Wildly Oscillating CO2 Spinamajig

[JoeStrout]

(Starting this as a new thread, as the other had gotten rather long.)

I got a chance this morning to load up your (Permeable Ceiling's) Powell colony to see what's going on with the CO2.




It appears to be basically a population oscillation, where the "population" in this case is the grass and trees.  When CO2 is high, these do great and grow like gangbusters (population boom), converting CO2 to O2... but then there isn't enough CO2, so they all wilt and stop photosynthesizing (population crash).  And then the people gradually convert the O2 back into CO2, starting the cycle again.

It's still surprising that the oscillations don't dampen out to an equilibrium, though... especially with the atmospheric regulator in there, which is supposed to counteract any imbalance.  There is a bit of a delay in those too though; it looks like somehow you've hit upon a resonant frequency of both the regulator and the plants (probably tied to the time step in the atmosphere simulation).


But what really cracked me up — and prompted me to post this screen shot — was the Squawker feed.  Your residents don't care about the growing/wilting plants too much, because they're too busy trying not to lose their lunch from the 3.9 RPM spin. 

[JoeStrout]

Hmm. Actually, no, the grass and trees have little to do with it. This problem is entirely in the atmospheric regulators.

The give-away is the CO2 production (or consumption) listed under "Buildings". Each regulator can produce or consume up to 1000 kg of CO2 per day. They set their production/consumption level once per day, based on the current CO2 levels using a simple look-up table that goes all-out if CO2 is under 0.03%, or over 0.05%. In between, they linearly interpolate (so they do nothing at 0.04%).

The problem is that this colony is so small, adding or removing 1000 kg of CO2 will remove ALL the CO2 from the atmosphere, or drive it from zero up to 0.16%. So, each day we bounce between "zero" and "too much." And that's with a single regulator (having two just makes it worse).

Clearing the atmospheric regulators eliminates the oscillations, but then CO2 slowly creeps up, as there aren't enough grass and trees here to balance out the people. Eventually that should reach an equilibrium as people find the CO2 too high, but it won't be a pleasant equilibrium. And to balance this out by adding more trees would be pretty hard (it takes about 16 trees per person — see this thread).

We really need the atmospheric regulators to be a bit smarter about how much CO2 they consume or produce, but for technical reasons that's not as easy as it sounds. But we'll come up with something... and at least now, we understand the problem!