RFUG + CO2 under gravel

I'm off on a new adventure! This week I will rebuild my aquarium, using a different method for introducing CO2. I will put a reverse flow undergravel filter under a substrate made of ZeoBest, zeolyte sand, and run an inline reactor in the canister filter outflow line, with that CO2 enriched water going into the RFUG. At the moment I plan this to be a lower light, CO2 tank.

First, I started building a RFUG that will do what I want. I don't want it to restrict the canister filter flow much, and I want it to evenly disperse the CO2 enriched water all over the tank bottom. I designed this as a manifold with outlet holes for water to exit under the gravel. Tom's idea of using PVC piping for this, with the exit holes at the bottom of the tubes, is what I adopted.

Manifolds work by slowing down the flow velocity, so they act as a source of fluid for all of the outlets. So, my first objective was to have the tubes of the RFUG have a total internal area much less than that of the manifold portion of the unit, and all of the plumbing up to the manifold having more internal cross section area than the filter output hose. The output holes themselves should be small to get a higher velocity exit water flow, but large enough so their total area doesn't restrict the canister filter output.

To make this easy to build I chose to use standard PVC fittings and pipes, not CPVC, with the smaller pipe sizes, but with a much more limited selection of fittings easily available. My canister output hoses are 5/8" ID, and 1/2 inch PVC pipe is just slightly smaller at .60 inch ID., so I used 1/2 PVC pipe up to the manifold. Then to slow the flow, I used 3/4 PVC pipe and fittings for the manifold, with an I.D. of .80 inch. The area change is the ratio of the diameters squared or about 1.8, dropping the flow velocity by about half. The perforated tubes which exit the manifold are back to 1/2 inch PVC, but with 5 tubes, each only carries 1/5th of the total flow, so those tubes also act as manifolds. To avoid the exit holes in the tubes reducing the canister filter flow too much I picked 7/64 inch diameter, and 50 total holes, with a total area of .60 X pi over 4. This compares to the canister filter output hose internal area of .39 X pi over 4, so it won't be the exit holes that determine the canister filter flow. If I had used 1/16 inch holes, as I had intended to do, those would have had a total area of .20 X pi over 4, which would have restricted the flow from the canister filter. Obviously, there are many combinations of hole sizes and numbers of holes that would work well, but I wanted my holes to be approximately on a square grid, and this does that.

Here is the RFUG before any joints are glued or the standpipe spray painted black:


And a closeup of the hole size and spacing:


You can see this is made of standard PVC fittings, connected by short stubs of PVC pipe. The brass valve is a drain valve to be used to bleed off the air anytime the canister filter has been turned off, for cleaning, etc.

Theoretically this will work great, but to test it I used a garden hose on the inlet, with the holes all pointed up, and all holes had about the same height "fountain" of water coming out.

 

I finished setting up the tank with this filter yesterday. If you want to follow the progress of it, go to: RFUG - A Blast from the Past - DIY.

So far this seems to be working fine. No burps of CO2 yet, but a few very small bubbles float around in the tank, and an occasional 1/8" bubble will pop up when the substrate is disturbed by a loach. My drop checker is still blue, even though I have a substantially higher bubble rate than before, when I was feeding CO2 into a small powerhead to chop up the bubbles, and had a consistent yellow green color. I'm assuming for the moment that I had very inconsistent CO2 distribution before, but high enough around the drop checker to turn it green, and now have much more consistent, but therefore lower over all concentration. This could be why I have had such a battle with BBA.

 

There are always surprises when we try something, but I think it is highly unlikely that this would work at all with a nutrient rich substrate like ADA aquasoil. And, the powersand would only make it worse, in my opinion. You can see that running the water up through the substrate would transfer a lot of nutrients in an uncontrolled way into the water. I don't think it would be worth the effort to even test it.