Jump to content
Pod Your Reef

Building Glass Tank from Scratch


zachtos

Recommended Posts

The tank... it LIVES!

 

I shall be reviving this project so update that subscription.

 

Parts have been ordered... I have discovered how to increase my output 2 to 3 fold which should bump the luxeon array up to the metal halide/T5 range. I am expecting anywhere from 10K-25K LUX, which would be similar to my overdriven T5 lights on my 240G tank (15K-50K LUX depending on location)

 

The new array will feature the same luxeon 3's, but now with special optical lenses which will focus the light and increase output by 80%. I also had been running them in parallel unknowingly at 50% current (40% output!). So by my calculactions I should get around 20-25K lux. That is right up there w/ metal halide and T5.

 

For all you others out there, this is a DIY PFO Solaris hood basically. A 'dumb' version w/o a microcontroller. Mine wil have the ability to control color temperature, but will require you to twist some knobs to do so. I could creat a PIC to control the array, but I am currently more concerned with output then anything.

 

I expect this to be the final revision... Revision 5. If I get unsatisfactory results, then I will be done with this array for good and will try to sell it on the forums.

 

*expect a new prototype posted in the next month*

Link to comment
  • Replies 138
  • Created
  • Last Reply

Results: Well, 8 hours of soldering and fussing with lenses later…

 

Watts consumed by LEDs and fan: 68W

 

Lux readings:

36” – 2000-2500

24” – 4000-5000

18” – 5000 – 6000

12” – 10,000-12,000

6” – 16,000- 20,000

0” – 70,000-100,000

 

*average LED output at water surface = 30,000 – 40,000

vs my overdriven icecap T5 lighting on my 240G

*average T5 output at water surface = 35,000 – 45,000

 

** Hotspots: these are average values, averages can change drastically moving just a few inches horizontally from the center point of one of the optic lenses. There are some very definite ‘hot spots’ created by the point source lighting. I would recommend keeping the lenses very close together if you were to create your own array.

 

 

DSCF2234.jpg

Final product. 15 whites, 7 blues and 2 greens. The reason for greens is because the whites are fairly lacking at the greens peak spectrum, so better safe then sorry. The only spectrum missing is 420nm, there are no LEDs that really reach that spectrum yet. There are ‘royal blues’ that hit 450nm but nothing at 420 really. I designed my array so the color spectrum is fully configurable. It made wiring much more complicated, but I think It’s worth the experiment.

 

DSCF2242.jpg

full lighting, I would call this ‘12K’ I feel

 

DSCF2245.jpg

Blues only

 

DSCF2247.jpg

Greens and some blues only

 

DSCF2246.jpg

attempt to dial to ‘14K’

 

 

 

DSCF2233.jpg

The optic lenses, these focus the light from 145degree spread down to 45degrees, increasing output by around 75-85%. Combine that with wiring them each at a full 700mA instead of 350mA like I mistakenly did last time, and you come up with nearly triple the performance.

 

DSCF2240.jpg

Hood rear view – all cables are detachable. You can see the color adjustment knobs next to each current regulator.

 

DSCF2232.jpg

The rat’s nest prior to optic lens mounting

 

DSCF2239.jpg

Hood internal view – I am now using optic lenses and more current regulators. The Middle optic was a real pain to mount because of the fan location. I ended up mounting it using a piece of a coat hanger. The rest are held down by 3 screws that squeeze the lens against the top of the hood. Believe me when I say, it was a total pain in the #%$& to get those LEDs to stay in position while I positioned the lens and screwed them down. Soldering was the easy part compared to that.

 

*Conclusion: The LEDs are just as powerful as T5, I don’t know about metal halide, but they measured the same LUX as my old 20K 250W DE HQI. The spread radius on this 6”x12” array gave about 18”x 12” coverage, quite impressive, but that is nearly $350 and 8 hours of work! Did I mention there is 0 heat transferred to the tank? The heatsinks get extremely hot but most of that heat exits the tank through the exhaust fan. I probably could add another fan or some small 5V fans to help the fuge lens, which is in the rear and runs much hotter. The heat will lower their lifespan if I don’t correct this. I feel that I could keep softies, LPS and SPS under these lights w/o problem given they produce the same lighting at the surface. You can keep them very close to the water surface as well. I am using an acrylic shield to keep them safe from salt spray. The LEDs should last 7-11 years based on our usage, but I’m sure that corrosion will rear it’s ugly head and ruin them prior to that. These are perfect lighting options for nano reefs, like I’ve said in the past, and great options for ‘office’ tanks that can’t have halide’s hanging above them.

 

*Future of this array and tank: I will add more photos, don’t worry. I will be drilling this tank and adding a 5G salt-bucket sump. I will use the LED in the rear for the refugium to export nutrients and also will create a mini DIY sulphur denitrator. The tank will be auto-topped off with limewater to maintain calcium/alkalinity. I will perform 50% water changes montly or weekly depending on need. No skimmer planned, but I do have a aquaCremora laying around. I will likely just swap the sump bucket w/ new salt water every water change using my 240G’s old water, which will be fine for this tank’s needs. I will likely pull some LR and zoanthids to start this tank, and will clip some SPS out of my main tank to try later. In the next month I’ll do all of this.

Link to comment
  • 2 weeks later...

DSCF2292.jpg

All the materials needed to drill the tank

 

DSCF2294.jpg

3 minutes of drilling later. The damn full of water kept the drill very cool. I'm so amazed that the $5 glass hole saw from ebay worked so well. Thanks Lau**

 

DSCF2296.jpg

Pre tank stripping

 

DSCF2299.jpg

tank is now stripped. I removed the background to gain more display area. I loved the background, but it served it's original purpose, to test if greatstuff foam would be safe and gain experience before I did the same thing for my full 240G reef. I also removed a divider from the back of the tank for a bigger refugium and spray painted the background flat black.

 

DSCF2305.jpg

DIY 1/2" spray bar painted black and a DIY 3/4" durso standpipe

 

DSCF2314.jpg

Here is the tank w/ the new spray bar and durso standpipe

 

DSCF2307.jpg

"But Zach, I'm scared and confused, why did you glue all the liverocks together using great stuff foam?"

-The reason I done this is because the rocks used to collapse all the time in the nano reef. When you have small rocks, they tend to not stay in place as well as they do w/ heavy rocks in a regular sized reef. I was hoping this solution would work...

 

DSCF2325.jpg

Ladies and gentleman... the amazing floating rocks.

 

*Up Next*

-trimming down the amount of foam in the live rocks so they stop floating... I hope.

-cover the exposed foam parts w/ glue and sand.

-add 1.5" of sand to the display, rubble rock, macro and heater to the refugium... of course water will come soon.

-finish the 5G sump using a 5G pail and a maxijet1200 for the return. (reason for this is for easy water changes and more stability)

-coral (zoas and sps) will come later. also will create a mini sulfur reactor for greater stability, and will attach my auto topoff device to add limewater to maintain calc/alk.

Link to comment

DSCF2333.jpg

Full tank shot with the new complete foam live rock. I coated the foam with superglue and sand to conceal the foam. It looks very real now. And yes, it no longer floats now that the foam is saturated with water. The flow from the closed loop setup is very nice, no dead zones that I can find yet. I calculate 35X turnover in this tiny tank!

 

DSCF2339.jpg

shot of the side w/ the refugium showing

 

DSCF2335.jpg

top down shot showing the overflow areas in the rear

 

DSCF2341.jpg

very simple ultra cheap sump. Includes maxijet1200 for a return and a autotop off float switch. To perform a water change, simply remove the lid, and swap the bucket w/a new one full of clean water. Nothing is inside the bucket, just more water volume to add stability.

 

DSCF2338.jpg

Shot of the top of the water, you can see the reflection of the LED optics... yes, there is a shimmer effect.

 

*soon I shall add zoa frags and a green clown goby.

Link to comment

Well, I have added a mini sulfur denitrator to my system. It basically is a simple device that acts as a low oxygen chamber for anerobic bacteria to convert nitrate to nitrogen gas, which simply bubbles out of the system. Read more about them here.

DSCF2347.jpg

 

I also added about 8 frags of zoanthids. I'm fairly new to fragging for myself, so I think I did a lousy job and many of the zoanthid colonies wont make it since they have glue on the top of their polyps. We'll see which ones survive. It's not a big deal anyways, my large 240G reef has plenty to spare, it's just a bother to get them out. The one SPS I put in there isn't happy, probably because the tank is having a mini-cycle. I'll post as things progress.

DSCF2346.jpg

Link to comment
the array cost me $200 out of pocket and about 20hours of work to date.

I am still not sure if my array was damaged from the overdrive period or if they just die early period. I seem to get one burnout every 10 days on average. It’s getting ridiculous and I am already getting sick of the project. If the array continues to die like this I may just convert to 70W MH... although that will probably be the same intensity, it will be less hassle. I'm definitely going to rebuild the array someday when more powerful 5mm's are available that are higher quality. I would just be happy if they stopped burning out though. By the way, the whites are the only ones that burn out, not the blues/reds/uvs. I did re-measure my array, the output is depressing but still very good for a nanocube. I don’t have the numbers at work but they were roughly

3" 8500lux

6" 7500lux

12" 6000lux

18" 5500lux

24" 5000lux

36" 4500lux

 

one side of the array is more intense then the other side for no apparent reason, I'm guessing one side was damaged by heat more then the other. I NEED MEASUREMENTS TO COMPARE THIS TOO still... I really wanted LUX readings off a 70w MH and 150w MH but I guess I will just have to assume its in that range and more powerful then 18W PCs. I'm still unsure of if doubling the powercompact bulbs actually increases LUX. that cant be possible can it? It can only give better coverage correct? So this still should give great coverage and intensity for a small tank. although I'm reaching the conclusion that a MH would be easier and about the same cost, although an ICE PROBE would be needed, it's not a big deal. LEDs are great for moonlighting and refugium lights but the jury is still out as a display so far. More follow-up research to come.

 

In my opinion youre probably experiencing LED death due to the lack of heat sinking in your setup. There is a difference between generating heat that will make your water get too hot (which isnt a problem here), vs the heat from each light which can damage the LEDs if its not managed properly. Usually LEDs (especially ones that are on for hours at a time) are mounted to PCBs that have heatsinks on them or the ground planes in the PCB itself will sink the heat out of its not too much. Mounting them on the plastic like you have done doesnt offer any thermal management at all.

Link to comment
In my opinion youre probably experiencing LED death due to the lack of heat sinking in your setup. There is a difference between generating heat that will make your water get too hot (which isnt a problem here), vs the heat from each light which can damage the LEDs if its not managed properly. Usually LEDs (especially ones that are on for hours at a time) are mounted to PCBs that have heatsinks on them or the ground planes in the PCB itself will sink the heat out of its not too much. Mounting them on the plastic like you have done doesnt offer any thermal management at all.

yep, the old 5mms suffered from excessive heat and thermal runaway. Several design flaws. It easily proved to be more cost/labor effective to build an array using luxeons high power LEDs.

Link to comment

IMO this looks better than the foamed-background.

I have actually been working on some LED designs,

 

I get my stuff from ledsupply.com, and they have 420nm 5mm purple LEDs. Not very high output, but the vawelength is there. Also, remember a "440 nm" LED is not just 440, it will probably have a range of 420-460 or something.

 

These are some notes I took comparing LED spectrums:

 

Peak nm:

chlorofyl b 453, 642

chlorophyl a 430, 662

 

Luxeon III - Royal Blue - 455nm <400-500>[440nm-460nm]

Luxeon III - Red - 627nm <600-675>[620.5nm-645nm]

Luxeon - Red - 625nm

5mm - UV - 400nm

5mm - Purple - 420nm

5mm - Pink - 440nm

5mm - Red - 630nm

5mm - Deep Red - 660nm

 

Also, about the spraypaint. The only "safe" paint I have heard of is Krylon Fusion. You said you used Rustoleum, didn't you have a couple of things mysteriously die? I'd stick with Krylon...

Link to comment
ihavenoclue

wow this thread is amazing!

 

He he he...While I was reading I completely forgot that this is all for fish and corals :D he he...

 

Sweet! keep doin what your doin!

Link to comment
  • 1 month later...
DSCF2341.jpg

very simple ultra cheap sump. Includes maxijet1200 for a return and a autotop off float switch. To perform a water change, simply remove the lid, and swap the bucket w/a new one full of clean water. Nothing is inside the bucket, just more water volume to add stability.

 

Well, I think I just figured out how I'm going to sump my nano. :lol:

Any chance of a plumbing diagram or photos?

Link to comment
Well, I think I just figured out how I'm going to sump my nano. :lol:

Any chance of a plumbing diagram or photos?

 

There is nothing to it. It is an empty bucket, one hose in, one hose out. There is a maxijet1200 inside the bucket that pumps water back to the display. The display is drilled as pictured, and has a external durso standpipe.

 

The tank is doing well by the way, it's fully cycled and has a handful of zoas that are growing very slowly inside.

Link to comment
  • 8 months later...

Well looks like you put a lot of work into this tank. I do like the "great stuff" wall you made. The display is a little small though. I would of personally put the fuge into the sump/bucket to make more room. It does look great though!!

Link to comment

Archived

This topic is now archived and is closed to further replies.

  • Recommended Discussions


×
×
  • Create New...