It is true that fermentation increases acid. But that says only half. What matters is which acid, and when. Good acids and off-flavour acids are born at different stages of the same fermentation.
There is a study that inoculated microbes one species at a time to separate which microbe makes which acid. The result was clear. Malic, lactic and acetic acid were detected only in the plots inoculated with bacteria, and were zero in the uninoculated control. Succinic acid was detected in the bacterial plots and in some of the yeast plots as well (Martins et al., Front. Microbiol. 2019). So the acid character of the cup is decided by who worked in the tank.
In the same study, the representative off-flavour acids, butyric and propionic acid (the onion-like smell), were not detected in any plot. That is also evidence the experiment did not reach over-fermentation. Put the other way, when these two appear, fermentation has already gone too far.
There is an experiment that followed sensory results at 0, 24, 48, 72 and 96 hours of fermentation. In natural fermentation the flavour became woody at 48 hours and chestnut-like at 72. In the yeast-inoculated series, caramel increased in the first 48 hours, fruit dominated at 72, and at 96 hours only herbal notes remained (Palumbo et al., Braz. J. Microbiol. 2024).
I think "only herbal at 96 hours" is the most important result in that study. Aroma did not keep increasing; the fruit disappeared and the coffee tipped toward green grass. Fermentation is not a one-way function; past the optimum it reverses.
In open washed fermentation, the 36-hour lot scored highest in the cup and both shorter and longer lots scored lower (Yuan et al., Fermentation 2024). The position of the summit differs by method, but that there is a summit is shared.
The quantitative side too. In an experiment following mucilage fermentation, while Brix fell from 15.8 to 8.45, acidity nearly doubled from 20.02 to 42.69 mg/g and pH fell from 4.47 to 4.05 (Plaza-Dorado et al., Fermentation 2024). Sugar and acid sit on a seesaw.
So "ferment longer and it gets sweeter" is backwards. Ferment longer and sugar goes down. What goes up is acid, and fermentation-derived aroma. If you want to keep sweetness, you have to stop partway.
If you use a sealed tank, one more caution. The final concentration of lactic acid was 5.06 g/L at the bottom, 4.70 g/L in the middle and 2.18 g/L at the top, more than a twofold difference within the same tank. Sugar remained even after 72 hours, and the study concluded that "extending fermentation time alone does not consume all the sugar" (da Silva Vale et al., Foods 2023).
So the top and bottom of the tank are undergoing different fermentations. Mix, or make the batches smaller. If you do neither, you can only sort with the variation in mind. That is why we do our anaerobic fermentation in bags.
For all our talk of managing by numbers, for over-fermentation the nose is faster. A cheesy smell means butyric acid; an onion smell means propionic acid. Remember these two and you can act before you measure pH.
Conversely, a good fermentation smells of yoghurt-like acidity and ripe fruit. When that smell is strengthening, you are in the middle of the log phase.
The total amount of acid increases, but the kind of acid changes partway. Within the right range, good acids such as lactic, acetic and malic increase; go too far and the off-side acids, butyric and propionic, appear.
Smell is the clue. Butyric acid smells of cheese or vomit; propionic acid of onion. In the cup they appear as sourness or a fermented smell.
It depends on the method. For open washed fermentation, 36 hours has been reported as the peak; for sealed anaerobic fermentation, fruit peaked around 72 hours and tipped to grassy at 96.
For judging by pH see Reading the fermentation pH curve; for temperature, What lowering fermentation temperature by 1 °C means; for the whole picture, The four stages of fermentation.
Follow this record. New measurements, stories and lot releases from LAOS — delivered to your inbox first.
New records and lot releases, delivered to your inbox first.
New lots, drying updates and field notes — every Friday, in one short letter.