DIY Aquarium Stand Completed

This project took a long time to complete.  I guess that's one of the distinguishing characteristics of a hobby.
 
I mostly finished the stand proper in the spring of 2011 (blog entry), but it took me another two and a half years to construct the hood and finish the trim.  In the meanwhile I replaced the original glass tank with an acrylic tank, switched from an all-Tanganyikan cichlids biotope to a heavily planted, low-pH biotope, and went through several iterations of lighting and auxillery equipment related to that migration.  This post will just cover improvements to the stand/hood since the first post.

Demise of the Glass Tank

I started with an inexpensive used glass tank from Craigslist.  After about 8 months use, I had my only overfilling accident: I was using a python to refill after a water change and neglected to pay sufficient attention.  The tank filled completely to the brim then spilled a few gallons on the floor.  The floors are concrete, so there was no visible damage, but some water probably did seep into the wall/floor joint immediately behind the tank.  That night, about 3 am, we woke up to a sound like a fountain.  One tank face had cracked near the top, the water level was down about 30 gallons, and the spray bar was making a lot of noise.

I suppose I'll never know exactly what caused the failure, in particular whether the stand proved insufficiently flat or rigid.  The location and shape of the crack suggested that outward bowing force from the water was the immediate cause.  After overfilling, I left the water level an inch or two higher than I previously had.  And perhaps there was some deformation somewhere due to water infiltration swelling.  In any case, after thinking it over, I decided to go with acrylic for the replacement, and ordered one from Tru-Vu aquariums.

Seating the Acrylic Tank

I was very concerned to minimize flexion stress on the new tank, so I seated it on top of a sheet of 1/2" styrofoam.   The filled tank has been on top of it for 2 years now, with no visible compression.   The acrylic tank rests on its entire bottom, so the weight is well spread out.  The glass tank rested only on its edge.  I had seated that tank on a strip of 1/4" heavy-weight neoprene, but when I remove the broken tank, I found that the neoprene had compressed to almost nothing.

Hood

The hood is very straightforward: a partial box of 1/4" plywood attached by hinges to a wooden frame that rests on top of the tank.  The hood's lower edge is about 1.5" below the top of the tank, so the water surface is not visible.  All wooden parts got a couple good coats of outdoor urethane.  In addition to splashing, the inside of the hood is a high-humidity area.  It's been in use for a year with no sign of moisture damage.

Lights

Originally I just thought to illuminate the fish in the evening; eventually I ended up with a hi-tech plant setup.  I've now got 4 48" (54 watt) T5-HO bulbs, each with a reflector, mounted to the inside of the hood.   The reflectors are screwed into plywood strips that are epoxied to the inside of the hood, so no fastening hardware appears on the outside.  I used the Glo T5 HO kit, and the ballasts hang behind the back of the tank.  I'm currently testing using a Finnex LED fixture for more intense area lighting, that just rests on top of the tank under the hood,

Ventilation

The back of the hood was left completely open, so air can freely circulate.  But when all the lights are on, it gets quite warm under there, and I was noticing temperature swings of 6-10 degrees F above the thermostat setting.   (The ambient room temperature can climb into the 80s during the summer, so it's not just the lights.)  I decided to try an evaporative cooling solution.  I covered 80% of the open space at the back of the hood with acrylic sheets into which four low-voltage computer case fans are mounted, that switch on and off with the lights.  When the lights and fans are off, there's enough passive ventilation to avoid moisture problems inside the hood.   When the lights come on, the fans blow outside air over them and across the water surface.   The acrylic enclosure ensures the cooling air passes through all of the hood volume.  On warm days I lose about a gallon to evaporation. The temperature is very steady within 2 degrees of the heater thermostat set point.  A possible enhancement would be to use another thermostat to control the fan speed, but that hasn't yet been a priority.

Bottom Trim

The final piece, installed last week, is the trim around the bottom of the tank to cover the plywood/styrofoam sandwich.  By waiting 2 years to install it, I learned something valuable, which is that uncovered gravel around the edges of a tank in a well-lighted room can contribute to noxious algae problems.  For a couple of months I experimented with taping cardboard around the bottom of the tank and found that shading the gravel was an important factor in eliminating a persistent cyanobacteria problem.  I ordered a single strip of 4" wide, 1/16" thick aluminum sheet from Online Metals, and screwed it into the plywood.  I was lucky enough to get access to a sheet metal brake to make the bends.

Under-tank space is cramped, but I've managed to stuff in two large canister filters, a 10lb CO2 tank and control system, a UV sterilizer (Aqua brand this time), and the lighting control timer.

Does a UV sterilizer kill algae? Yes (mostly, and so far).

About 6 weeks after fish were added to my new tank, the water turned green.  It's in a room that gets a huge amount of natural light, the tank even gets a few minutes of direct sunlight each day, and that's not going to change.  It's an African cichlid tank with a high pH, and I'm not sure plants are going to be a workable option, though I plan to try it some time in the future after I have a hood and lights installed.  I read varying reports on the internet about whether a UV sterilizer can cope with this problem: the argument against is that given adequate nutrients and sunlight, algae will find a way to grow; the only real solution is to remove the nutrients (e.g. by growing other plants).  Two weeks after installing a Turbo Twist "Pond" (yeah, right) sterilizer, the problem has been reduced to an unobtrusive but still detectable level.  The shots below are taken from the side, looking down the 7 foot length of the tank.  It's only 16 inches deep, so from the front the water's now quite clear.  The water was most green 1 day after installation because the color was varying more strongly then day to day, depending on weather and light.

Day 1

Day 2

Day 3

Day 5

Day 9

Day 14

Details: Tank is about 135 gallons, with 2 Eheim 3e 2078 canister filters each rated at 400+gal/hour flow.  Each filter has an inline heater in the outflow line.  I added a Turbo Twist 6X to one of the outflow lines which visibly reduced its flow, compared to the other, so maybe it's in the 250-300 gal/hour range now.  There is still algae growing on the rocks and gravel, which is kept under control with algae eating fish.












Update: 2013/01/03
The Coralife TurboTwist UV sterilizer failed after about 6 months.  It was troublesome from the start: the plastic parts are brittle and easily broken, which was the final failure mode, but the bulb also failed to seat well in its socket making its performance flakey.  I don't recommend this unit as it proved to be low quality and non-robust.

For the time being I'm doing without a UV sterilizer, though I may try another model in the future.  I've added plants to the tank and increased the light greatly.  I'm going through various algae blooms again, however not green water this time.  I've read elsewhere on the web that green water will burn itself out and clear after a few weeks with consistent tank care and no UV sterilizer.  I don't have much evidence one way or the other, but it's possible that my positive experience with UV sterilization was at least partly fortuitous timing with a natural cycle.

Update: 2014/01/20
Adding plants to an aquarium makes the algae control problem much more complex, because one needs to maintain some level of nutrients in the water.  Last spring I decided to add another UV sterilizer, due to very bad outbreaks of BGA (cyanobacteria).  It didn't seem to help much with that problem, but I"ve kept it running on the suspicion that maybe it helps slow some algae and pathogen outbreaks, more than it may interfere with some beneficial processes.  Who knows.  In any case, this time I installed an Aqua brand UV sterilizer which I can recommend as a well designed and constructed unit.  Mine has been trouble-free for close to a year.

Winter Is Coming

All the rain that falls on my roof or yard ends up in a pair of sumps by the side of the house, from where it's pumped uphill then out into the street swale.

If the pumps fail, the house floods.

Last weekend I did an inspection and discovered that 2 of the three sump pumps were inoperable, plus the covering deck was starting to crumble from rot.  I tore up the old deck and got a plumber in to replace the pumps.  There's now one in each sump, with a crossover connection near the top.

Then I rebuilt the deck on top.  I thought about using composite, since it ought to be very rot resistant, but wood's still a lot more comfortable to walk on in bare feet, when it's in the hot sun.  So here's another redwood decking job.

I finished it with penofin verde, an environmentally friendly oil that cleans up (sort of) with soap and water.

DIY Aquarium Tank Stand

First things first: here's what the fully assembled stand looks like with the empty tank mounted.  (See completed project here.)

The Backstory

While thinking about how to move a couple of smaller tanks full of fish into a new house, it occurred to me to get a new large tank, set that up first, move all of the fish into the new tank, then move the empty old tanks.   After scanning Craig's list and making a couple of phone calls, the next thing I knew I had 200lbs of empty glass tank sitting on my living room floor and a major design/build project taunting me.  I ended up moving the small tanks with a few inches of water and fish in the bottom, months before finally getting the new tank set up.  It turned out to be much more difficult than I anticipated to come up with a functional tank stand I thought I could live with for the next decade.


Design Constraints

The aquarium is going to be located in a large room with a modern design aesthetic: the floors are polished concrete, while the walls and ceiling are a combination of glass, brushed aluminum, wood and stucco with some prominent exposed steel structural elements.   

The tank is 125 gallons: with water and rocks the dead load to be supported is approximately 1500 lbs.  Also, the house is located about 1 mile from the San Andreas fault, so the stand needs to be quite strong with respect to lateral shear forces.

Design Process

I spent close to two months iterating on design.   I initially considered conventional wood frame/plywood sheathing designs, but didn't like the look.  I wanted to get away from the 70's rec-room vibe it conjures up for me.  I also wanted to avoid a hulking appearance, so I played around with leaving parts of the stand unenclosed.  I figured metal framing would allow the most design possibilities.  


As usual, all of the difficulty came in trying to strike the right balance minimizing cost (roughly amount and kind of material used) while maximizing design capabilities (where more and more expensive material helps).  


Update: I originally intended to say something about the technical design process, but that was almost 3 years ago.  The short story is that it's possible to estimate the deflection of a beam under load by standard equations and engineering specs for your material, but very difficult, so far as I could learn from the web,  to estimate how much stress a joint can bear without failing.  Accordingly I probably over-sized the beams in use, and put a lot of effort into designing sheer panels to prevent the frame from collapsing laterally.

Here's the final sketch I emailed to 80/20, who converted it to a CAD drawing.

Here's the kit from 80/20 laid out and ready to assemble.  I was impressed by the accuracy of the cuts and other machining.  It was better than the tolerances promised.  The frame went together very square without any trouble.

Starting to attach the cross pieces to the front/bottom beam.

Bottom together, but still only loosely.

First vertical columns going up.

Perimeter beams & columns set up, cross pieces going in loosely.

Cutting the plywood shear panels.  I used 11/16 plywood, and glued two together for a final thickness of just under 1.5", which gave a slight inset from the aluminum framing on the faces.  I cut the panels 1/32" oversize, then tightened the frame around them on installation.  Since the 80/20 machining accuracy was very high, this worked out quite well.

The outer panel faces were stained before cutting and gluing.  After gluing I did a touch-up staining, and then several coats of urethane for water resistance.

Two pieces of shear panel clamped after gluing.

The top is a single thickness of 11/16th plywood.  There's an invisible back shear panel that leaves space for plumbing/electrical at the sides.  Here I'm fitting both panels prior to finishing and attachment.

The finished shear panels that are externally visible are attached by friction (tightening the frame around them), and also by internal bolting to the frame.  My design guess is that this increases shear strength of the frame considerably.  However, I don't plan on any destructive testing to confirm it.

Front view of the three visible shear panels.

Back shear panel installed.

Fitting the equipment cabinet door.  I left 1/16th inch clearance around the edges, and wedged it into position with cardboard while marking where to drill holes to fasten the hinges.

Internal view of where the door hinges will attach.   Structurally, this is an inset frameless cabinet door.  The usual hinge would be a European style with a mounting cup drilled out of the door face.  I didn't want to buy extra gear just to drill a couple of holes (and don't have a drill press either), so I looked around a lot and finally found some nice self-closing hinges of with the right properties, online at San Diego Hardware.

Hinges installed.