I thought there had to be an easier way. Velcro was the answer for me. :)

Here is a picture from today. I cleaned the screen on Thursday, 4 days ago, because I had scrubbed the rocks in my tank to get rid of the majority of the hair algae and it clogged up my screen real quick so I had to clean it earlier than expected. I used a toothbrush on my first cleaning and got the screen almost new looking again. I know, your not supposed to clean it that good the first time but it was mostly slime algae and I figured the more I get rid of it off the screen the less will grow back. Second cleaning on Thursday there was a lot more green hair type algae vs slime algae.. and today even more green vs brown so the slime phase, which I also had on my first ATS, is going away much quicker the second time around.

4 days of growth from a clean screen. LEDs do work.. and they work GREAT IMO! :D
http://farm3.static.flickr.com/2788/5705001545_03d4930c17_b.jpg

[wgood33]

Ace,
Pleasure to see you here, its been awhile. My ATS (your inspiration) is working great. Your new build here is exactly what I have had in mind for my next system this year. Will be reading this entire thread in the coming days. Regards, Bill

[Ace25]

Thanks Bill! Long time no talk. As of right now, I would say this ATS kicks serious butt over my first one. I have designed enough things for this hobby in my head and then built it only to find out the many different ways reality can show you things you didn't think of during the planning stages, my main LED light was certainly one of those. But this design for the ATS has actually worked exactly like I hoped it would when I designed it. I am still shocked by that myself because nothing ever works 100% like you think it will when your designing it.. this was one of the super rare instances it did work out. Actually, I can't think of anything else reef related that I designed that didn't require weeks of tweaking after the fact to get it working like I had hoped. To say I am happy as a pig in $*** would be an understatement. ;)

Now I am working on my electrical mess ... I have a new Uber Arduino controller (Reef Angel + pH probe + 3x Temp probes + WiFi + 2x 8 Port power blocks + Custom made 8 Channel PWM controller). I am just getting it all mounted on some of that white particle board stuff along with 4 Mean Well PWM drivers. Once that is done my next project is removing my canopy, using clear netting for the top, and hanging the LED light above. Always something to do in this hobby.. but for the past year growing corals have been my lowest priority. Once my hardware projects are all done I can start to focus on that again, but for now I have plenty of corals in my 29G under LEDs that are growing good, even SPS corals, so when the time is right I can put some back into my 75G which is pretty bare, just the giant Duncan colony, couple other LPS corals, and a couple super hardy SPS corals like the Blue Milli and Psammacora.

[wgood33]

Haha, Nothing like mission acomplished on a big project with unknowns. I too will be going enclosed with LED's, likely to copy cat ya here. Glad it was a success, I know how that feels... Great!

[vicbay]

alright. gratz on the green screen.

good to see its doing well. and meeting expectations is truely sweet.

anyhow. been playing a little and my build is not quite up to par(lol) with yours. i am only using xr-e warm whites. a meanwell non-dimmable constant current driver at 700ma. i have three lined up about 1" apart to just test things out. my discrepancy is the par. i get bout 1500 one inch away from the led on center. if i maintain 1" away but move about 1" to the side it drops to bout a 1000.

yours looks about 2 inches from screen and your saying you have even coverage correct? i assume you mean 1500 par plus or minus 100. is that a safe assumption? or are you seeing the kind of drop i am?

just curious if maybe my leds are underdriven or perhaps not warmed up completely(only on bout two mins). still wish we had par numbers from an sm100 or at least a 2700k t5 with quality reflectors.

moving on. thanks for the updates and sharing and keep them coming.

btw heres spectrum from 23 cfl(top). and xr-e warm white led(bottom)

[Ace25]

I think the difference is I am using a Mean Well 1050 driver, I thought it was the 700 but realized I had that somewhere else. As for spread, I have them spaced out more and facing down at a 45 instead of straight on. I think those 2 things spread the light out more, and the 1050 driver pushes up the PAR, but fans are required to keep them cool.

[vicbay]

ahh. ok. that makes me feel alot better. thought it might have been something i had done. ok. so ill just shoot for 1500ish on screen and see how that goes. thanks for clearing that up. cant wait to see next week pic. should have some bragging rights type growth. lol

[SantaMonica]

You can get a nova extreme model 1124 or 1127 and replace the bulbs with 3000k, and measure it for par.

[Ace25]

I have that exact light fixture on my hospital tank, but I don't have the exact bulbs (3000k) to do testing on. Do you have a prefered vendor for those bulbs Santa Monica?

[Floyd R Turbo]

I get these

http://www.aghydroponics.com/T5-2Ft-Gro ... l24830.htm (http://www.aghydroponics.com/T5-2Ft-Grow-Bulbs-24W-HO-3000k-p/litbulbt5-pxfl24830.htm)

[vicbay]

I have that exact light fixture on my hospital tank, but I don't have the exact bulbs (3000k) to do testing on. Do you have a prefered vendor for those bulbs Santa Monica?

i believe these would work.

http://www.1000bulbs.com/category/f24t5 ... bes-3000k/ (http://www.1000bulbs.com/category/f24t5-high-output-fluorescent-tubes-3000k/)

[Ace25]

Ok, I will pick up a couple bulbs and do some PAR testing. That spectral output picture comparison was great. I hope it didn't get overlooked being on the end of the last page because that is a great comparison shot for spectrum between CFL and LED. I have that same type of spectrometer at home, blue triangle shaped box, and it is good for what you just showed, just not good enough to really get good details.

Picture from today just for the heck of it...
http://farm4.static.flickr.com/3312/5708330607_ddc16df2f5_b.jpg

This is my next project, new controller.
http://farm4.static.flickr.com/3492/5708387391_dda5eeb470_b.jpg

[vicbay]

yup. thats the one. and i only use it for comparison. claims you can calibrate it using a fluorescent tube and setting the green bar at 546nm i believe. but the scale on mine doesnt move or i have yet to figure out how to move it.

anyway. it works enough for what i needed. but like you said. not great on detail.

ahh. another project. going to do a build thread somewhere on it?

and im biting nails waiting on par data from those bulbs in that fixture. from other post im guessing its roughly 1000. but thats from tanks using a similar fixture but different bulbs. thanks for doing this and sharing.

[SantaMonica]

Those darker areas of growth are needing more PAR.

[Ace25]

Looking at Day 15 next to Day 16, you can see the "dark areas" going away daily. The dark areas are the brown slime stuff, which has been a typical part for me for a new screen maturing.

Notice the dark areas at the end of the screen and notice it going away the next day.
http://farm3.static.flickr.com/2788/5705001545_03d4930c17.jpghttp://farm4.static.flickr.com/3312/5708330607_ddc16df2f5.jpg

[Ace25]

Day 18, 3rd cleaning, 7 days since last cleaning. Still some slime left but after this cleaning there was still a lot of green algae on the screen whereas before when I cleaned it, it was almost new looking. So it appears to get getting a really good base of algae down now.

Side 1
http://farm4.static.flickr.com/3306/5713993327_8c8e6136de.jpg

Side 2
http://farm4.static.flickr.com/3301/5714553882_7c85f15462.jpg

[SantaMonica]

Congrats... looking good.

From a growth standpoint, I can tell that it could handle more light. So if you can turn it up, now would be the time.

[Ace25]

I agree with you SM. I am thinking a couple more LEDs will help out also. I think I can push 14 per driver and just drop the current slightly. It should give me more light. I can also experiment a little and put different LEDs on each side to test which LEDs do better than others. Now that I have a good baseline for what my current LEDs are doing I can tell if adding blues and reds help or hurt growth.

[Ace25]

This weekend I will be swapping out 2 neutral whites on one side with 2 Royal Blues. Here are pics from Day 26 before and after cleaning. I think the growth has stalled in the last week and I am pretty sure it is due to the lighting, so time to experiment a little to fine tune my setup, first by just doing one side so I can have an easy visual comparison.

http://farm3.static.flickr.com/2769/5742170996_3168e8f24e.jpg

http://farm6.static.flickr.com/5143/5741617781_3087a89dc6.jpg

http://farm3.static.flickr.com/2167/5742171040_581550d5dc.jpg

[vicbay]

nice growth. before ya swap out i have to ask. are ya feeding the same? any other changes that might contribute to the change in growth? just trying to cover all bases lol

thanks for update.

btw. have you by chance had any time to fiddle with par readings from the 1147 fixture with proper bulbs? and hows the other controller project going?

[Ace25]

Actually, I was feeding a little less the week the growth slowed down.. didn't really think about it until you mentioned it so I upped the feeding again for the last week. I didn't switch out any LEDs yet because I wanted to see if the feeding made a difference, and it did, in a positive direction. I will order the T5s this weekend so I will be able to give some PAR readings next week. Too many projects going on at once (having 3 tanks = lots of projects going on all the time, never a dull moment, even 17+ years into the hobby). Oh ya, this is day 34 from a brand new screen.

Pics from this weeks cleaning.

First glance from these first 2 pics doesn't seem like much, lighting/camera I have doesn't show much.
http://farm3.static.flickr.com/2512/5768860564_fd0c846a5a_z.jpg

http://farm3.static.flickr.com/2012/5768317919_2dc108bb6c_z.jpg

These side pics show it much better
http://farm6.static.flickr.com/5303/5768860630_fcf29e0168_z.jpg

http://farm4.static.flickr.com/3505/5768860650_3ab864b8e2_z.jpg

And since I haven't posted a pic of this yet, here are the lids/fans over my LED light boxes.
http://farm6.static.flickr.com/5267/5768317993_3bb0d090f3.jpg

[SantaMonica]

Much better. As suspected the led's have to be at a distance to overlap.

You can keep turning up the power until you start getting yellow.

[Ace25]

So I ordered 8 660nm LEDs from RapidLED (http://www.rapidled.com/servlet/the-172/Osram-660nm-3W-Red/Detail) today. After much debating about which way to go, blues or reds, and reading Chipper's post a few days ago regarding reds, I decided to try this route. I also ordered another 700ma driver. I will replace my 1050 driver I have now with 2 700ma drivers and run 10 LEDs per side, 4 reds, 3 warm whites, and 3 neutral whites. I am thinking that the neutral whites will have more than enough blue in them for algae so I shouldn't need to add any royal blues.

I also ordered 2 Philips 3000k 24w T5HOs (http://www.1000bulbs.com/product/4688/FHO-24T5830P.html) from that link vicbay provided. When they arrive I will take PAR readings/pictures/videos to show everyone.

[rygh]

Those Osram 660nm reds look very interesting.
A lot cheaper than the LedEngin ones.
I can hardly wait to see your results.

Although - given how you have those vertical streaks of extra growth, that might imply you
have a bit lower flow than needed, and not an lighting issue.

The theory being that you have uneven flow on the screen, which would of course be vertical like that.
Where the flow is higher, it grows better.
If it was lighting, the impact would be spread out, or even in circular patterns under the LEDs.

[Ace25]

I hope the LEDs are in the mailbox when I get home.

One thing I noticed about the LedEngin red LEDs is they were 5w, not 3w, and I have no idea how a 5w LED works (ie. can you put them in the same string with 3w LEDs?). The cost of those along with the questions I had, it was a much easier decision to go with the 3w LEDs once I saw that rapidled sold them.

You could be 100% right about the flow/streaks.. we will see what happens in the next couple weeks. Adding the red LEDs should answer that question if it is flow or lighting related, but there is a strong posibility that you are right.

[rygh]

One thing I noticed about the LedEngin red LEDs is they were 5w, not 3w, and I have no idea how a 5w LED works (ie. can you put them in the same string with 3w LEDs?).

Yes, it will not be a problem.

Current handling of the LedEngin exceeds normal 3W.
The forward voltage is basically the same.
And under-driving LEDs is fine. They will work even at very low currents.

[Ace25]

Got my reds in today.. I decided to put 2 royal blues on one side to see if it makes any difference. So one side is 3 Warm white, 3 neutral white, and 4 660nm reds, other side is 2 warm, 2 neutral, 2 royal blue, and 4 reds.

http://farm3.static.flickr.com/2003/5796023176_bf2747bb21_b.jpg

http://farm4.static.flickr.com/3388/5795464037_f6041312d6_b.jpg

Thought this pic looked cool, camera couldn't focus good on the lights and this is how one picture came out
http://farm4.static.flickr.com/3158/5795464055_23f3433716_b.jpg

[Ace25]

Couple pics of what it looks like installed.

Side by side old / new
http://farm6.static.flickr.com/5267/5768317993_3bb0d090f3.jpg http://farm4.static.flickr.com/3626/5797579584_d1a62524ec.jpg

From above
http://farm3.static.flickr.com/2443/5797017295_a1d28043d4_b.jpg

[rygh]

Looks very cool!
I hope you get some good data.

[Ace25]

This weeks cleaning pictures:

Red/White Side
http://farm6.static.flickr.com/5221/5821847339_9cf82d0499_b.jpg

http://farm3.static.flickr.com/2740/5821847391_1e2edd1c1d_b.jpg

Red/White/Blue Side
http://farm3.static.flickr.com/2366/5821847373_4d6fa37d75_b.jpg

http://farm4.static.flickr.com/3257/5822411496_16f9758240_b.jpg

Harvest (wet: 38.2G on Red/White side (left), 37.9G on Red/White/Blue side (right))
http://farm3.static.flickr.com/2718/5821847465_94db6ccbd3_b.jpg

[salty joe]

So that's a 60 W scrubber, is that right? Is a spectral graph available for the different LEDs? I would have expected a greater difference in the color as well as the yield of algae. Thanks for posting your results.

[Ace25]

Yes, about 60w of LEDs @ 750mA. As far as a spectral graph, Cree has them on the web for the whites, and for the blue and reds, they are monochromatic, so blues are 450nm and reds are 660nm.

One very minor difference was visually the side with only red/whites seemed slightly darker in person. The side with the blues seemed a tad lighter in color and also had slightly more "slime" in the algae. Since this was only 7 days since adding reds/blues I can't draw a good conclusion quite yet, especially since I cleaned the screen to near new looking right before I added the new colored LEDs. This time I just used an old credit card to clean (no toothbrush scrubbing this time) so I will see how that changes things for next cleaning. I suspect I will have more weight when I harvest next week. So far, I do think the reds are helping, but too soon to say for sure.

[SantaMonica]

Since there is no yellowing, it is still ready for more power. Double the power will probably put it into the green hair within a week.

[Ace25]

Another week, another harvest. Harvest weight (wet but squeezed as much water out as I could): Red/White side=50grams, Red/White/Blue side=46grams. PAR is higher at the screen than a SM100 scrubber, so I am doubting that lighting is an issue, but I still think I can refine it more (possibly use optics/switch out the whites for more reds/blues to maximize LED potential). I also added a Maxijet 1200 going up through my emergency output side so I have about 500GPH coming in from the overflow side and 300GPH coming from othe opposite side. It seemed to help out. Now I am thinking my slot it just a tad too tight... when all clean it is loose, but any little bit of algae gets stuck in the pipe and it causes flow problems in narrow spots on the screen, as can be seen by the yellowing areas on each side which corresponded with less flow over those spots.

Red / White side:
http://farm6.static.flickr.com/5117/5845038593_8bf58e0909_b.jpg

Red / White / Blue side
http://farm4.static.flickr.com/3376/5845590776_4181d9b98e_b.jpg

Better picture to show true coloring of the screen, pics above don't come out showing the true color due to angle, lighting, and really crappy camera I have.
http://farm6.static.flickr.com/5069/5845613046_b8c51fbcc5_b.jpg

[Floyd R Turbo]

This is one of the reasons why I decided to do my experiment. I don't think you can compare growth from one screen on one system to another screen on another system quantitatively. The growth from any system I believe is affected by a variety of parameters, from feeding, water quality, top off water quality, bio-load of fish, even corals, amount of live rock, etc. I guess I will have better insight when I build scrubbers for the 2 other tanks that I maintain, because then I will have 2 tanks running T5HO scrubbers.

As for your scrubber, if you are getting algae pieces stuck in the pipe from the inside, yes that is a problem and your slot may need to be widened so that particles make it through, but if you're talking about algae growth on the screen blocking flow, keep in mind that as algae grows into the slot it will get partially blocked and then the water flow will balance eventually to compensate, causing growth in other sections of the slot which will block flow and cause an increase in pressure that will eventually all balance out (probably by the time you need to clean it again!) so if you get good even steady flow across the whole screen after cleaning, you're probably OK.

[Ace25]

I agree, there is no way you can compare identical scrubbers on 2 different tanks due to the reasons you described. It would be near impossible to have identical results on 2 different display tanks.

As for the slot, yes, it is algae coming from my display getting stuck inside the pipe. I have this fern looking algae growing in my display that I pull out handfuls each week. I pull out as much out of the display as I clean off the screen each week. I try and remove as much as I can but a few pieces get pulled into the overflow, in turn getting stuck in the pipe and causing flow issues. Thinking about it more, I don't see how making the slot a little bigger would help because these are large chunks of algae getting in the pipe and I would have to widen the slot way to much in order for this type of algae to pass through, so I will probably just have to wait it out until the algae in my display is gone, then I will not have any more flow issues. When I clean the screen and pipe each week the flow is perfect over the entire screen right after I clean it.

[Floyd R Turbo]

Take a paper clip and straighten it out, then bend it so that you can then insert one end into the slot between the screen and the edge of the slot on one side, and drag it back and forth. Then go to the other side. That should loosen any clogs, or at least push them to the end of the slot tube.

[vicbay]

Another week, another harvest. Harvest weight (wet but squeezed as much water out as I could): Red/White side=50grams, Red/White/Blue side=46grams. PAR is higher at the screen than a SM100 scrubber, so I am doubting that lighting is an issue, but I still think I can refine it more (possibly use optics/switch out the whites for more reds/blues to maximize LED potential). I also added a Maxijet 1200 going up through my emergency output side so I have about 500GPH coming in from the overflow side and 300GPH coming from othe opposite side. It seemed to help out. Now I am thinking my slot it just a tad too tight... when all clean it is loose, but any little bit of algae gets stuck in the pipe and it causes flow problems in narrow spots on the screen, as can be seen by the yellowing areas on each side which corresponded with less flow over those spots.

What is the par on your screen? Is it uniform? If not what does it range from?
Did you ever have a chance to measure par on a fixture like that used on an sm100?

I tested my leds a lil more and get 1250 par at an inch and one eighth away. However to keep it that uniformly across the screen I'd have to put an led star ever 1.1 inches. That's too expensive.

I did check my t5 48" fixture over my display. With a 3000k bulb I get about 1050par sticking sensor to the bulb. With an ATI blue plus bulb. I get about 1300 par. Sticking sensor to bulb. So that suggest that 1250 is too high for my led scrubber. Or at least not needed. So I started looking for smaller leds. Seen ml-b cree leds. They're 1/4 watt. But would spread light better. Just lots more work. Anyone had any experience with these?

And does anyone have any par measurements from the xp-g warm whites or xm-l warm white crees?

[Floyd R Turbo]

When you say 'sticking sensor to bulb' does that mean you're putting it right against the lamp?

[Ace25]

The fixture on the SM100 scrubber is around 440 PAR on average at the screen on each side, assuming it is right about 2" from the screen (don't have one in person to get an exact measurement).

Right now, the PAR on the side with Red/Whites is 500-600 PAR at the screen without optics running @ 700mA. The side with the royal blues is 800-900 PAR over the entire screen, it is pretty uniform in coverage on my setup. I do have a star every 1" now, 5 in a row, 2 rows so 10 leds per side.

I just tested my XR-E warm whites on my main display light, not exactly what you need to know though. Without optics, 2" from LED, it is about 500 PAR directly under, but it tapers off quickly as you move to the side. With 60 degree optics, it is 2000 PAR 2" from the LED, those were running at 1000mA though.

This was how I measured the light for the SM100 fixture.
http://farm6.static.flickr.com/5229/5813212337_6cbf09dae3.jpg

[vicbay]

When you say 'sticking sensor to bulb' does that mean you're putting it right against the lamp?

Lol. Yup. The sensor is touching the lamp. I know it's unorthodox but its easy to compare lamps that way.
But if anyone is aware of a problem with that kind of measurement please let me know. Thanks.

[vicbay]

The fixture on the SM100 scrubber is around 440 PAR on average at the screen on each side, assuming it is right about 2" from the screen (don't have one in person to get an exact measurement).

Right now, the PAR on the side with Red/Whites is 500-600 PAR at the screen without optics running @ 700mA. The side with the royal blues is 800-900 PAR over the entire screen, it is pretty uniform in coverage on my setup. I do have a star every 1" now, 5 in a row, 2 rows so 10 leds per side.

I just tested my XR-E warm whites on my main display light, not exactly what you need to know though. Without optics, 2" from LED, it is about 500 PAR directly under, but it tapers off quickly as you move to the side. With 60 degree optics, it is 2000 PAR 2" from the LED, those were running at 1000mA though.

Ahh. Ok. Thanks for the information. Now that I have some numbers I guess its time to try a build finally. Lol.

Keep the updates coming ace. I'll sure keep my eyes out for them.

[Floyd R Turbo]

I would think you would not get a reliable measurement by doing it this way. You should measure PAR at the distance where the screen will be. I have never heard of PAR measurement being taken at the lamp surface.

[Ace25]

Didn't want to make a new thread just for this since I doubt I will be updating info on this tank... just thought I would share.

I got a free tank, stand, and 20G sump from fellow reefer. Tank was old and dirty, so first thing, clean it up and let run with a gallon of vinegar for several days.

http://farm4.static.flickr.com/3248/5874792885_a34e2f9742.jpg

After cleaning, painting, and cleaning up the stand.
http://farm7.static.flickr.com/6007/5910447882_78bef9f4c5.jpg

This tank I plan on going solely with an ATS for all filtration. So I built this ATS which uses a full size screen of 10.5" x 13.5". The screen will be 10.5" in the slot and have a 600gph pump with a valve to adjust it down if needed. I will be using 42w 2700k CFL bulbs.
http://farm6.static.flickr.com/5120/5910380562_5b9217ea52_b.jpg

And this is what it looks like in the sump.
http://farm7.static.flickr.com/6009/5910380596_603228e88f_b.jpg

Still a bunch of work to do before the tank it up and running in a couple days.. one thing is build a light/splash shield to cover the top part of the screen and make sure nothing sprays out of the sump.

[itzrulez]

dont forget to update about this tank!
i want to see a tank goes with ATS from zero!

[SantaMonica]

Also you might want wider reflectors... those are big CFL's and need to have the light spread out.

[Ace25]

I agree, I need to find some good ones.. I checked all the local hardware stores and I couldn't find a "good" reflector. I need one that is about 8" instead of 5". With the reflectors I have now the 42w bulb sticks out past the reflector at least 1" so I know I am losing light out the sides.

[Ace25]

Here is something a little different that I noticed this week on my LED ATS... it has been warm here so all week I have put a fan over the top of the ATS/Screen to help keep the tank cool so my chiller doesn't turn on. Well, turns out that adding airflow over the screen has made the screen grow faster. Nothing else has changed between the previous week and this week, same feedings, no water changes. Previous weeks I could tell by the end of day 6 that the screen was getting full because my emergency overflow would start to take effect (and make gurgling noises telling me this). This week with the fans the screen has filled up in 4 days and I started hearing the gurgling noises on Wed afternoon instead of Friday night. I hadn't thought of how airflow over the screen may help growth but thinking about it now it makes sense. Anyone else notice if you put increased airflow over the screen you get faster growth?

[rygh]

Here is something a little different that I noticed this week on my LED ATS... it has been warm here so all week I have put a fan over the top of the ATS/Screen to help keep the tank cool so my chiller doesn't turn on. Well, turns out that adding airflow over the screen has made the screen grow faster. Nothing else has changed between the previous week and this week, same feedings, no water changes. Previous weeks I could tell by the end of day 6 that the screen was getting full because my emergency overflow would start to take effect (and make gurgling noises telling me this). This week with the fans the screen has filled up in 4 days and I started hearing the gurgling noises on Wed afternoon instead of Friday night. I hadn't thought of how airflow over the screen may help growth but thinking about it now it makes sense. Anyone else notice if you put increased airflow over the screen you get faster growth?

I do remember some old thread where bubbles were added, and it improved growth.
And recently with some pH issues, I am have been looking closely and seeing a very significant pH spike
when the scrubber goes on, especially when overlapped by the display tank lights.
The only real possible cause for that pH spike is a sudden drop on CO2.
And in fixing that pH issue, I have seen a bit of increased growth as well.

SO: It is quite possible that scrubber growth can be CO2 limited, and you improved airflow helped.

Note that on planted freshwater tanks, it is very common to add CO2.

[SantaMonica]

Fans have been used by many people for cooling, then removed, then added back for summer, etc. This is the first I've heard of it being correlated with more growth. Remember, screens have up and down weeks of growth anyway.

There is a very large amount of CO2 going into the water already from all the respiration going on. Only autotrophs remove CO2. Everything else.. fish, corals, bacteria (including nitrifying and denitrifying), pods, the stuff in live rock, microbes, and all other inhabitants, put CO2 into the water.

Planted tanks, of course, are like a giant autotroph with few inhabitants. Thus, it needs CO2.

[mlogsdon]

I agree, I need to find some good ones.. I checked all the local hardware stores and I couldn't find a "good" reflector. I need one that is about 8" instead of 5". With the reflectors I have now the 42w bulb sticks out past the reflector at least 1" so I know I am losing light out the sides.
Ace home depot in lompoc has the larger ones

[Ace25]

Video and picture .. I really wish I had a better camera than this 10+ year old one I have, but so many reef items take priority for me.

3 days worth of growth. Lighter spots on screen are flow related issues I need to resolve with my slot.

http://www.youtube.com/watch?v=hRtD38lc1OQ

http://farm7.static.flickr.com/6128/5940915073_653a7308ae_b.jpg

[Floyd R Turbo]

Looks very similar to the growth I am getting. The streaks of light green vs darker and all. I don't know if it's a flow issue, SM said the darker areas are likely diatoms which grow the fastest. When I rinse my screen off the water is brown and all the green sticks to the screen, and looks in my hand like it does in your pic. I wouldn't fret too much over that, excellent growth.

[Ace25]

On mine I can see the difference in flow on the screen and it is significant, at least 50% less water going over the lighter spots, and those spots are always in the same spot on the screen, even if I turn the screen 180 after I clean it. I have very tiny little bumps on the slot in those spots .. so tiny I really didn't think it would have any effect but apparently it does.

In regards to LEDs, the conclusion I have come to is that reds do help a lot, blues do not help at all. There is no 1 size fits all answer to the question of how many LEDs does a person need, there are several factors involved with that answer like screen size, distance to screen, and mounting options. So the answer is yes, LEDs will work for an ATS, but it does not come cheap and takes some work. For pure ease of setup a CFL setup is by far cheaper and easier in the short term. If you use 23w CFLs and get them for $1ea (which I have seen them when on sale) it will take many years before you recoup your cost of the LEDs. Power savings, your only talking about a 20w difference by going with LEDs, so maybe $1-2 a month in power savings. So unless your very space limited for lighting and must go with LEDs, I really wouldn't recommend them for use on an ATS right now. LEDs are a great MH replacement over a display but for the purpose of an ATS light source you really do not get your moneys worth by going with LEDs vs CFLs in my opinion. That is how I feel today.. 6 months.. year from now.. LEDs may be 25cents ea instead of $3-5ea and at that time it may be more cost effective to try LEDs as your lighting option.

[SantaMonica]

Very nice growth. Yes the bands are from flow differences.

Pretty much what I though about led's. Unless you can build it yourself, it just not worth making a scrubber that's going to cost $2000.

[simon wright]

excellent build there , i have led unit over my reef , and more than happy with leds .
i sourced my led unit for my tank from china myself , they are ready made units .
sure i can get them to build me a ready made unit for my scrubber , .
do you think i should use 3w leds ?

[Ace25]

I can't recommend china LEDs. To many bad experiences with them. I ordered a batch of 100 and 25 were DOA, and another 50 died within 30 days. These are the 3w Prolight LEDs. On my scrubber LED light, when I put in 2 royal blues, one of them was Cree, the other Prolight, the prolight again only lasted 30 days, all other LEDs are going strong.

For a scrubber, I do think 1w LEDs with reflectors (using 3x as many LEDs as you would with 3w) would probably work better than 3w LEDs because you will get more even coverage but keeping them water proof and still allow for cooling is the biggest issue I see with that idea. I have no idea how it could be done but I am sure there is a way. Also trying to find non-china 660nm 1w LEDs will probably be difficult.

[simon wright]

[attachment=0:a2abot9y]luminare main advert pic.jpg[/attachment:a2abot9y]
this is what i have over my tank now , it is the 100w version all 3w leds mix of white 12000-15000k blue 455nm .
was thinking of small version built to my speck all white 4000-6500k 3w leds , my scrubber is horizontal built into my sump so there would be no issue with it getting wet and the underside of the unit is splash proof anyway .
the leds are epstar seem to be good leds , or like you say could go 1w leds , i have no problem with heat at all on these light units as the heat sink does a great job . got to give it to the chineses they are very clever at what they do .
at the moment i am trying out a 10w led 4000-6500k it is like a secruity light version and is equivellent to 100w output , so will see how that goes first.

[simon wright]

this is the one i am trying now,
IP65 Waterproof, Use for inside and outside
Low heat, no UV or IR light radiation
1 x 10W high power LED, >900LM
Color temperature: 6000K (Cool White)
LED Life: > 50,000 hours
[attachment=0:fg1gyfkd]41b8E4wf8iL__SL500_AA300_.jpg[/attachment:fg1gyfkd]

[rossigeologo]

very interesting, but... what about the cooling? could you post where did you find it? Thankyou :)

[simon wright]

very interesting, but... what about the cooling? could you post where did you find it? Thankyou :)
which one , do you want to know about ? heat sink takes care of the cooling on both.

[Ace25]

I have built my share of LED lights in my time and I would never make an LED light without a fan. A heatsink acts like a blanket as far as heat, it traps it. Sure, it pulls it off the LEDs but only a few microns away and then the heat just builds up on the heatsink. Without air movement to push the heat away from the heatsink you still end up cooking the LEDs. A good example would be a road on a hot day.. no airflow you see the "mirage effect" from the heat radiating off and you will burn your skin if you touch it.. same amount of heat but with a 10-20mph wind, no more heat radiating off that you can see visually with the mirage effect and the asphalt it safe to touch (yes, heat is still radiating off but the air is pushing it away as soon as it does).

Trying to understand why anyone would use white LEDs for an ATS? I used them just because that was all I had laying around at the time but as of yesterday (when I noticed my Prolight LED blew) I have now replaced all my LEDs with 660nm reds. Yes, they will work, but If your going to use LEDs to get the best possible spectrum, go with 660nm LEDs, otherwise there is no difference between going white LEDs vs white CFLs. I forgot you were going horizontal so you have a huge benefit in regards to mounting lights vs vertical scrubbers. I was thinking of a vertical scrubber + 1w LEDs being very difficult to water proof and cool, 3w/5w/10w are all easier to cool because they are on stars instead of poking anode/cathode legs though the backside to connect.

[simon wright]

yes i was having a good read today on red leds for growth , it is also the case in growing phyto plankton from what i have read .
i think i might try the light i have now over the scrubber ,but with the red led at the nm you have mentioned .

[simon wright]

also i have been thinking a lot about going vertical with the scrubber , my nitrates have never been zero , not high though .
think i will have to go vertical to get zero .

[rossigeologo]

thnx Simon, I was referring to the last picture. There is a fan on the heat sink? i believe it is necessary for leds, otherw. their life time is shorter than usual.

[SantaMonica]

I updating all the example pics of scrubbers. If you would like to have any special pics of yours for an LED example, we could do that since yours is the only one that has grown anything.

[Floyd R Turbo]

Don't forget about Rygh's scrubber!

[Ace25]

I agree with Floyd, Rygh's scrubber was certainly before mine and I used his as a starting point when building mine. I do think I am getting much more growth than he is though, actually cleaned my screen yesterday and each 4 days it is more and more full, felt like several pounds on the screen yesterday when I pulled it out to clean (obviously a lot of water weight, but still, each time I pull it out to clean it feels significantly heavier than the time before).

I don't know what other types of pictures you would like, but if there is a specific picture you want just ask and I will take one for you. If you want a screen picture (full) it will be a few more days until the next cleaning.

[SantaMonica]

Yes some full-screen pics, next to the LED's would be nice.

Growth is the important part, so if Rygh has some growth pics too we can use them.

[Floyd R Turbo]

He has some on page 4 of this thread

viewtopic.php?f=3&t=645 (http://www.algaescrubber.net/forums/viewtopic.php?f=3&t=645)

But not a more recent one. Will bump his thread

[Ace25]

Just a quick update since I see more and more people interested in LEDs for an ATS these days. My conclusion on an earlier page and how long it may take for me to change my opinion on when I think LEDs will become a cost effective replacement lighting seems to be happening much sooner than I expected. Only 3 months. With a variety of LED options out now and being tested, as of today, I actually think a Red LED setup, whether it be an eshinesystems LED light or a DIY light, seems to be at a point that is comparable to T5HOs. Still no where near cheap enough to replace CFL/Reflector setups, but I don't think CFLs are a very efficient lighting method (and I am running a dual 42w CFL light on my 60G, it takes 3 weeks to get 1/2 the growth that my 75G with red LEDs gets in 3 days, BIG difference). Right now I am only running 4 660nm Red LEDs per side, and it is working great, but I do think 6 3w LEDs per side on a 13"x10" screen is the magic number and I ordered 4 more LEDs today so I can have 6 per side.

Here are my thoughts on amount of LEDs today with what I have first hand experience with. When Floyd gets his custom LED light made, the choice of spectrum will probably change slightly to incorporate more 630-640nm light vs what I am saying today, I just know 660nm works very well on its own so that is why I am making the recommendation. If I find out a mixture of different spectrum LEDs work better, I will again revise my opinion, so please don't take my opinion as the final answer to "LEDs on an ATS", this is just one of dozens of variations we will discover that work well for our purpose. I am only speaking about 3w LEDs, not 1w, 5w, or 10w, those are a whole different can or worms and I have no experience with any other type of LED for an ATS other than 3w LEDs. I know the other types will work, I just don't have a clue what the answer would be to how to design a system around them at this time.

6 3w OSRAM 660nm LEDs per side, 2 rows of 3 staggard, will give you plenty of coverage, intensity, and spectrum to light one side of a screen. So for a standard 13"x10" screen, 12 3w LEDs total and 1 700mA constant current driver along with heatsink (and fan if heatsink isn't big enough) is what I feel is required to get VERY good growth out of a standard size screen ATS.

http://www.rapidled.com/servlet/the-172 ... Red/Detail (http://www.rapidled.com/servlet/the-172/Osram-660nm-3W-Red/Detail) (12 LEDs = $66)
http://www.rapidled.com/servlet/the-3/d ... ers/Detail (http://www.rapidled.com/servlet/the-3/driver-700ma-48v-drivers/Detail) ($18)

So the 2 main components will run you $84 + shipping/tax in california, and add about $26 more for thermal paste and some type of material for a heatsink, and possibly a small fan. Your looking at around $110-$120 for a DIY LED light that will work fantastic for a 13"x10" screen as long as you make sure to protect the LEDs from saltwater (like I did by making a glass box with a lid.) I would say a good distance for LEDs from the screen would be about 3". Remember, most 3w LEDs without optics usually have a very wide viewing angle (120-140degree), not a tight cone shape, optics give you that. No need to add optics for a scrubber with LEDs so close, you want the beam spread out anyway to get good coverage.

I am now designing ATS v3 to replace v2. With more lessons learned from version 2/LEDs, I believe I can now shrink my ATS, put a 20G tank under my stand, and run only the ATS w/ LEDs as my only filtration and be able to get rid of all the equipment off to the side of my tank like the pictures show in the first page.

[jnad]

Hello!
It seems very interesting to use lead.
Because I now make a nano scrubber to be placed on top of the aquarium, I have a questions regarding distance from the leds to the screen:
You propose 3 inches as a suitable distance, but unfortunately this is too far when it comes and make a small nano scrubber. Is a 1 inch distance between 3W leds and the screen too close? Will this lead to burning algae?

jnad

[Ace25]

If you want to put the LEDs closer than 3" I would put more LEDs per side to make up for the coverage and run the current at 500mA or less on the LEDs. Since your talking about a nano scrubber you still shouldn't need more than 6 per side. I just looked at your nano scrubber, single sided, so ya, 5 or 6 LEDs at 1" distance on your scrubber would be my recommendation and run them at 400-500mA, which means you would want to get a dimmable driver like a Buckpuck that handles up to 6 3w LEDs.

Driver - http://www.ledsupply.com/03021-d-e-700.php
Power Supply - http://www.ledsupply.com/12vdc25a.php

Or get a bigger driver and it has a built in power supply
meanwell driver - http://www.rapidled.com/servlet/the-41/ ... ble/Detail (http://www.rapidled.com/servlet/the-41/Mean-Well-ELN-dsh-60-dsh-48D-dimmable/Detail)
knobs to dim it - http://www.rapidled.com/servlet/the-116 ... ter/Detail (http://www.rapidled.com/servlet/the-116/10K-Ohm-Linear-Potentiometer/Detail)

[SantaMonica]

You would not want to dim a scrubber; you'd want to reduce hours instead.

[Ace25]

By dimming you achieve 2 things when talking about putting 3w LEDs within 1" of the screen material. First, the LEDs won't be as intense so there is less chance of "burning" happening and second, the LEDs will last longer and run cooler. We are only talking 200mA less than I run mine. In theory, by me running my red LEDs off a 700mA constant current driver, I am in effect running the LEDs "dimmed" because they are designed to run at 1000mA.

Just my opinion, but I think running them at 500mA is the correct answer on jnad's nano scrubber setup from looking at his current design/build thread if using 3w LEDs. I think a 1w LED setup would probably be a better solution overall for his scrubber, but building a 1w LED light to fit his scrubber is seems harder to do.... well.. maybe not. I just found this link. This would be my first option if I were in jnad's shoes. I would get these LEDs and the buckpuck driver and fit as many as you can into the light section of the scrubber on a piece of aluminum plate and a small fan like a small computer fan blowing down into the light chamber to keep the LEDs cool. I would paint the outside of the scrubber black though to block out the escaping light. Bright red light spilling into a dark room can get old real fast.

http://www.ebay.com/itm/10x-Red-1W-1-Wa ... 3004wt_905 (http://www.ebay.com/itm/10x-Red-1W-1-Watt-High-Power-Led-Bulb-Lamp-Light-660nm-/360397458042?pt=LH_DefaultDomain_0&hash=item53e95cca7a#ht_3004wt_905)
http://www.ledsupply.com/03023-d-n-350.php (each buckpuck can handle up to 6 1w LEDs, you would need 2 + a power supply for 10 LEDs if you could fit that many, 5 LEDs per driver.

I think of it like this, is it better to run a 42w CFL or 2x 23w CFLs for a 13"x10" screen? The 2x 23w will give you more coverage and that could be considered a form of "dimming" a 42w CFL. That is my thinking when it comes to LEDs and running more LEDs but running them at 1/2 the amperage. It isn't the most cost effective route with LEDs but in certain scenarios the most cost effective route isn't an option if you want to use 3w LEDs.

[rygh]

Dimming is a good idea.
Not only do they last longer, but they are more efficient as far as lumens/watt.
And yes, that is a good way to spread out the light.
Of course, you pay more. Dollars per lumen goes up.

As far as my LED system, I am still quite happy with it.
Well, except that recently it switched to a tougher algae. I thought it might be bryopsis,
but looking more, I think it is just a mix of turf.
I don't get a ton of algae, but that is because I don't feed much either.
At one cube a day, plus a few flakes, you can hardly expect pounds of algae every week.

[SantaMonica]

I was talking only in terms of algae, not electronics.

[jnad]

If you want to put the LEDs closer than 3" I would put more LEDs per side to make up for the coverage and run the current at 500mA or less on the LEDs. Since your talking about a nano scrubber you still shouldn't need more than 6 per side. I just looked at your nano scrubber, single sided, so ya, 5 or 6 LEDs at 1" distance on your scrubber would be my recommendation and run them at 400-500mA, which means you would want to get a dimmable driver like a Buckpuck that handles up to 6 3w LEDs.

Driver - http://www.ledsupply.com/03021-d-e-700.php
Power Supply - http://www.ledsupply.com/12vdc25a.php

Or get a bigger driver and it has a built in power supply
meanwell driver - http://www.rapidled.com/servlet/the-41/ ... ble/Detail (http://www.rapidled.com/servlet/the-41/Mean-Well-ELN-dsh-60-dsh-48D-dimmable/Detail)
knobs to dim it - http://www.rapidled.com/servlet/the-116 ... ter/Detail (http://www.rapidled.com/servlet/the-116/10K-Ohm-Linear-Potentiometer/Detail)

Hello!

Thanks for your time looking in to my question, and scrubber setup.

I am setting up a new nano reef this fall, this nano reef will be about 120-200 liters of water. I am going to build a new scrubber for this nano reef and thinking i will use leds in this scrubber, thats why i asked about the distanse between leds and screen.

When it comes to my current scrubber setup for my 63 liter nano reef i am just about finished with a new nano scrubber for this tank. This scrubber is lightned with a 1xPL-24W bulb and eksternal driver, I will post a building tread for this new scrubber this week.
Here is a link to the bulb i am using:
http://download.p4c.philips.com/l4bt/3/323037/master_pl-l_4_pin_323037_ffs_nor.pdf

jnad

[Ace25]

Quick update showing how it is working today.. This is a 5 day cleaning, I cleaned it on Sunday and it is now Friday and this is what it looks like before I cleaned it. Pictures of the screen don't show how it truely looks in person unfortunately. It was very thick (3D type growth) and very heavy holding the screen. I see more growth with 6 reds + 1 blue than I was getting with just 4 reds. Not sure how much the blue helped though since I made 2 changes at once. I will say one thing, I get less growth since I took off a big fan I had over the screen. Now I just have 2 computer fans cooling the LEDs. So in my opinion, there is some limiting factor with algae growth that is overcome by adding a lot of airflow over the screen, although the downside to that is lots of evaporation and your heaters running a lot more, running up the power bill. I decided the extra cost in power wasn't worth the extra growth.

http://farm7.static.flickr.com/6033/6267306871_bedc1fa04a_o.jpg

http://farm7.static.flickr.com/6046/6267306909_8569edc301_b.jpg

http://farm7.static.flickr.com/6115/6267832234_37bf1abf5d_b.jpg

http://farm7.static.flickr.com/6038/6267306937_a5cb66fcd2_b.jpg

http://farm7.static.flickr.com/6106/6267832276_e6a2f14713_b.jpg

http://farm7.static.flickr.com/6237/6267832290_b45a918e77_b.jpg

[SantaMonica]

Best LED growth yet

[SPOTTER]

can you show me what you used to mount the heat sinks? it looks like they are on some kind of bracket that allows you to tilt them.

[Ace25]

I just used standard 2"x1" C channel aluminum. I put some stick on rubber feed on the bottom ones to be able to tilt them and have them stay. The top one I just riveted a small piece of aluminum wrapped in electrical tape.. cheesy I know.. but it works. ;)

http://www.onlinemetalstore.com/files/Image/oms_product/medium/ALUM_C_CHANNEL_ARCH_063_3000.jpg

http://farm3.static.flickr.com/2003/5796023176_bf2747bb21_z.jpg

[kerry]

Mine looks just like that but I used material about 1/2 that width.

[Aquayne]

It is likely that the increased airflow was providing more CO2.

[wgood33]

Ace your scrubber design is awesome. I too will be radically modifying my ATS setup and wanted to ask you a question. I believe our screens need to be tougher in order to last longer. My 1st ATS used 2 of the typical screens sandwiched together with wire ties. After about 10 months the algae grew in between the 2 screen making the unit too thick for the slot. So I removed 1 screen and went to a single screen which is now torn in a few places. We need tougher screens, thinking of building one out of thin acrylic. I actually preferred the sandwiched method because you could get aggressive on the pruning and there would be existing algae in between the screens to easily grow back. Comments? Bill

[SantaMonica]

The problem with sandwiched screens is 3D growth; it will shade much sooner and kill the roots. But it does grow back quicker. They are best used where frequent cleaning is going to happen anyway.

[Ace25]

The only tool I use to clean my screen is an old credit card. Takes about 30 seconds per side and a good rinse afterwards. This leaves a lot of algae still inside the holes to grow back quickly. I have been getting 1 year out of each screen before it needs replacing, and they only cost 33cents each at the local craft store. I bought 10 screens when I first setup an ATS and I still have a few spares left today. I noticed when using any type of brush to clean the screens it could clean them too well because it cleans the algae inside the holes and takes longer to grow back. I've done both dual layer and single layer screens and I think I will stick with single layer from now on. Lesson I learned is you don't need to do really aggressive cleanings once the screen is established with green algae, just a quick swipe with a plastic card to trim it is all it needs, like mowing the lawn.

[SantaMonica]

Unless you have black growth, which really should be removed, even if in the holes.

[wgood33]

All good points. The 2nd best thing about my ATS is the stability it gives to my tank. Had a fresh-water friend over yesterday and he asked me if my tank was the best one on the coast, I said no but it was the most stable tank I know of (ty ATS). So anyway going through a screen change 1x per year, where I have to start anew with the carpet growth is something I want to solve. Not going to hijak your thead here (awesome thread), just thought I'd ask if anyone is using a more hearty screen.
btw- best thing about my ATS is the reduced algae in my display :P

[kerry]

All good points. The 2nd best thing about my ATS is the stability it gives to my tank. Had a fresh-water friend over yesterday and he asked me if my tank was the best one on the coast, I said no but it was the most stable tank I know of (ty ATS). So anyway going through a screen change 1x per year, where I have to start anew with the carpet growth is something I want to solve. Not going to hijak your thead here (awesome thread), just thought I'd ask if anyone is using a more hearty screen.
btw- best thing about my ATS is the reduced algae in my display :P
When you are in need of a new screen wait until cleaning day and rub the old screen all over the new one and rinse, works great.

[kentth]

So how do you tell if you need a new screen?

Kent

[kotlec]

Ace , what is your usefull watt per square inch at the moment ? Im fighting too brown --- too yellow screen.

[kerry]

So how do you tell if you need a new screen?

Kent
When you have big tears, holes, or the screen is to thin to rough up any more. Not sure if others re-roughen their screens or not but, I hit mine over again about every 30-60 or so days when it does not feel as prickly or if the algae is not holding on as good.

[kevin.w.brake]

6 3w OSRAM 660nm LEDs per side, 2 rows of 3 staggard, will give you plenty of coverage, intensity, and spectrum to light one side of a screen. So for a standard 13"x10" screen, 12 3w LEDs total and 1 700mA constant current driver along with heatsink (and fan if heatsink isn't big enough) is what I feel is required to get VERY good growth out of a standard size screen ATS.

http://www.rapidled.com/servlet/the-172 ... Red/Detail (12 LEDs = $66)
http://www.rapidled.com/servlet/the-3/d ... ers/Detail ($18)

So the 2 main components will run you $84 + shipping/tax in california, and add about $26 more for thermal paste and some type of material for a heatsink, and possibly a small fan...

Hello Ace,
Thanks for all the detail you have provided on this project, it is very interesting and helpful.
Has your opinion changed about this?
These 6 LEDs on each side of a 13x10 screen should create a desirable growth rate for a scrubber?

I am planning to make a new ATS and would like to try using LEDs.

Thanks
Kevin

[Floyd R Turbo]

knock on wood, but I have never had an algae detachment issue. I have only re-roughed my screen once in a year and that was about a month after a partial-death due to blocked pump intake. I have scrubbed the screen down extremely well a few times in an attempt to rid the screen 100% of red turf, but it keeps coming back to no avail, but I don't know if that matters that much. I have considered replacing half the screen to see what happens as far as type of growth on the screen.

[kerry]

Never had a detachment issue yet, I just see spots that the algae come off easier then other areas so I hit them with the saw again to be sure I do not have issues. I also hit the red algae spots with the saw every now and then to keep it knocked down.

[Ace25]

Hello Ace,
Thanks for all the detail you have provided on this project, it is very interesting and helpful.
Has your opinion changed about this?
These 6 LEDs on each side of a 13x10 screen should create a desirable growth rate for a scrubber?

I am planning to make a new ATS and would like to try using LEDs.

Thanks
Kevin
Yes, today I still think those LEDs will provide excellent growth on an ATS. Do I think it is the most ideal lighting for algae? No. Do I think it is better than white lights (CFL or LED), BIG YES. We are all hoping Floyd will be able to really give us the answer on what combinations of LEDs is the most ideal, but regardless of what he learns, white light is not going to be part of the answer. The only real question now is how many reds/blues, what wavelengths, and what ratios to put them in. Today I am thinking it may end up being something like 4x 630nm/6x 660nm/1x 425nm/1x455nm but my "guesses" seem to change monthly so don't take that as the final answer.

Until we figure it all out all I can say with absolute certainty (since I am running it) is that 6x 660nm 3w LEDs per side will give very good growth on a 10"x13" screen, much better than a 42w CFL bulb.

For the other questions, watts per sq in is really not something I would even attempt to figure out with LEDs. I know it would be nice to have an easy answer but with LEDs that is just not possible. LEDs are so different in so many ways you just can't use watts as a measurement tool. Things like wavelength and amperage draw are 2 of the biggest factors and reasons why using watts won't work. Even Lumens doesn't work because we are dealing with monochromatic light sources. The lumens coming off a Red or Blue LED will be a small fraction of what comes off a white LED, but the wavelength coming off the Red LEDs is 100% usable light, not so on white LEDs where a lot of the spectrum is just wasted light. This goes back to the "intensity vs spectrum" debate. I can put 1000 lumens of white light over an ATS screen and still not get 1/2 the growth as 100 lumens of Red light because with the white light 950 of the lumens are outside of the spectrum that is ideal for growing algae and the 5% that is within the spectrum is mostly in the blue region, not the red region.

Same thing with PAR, a Cree XP-G Cool white running at 1000mA will put out 200 PAR 2" under the LED by itself, where as a Red LED only puts out about 40 PAR, but that entire 40 PAR is usable light by algae where as out of the 200 PAR in the cool white LED, only about 50 PAR is usable for algae, and out of the 50 PAR, 45 of that PAR is in the blue spectrum, none in the upper red spectrum. Only Warm Whites have some noticeable red spectrum, but still do not reach up into the 660nm area.

As far as screen cleaning/roughing, I have never re-roughed up my screens. They always hold algae and once the algae is established as long as you don't go overboard cleaning the screen, you will always have a base of algae stuck to the screen, which to me tells me I have no reason to want to re-rough up the screen. It is like saying "well, the tree is rooted and growing good, but the ground around it is too hard so let me dig up all the roots in order to soften the ground for the roots".. well.. destroying the roots is worse for the tree than giving it softer ground, which is how I think of the screen/algae. Once it is rooted to the screen, it is there to stay unless you do something harsh to remove it. Only when the screen gets too thin and starts breaking do I think about replacing it, which is usually about 1 year, but that is when it first starts to get thin, I bet I could easily go 18 months before it really needs replacing but since they are so cheap (screens) I just replace them yearly.

[kotlec]

Until we figure it all out all I can say with absolute certainty (since I am running it) is that 6x 660nm 3w LEDs per side will give very good growth on a 10"x13" screen, much better than a 42w CFL bulb.

Using 3W leds at 700mA makes them actually closer to 2W. Thats why I asked about total wat. Just to be sure. I also use 660nm LEDs.

So we get 12W of 660ers for 130 sq/inch - or 1W per roughly 10 sq/inch. Right ?
Now I have 8w spread over 3" x 4 " screen. Omg its 8w per 12 sq inch when I actually can go with 1,5 wat. Auch.

[kerry]

I have fast growth with the 660nm LED's. I fire them with 660mA's and have them about an inch away from the screen. Its still brown growth but its getting the nutrients to zero-ish.

[Ace25]

One other reason "watts" isn't the best measurement is due to distance the LEDs are from the target. 6 LEDs (and your correct, I am running mine at 2.1w) may be perfect on my setup because it is 2" from the screen, but put those same 6 LEDs 4" from the screen, and you may only get 1/2 the growth, so in that case you may need 12 LEDs if you have to place them further from the screen. This is why I find it so hard to give anyone a "one size fits all answer" when people ask me about LEDs. There are too many variables you have to consider before you can give a # to the amount of LEDs one would require.

Your math sounds right to me though basing it on my working setup, 12.6w of 660nm LEDs for 130sq/in IF the LEDs are 2" from the screen.

[kotlec]

My LEDs are actually 1" from screen. Ouch.

[wgood33]

Ace, you wrote: "We are all hoping Floyd will be able to really give us the answer on what combinations of LEDs is the most ideal, but regardless of what he learns, white light is not going to be part of the answer. "
Can you post a link for me to this string? I too will convert to LED on my ATS, so this info will be great. Sorry, just starting to follow along in this forum. Thanks, Bill

[Ace25]

He has posted about his experiment in several threads, but hasn't made one specifically for the light test. I am sure when he gets to that point he will create a thread for it.

http://www.reefcentral.com/forums/showt ... 045&page=4 (http://www.reefcentral.com/forums/showthread.php?t=2028045&page=4)
http://www.reefcentral.com/forums/showt ... 0&page=108 (http://www.reefcentral.com/forums/showthread.php?t=1977420&page=108)

Off topic, here is a display light I made for Johnarky's 60G cube tank. 14 Cree Royal blues, 12 Cree Cool whites, and 4 moonlights + a BoostLED Typhoon PWM controller.

http://farm7.static.flickr.com/6223/6319093797_2c73fa3635_z.jpg

http://farm7.static.flickr.com/6111/6319093945_ae6437db2d_z.jpg

http://farm7.static.flickr.com/6215/6321113577_e364345deb_z.jpg

http://farm7.static.flickr.com/6217/6321637372_955d3b5da2_z.jpg

http://farm7.static.flickr.com/6239/6321637524_6458d067a6_z.jpg

[rossco838]

How do you like the reef angel controller?...looking into buying myself an early xmas present

Sent from my SGH-I897 using Tapatalk

[rygh]

How do you like the reef angel controller?...looking into buying myself an early xmas present

Sent from my SGH-I897 using Tapatalk

FWIW, I have used one for about a year.
I love how it is fully programmable.
Also very cheap.
Lots of add ons.
Great support and forum.
Downside : Well, its built a bit cheap.

[salty joe]

First, thanks for posting your experience with LED scrubber. Do you know at what amperage your LEDs are the most efficent? Down the road I'm going to try an LED scrubber and see what I can do with 100 watts of power. My plan is to run two alternate scrubbers at 100 W each. When one is lit, the other will be dark. I want to light them intensely to push the algae to photoinhibition, at which point the light goes out and the other scrubber gets lit up. The downside is an O2 sensor is a little pricey, but I don't know how else to tell when photoinhibition occurs. I read a paper awhile ago that showed marine algae photosynthesis rates and it was real fast until photoinhibition occurred, then it dropped off like a rock. If anyone is interested, I can try to dig it up.

[SantaMonica]

You are not taking into account the adaptation of the algae. Whatever lighting regime you provide, the algae will re-grow to make the most of it, keeping it's physiological boundaries away from "the edge". The studies you read probably don't take that into account, because it was not their purpose to see what "replacement" algae would pop up.

[salty joe]

You are not taking into account the adaptation of the algae. Whatever lighting regime you provide, the algae will re-grow to make the most of it, keeping it's physiological boundaries away from "the edge". The studies you read probably don't take that into account, because it was not their purpose to see what "replacement" algae would pop up.

By "the edge", do you mean photoinhibition? It was a paper on Cheato, but it was mentioned that most marine algae react in a similar manner. By "replacement" algae, I assume you mean new growth of the same algae, is that right?

Basicly, it was shown that if hit with intense light there is rapid photsynthesis for a period (about 4 hr IIRC) then photsynthesis drops off a cliff until the algae goes through a dark period. My intention is squeeze as much growth from every single watt by taking advantage of that.

[SantaMonica]

No, a different algae that is more light tolerant. Probably lighter in color.

[salty joe]

I am thinking of nailing the screen from the getgo with relatively short periods of light with a similar intensity as the sun. Whatever grows is what I'll be stuck with. Hopefully something will grow. Once the screen has algae established, I'd check the scrubber drain water with an O2 sensor and hope to find a point where photoinhibition occurs and use that time period to alternate between scrubbers. Will it work? I hope so. Lots and lots of hoping. We'll see.

[kcress]

Really nice build you have here Ace! This is exactly where I'm going. I have a few questions I hope you can provide answers for.

1) Any reason not to use silver or white paint? My thinking is why not ricochet the light around until it's intercepted by the screen?

2) How's the water getting out of the screen side? I presume there is gap at the bottom of the partition. How tall is the gap?

3) What flow are you running?

4) Is your partition acrylic?

5) Any little niggling issues that would cause you to do some detail slightly differently?

Thanks!

[Ace25]

Do you know at what amperage your LEDs are the most efficent? Down the road I'm going to try an LED scrubber and see what I can do with 100 watts of power. My plan is to run two alternate scrubbers at 100 W each. When one is lit, the other will be dark. I want to light them intensely to push the algae to photoinhibition, at which point the light goes out and the other scrubber gets lit up. The downside is an O2 sensor is a little pricey, but I don't know how else to tell when photoinhibition occurs. I read a paper awhile ago that showed marine algae photosynthesis rates and it was real fast until photoinhibition occurred, then it dropped off like a rock. If anyone is interested, I can try to dig it up.
Not exactly.. most 3w LEDs are at peak efficiency at around 700mA, which is around 2.1w, maybe a little less. I am running a string of 14 LEDs on a single Meanwell 700mA constant current driver right now. Above that you do not get equal output vs power in, it takes more power to get less lumens. I like the sounds of your experiment. Sounds like a great idea to test out that could lead to even better working scrubbers.


Really nice build you have here Ace! This is exactly where I'm going. I have a few questions I hope you can provide answers for.
1) Any reason not to use silver or white paint? My thinking is why not ricochet the light around until it's intercepted by the screen?
2) How's the water getting out of the screen side? I presume there is gap at the bottom of the partition. How tall is the gap?
3) What flow are you running?
4) Is your partition acrylic?
5) Any little niggling issues that would cause you to do some detail slightly differently?
1. That sounds like a great idea. At the time I was more worried about light proofing it and wasn't too concerned about reflecting light since LEDs are so directional but I still get a lot of hair algae growing in that chamber in the water (which I don't mind, just helps filter more) so adding silver paint first to act as a reflector, then black on top of it to make the chamber black from the outside would be a good idea.

2. 3 baffles, 6" tall on #1 and #3, #2 is top down baffle and is bigger. All baffles 1" apart from each other and are black ABS plastic with a sheet of glass behind each ABS sheet for support (since ABS doesn't bond well to glass, I wanted a glass to glass bond). The first baffle I cut teeth in the top of the ABS like an overflow box and it catches almost all the algae that tries to make it through. To separate my 2nd and 3rd chamber I used a 9" tall piece of glass so the first 2 chambers of the sump are 9" deep (level needed for my skimmer) and the 3rd chamber has the return pump and the ATO float switch, this is the only chamber where the water level changes with evaporation and it is only 6" from baffle to glass, so my top offs happen often and tiny amounts each time, few cups worth, which keeps my salinity levels really steady.

3. Right now I am using my overflow (which I tested and am getting 600GPH) and also a maxi-jet 1200 coming from the other side, up the emergency overflow pipe. Guessing around 200GPH out of that so 800GPH total over the screen.

4. ABS + glass

5. Design the light box a little better. I made it too tight, barely enough room to set lights in them. Polish the edges of the glass on the light boxes, you don't know how many times I have cut myself on them, they are razor sharp. I also made my screen about 2x the size needed for my tank, always thinking "bigger must mean better" but now I am starting to re-think that. I am now thinking bigger may not be better in the case of screen size because if you have one sized for 2x the tank size (or feeding by the new guidelines) and you get the ATS working great, even though it shouldn't, I think you may be adding a lot more DOCs back into the tank than the tank can handle. If I cut the screen in 1/2 I would still get great filtering but more along what the feeding guidelines say I should have size wise. With 1/2 the amount of algae that should mean 1/2 as much DOCs getting released into the water as well. Just my thinking for this week.

[kcress]

Those dividers are a lot more involved than I thought - looking at the pics!

Thanks for the detailed answers Ace.

Now where's my glass polisher....

[kotlec]

I am now thinking bigger may not be better in the case of screen size because if you have one sized for 2x the tank size (or feeding by the new guidelines) and you get the ATS working great, even though it shouldn't, I think you may be adding a lot more DOCs back into the tank than the tank can handle.

Does reducing light on hours can do anything to that ? If it would reduce DOC production , you would have big screen that you need to clean only half often.

[salty joe]

[quote="
5. Design the light box a little better. I made it too tight, barely enough room to set lights in them. Polish the edges of the glass on the light boxes, you don't know how many times I have cut myself on them, they are razor sharp. .[/quote]

Just get a small rock and gently rub the sharp corners, they will dull up in a hurry.

[fajars2003]

if there any different on algae growth with the red grow Led, and red led ?

[fishbrain888]

Ace - Can you or someone who also might know comment on the amount of heat that is generated by the LED's vs. T5HO, specifically the heat being projected in the same direction as the light?

[MorganAtlanta]

With LED you get very little heat as compared to fluorescent, both because the LEDs are more effecient at turning electricity into light, and because you need so much less wattage to get the light you need. I went from 150 watts of T5HO to just 30 watts of LED and am getting much better growth than before. The heat sinks on the LEDs don't even get warm to the touch, even without fans. My display has roughly 200 watts of LED light, and I need fans on the heat sinks, but that replaced 750 watts of MH lights.

[salty joe]

MorganAtlanta,
Can you elaborate on your LED scrubber? What kind of LEDs, how far from screen, photoperiod,etc?

[fishbrain888]

Thanks Morgan. Sounds like I'm sold on going with LEDs on the scrubber.


With LED you get very little heat as compared to fluorescent, both because the LEDs are more effecient at turning electricity into light, and because you need so much less wattage to get the light you need. I went from 150 watts of T5HO to just 30 watts of LED and am getting much better growth than before. The heat sinks on the LEDs don't even get warm to the touch, even without fans. My display has roughly 200 watts of LED light, and I need fans on the heat sinks, but that replaced 750 watts of MH lights.

[Ace25]

So I am now working on Version 3 and 4 of my scrubbers for both my display tanks. I have several different ideas to go with and still working out final designs but I got more LEDs for them. The ATS in this thread will be replaced by one that is much more compact, instead of using a 55G tank outside the stand as a sump I can use a 15G tank inside the stand for the sump and still have a great ATS on it (which was my original sump for the tank, I replaced it when I got into scrubbers). My newest design is a mixture between the one seen in this thread and the SM100 scrubber. It will be modular, but it will fit snug within the top rim of the sump, unlike the ATS in this thread which is not modular at all. I have been anxiously waiting to see Santa Monica's new design but in the mean time my brain doesn't turn off and I am always trying to think of new designs, and always striving to find the "perfect" light. One thing I want to test is to see if the cheap china LEDs differ any in growth from the OSRAM 660nm LEDs. At less than 1/2 the cost per LED if the China LEDs work out that will be a big cost savings for others.

I just got these this week, 3w 700mA LEDs. 10x dual band 630/640nm, 10x dual band 650/660nm, and 10x 455nm LEDs from FEDY (China). They only cost $1.80ea. The 430nm were very expensive, $8ea, so I decided against trying those at this time due to the extreme cost vs what I think will be little benefit.
http://farm8.staticflickr.com/7149/6569941265_a93d757fcf_z.jpg

http://farm8.staticflickr.com/7165/6569941509_13f19bb296_z.jpg

[kerry]

Real nice, I am going to have 20 LED's coming here shortly as well to play with until SM shows the new stuff. I am ordering 10 of 660nm and 10 of 15000K for the display lighting but, I have not decided on which tank to do first with the 15000K, the 75G African tank or the 40G SW. I already have my 10G SW running 12000K and its real nice.

[Ace25]

Seems kind of a high kelvin for a African tank (I am assuming your talking about freshwater). I think 10k-12k LEDs all by themselves look great over a FW tank. I would use the 15k over the 40G.

[kerry]

I know have 18000K t-8's over the 75G African tank now and it looks real nice, the blue and yellows really show. I only tried this because my 10,000K burned out so I stuck in my only one left which was the 18,000K for SW tank. I am leaning toward the 40G as i would like to get the halide off of it.

[kerry]

Also to note while using the 18,000K for freshwater, it makes most of the algae water born so your filter pics it up and not the surfaces of the tank and contents. This was a real nice surprise!!!! I would say I only have about 2-5% of the algae growth with the 18,000K verses the 10,000K but the filters are alway green now when I rinse them as with the 10,000K the filters where always rinsing a brown color when rinsed.

[Ace25]

So here is a quick and dirty MS paint of my idea for my next scrubber.

To explain what that simple thing is, pretty much a mini Santa Monica 100 that uses LEDs and will fit inside the rim of the sump on a 15G tank. The ATS box will sit 1/2 above the rim and half below the rim to keep it compact in my short stands. It will be all black, I just used different colors in the drawing to show the different pieces. Screen will be 6"x6" and using 6 LEDs per side, 2x 455nm, 2x 630-640nm, and 2x 650-660nm per side. The box will be made out of black ABS plastic ordered from tapplastics.com, cut with a table saw, sanded, and bonded with weld-on. There will be a 90 degree barb fitting for the input side and a drain on the bottom, there will be 3 independent lids, led light boxes using lexan as dividers between the screen so ensure no warping/melting from the LEDs being so close. The whole thing will be solid, water tight, and modular, simply unplug the pump feeding the ATS and plug for the LEDs and you can pull the entire thing right off if you needed to. There will be baffles in the sump, I just didn't draw any.

This drawing depicts a 15G sump (24"x10"x13"tall), so the ATS part is really small, about 9" long x 8" deep x 8" tall for the box part.
http://farm8.staticflickr.com/7174/6692455341_768d738012_z.jpg

[Ace25]

Forgot to mention, I am going to go with a new route on light times with my new builds. Since I will be blasting the screens with the correct spectrum, and lots of it, I am going on the theory I will be reaching photo-inhibition rather quickly (just a wild guess, no science behind it other than my positive experience switching times from 18/6 down to 12/12 on my 42w CFLs) so I am planning on running a 4hrs on / 4hrs off light schedule to see if that greatly increases the algae usefulness. In theory it should be getting 3 days worth of growth every day, which means the screen should be full every 3 days, which is exactly how long my 5G ATO bucket lasts. Hopefully it lines up where once every 3 days I refill my ATO with Kalk water and clean the screen and swipe a mag float across the glass and that will be the extent of my maintenance routine (besides feeding).

Also the reason I am still going with royal blue LEDs is because I still believe that blues are useful for what we are doing. I don't think blues do anything for growth of algae, so if you put blues over algae and expect growth you will be disappointed. What the blues will do is provide light needed for respiration, which will in turn make the algae stronger as well as remove more CO2 from the water. It is very hard to see the blues providing any benefit with your eyes though which is why I originally thought they were not helpful, but after running all reds and then adding a single blue back to each side the algae does seem to be stronger/thicker on the screen and not as stringy and weak like it was when I only ran 660's.

[Floyd R Turbo]

Very interesting take on the blue LEDs. SM noted in his 'test' using the eshine fixture that he did not think it was doing anything. Maybe this is why.

I can't say that I agree that you will get any more growth by doing 4/4 vs any other cycle. Algae can only grow so fast, can't fool mother nature. I would think at best you would be able to get 1.5 to maybe double the growth. But I guess we will see!

[kerry]

Yes, great data. As I am reconfiguring my 10G LED horizontal screen to vertical I will leave room for a blue that I do have at this point but, I do have to order more for a display project in a week or so and I will be sure to get a few to try myself.

[Ace25]

To clarify a little better to what I was saying about blues. When I ran all 660nm LEDs, I got great growth, but when it came time to clean I could literally take a credit card and with a single swipe clean a section of screen, like using one of those car window cleaner things, one swipe wipes the water off the window. Since adding a blue back to the light, now it is more like removing paint from the wall, short 1" strokes with the credit card to really dig in and break the algae free. It really is a night and day difference on how easy/hard the algae comes off the screen.

As for the 4/4 light, ya, completely unknown how it will actually work. I suspect your right though Floyd and I won't see 3x the growth, but I think your right that 1.5 to maybe 2x the growth could be possible. We will see.

[SantaMonica]

It could be the opposite of what you think. Ease-of-scraping is from the roots dying. Dying is from lack of light and flow. It is possible that the all-reds actually caused more growth, which blocked more light and flow from reaching the roots.

[Ace25]

Certainly an interesting theory, and I may agree with it if it weren't for the fact I was running 6x 660nm for months, then the only thing I changed was I added a single royal blue to each side, going from 12 LEDs total up to 14 LEDs, 7 on each side, I didn't remove any reds. So your theory doesn't seem to fit my scenario unless your saying that by adding a blue it somehow reduced the effectiveness of the reds already running.

[Floyd R Turbo]

I think Ace is right based on that. And that is certainly a very, very interesting observation.

[bucks448]

Ace, I may have missed this. Did you add the blues to the same driver you were using for the reds? If so, wouldn't this reduce the current to all LEDs? Unless you had the REDs dimmed down before adding the blues and were able to compensate for the additional LEDs. Just thinking...

[kerry]

If you have your driver maxed out it that might be an issue. I have added LED's to my display and not seen any visual brightness difference but I had plenty of room to add more.

[Ace25]

Nope, a Meanwell driver can handle up to 14 LEDs per driver at its max output, so a 700mA driver will handle 14 LEDs @ 700mA, an ELN-60-48 will handle 14 LEDs @ 1300mA, or 28 LEDs at 650mA if you put 2 strings of 14 in series. Most people run 12 LEDs just so they are not maxing out the driver, but meanwells are solid and can handle it without problems, cheap China drivers on the other hand can not handle it at all (seen lots of them blow in less than a month trying to run them at max).

[kerry]

I have a cheap china one I have been testing on my 10G QT tank, its been running 9 12,000K 3W LED's powered by a 25V@660mv. That's two extra over the max. Its been going for about at least 3+ months. I turn it on at 5:10am when I get up and turn it off at 9:30-ish pm when I go to bed. So far I have been pretty impressed. I figured the LED's and driver would have had a failure buy now but all it good, so just last week I ordered $110.00 more of this china stuff from the toplightled place on eBay. He's in California but by his emails I am sure he and his products come from the china area. I do a lot of 5 volt digital (gate stuff) and 12/24 Volt (counting and calibration devices of inputs for re-mapping of OEM automotive computers to allow the use of High Performance Upgrades) stuff for my job and have an electro/mech schooling back ground as well. I had the cover off of the driver before I hooked it up, its not substandard in the building technology. I cannot speak for the brand name of the parts but, they are built right. I think they could use a bit more heat sink but other then that its been smooth sailing with these. They even have a pretty good filter in them. I am not trying to take anything away from Meanwell at all. I just wanted to share my experience I have had with them. Who knows, these other ones I have ordered might be junk?

[Ace25]

I was thinking more of the large FEDY LED drivers from China, ones that are made to run 80+ LEDs per driver. They don't run that many, well, at least not for longer than a month. From my experience it seems if a FEDY driver says it can handle 80 LEDs @ 700ma, the reality is it will handle 60 tops, any more and it seems to overdrive the driver and they get super hot and blow.

Smaller drivers like your describe are different, not sure on quality of those, never tried small ones, but I suspect they are not overstated in their specs like the larger ones. Here is a huge thread on RC where everyone started jumping on the FEDY driver bandwagon due to the price, but about 1/2 way into the thread you can see them failing and people dropping them like flies and going with Meanwells instead.

http://www.reefcentral.com/forums/showthread.php?t=1950575&page=30

[kerry]

That's good info to know. I have not bought any huge driver yet. I will say that from past experience I have not been pleased with meanwell. I am a radio operator and do a lot of CB and HAM radio operation. In the past I have bought many meanwell power supplies and had to add extra parts ( filters ) to them so they would not cause reception interference as they where noisy in my receive part of the radio. However as of today I put inline a china driver and guess what, it bleeds the receive of my radio. It is a small one, rated up to 10 LED's @ 700ma and I am driving 11 because I have 7 of them which are 12,000K and the other 4 are 660nm which require less power. I figure this makes it to the max. Well so much for them having good filters LOL!!!! I have a meanwell coming from rapidled and will put it in place of the noisy china unit and see if it makes a difference.

[salty joe]

Could this problem be solved by enclosing the driver with a Faraday cage?

[kerry]

That might capture the EMI but I have not tried or thought of that. I figured I would just a diode and a capacitor to the circuit so it filters better or improve the filter they all ready have. Its loud EMI, My radio antenna is about 60-70 feet away out in my shop and I get 7 S-Uinits on my meter, thats 42 decibels!!!!

[Ace25]

Wow... I am not that into the electronics side of the drivers, but I know what your saying is actually very bad. Meanwells even have a metal shield covering most of the parts inside once you take the plastic cover off. I know Meanwells are certainly not the most efficient driver (one of my pet peeves, there is too much overhead in wasted electricity by the driver) but they are built very solid/reliable, but I can see how to people like you that much interference is a major issue and a big reason to look at other options.

[SantaMonica]

RF shielding is probably not a concern at all with driver makers.

[kerry]

RF shielding is probably not a concern at all with driver makers.
LOL, As I would say on the CB radio thats a BIG 10-4! I can deal with the noise by making high and low pass filters.
Never did I think my fish hobby would interfere with my Radio hobby. I noticed this first when I got my Odyssea halide about a year ago. The transformer for the halide side made wicked noise in my radio. The florescent transformer is fine for the t-5's. I guess I will be making some more filters.

[wgood33]

Ace,
Hope all is well with you, it has been awhile. Your thread has inspired me, just like you did a couple years ago to try the ATS. My scrubber is working great, time to improve it with LED. I am going to build an enclosure to go on top of my existing sump and was hoping you could spec the parts for my lighting;

2ea panels that will be 14" x 5"
Panels will be close, 4" from the turf screen.

Please tell me what YOU would use, and thanks again. You have really helped me in this hobby.
Regards, Bill

[Ace25]

Are the panels going to be right next to each other so you can make 1 long LED strip per side or do you need to have separate LED lights for each section of screen?

Just going off what I am guessing, a total of 28" long and 5" tall, I would think a single strip would work per side, 14 LEDs per side (2 LPC-35-700 drivers) or if your going to go with China LEDs you can use one Meanwell ELN-48-60 driver and put each side on a string and then wire them in series. You have to crank the meanwell driver to max (should put out at least 1300mA) so you can run each side at 650mA, which is perfect for "China" LEDs, anything that is 700mA. Cree LEDs today can be run up to 1500mA.

I have not seen any reason to not go with china LEDs to save some $. The only thing with China LEDs is their failure rate is MUCH higher than Cree/OSRAM LEDs. Out of 100 "china" LEDs, you will be very lucky if at least 10 don't burn out the first week, so order extra LEDs if you go that route.

For home/DIY style stuff, China LEDs are fine, if I were to make LED lights for other people, I wouldn't even consider using them because the headaches I would have to deal with every time an LED burns out. If I did this for someone that isn't local, I would imagine the first time it needs to be shipped back/returned for repair would completely offset the cost of the higher quality LEDs, but if you can DIY yourself and have no issues replacing LEDs, then you're fine.

So, just comes down to budget. If it were me, and the budget was there, I would go with OSRAM LEDs because they have been perfect for me. Even in very wet/humid conditions they have worked flawless since day one and going on close to a year running 12 hours per day.

Only issue I can think of is if you wanted to "experiment" with different types of LEDs. When making a "strip" type light, it doesn't really allow for much customizing of LEDs vs putting LEDs in a square pattern, but that really shouldn't be an issue because 660nm LEDs all by themselves seem to do the job just fine without getting uber fancy and adding different types of blues and reds to the mix.

Getting looong now.. but I just re-read your question again. If you really want to know what *I* would do, I would not make a long strip ATS, I would go with something more square like vs rectangle like. Then I would be fancy and add the blues (420nm and 455nm) and reds (630nm-660nm), but even I think I am crazy and over thinking it. So, there you go, 2 answers, one "what would I do in your situation" and one "what would I do in my situation".

[wgood33]

I agree that more square would be better. My cabinet is not so tall inside, maybe I can go with 12''x6" light panels for my double sided screen. I don't like maintenance or repairs, CREE for me. Should I buy 2ea of the meanwell drivers you listed? How many reds/blues for each panel and how far should I space them apart?

[mess7777]

Ace, i saw some pictures of your tank, looks great!

What skimmer are you running? Would you ever consider going without skimmer since your ATS is kickin but big time?

[Ace25]

I have a bubble magus NAC7 skimmer, but it is rarely used. I have used it maybe 3-4 days in the past year. Under normal circumstances the skimmer is not needed at all, only when I dump in 30G of phyto water or do a really good cleaning on the rocks/sandbed do I run the skimmer for a day. My other tank I have never used a skimmer on it, only an ATS with purigen and occasionally a little GFO.

[Ace25]

Just tossing out ideas.. thought of this one and sketched it up.

My idea of a HOB "Super" UAS filter. Left side is the intake for it, pump is under the box externally. There would be a surface skimmer box around the intake tube. I never thought the concept wouldn't work, just thought it was more work than the older method for people that have sumps.

Water gets pulled through the surface skimmer and the pump injects air to create the bubbles are is attached via a bulkhead, then the bubbles come out a slot in bottom that has a piece of acrylic drilled like a bubble plate inside. The slot is big enough for the screen to fit very loosely in the slot. There lid that has 2 angled pieces glued to the lid to provide a path for the bubbles to go into a collection cup but when you pull the lid off it makes it easy to reach in and pull out the screen, when the lid is on the screen is held firmly in place. There will be an exit path for the water on the right with another baffle to block bubbles. Then it will drain through a media chamber where you can put in carbon, purigen, GFO, etc, and down the bottom and back up to the return line. There will be acrylic windows and I will use 7 LEDs per side. Box dimensions are 9" wide, 8" tall, 3" deep, screen size is 6" tall x 4" wide.

I think I am going to build one in the coming weeks just for the heck of it.. should only cost me a few dollars in plastic, I have everything else to do it.

http://farm8.staticflickr.com/7064/7137827911_e4315aff1a_z.jpg

http://farm8.staticflickr.com/7233/7137827937_88bfe9c74d_z.jpg

http://farm8.staticflickr.com/7259/7137827953_11d877f0f8_z.jpg

http://farm8.staticflickr.com/7203/6991746168_6e785ecac0_z.jpg

On top I would put a CPR Backpack skimmer cup with an O ring to hold it in place.
http://farm9.staticflickr.com/8144/6991746216_730f83eab2_o.jpg

[kerry]

Thats a slick idea!!! I am fighting micro bubbles on my HOB I rigged up a couple weeks ago.

[jbowser]

nice design!

I thought that ats (or uas for that matter) and skimmers didn't work well combined on the same system? or isn't that what you are going for?
I'm also thinking about using HOB uas on my upcoming 15g, but I haven't figured out how exactly...

Jb

[kerry]

You can use a skimmer if you have fish only, with corals you should not so you dont remove the food from the water column that they would normally feed on.

[Ace25]

True, in a reef tank, a skimmer is really not needed at all. It would be much more beneficial in a Fish only system like Kerry stated. Skimmers do not compete with algae scrubbers though, bio-pellets compete with algae scrubbers. For a freshwater version you would just make the lid solid and not have a collection cup on it (since skimmers don't work for FW). If you have a fish only tank though, more than likely you are not going to want a sump, and you are not going to want multiple pieces of equipment and powerheads in the tank (1 for skimmer, 1 for media reactor, 1 for UAS), so I thought the best idea would be to combine them all and use a OTP-1000 needle wheel pump like is used on the Reef Octopus BH-1000 skimmers. Still, the protein skimmer part is just a way to get the most use of the the bubbles as possible. I am certainly not expecting that part of it to work like a $200 stand alone skimmer by any means, but even in a reef tank set to skim very low amounts, I still think skimming a little isn't going to hurt anything (when I say a little I am talking 2+ weeks to fill that little cup).

Of course all this works perfect in my head while thinking about it... how it will work in reality may be the same or could be a complete failure, I have no idea until I try it.

[SantaMonica]

I like it.

[mess7777]

I like it.

he likes it! Hey Mikey! :)

[srusso]

nice work Ace! Fantastic design!!

[Ace25]

Thanks for the compliments. I found a way to make a prototype today. At my work we have 2 $80,000 3D printers (the size of a refrigerator). :D I can make one using the printer. Woohoo!!!

[Doompie]

Thanks for the compliments. I found a way to make a prototype today. At my work we have 2 $80,000 3D printers (the size of a refrigerator). :D I can make one using the printer. Woohoo!!!

Whaaaaat!! how insanely cool is that..*i'm not jealous..i'm NOT jealous*
Very curious to see that!

[kerry]

I have heard of those 3d printers. Very KEWL tech.

[Ace25]

You can get small 3D printers for home ranging from $500 - $2200 depending on various options, the $2000 makerbot is a little to small for me (does 8.9 x 5.7 x 5.9 inches), the RepMan is a little bigger at 10x8x8 and cost $2200. Prices are going to start dropping on 3D printers soon though for home use because there is a huge groundswell of startups trying to get into the game now, so 1-2 years from now I can see the $2200 going for $500. In the mean time I will use the industrial ones I have access to at work. I am re-drawing my prototype in SolidWorks this weekend so I can print it next week.

www.makerbot.com
http://www.bitsfrombytes.com/usd/store/rapman-32-3d-printer-kit-education

[Ace25]

Just an FYI, this is the printer I have at my work. :D It can print 10x10x12 peices, so I am now making the HOB box a little bigger, maxing it out at 12x10x3 for the box to allow for a bigger screen and slightly bigger media chambers. I want to make a filter suitable for a 55G that is HOB, I can always shrink it for smaller tanks, but for now I want to make one for a tank I have running.

http://www.dimensionprinting.com/3d-printers/3d-printing-1200es.aspx

I am finishing up the final tweaks to the drawing over the next couple days in SolidWorks in order to print it in 3D. Hopefully I will have a working prototype in the next couple weeks. I decided to go with the Maxi-Jet 1100 utility pump. It doesn't have a pinwheel but with a venturi it should still provide plenty of bubbles without going overboard and turning the box into a white froth mix that light can't penetrate.