A reactor always speaks, you just need to understand it
Overheating, odor, substrate structure, worm behavior, moisture, condensation. On its own, each sign tells you almost nothing. Together they describe the state of the whole system.
Core temperature
The main indicator of fermentation. What matters is not the number itself but how fast it rises and the difference between the core and the surroundings: that is what shows whether the system is coping with its load.
Moisture and condensation
Water decides whether the substrate breathes or suffocates. Droplets on the walls and a wet top layer tell you about the balance of heat and air before a person would notice it.
Odor
Odor is the microbiome’s fastest report. A sour smell means the aerobic process has given way to an anaerobic one and the pH is already dropping. A smell of ammonia points to too much nitrogen in the feed.
Population behavior
The population reacts to conditions faster than any instrument. When the worms move down, climb the walls or stop feeding, they are responding to a change in their habitat.
A sign is not yet a diagnosis
The same sign means different things under different conditions. It only gains meaning in context.
How context is read27 °C is not yet a diagnosis
The working range is 15–25 °C; from 30 °C you are in the zone where there are 6–12 hours left to act. 27 lies in between, and on its own the number means nothing. Direction, moisture, feed and the cycle phase give it meaning.
27 °C on the way down
If the temperature had been falling for three days and moisture held at around 74%, this would be the usual tail after feeding. Vermix would suggest nothing: a day without alerts is what normal looks like.
27 °C on the way up
Here it is rising, moisture is already above 80%, and feed went in 36 hours ago. At this pace the critical 30 °C is a matter of a day, so you need to act now, not after the fact.
How an observation becomes a protocol
Every conclusion goes through the same mechanism: an observation passes through diagnosis, risk assessment and the state model, is checked against the scientific base, and only then becomes an action protocol. Every link is visible, so you can check a conclusion instead of taking it on faith.
A living system leaves a digital trail
Temperature, moisture, odor, substrate structure, population behavior, feeding history: separately they are scattered observations. Together they become a digital description of the reactor’s state.
The models behind the conclusions
Not “the neural network decided so”: behind every alert is a pattern written into the code as a number. Thresholds are stored as constants, and each one is guarded by an automated test, so text that drifts from the science simply won’t ship.
C:N balance
The carbon-to-nitrogen ratio of the diet determines whether the core heats up. The optimum is 25–35:1. Below 20 ammonia appears, above 40 the process stalls; the calculator uses this scale to work out how to correct an imbalance.
Thermal delta ΔT
The temperature difference between the mass and the air shows whether the system is coping with its load. The threshold is 5 °C: up to it, all is normal; above it, attention is needed.
pH and oxygen balance
pH 6.5–7.5 and access to oxygen keep the microbiome aerobic. A shift in either one is an early warning of acidification and odor.
Checking the forecast against reality
After feeding, the system forecasts heating and checks it against your next reading. If reality has parted ways with the forecast, Vermix says so plainly, “hotter than expected”, instead of fitting an explanation after the fact.
Where the models come from
More than ten years of our own experiments. The protocols are written from the overheating and acidification we went through ourselves on our vermifarm.
How Vermix beganWe don’t replace biology with a machine, we make it understandable
The model calculates patterns; the decision stays with the person. Our job is to show the state of the reactor while the process can still be changed.