I decided to add a few aquariums to my collection and wound up adding a tower configuration to my apartment: a 20-gallon system sandwiched between two 10-gallon systems. Granted, it is not a very fancy, but the tower configuration is suitable for apartment living. As I worked on this project, I started to re-think some ideas I have had in the past about aquarium-keeping. First, I still love UGF's (undergravel filters), but this project helped me realize why so many people hate UGF's and what it is about the way I use UGF's that makes them work for me. Second, as much as I love UGF's, I decided to try a new filter design on two of the tanks (the bottom 10-gallon and the 20-gallon (middle)). Third, I also did a DIY aquarium background this time, so I'll spend a little time on what went into that. Finally, I am going to revisit cycling of the aquarium, silent cycling in particular.
As I said, I love UGF's: they are simple, easy to maintain, and are fairly inexpensive. However, many people loathe UGF's with a passion. For the longest time I could not figure out why people hated UGF's so much. I have heard people complain about them clogging up, being inefficient, and downright awful. I think these accusations stem from how most UGF's are designed. That is, most UGF's are either raised-plate designs or they are directed-flow designs.
By contrast, directed-flow UGF's (right; copyright not owned, used for critical evaluation only) are filters that were designed to combat the uneven distribution problem associated with simple raised-plate UGF's. Directed-flow UGF's overcome the uneven distribution of water flow by restricting the size of the holes near the uplift tubes. The problem with this is that it significantly reduces the surface area through which the water can flow--instead of the entire plate being permeable, directed-flow UGF's restrict the water flow to just a few narrow slits. This reduction in permeable space, though it nullifies the uneven distribution problem, makes these filters much more susceptible to clogging. What is more, many of these types of filters have multi-level plates which inhibit the effectiveness of gravel-siphoning as a means of removing the clogged debris. Personally, I have to agree with the general consensus out there: directed-flow UGF's really are terrible filters. They are just too susceptible to clogging and they are not very maintainable. Moreover, when these filters clog, they are apt to produce anaerobic zones that could produce sulfuric compounds which could kill an aquarist's fish. In short, I agree that directed-flow UGF's (as opposed to UGF's in general) are bad filters.
The more general problem, I think, is that UGF's are innately purely biological filters and they consequently tend to take a while to cycle. Thus, for people who want to keep a full tank of fish sooner than later, UGF's do little more than try their patience. UGF's take a while to mature and can be prone to mini-cycles during the first few months, hence most aquarists avoid them. Despite this slow start-up, they are quite capable filters. For example, I have a 40-gallon tank that is only half-filled with water (I keep newts in it that need to come out of the water from time to time) but, despite that it is only half-filled, I keep a full 40-gallon bioload in it (over 50 small fish (0.5 to 1.5") and about two dozen medium-sized fish (2 to 3")). The nitrate levels are a little high because of this extra load (around 40ppm--I prefer 10 to 20ppm), but otherwise the system is doing great. My point is that UGF's can handle high bioloads, but you have to work them up to it.
Nonetheless, even though raised-plate UGF's are manageable, I prefer to make my own UGF's. The reason I make my own is to correct the typical problems seen with UGF's. I usually make my undergravel filters our of four primary ingredients:
1) 3/4" PVC tubing and suitable connectors,
2) egg crate plastic diffusers (more often used as a industrial ceiling cover),
3) craft mesh (often sold at craft shops next to the embroidery supplies), and
4) small zip ties.
From these ingredients I can make sustainable UGF systems that do not clog very easily and have a fairly even distribution of water flow. As an illustration of what I mean, the reason I use 3/4" tubing is that it produces a fairly wide gap under the filter plate. Since the gap is wider, the flow rate is reduced and there are fewer problems with channeling. Also, I use a closed tubing design so that I can control where the water enters the uplift tube(s). That is, I make my UGF's such that the primary openings are toward the center of the plates rather than near the uplift tubes.
Making my version of a UGF is relatively easy provided that one has a measuring tape, a graphite pencil (most are graphite), a hack saw, a hefty pair of scissors (or shears), and a place to do the cutting. I managed to make all of my UGF's in my very small apartment and they all turned out well. I simply made my tubing frames about two inches shorter than the base of the respective tanks (with regard to both width and length), placing appropriate openings via PVC connectors such that the water flow would be directed away from the uplift tubes. As you can see (adjoining photos), I capped two of my filter plates rather than making them into ordinary UGF's. I capped them because I still wanted to be able to use them later yet also desired to use them as plenums in conjunction with a primary filter (I mentioned in the introduction that I decided to use some internal canister filters this time). After cutting the tubing, I cut the egg crate to size using my shears (wrapping a towel around the shears and cutting on top of an open bucket helps reduce the amount of stray debris), making suitable hole for the risers and uplift tubes. Next, I cut the mesh down to size, joining shorter sections together via the zip ties, and then zip-tied the mesh to the egg crate. That's all there is to it: simple UGF's that can be made at home which are superior to most of the commercial models.
As I said, this time I decided to try another filter design (we'll see how long I stay with it); I decided to employ internal canister filters in two of the tanks (the 20-gallon and the bottom 10-gallon). These filters were extremely easy to make and only require a few basic ingredients:
1) 4" black plastic piping,
2) 4" pressure-testing caps (full-pressure caps are expensive),
3) suitably-sized pond pump,
4) two 3.75" mesh disks (cut from the mesh described earlier for the UGF's),
5) 1/2" o.d. tubing,
6) 1/2" elbow joint,
7) one or two types of filter media (I chose red lava and aquarium gravel).
Instead of describing these filters in boring detail, I will simply say that the first mesh disk separates the pump from the lower media while the second disk is used to separate the different types of media (if applicable), that I filled in the top of my canisters with gravel despite the fact that I did not show such in the pictures, and let the pictures do the rest of the explaining:
For most of the coats of concrete, I mixed it to be paint-like (house paint, not acrylic or water colors) in consistency. However, I used the concrete directions to obtain a paste to apply to the back of the background to weigh it down (about 3/4"). Despite this extra weight, my background still tends to float a bit, but I can control this by keeping the water level an inch or two below what would normally be considered a full tank. In the future, instead of using a 1" backpiece, I plan on only using a 1/2" styrofoam backpiece--this should do the trick for a 10-gallon aquarium (I really don't want to have to adhere the background to the tank).
Lastly, I have not had any chemical problems as a result of the concrete. It seems that once it is cured (48 hours), it is quite fish-safe. Though it is a tad brittle, it seems to be handling the immersion and water flow quite well.
Most of the time I strongly advise against cycling with fish. That said, I am going to try to silently cycle my three tanks. The idea behind silent cycling is to keep the bioload so small at the beginning that the ammonia and nitrite levels, even when they spike, are only present in trace amounts. The process of silent cycling has the advantage of allowing an aquarist to keep fish in the tank almost immediately (within 48 hours of set-up), but there is a trade-off: stocking the tank is a very slow process when using a silent cycle (sometimes more than 3 months to reach a full stocking level). To cut down on this time requirement, I am seeding my new tanks with some bits of lava from my 40-gallon system. I will post more information on this silent cycle as time goes on. For now, every tank has one ghost shrimp and one adolescent male guppy. I will keep it at this level for one to two weeks.