The usual check: a look, a sniff, a couple of numbers in the log. Today they will turn out to matter.
Lid open: the smell is neutral, there is more condensation on the walls than usual. The worms are slightly lower than yesterday.
Core temperature 26.8 °C, moisture 82%. Two numbers and a note about the condensation, that’s all.
The reading fits into a trend: moisture has been rising for three days and the core is warm. On its own, each sign is not yet a crisis.
For now it’s just a morning. In four and a half hours these numbers will add up to a signal.
Not “something’s wrong”, but a breakdown: what happened, why the system decided so, what to do and what comes next.
Without the log this day would have looked ordinary: the smell and the worms leaving would have shown up two days later, when recovery takes weeks, not minutes.
Vermix did nothing by itself: it explained what to do and why. A person does the hands-on work.
With a fork, a hand’s depth down: get oxygen back into the core and release the heat. Seven minutes of work.
A note in the log: the next feeding is pushed back 24 hours, so the heat source is gone.
ΔT 3.1 °C, below the 5 °C threshold. The moisture trend has turned around. Protocol closed.
A shortage of available organic matter slows processing down. The population stays alive but stops growing and starts using up its own reserves.
The protocol is built not on general advice but on the ΔT and trends of this particular reactor.
How conclusions are builtThe day took fifteen minutes. The episode stays in the reactor log along with the readings and actions.
A trio of signs: the moisture trend, a warm core, a recent feeding. On its own, each is not yet a crisis.
Loosened the layer and pushed feeding back. Fifteen minutes instead of a lost population.
Start with one reactor: the first 14 days are free, with no card and no sensors.