I ran out of CO₂ gas today for the Fireplace Aquarium and swapped out the old CO₂ supply cylinder for a new one.  I get 600 g CO₂ at a time in disposable cylinders used for MIG welding, and sure enough, the empty cylinder weighed 1200 g and the new cylinder weighed 1800 g.

End of tank dump

I few weeks ago, I noticed the CO₂ bubble rate had increased substantially due to increased flow rate through the regulator on the CO₂ tank.  Regulators are the first step in controlling aquarium CO₂ and regulators work best when the pressure inside the CO₂ tank is constant.  Most of the time is this easy because inside the cylinder the CO₂ exists in both a liquid and gas phase in equilibrium, and so long as there is any liquid CO₂ at all, the equilibrium maintains a constant pressure in the tank of around 860 psi.  When the CO₂ is running out, however, there comes a point when there is no liquid remaining in the tank at which point the pressure of the gas phase CO₂ gradually drops as it is exhausted.  Single stage^* regulators “notice” the pressure in the supply cylinder is dropping and attempt to compensate by opening more fully, however they inevitably over compensate resulting in increased gas flow even though the source cylinder is at lower pressure.  In some cases this failure of regulation can happen quickly and dramatically and can “dump” all the remaining CO₂ gas into the aquarium, poisoning the animal residents (the plants won’t be bothered however).

With my set-up the regulation failure is not so severe, and can be easily mitigated by awareness a problem might be coming and by adjusting the flow through the regulator down a little on a daily basis until the cylinder runs out gas completely.  I get more than three months of constant pressure CO₂ and then a couple weeks of manageable instabiliy at the end.  I’ve set a 3 month reminder for myself to remind me when to start carefully monitoring the end of this new cylinder.

^*Dual stage regulator alternative

A dual stage regulator is designed to maintain constant flow even if the source pressure changes and can avoid the end of tank dump that can happen with single stage regulators.  Dual stage regulators are much more expensive than single stage, can be easily and erroneously confused with “dual gauge” regulators, and don’t seem to be designed for the disposable CO₂ cylinders I like to use.

Today I shifted the location of the CO₂ diffuser from the left wall of the tank to behind the moutain and underneath the powerhead.  The atomised CO₂ gets sucked up directly into the flow and actively pushed around the tank.  Previously I had the diffuser on the opposite side of the tank hidden under the plants and whilst it seems sensible to put the CO₂ where the plants are, because the flow pattern is circular, that puts the plants “upstream” of the diffuser last in the queue – the water has to do a full circuit to get back to them and by that time all the microbubbles have already reached the surface.  Those plants can still benefit from the dissolved CO₂ of course, but they won’t get any microbubbles trapped under their leaves.  The diffuser is also now hidden behind the mountain, which is good, and is shaded by the powerhead to reduce algae growth on the diffuser, also good.

While I was moving things around and doing a cleanout I also took the opportunity to clean the diffuser by removing it from the tank while still active, layering on a few drops of liquid carbon, and letting that go for 10 minutes or so.  Back in the tank this increased the CO₂ flow rate from 120 bubbles per minute to 130 bubbles per minute, so that worked well.

The fish seemed pretty happy about the whole thing but it will take a few weeks to get an opinion from the plants and algae.

Notes:  Cute video of a 5-banded barb swimming through the tunnel under the mountain.  The plant growing on the mountain is Bucephalandra caterina.  The carpeting plant at the base of the mountain is Marsilea hirsuta.