[uncskainch]

Thanks for the detailed reply, oqsy. A few follow up "thinking out loud"questions:

(1) I realized that what's in the bell is not pure CO2, and to counter the possibility of the small bubble traveling from hose to bell having too much gas exchange, I placed the bell very low in the aquarium (also for aesthetics). So would that mean that the CO2 trapped in a bell that is placed very low in the tank and very close to the output hose from the yeast system would be more pure, in theory, than the gas trapped in a bell higher up in the tank?

(2) With this placement, since the big "burp" of overflow CO2 (sometimes a big bubble, sometimes very small ones like the ones coming from the hose) has to travel quite far to the top of the tank, I'm guessing that the negligable benefit of moving the bell upward, if there is any, is probably offset to some degree by the fact that overflow bubbles (even though they would be less pure than the bubbles coming straight from the hose) have a long way to go and would diffuse along the way?

(3) If the gas trapped in the bell is a mixture of CO2 and other gases, would the other gases build up at a rate higher than the CO2? In other words, over time, does the bubble in the bell get less and less pure? And if so, would it make sense to "dump" the bubble every so often and let it start over, accumulating a new bubble that is more pure? Or do the big burps of overflow more or less accomplish this, and I shouldn't worry about it?

And finally...

(4) I'm getting a surprising rate of CO2 production from this system. KH is 11 or 12 and PH has dropped from a solid 8.0 to a stable 7.0. PH sometimes goes from 7.1 or 7.2 during the day back to 7.0 at night, but that range has stayed pretty much rock solid since the initial downswing from 8 when the system was added. How long would you continue checking PH levels daily?

If this system really is more stable and less dependent on the relative strength of the yeast mixture at any given time, as I surmised above, it seems that I could, at this point, just let it keep going until the bubble rate can't keep up with the diffusion of the big bubble in the bell (ie. until the bubble in the bell starts to shrink and ceases to overflow ever). Then when I do add a new bottle of mix (probably inline with the first one so the dying bottle #1 and new bottle #2 work together), I theoretically should not really have to worry about PH swings from any shift in the mixture, right?

Ultimately, I'll probably keep obsessively checking PH levels through the first couple of bottles of yeast mixture just to be on the safe side and test out this stability theory, but I thought I'd ask if my reasoning is correct and that I shouldn't have to worry too much about big swings with this kind of system.