Mountains of sawdust (360g plywood, LED, Arduino build)

[der_wille_zur_macht]

Taqpol,

My tank is basically two of yours front-to-back. I'll likely have between 180 - 200 LEDs. IMHO that's probably a little lower than many people would do on a tank this big, but I have a few justifications:

1) The eurobrace is 4" wide. I'm not planning on putting any livestock underneath it, so the "true" area I'm lighting is more like 66" x 40", not the full 72" x 48".
2) I'm using XP-G for cool whites, which are ~30% more efficient than the typical XR-E, which means 30% fewer cool white LEDs required.
3) I will be carefully placing the LEDs so I'm not wasting a bunch of power lighting any "open" areas, i.e. empty sand!
4) I will deliberately be building the LED rig to give high intensity near the top of the rockwork and lower intensity near the bottom, because I WANT a subtle progression of coral types from the "top" of the reef to the bottom.

I'm also likely not using optics, though many probably would on a tank this deep - mainly because of #4 above. Plus, on a larger tank, a smaller percentage of the LEDs are in danger of spilling outside the tank without optics. There just aren't any good, cheap wide optics for the LEDs I'll be using (XP-G, XP-E and potentially Rebels) anyways.

That'll be it for the main lighting on the tank. There will also be two or three other small, separate LED fixtures associated with the tank - one as a moonlight, and two or more as "special effects" during the day.

 

[Taqpol]

Do you worry about spotlighting your rock islands but leaving open sand looking too blank? I've heard a few plans for getting around this (Having a few tightly packed blocks over your corals and some linear 6-12 LED strips spread out over the sand bed) but do you have a specific plan yet? I was also considering keeping 2-4 of my current 48" T5's and using them around the edges to fill in whatever the LEDs don't cover.

Originally I was thinking of using three 24 LED blocks in place of where most people would put MH's, but 72 LEDs doesn't seem to be enough. Now I'm thinking of using four 24 LED blocks and clustering them over two islands I will have on opposite ends of the tank. My problem is that I want to be able to keep my current derasa and Squamosa clams on the sand bed. Given that of the four these are the two "lower light needing" clams, how many LEDs and in what configuration do you think would be best for that kind of penetrance?

 

[der_wille_zur_macht]

It's hard to describe some of this in words, but no, I'm not "worried" about spotlighting. I specifically intend to create it, in fact - I WANT there to be a large variation of light throughout the tank.

Diving on real reefs, the first thing I'm always overwhelmed by is the overall brightness. But right behind that, you immediately get a sense of contrast, shadow, and high variation in light. When you're down on the face of a slope or other big feature, there are shafts of light filtering down, dark caves, and TONS of shimmer. That's the sort of look I'm going for. I don't WANT 100% uniform lighting from edge to edge, to me that just looks bland and phony.

That said, I think I'm in the rough ballbark, overall intensity-wise, where I COULD create the consistent look you'd get from T5 or MH if I wanted to.

 

[der_wille_zur_macht]

But let me clarify. When most people talk about "spotlighting" it's because they used really tight optics and big spacing, and they get "cones of light" in the tank. I'm going to be "spotlighting" in the sense that a certain rock feature might have 20 more LEDs over it, vs. a lower overall density elsewhere in the tank, etc.

 

[username in use]

I like where you are going with the lighting. Should yield a very dramatic affect. Cant wait to get some photos of this.

 

[CyclistMT]

So DWZM I thought it was time to bump this back up to the top because...

1 - There hasn't been an update for quite some time. Come on man! Get some glass in that sucker and water test it!

and

2 - I have a couple of questions.

I've been following your build here and over on Fingerlakes and things look to be coming together nicely. Great job! So, I'm comtemplating building my second wood tank. The last one was about 150 gal and I used the Sweetwater Epoxy paint. The size of that tank made it really solid with no deflection so the epoxy paint worked just fine. However, with this next tank I'm basically going double the size and will be using 3 panes of glass like Kent did over on Fingerlakes. My dimensions will 60x48x24 (maybe 26 to account for waves, I don't know yet).

So on to my questions. First I'm interested to find out how much deflection you get since you didn't use any cross bracing. I would like to avoid that too if I can but the jury's still out on that. Maybe you're not that far along yet.

Second, I'd like to know more about your filleting technique since you decided to not use the 45's. I'm hoping to avoid these too becuase I think it makes it overly complicated, however, I will have 3 larges panes of glass so maybe I'll have to do it anyway. Anywho...by filleting, do you mean you just poured epoxy in each corner to a certain depth with no other reinforcement? That's how I'm understanding it but it's kind of hard to see in the pictures.

Thanks for your help and good luck with your build!

 

[der_wille_zur_macht]

So DWZM I thought it was time to bump this back up to the top because...

1 - There hasn't been an update for quite some time. Come on man! Get some glass in that sucker and water test it!

Sorry. All my "hobby time" has been sucked up otherwise lately. In the middle of fiddling with some of the electronics right now. Also we're in the middle of selling one car and buying another, so spending hundreds of dollars on glass is on hold for another week or two while things settle financially.

So on to my questions. First I'm interested to find out how much deflection you get since you didn't use any cross bracing. I would like to avoid that too if I can but the jury's still out on that. Maybe you're not that far along yet.

Not that far yet, but with the super-beefy eurobrace, I'm not expecting anything significant. I mentioned this earlier, but I can walk/jump on the rim of the tank with no issues. When I do fill the tank, I'll have a measuring device of some sort on the top so I can monitor deflection. I was thinking about making a "telltale"stick - clamp a stick to the back of the tank, laying across the opening on top, and put a pin in it sticking down such that the pin is a tight fit against the front eurobrace (if that makes sense). Then, as I fill it, I can check for a gap between that pin and the eurobrace as the eurobrace moves outwards. I'm hoping this measurement will be on the scale of using feeler gauges, not a measuring tape.

Anywho...by filleting, do you mean you just poured epoxy in each corner to a certain depth with no other reinforcement? That's how I'm understanding it but it's kind of hard to see in the pictures.

That's exactly it - so basically I end up with a solid epoxy "45" in each corner. This is a very brute-force method from a composites point of view, but we're not really worried about strength/weight ratios, just ultimate strength. I know a bunch of people have done the wood 45's without a single issue, but it makes me nervous to mix thick plain dimensional lumber with plywood like that. Plus, since I didn't have to worry about air bubbles making a leak in my fillets (they're way to thick for that to be a possibility, and the corner was already sealed from the pours on the two adjacent sides) I could pour the fillet in one go, which cut down on time and effort compared to pouring several layers over a wood 45.

 

[Soultwater]

Interesting build... subscribed
I could see myself doing one of these

 

[salty joe]

I like where you are going with the lighting. Should yield a very dramatic affect. Cant wait to get some photos of this.

Same here!

I also very much like the idea of creating a sense of not being able to see everything in the tank at once. I plan on that effect, but I'll use a couple big rock structures and have more little fish and creatures that hide. Kind of the opposite of an in your face approach with lots of big showy fish.

You do nice work.

 

[der_wille_zur_macht]

Same here!

I also very much like the idea of creating a sense of not being able to see everything in the tank at once. I plan on that effect, but I'll use a couple big rock structures and have more little fish and creatures that hide. Kind of the opposite of an in your face approach with lots of big showy fish.

You do nice work.

Big rock structures and lots of little fish planned here, too. IMHO when you have big fish in even a big tank, the sense of scale is a little lost. When you have lots of tiny fish in a big tank, the difference in size between the fish and the tank make the tank look even larger than it is. Again, personal preference but that's my take on it.

As such, I'm planning on two types of small fish. First, a small schooling fish (some type of cardinal maybe). And, LOTS of small fish that hide - blennies, gobies, that sorta thing. Then, a few medium to medium-large fish just to balance it out.

 

[widmer]

It's hard to describe some of this in words, but no, I'm not "worried" about spotlighting. I specifically intend to create it, in fact - I WANT there to be a large variation of light throughout the tank.

This is going to be stunning.

But after reading this, I'm particularly interested in why you aren't so interested in buying the heatsink strips from mpja as we discussed in the DIY LEDs thread. You could put a dozen LEDs on a heatsink or two and with a meanwell per, and with a really simple bracket/rail, it would be very modular, easily-modifiable light distribution system.

BTW how many LEDs are you planning on at this time? I got the part about no lenses but wasn't clear on the number?

 

[der_wille_zur_macht]

160 - 200 LEDs depending on which LEDs I go with and drive current, which I haven't settled on yet. And no meanwells, I'm DIY'ing the drivers. You're right, the strips of heatsink would be really nice, but I'm not totally convinced it's better than strips of extruded angle. I know the heatsink material would give me more LEDs per piece, but I don't need/want them that close together, really, and the angle is just so much cheaper. I'm not ruling it out, just not convinced it's the best.

When I get closer to building I'll draw something to show what I'm planning and maybe then it'll be more clear.

 

[widmer]

How many LEDs is each of your drivers going to push? And what's the voltage input to each of them (household AC or DC fill_in_the_blank volts)? I'm afraid I wouldn't be able to wade through any of your arduino/driver threads here...

 

[der_wille_zur_macht]

Depends on which day of the week it is. I've gone through a few different designs. The current design is very buckpuck-like. It'll take ~24v and drive 6 LEDs. I'm able to fit three of these circuits on a 5cm square PCB, so I'm basically driving 18 LEDs from each PCB. The cost to drive each LED is around $1 - $1.15, whereas if you assume a meanwell ELN costs $30 - $40, it costs $2.50 - $3.33 per LED. Buckbucks are around $3. Those prices include driver and power supply, if required.

 

[thenewone]

good to know

 

[crvz]

Depends on which day of the week it is. I've gone through a few different designs. The current design is very buckpuck-like. It'll take ~24v and drive 6 LEDs. I'm able to fit three of these circuits on a 5cm square PCB, so I'm basically driving 18 LEDs from each PCB. The cost to drive each LED is around $1 - $1.15, whereas if you assume a meanwell ELN costs $30 - $40, it costs $2.50 - $3.33 per LED. Buckbucks are around $3. Those prices include driver and power supply, if required.

Gotta question about the drivers, if you don't mind. I believe the meanwell's can take a 120V input, right? But the buckpucks require the 24V input via a power supply. I assume you can run more than one buckpuck on the same 24V power supply (which is plugged into a 120V source), yes? Or is that a bad assumption.

the reason behind my question is that I want to be able to turn on ~6-8 LEDs at one time (as they're being used for dawn/dusk), but I'm thinking if I run buckpucks on one power supply, I lose my ability to turn them in such small intervals via the 120V Neptune DC8 controllers. Based on these thoughts, I'm thinking it would be cheaper to buy meanwells and run 8 LEDs on them instead of buying buckpucks each with a dedicated 24V power supply. Make sense?

 

[widmer]

DWZM I'm going to return the favor and send a friendly reminder to keep this thread on topic

crvz I took a stab at answering your question in our designated general LED Q/A thread:

http://reefcentral.com/forums/showthread.php?t=1678127&page=83

 

[salty joe]

So I was laying in bed last night and thought, Wow that would suck if a plywood tank had an overflow or got a big splash and water seeped between the bottom of the tank and the stand. Assuming the stand is covered with a sheet of plywood. Do you think it's worthwhile to epoxy the bottom of the tank and maybe even the top of the stand?

 

[der_wille_zur_macht]

I suppose it would be worthwhile if you were ultra-paranoid, but my approach is that the "outside" of the tank, the stand, etc. should be handled the same as any other system with wood near a saltwater tank - i.e. protect it, but you don't have to go nuts.

The paint I used on the outside is lifetime-warranty exterior house paint. If it's supposed to last a few decades of wind, rain, snow, and sun on the outside of my house, I'm sure it can stand the occasional splash or puddle of water.

FWIW, the "top" of the eurobrace got a nice thick layer of epoxy, and I let it drip/bleed down the outside of the tank a few inches, so that splash-prone area has extra protection. Same around the bulkhead holes - I painted on epoxy for a good 4 - 5 inches around the outside of each hole, in addition to sealing the cut inside surface of the hole.

When I do the stand, it will probably get the same paint I used on the outside of the tank. The "floor" in the stand will get something a little more durable (spar urethane maybe - I'm out of epoxy) and will be sealed to form a "tub" to catch overflow or other potential disasters from the sump. And the controller I'm building will have a "spill sensor" in that area in case that does happen - it'll be able to sound an alarm and potentially text/email me.

 

[shunter]

Great build! i am really enjoying this, keep it coming...