Cloudy, Fizzy, or Fuzzy: How to Read a Ferment

Walk into our fermentation room on day four of a lacto batch and the first thing you notice is that there's almost nothing to notice.

The airlocks are working. That's the sound: a slow, irregular tick as carbon dioxide pushes its way out and nothing gets back in. The brine has gone from clear to hazy. And the smell, which everyone expects to be the dominant experience of a room full of fermenting chilis, is barely there. A little green, a little bright, nothing you'd call sour yet.

That's a healthy ferment. Boring. Active, and boring.

Ten years in, most of what we know about reading a ferment comes down to telling apart two things: what looks alarming and is fine, and what looks minor and means the batch is finished. People confuse them constantly, and it costs in both directions. Good food gets thrown out. Bad batches get packaged.

Here's how we read ours, and how you can read yours.

The short version: A white film on the surface is almost always kahm yeast, and kahm is a flavor problem rather than a safety one. Anything fuzzy or colored is mold, and mold ends the batch. Cloudiness and bubbling mean the ferment is working.

What a healthy lacto ferment looks like

Our lacto sauces are sliced Red River Valley chilis in a salt brine, held under weight, minimum 21 days. No mash. The lactic acid bacteria we need are already living on the peppers when they come out of the field, and we add a known starter culture on top of them as insurance, so a batch never stalls waiting for the wild population to take hold.

In the first few days those bacteria get to work on the sugars in the fruit and start producing lactic acid. Three things follow, and all three look like trouble to someone who hasn't seen them before.

The brine goes cloudy. That haze is bacterial population growth plus fine particulate coming off the peppers, and a clear brine on day six worries us more than a cloudy one. It bubbles, because carbon dioxide is a byproduct of fermentation, and active airlocks tell us the culture established. Silence in the first week is the thing to chase down. Sediment collects at the bottom as spent yeast and bacteria drop out of suspension. Normal, expected, harmless.

The aroma keeps its own schedule. Early on it stays quiet. By the back half of the ferment it rounds out and turns faintly sour, and that shift is one of the ways we know a batch is getting close. No timer makes that call. The crock does.

We check pH on the finished sauce rather than tracking a curve through the ferment. That number tells us the batch did its job. Lactic acid bacteria acidify the brine as they work, and once a ferment drops below roughly pH 4.6, the environment stops supporting the pathogens that matter. That's the published safety threshold, and it's why properly made fermented vegetables have such a durable record.

Our sauces finish below pH 3.5. We don't strain to hit that. It's what a healthy 21-day ferment does on its own when the conditions are right, and it sits well under the line.

Apple cider vinegar goes in after the ferment. It's there for consistency batch to batch. The fermentation already did the preservation.

Takeaway: Cloudy brine, active bubbling, and settled sediment all mean the ferment is working. Quiet and clear in week one is the combination to investigate. A finished lacto sauce should land well under pH 4.6. Ours finish below 3.5.

Kahm yeast is a flavor problem, not a safety problem

This is what new fermenters panic about most, by a wide margin.

Kahm is a thin white or cream film that forms flat across the surface of the brine. It's not one organism but a group of wild aerobic yeasts, commonly Pichia and Candida species, that show up where brine meets air. They produce no toxins. They don't harm the lactic acid bacteria doing the fermenting. In the published literature and in our crocks alike, they're a nuisance.

How to know it's kahm:

• Flat and matte, often wrinkled like a skin, never fluffy
• White to cream, one color, with no green, blue, black, or pink
• Spread across the surface, sometimes evenly, sometimes in patches
• Smell stays clean and sour, never musty

It shows up when oxygen reaches the surface, which usually traces back to warmth, a loose seal, or something floating above the brine line.

We get it occasionally. Not often, but it happens. When it does, we skim the film and take a percentage of the brine with it before the batch moves toward packaging. That's more aggressive than the skim-and-continue advice written for home jars, and deliberately so. Kahm carries flavor, and flavor is the whole product. We'd rather lose a little brine than ship a batch that tastes like a compromise.

At home: skim with a clean spoon, wipe the rim, get everything back under the brine, and use the batch sooner rather than later. Once kahm establishes, it's hard to fully stop.

Takeaway: Flat, white, and wrinkly is kahm. Skim it, resubmerge, keep going. The batch is safe. It just loses a little nuance.

Mold means the batch is over

Different organism, different answer, no negotiation.

Mold is fuzzy. It has height and texture, cottony or hairy rather than filmy, and it arrives in color: green, blue, black, gray, pink. The smell turns musty.

What ends the batch is what you can't see. Molds put down hyphae, microscopic root structures that grow down into the food well past the visible colony, so scraping the top off doesn't remove the organism. Some molds also produce mycotoxins that spread through the batch and survive heat, which means cooking won't fix it. There's no sensible version of salvaging it.

See fuzz, discard the batch, then clean and sanitize the vessel and the weights before starting again.

We lost batches to mold in the early years. Haven't in a long time, and the fix wasn't sophisticated: keep everything under the brine. Every mold problem we ever had traced back to material breaking the surface. Oxygen at the top of a jar is what mold needs, and submersion takes it away.

Our weights are heavy food-safe glazed ceramic plates sized to the crock. Sanitizable, and heavy enough that they don't need help. At home, a purpose-made fermentation weight, a smaller jar nested inside, or a brine-filled bag all do the same job. The tool doesn't matter. Nothing floating is what matters.

Takeaway: Fuzzy, colored, or musty means discard. Mold grows roots into the batch you can't see, and its toxins survive cooking.

The false alarms

Two more things set off new fermenters, and both are fine.

Garlic turns blue. If you ferment garlic alongside your peppers, expect a startling blue-green, particularly early. It's a pigment reaction between sulfur compounds in the garlic and amino acids in the mix, it's cosmetic, and it usually fades.

Peppers change color. Bright reds dull, greens shift olive. That's pigment breaking down in acid.

The tell that separates every false alarm from a real problem is texture and smell together. Slimy peppers plus an off, putrid, or musty aroma is a real problem. Odd colors with a clean sour smell and firm texture is a ferment doing what ferments do.

What cost us batches

The failures that taught us something were never the ones people worry about.

Floats, early on, before we standardized weights. Every one of those was preventable.

Brettanomyces. Brett is a wild yeast, and in the wrong vessel it's a contamination event rather than a character note. It got into batches and produced flavors we wouldn't sell. At home the stakes are one jar. At production scale it's a vessel and process problem.

And water pH. This is the expensive one.

For a stretch we couldn't get consistent lacto establishment. Batches that should have taken off would sit there, sluggish or stalled, and we couldn't find the variable. It was the water. Our source water ran too alkaline, and starting a ferment from too high a pH slows the acidification the whole process depends on. Lactic acid bacteria need to outrun everything else in the vessel. We were sending them out without the head start.

We should have tested our water from the first batch. Not testing it cost us tens of thousands of dollars in product.

Everyone tests salt ratios. Almost nobody tests the water those ratios go into.

Takeaway: Test your water. Salt percentage means nothing if the water it dissolves into starts too alkaline for the culture to establish.

The variables you control

Reading a ferment is a diagnostic skill. Not needing to read it is a process skill. Four things do most of the work.

Submersion, first and most. Nothing above the brine, ever. That one prevents mold outright.

Salt, enough to favor lactic acid bacteria over everything else competing for the same sugars. We run a 5% brine on the lacto line.

Temperature. We ferment between 72 and 74°F in a temperature-controlled room. Warmer moves faster, and speed is a trap: high temperatures are hard to hold steady, they tilt the balance toward yeast activity, and they cost you complexity and clean flavor. We'd rather wait.

Water. Know what you're starting from, and learn it cheaper than we did.

Two ferments, two sets of signals

Everything above describes our lacto line. Our #KojiHachi line reads completely differently, because it's a different kind of fermentation.

Koji, Aspergillus oryzae, is the culture behind miso, soy sauce, and sake. Working on a single pepper over a minimum of eight months, with reserve batches running two to four years, its enzymes break proteins down into free glutamates. That's the source of the deep savory weight in those bottles. Enzyme activity and time, rather than lactic acid and bubbling, so we watch different signals on a timeline measured in seasons.

Same room, opposite ends of the fermentation map. The longer version is in why we wait at least eight months for #KojiHachi and what separates the two methods.

At a glance

Most ferments that look like they're failing are fine. A few that look fine aren't.

Ten years of crocks taught us that patience is a set of controlled conditions held steady long enough for something slow to happen properly. You can't cheat time. We never tried.

If you'd rather taste the result than run the ferment yourself, our lacto-fermented sauces start with sliced Red River Valley chilis, salt, water, and 21 days minimum. Both Scovie Award winners, Habanero #HUSTLE and Habanero #CrazyAsAPeachOrchardBoar, came out of exactly the process described above.

And if you're running your own crock this summer: test your water. Please.