Building Glass Tank from Scratch

[zachtos]

I do not doubt that the LEDs are 80,000mcd. But the specs are totally wrong. I must find the actual maximum current WHICH IS VERY @#@% difficult without specs. from the manufacturer. I emailed them many times before I ordered. They only have specs for 5mms other than the non detailed website specs. I suppose I can just keep cranking the current until I pop em and run at 75% of that current. I will try to fill the holes w/ UVs but this will cause major growth issues w/ checkerboard lighting. I think the problem is the narrow viewing angle of only 12degrees instead of the 5mm's 20 degrees.

 

I'm debating ordering 350 whites, 50 uvs, 50 blues in 5mm form and running at 25mA. tank is 9" deep, display is 5"x12" and I am unsure of how much I need to replicate 70W MH w/o checkers. Cost is not an issue and I dont mind the soldering. I DO mind the drilling. Its too hard to drill that many holes without cracking the material. I need to find an ALTERNATE material. (rubbermaid tub? balsa wood? some form of thin wood...)

[Haagenize]

Full tank w/ 1-13WPC bulb

 

lol your cat is barely on the desk.

 

 

Can anyone explain to me what the foam wall does? because its like taking up 1/5 of the tank

[zachtos]

Full tank w/ 1-13WPC bulb

 

lol your cat is barely on the desk.

Can anyone explain to me what the foam wall does? because its like taking up 1/5 of the tank

 

I didnt intend for the background to take up that much space. The foam triples in size after you apply it. The background makes a great place to mount frags and will eventually cover w/ corraline algae and become part of the tank. You wont know that it is fake in about 6-10months. This is all an experiment tank. Experimental lighting and background basically.

[Haagenize]

oh okay =D

[zachtos]

my perforated plastic sheet has arrived. it has 34 holes / in^2. enough to make my array as dense as I please. although I do need to redrill every hole I intend to use from 4.7mm to 5mm. It is still a time saver versus slow speed drilling, this material will not crack.

[Samoyed]

WOW..im impressed, how thick is the glass?

[zachtos]

3/16" thick glass and 1/4" thick base.

[Haagenize]

we want more pictures =D

[aquaman7]

I added 2 tsp of human urine and 24 hours later I'm at 2ppm ammonia... so the cycle has begun....

 

...intense

[zachtos]

nothing exciting worth posting. LEDs will be here later this week maybe, then I will post. I did add one xenia frag, GSP frag, emerald crab, 2 ceriths, 1 astrea, 2 nassarius, 1 trochus snail. no fish yet, but would like to get a green clown goby. xenia seems to be having water quality issues, working to correct that.

 

pics of intersting stuff later this week hopefully.

[zachtos]

working on my array. It is very bright. I anticipate being done this week w/ new pix. Xenia died for unknown reasons, but GSP are fine.

[guntercb]

Zachtos,

 

Awesome little tank. I really like your trying new things. Cool background.

 

Can you tell me a little about how you wired your float switch? Did you just splice it into the extension cord? Or did you use a relay? If you just spliced it into an extension cord are you worried about the 120 V close to salt water? I see you are an electrical engineer, so I know you are not doing anything dumb with electicity.

 

Thanks in advanced.

 

Cheers,

Chris

[zachtos]

I just spliced the float switch into the wire. As long as you stay 80% under the float switches rated wattage, you should not have any failures. If you havent noticed, your power heads, submerged heaters and pumps are running 120Vac in your tank also. If your'e woried that the wire may become frayed or exposed, you can heat shrink the portion in the tank and place silicone around the base of the wire. otherwise you can get a grounding rod to protect from stray voltage or a GFI protected outlet. A relay will drop down the voltage entering the tank, but is no less dangerous to the inhabitents in my oppinion.

 

I wired up my LED array last nite for the display. It's very fancy if I don't say so myself. 350 LEDs, 4:1 white:blue. Insanely bright. I ordered a LUX meter that should be here this week to measure the intensity and compare to PCs and a MH 250W.

 

Features include:

Daylight mode

Nite mode

surge protection

current limmiting

proffesinal grade mounitng and wiring

 

Details and Photos this week, I promise!

[guntercb]

Zachtos,

 

I like the simplisty of wiring it directly to the extension cord, not to mention it makes it a bit cheaper. The 80% advise sounds good.

 

The LED project is awesome. Let us know the LUX readings when you get them.

 

Chris

[zachtos]

my beautiful array, note the use of a 1.5A fuse to protect from being dropped in the tank. I also have a capacitor on the power supply to help clean up the output voltage. I used 91.5ohm resistors in series w/ each 10LED strand. Each LED drops about 3.4Vdc and the resistor may vary between 2.2V and 2.5V depending on the temp of the LEDs giving 24-27mA, LEDs are rated for 20mA and 30mA max. I did not drill all of those holes, I ordered a sheet of perforated polycarbonate from usplastics.com. I did have to quickly widen all 400 holes of the 4.7mm holes w/ a 5mm drill bit. It really only took 20 minutes to redrill, it cuts like butter. I spent alltogether maybe 10 hours soldering/drilling/trimming/bending the LEDs and maybe 2 hours mounting them and making final connections/cleanup.

 

The array is secured w/ hexbolt standoffs and then is hot-glued onto them (couldnt twist the lugs/design error due to big fingers) cooled by 2 x 5Vdc fans, protective acrylic 1/16"sheet to shield LEDs from salt creep.

 

30 UV LEDs, these stay on after the power switch is flipped. The 36Vdc supply is made by running a 12Vdc supply in series w/ a 24Vdc supply. These lights are always on, but stay on after the main Lights go off. The 12Vdc is going throughout the entire array at nite but can not lite the LEDs at such a low voltage, but the UV's positive terminals are tapping the positive leg of the 12Vdc supply to keep them running at the correct voltage. This allows me to run my moon lights using only one 12Vdc supply instead of TWO!

 

before 13WPC, I like the colors better before... The LEDs seem to "wash out" colors... you decide.

 

after, 350 white/blue LEDs, 40,000mcd whites @26mA ea.

looks brighter to me, LUX reading to be delivered this week late,

GSP is the only coral I got (for now), and it is opening.

 

cliche quote "this moon lite pic does not do it justice!"

30 - 410nm UV LEDs @ 21mA

I cant stress how gorgeous the green star polyps are under the 410nm UV LEDs... the greens/yellows really glow. I really dont know how different they look then your standard PC/VHO actinics since it is a narrow band of UV, it may only fluorescence certain colors.

 

My intended goal: Simulate 150W MH light

• create lights that outlast the tank (these are supposed to last 50,000 hours until they reach 50% intensity, IE 13 tank years at 10hour days)
  
• create low heat lighting to avoid cooling costs/issues (they run about as hot as a 7W PC, warm... but they wont burn you, they do require fans to keep the current from getting too high)
  
• save power (these LEDs consume 42watts... 72% less power then a 150W MH)
  
• do it cheaper or equal (150W MH retail would be around $200 for a DIY kit, these cost me about $150-$200 minus tools)
  
• be equal to or as powerful as 150W MH (not sure, LUX metere is on the way!)
  
• prove that LEDs can grow coral (the spectrum should be there... I'll keep a log of my findings/growth starting with SOFT CORALS, then SPS/clam)
  

NEXT on the list... convert my fuge light into pure blue/red/uv LEDs, maybe a touch of white to balance out the purple glow. I already have the LEDs. Will wire w/in the next 7 days.

[neanderthalman]

NICE JOB. Very tidy, very even. Did you notice any checkerboard effect in the lighting, or are the LEDs closely packed enough to prevent it?

 

Odd that it seems to you that the colors are washed out. The pic doesn't demonstrate it well, as the lighting is bright enough to saturate the CCD. I found my colors to 'pop' under the LEDs, with identical coloration under 20K MH.

[zachtos]

The LEDs do resemble my old 250W 20K MH, but they are a bit less purple. No checkerboard pattern at all, even mix of white/blue. the 10mm's were garbage. the 5mms are great.

[zachtos]

I finished wiring my fuge array. I consists of 96 LEDs. 24blue, 24UV, 48Red. 620-630nm RED, 410nm UV, 465nm BLUE. I chose these instead of white because chlorophyl A&B do not use green light. They react to 640nm and 400-470nm. So these lights should grow plants much more efficiently with no wasted light spectrum, perfect for a refugium without coral. I plan to run these 20 hours a day to grow chaetomorpha marco algae which seems to be struggling to survive for the last 4 weeks with my 7W 2700K PC bulb.

 

4hours to solder/construct my 4"x4" array. running on 24Vdc @ 22mA, 120ohm current limitting resistors. On off switch and capacitor for voltage cleaning/stability. Tomorrow I plan to wire in a 500mA fuse for protection.

 

Behold the macro mega light 2000!

 

 

LUX readings this weekend.

[zachtos]

LUX readings have arrived. I dont have any powerful lighting to test right now... So tell me if these results are good or not compared to MH

 

 

 

left out 0" because that measurment skews the graph, and is not practical for real life application.

 

are these numbers good enough for SPS/clams??? don't be biased and say "you need MH etc blah blah blah". Use the posted results please.

[zachtos]

 

measured the intensity of my 4mo old 20,000K AB 250MH bulb today... keep in mind, this is a small array versus 250W!!! This is 89% average intensity of MH and seems to be able to penetrate deeper then MH can w/ less drop off at great distance.

 

LEDs current PROs

 

1. gives off very low heat, does cause some heat but it is less then PC bulbs.

 

2. lasts longer then your tank will (should last atleast 6 years, max maybe 11years until 50% intensity)

 

3. due to low heat generation, may be mounted very close to water surface, this gives them an advantage because they can get much closer then MH, thus giving them the winning edge in my book

 

4. Less power used, 40W vs 250W, means less electric bill

 

5. Fully customizable lighting effects

 

LEDs current CONs

1. cost is about twice as much as MH DIY kits and about 50% more then complete setups.

 

2. requires a lot of man hours, but may be remedied using the 2" bulbs shown above.

 

3. intensity seems to be between 175W MH and 250WMH, doesn't look like it can compete w/ a 400W MH... yet. (but close)

 

I think LEDs win in my book. Atleast until you get to tanks over 120 gallons. Then you really should just get the MH. The array will be too big and costly at that time and it probably wont be able to give you enough LUX... not sure yet, no 400W comparisons availale.

 

I'm thinking of starting array in the near future for a 90 gallon tank. I am considering paying an unemployed college student to build it on his free time. It would not take him more then a few 40 hour weeks maximum.

 

$1600 for 4600LEDs, 6.4/in^2, at 45"x16"

I think maybe $1000 labor charge would be fair. (130 labor hours at $8/hr, maybe cut the array in half and do half on my freetime or just do it all myself on spare time)

$150 power supplies, 10x12Vdc psupplies135 strands of 34LED long strings at 120Vdc

 

so maybe about $1800 to do it myself or $2800 to pay someone... still a bit expensive... but then again so is 2x250MH and 2x110W VHO actinic... about $1000 for those then add in that lovely $1000 chiller unit and all that increased evaporation and electricity and yearly $280 bulb changes... I stay its more expensive to go MH w/in 2 years easily...

[zachtos]

to answer various inquiries about the power supply once and for all:

24Vdc power supply

12Vdc supply

these are wired in series to provide 36Vdc which allowed me to run longer arrays of LEDs. I ran 24Vdc to my fuge array on seperate powersupply. I couldv'e shared it w/ the display but I didn't want to overload it, and wanted it to be on a different time schedule. I could've used a PC powersupply to get 12Vdc but those are bulky and fairly loud and oversized for my project. Also I'm not sure How it would work if i connected the outputs in series to make 36Vdc, or how safe that would be.

[Seldon]

Im pretty sure you couldnt get 36v because you would need 3 seperate voltage rails and most cheap supplies only have 1 12v rail and high end supplies have only 2 12v rails so you would only be able to get 24v total. Maybe a few of the new high end ones have 3 12v rails, but then you are talking 500 or more dollars for a psu.

 

seldon

[Newbie527]

very interesting project...

so your findings are saying that the LEDs will reach certain light spectrums and intensity that MH,VHOs, and T5HOs reach?

 

also you talk about price of creation...my only wonder is some place in china could have equipment that could create a LED board in one fell swoop such as they can with a circuit board.

 

But i think you are right ...realistically, if your findings are true, these LEDs would make more sense on smaller footprint tanks that are not so deep.

 

good luck and very interesting stuff.

[zachtos]

actually as per the numbers, these lights work better in very deep tanks then MH's do. they taper off less as shown by the numbers. But for the money and time required, small foot print tanks are the best options for LEDs...

 

*I have corraline growth spots all over my top half of my foam wall and powerhead from only 5 days of LED lighting, my macro is growing at a noticable increased rate w/ the reds/blues also.

[Haagenize]

woah, look at all those lights