Sourdough Starter Without a Mixer: Method, Markers, Troubleshooting

A starter is ready when it doubles predictably in four to eight hours after a feed and its pH has dropped below 4.5. That usually takes ten to fifteen days, not the five most tutorials promise. Somewhere in between, nearly everyone hits a false start: a dramatic surge of gas around day two or three, followed by several days of nothing at all. That surge is not your starter waking up, and the silence that follows is not failure. It is the single most common reason people throw away a perfectly healthy culture.

The rest of this page gives the real timeline, the two measurable markers that replace guessing by eye, what fermentation genuinely changes nutritionally, and the one safety rule that matters.

What a starter actually is

A sourdough starter is a culture of lactic acid bacteria and wild yeasts kept alive in a mixture of flour and water. The bacteria produce lactic and acetic acid, which acidify the mixture and give the bread its flavour; the yeasts produce the carbon dioxide that raises the dough. The two populations coexist and sustain each other, with the acidity the bacteria generate keeping most competitors out.

Those micro-organisms come mainly from the flour itself, not from the kitchen air as is often claimed. That is the practical reason wholemeal or rye flour starts faster than white, and the guide to alternative flours covers what each one brings: the bran layer, removed during milling, carries most of the microbial load and the minerals that feed the culture.

Starters kept at room temperature with frequent feeds, which is what home practice amounts to, are characterised by fermentation times not exceeding 24 hours and favour micro-organisms that grow quickly on flour-based media, chief among them Fructilactobacillus sanfranciscensis. Liquid starters held for long periods at high temperature select for different, acid-tolerant lactic acid bacteria instead. In other words, the way you maintain your starter chooses which population lives in it, and therefore what it will taste like.

Equipment

A straight-sided glass jar of about half a litre, a lid simply rested on top or a cloth, a scale reading in grams, and a spatula. That is the whole list.

A straight jar rather than a shouldered one lets you see the volume actually reached. An elastic band placed at the level of the starter after each feed turns that jar into a measuring instrument, and it is the single most useful trick on this page.

The advice to avoid metal, repeated everywhere, deserves some qualification. It dates from a time when utensils were iron or copper, reactive metals that an acidic medium attacks. A stainless steel spoon, a stainless bowl or a knife blade cause no problem at all, including prolonged contact. What genuinely should be avoided is bare aluminium and copper.

The real timeline

Days 1 and 2

Mix 50 g of wholemeal rye or wholemeal wheat flour with 50 g of water at room temperature. The consistency should resemble a thick pancake batter. Cover without sealing and leave at 20 to 25 degrees.

On day two, simply stir, adding nothing. Usually nothing visible happens.

Day 3: the false start

This is often where dramatic activity appears: the mixture swells, bubbles, sometimes overflows. The smell can be frankly unpleasant, reminiscent of cheese or fermented vegetables.

That surge is not the work of the lactic acid bacteria you are after. It comes from other bacteria, present from the start, which thrive while the medium is still close to neutral. They will disappear on their own as acidity rises.

From this day on, begin daily feeds: discard all but 50 g, then add 50 g of flour and 50 g of water.

Days 4 to 7: the lull

Activity drops off, often completely. The mixture looks dead. This is the point at which most people give up.

Nothing visible happens because something invisible is happening: the medium is acidifying, the false-start bacteria are dying off, and the slower lactic acid bacteria are taking their place. In starters followed in the laboratory, pH falls below 4.5 by around day three, and that acidification is precisely what does the sorting.

Keep feeding daily, identically. Change nothing, and add no sugar, fruit or commercial yeast. Those additions feed the competitors, not the culture you want.

Days 8 to 15: the return

Activity comes back, but different: finer, regular bubbles, a predictable rise and fall, and a smell that shifts towards yoghurt, apple or light beer. The starter becomes consistent, which is the real marker of maturity. After around ten days of successive feeds, a laboratory-monitored starter carries on the order of 9 log colony-forming units per gram of lactic acid bacteria, roughly a billion per gram.

The two markers that replace judging by eye

The timed doubling. Put an elastic band at the level of the starter right after a feed and note the time. A mature starter doubles in four to eight hours at 22 degrees, then falls back. What matters is not the height reached but the regularity: three consecutive cycles with the same timing are worth more than one spectacular rise.

The float test. Take a spoonful of starter at the peak of its activity and drop it in a glass of water. If it floats, it holds enough gas to raise a dough. This test has a limitation worth knowing: a very wet starter can sink while being perfectly active, and a starter loaded with acetic gas can float without having the strength required. It confirms, it does not decide.

A third marker exists if you want precision: laboratory pH strips, a few francs at a pharmacy or drugstore. A mature starter sits around 3.5 to 4.5. It is the only genuinely objective criterion, and it settles most beginners’ doubts.

What fermentation genuinely changes

This is the area where the most claims circulate, and it is worth separating what has been measured from what has not.

Phytates and iron: solid mechanism, uncertain clinical effect

Wholegrain cereals contain phytic acid, which forms insoluble complexes with iron, zinc and magnesium and reduces their absorption. Enzymes called phytases break those complexes down. They are present in the flour itself and can also be released by yeasts and bacteria during fermentation. Their activity optimum sits between pH 4.3 and 5.5, which is exactly the range sourdough acidification brings the dough into. That is the central mechanism, and it is well established.

The absorption figures are striking. Two trials in healthy women, using radioactive iron isotopes and measuring incorporation into red blood cells, compared identical wholegrain rye bread baked with or without phytate degradation. From an 80 g portion, non-haem iron absorption went from 7.0 % to 19.1 %, that is 2.8 times higher. From 120 g, from 4.6 % to 15.0 %, 3.5 times higher, with the effect more pronounced the higher the starting phytate content.

But a systematic review of eight human intervention studies, conducted following EFSA’s guidance for health claims, concludes that the results remain inconclusive. Postprandial studies do show the rise in bioavailability. Long-term trials, however, did not improve iron status, and in one case worsened it: in a group eating low-phytate sourdough rye bread, ferritin fell from 32 to 27 micrograms per litre and total body iron from 6.9 to 5.4 mg per kilo over twelve weeks. The authors note that no human study addresses the question with appropriate controls and methodological quality.

The honest reading is therefore this: sourdough improves the iron availability of a meal, that much is measurable, and nobody has shown it translates into better iron status over time. The page on plant iron absorption covers the levers whose effect is better established.

FODMAPs: the most likely explanation for the digestive comfort

Many people report tolerating sourdough bread better. The best-supported explanation is not gluten but fructans, fermentable carbohydrates of the FODMAP family, of which wheat is a major source. Sourdough cultures and yeasts break them down, with fructan reductions above 90 % reported in vitro using selected strains, and FODMAP reductions of more than 65 % compared with wholemeal flour.

This dovetails with a striking observation about non-coeliac gluten sensitivity: a review of ten randomised trials reports that only 16 % of people who describe themselves as gluten-sensitive show gluten-specific symptoms under a double-blind, placebo-controlled challenge. Several lines of work point to fructans rather than gluten. If you digest sourdough bread better, it is more likely that you react to fructans than to gluten.

Two important caveats. Those reductions come from in vitro trials with chosen strains, not from a kitchen starter of unknown composition. And duration matters: long fermentations degrade more than short ones. The subject is developed on the page about the low-FODMAP diet.

Gluten: what sourdough does not do

Sourdough partially degrades gluten proteins, and that is probably one reason a sourdough loaf feels lighter. But the reference review on fermentation in breadmaking is explicit: characterising and selecting micro-organisms able to do this reliably still requires standardisation, and the clinical implications are not firmly established.

The practical consequence allows no ambiguity: sourdough bread is not gluten-free and is not suitable for anyone with coeliac disease. No home fermentation time makes a wheat loaf safe in that case.

The one safety rule that matters

A healthy starter smells acidic, fruity or beery. It may develop a greyish liquid on top, often called hooch, which simply signals hunger: remove it or stir it back in, then feed more often.

Visible mould, on the other hand, coloured or fuzzy, green, black, pink or cottony white, means throwing away the entire starter, jar included or jar washed in very hot water with detergent. Do not scrape the surface, do not rescue the bottom, and do not use the starter on the assumption that baking will fix it.

The reason is documented. Many moulds produce mycotoxins invisible to the naked eye. These toxins are heat-stable: heat kills the moulds but the toxins remain, because they are chemically very stable. Cooking, baking and frying therefore generally do not destroy them. The German Federal Institute for Risk Assessment recommends discarding mouldy foods entirely, specifying that cutting off or scraping away visibly affected areas is not sufficient, since mycotoxins may already be distributed throughout the product. That recommendation explicitly targets soft or water-rich foods, of which bread is one, and a starter is more so.

Losing a two-week-old starter is annoying. Rebuilding it costs a hundred grams of flour.

Keeping a starter

The feeding ratio

This is the variable most recipes leave unmentioned even though it governs everything. A feed is expressed as starter to flour to water.

A 1:1:1 ratio, meaning 50 g of starter with 50 g of flour and 50 g of water, gives a fast peak, four to six hours at room temperature. It suits using the starter the same day.

A 1:5:5 ratio, 20 g of starter with 100 g of flour and 100 g of water, stretches the cycle over eight to twelve hours. That is the right choice for feeding in the evening and baking the next morning, and it yields a less sour starter.

Diluting further delays the peak; concentrating brings it forward. All schedule management lives in this setting, not in feeding more or less often.

Temperature

It does more than speed things up or slow them down. It selects different populations, and therefore the flavour. Around 20 to 24 degrees, the balance favours lactic acid and a mild, yoghurty profile. Warmer, acetic acid gains ground and the starter turns sharper, more vinegary.

Above roughly 30 degrees, the microbial composition can drift with nothing to signal it. One trial kept starters at 23, 30 and 37 degrees: by day four at 37 degrees, the usual monitoring parameters detected no change, with pH well below 4.5 and total lactic acid bacteria counts virtually unchanged, even though the population had shifted. If your starter spent a summer at 35 degrees and you dislike the taste, it is probably not your imagination.

The fridge

For occasional use, a starter keeps in the fridge at 4 to 6 degrees, fed once a week. Feed it, let it get going for an hour or two at room temperature, then move it to the cold.

To wake it up, take it out and run two or three feeds at room temperature over twenty-four to thirty-six hours before relying on it. A starter taken straight from the fridge and used as is raises dough poorly; this is a very common cause of flat loaves.

A starter forgotten for several weeks in the fridge, covered in grey liquid and smelling strongly of alcohol, is almost always recoverable. Discard everything but a spoonful from the bottom, and feed at 1:5:5 for three days.

Backing it up

Spread a thin layer of mature starter on baking paper, let it dry at room temperature for two to three days until you have brittle flakes, then keep them in an airtight jar away from light. They reactivate in three to four days through successive feeds. It is insurance that costs half an hour and saves starting from scratch.

A sourdough loaf without a mixer

For a loaf of about 850 g: 500 g of T65 or T80 wheat flour, 350 g of water, 100 g of mature starter at the peak of its activity, 10 g of salt.

Autolyse. Mix the flour with 330 g of water, no starter and no salt, until no dry flour remains. Rest for 45 minutes. The flour’s enzymes begin working and the gluten organises itself without kneading, which cuts the mechanical work needed by the same amount.

Incorporation. Add the starter, mix by hand pinching through the dough, then add the salt and the remaining 20 g of water. Salt goes in after the autolyse because it tightens the gluten network and slows hydration.

Bulk fermentation with folds. Ferment for 4 to 5 hours at around 22 degrees. During the first two hours, do a fold every thirty minutes: with a wet hand, grab an edge of the dough, stretch it upwards and fold it to the centre, turn the container a quarter turn, repeat four times. These folds replace mechanical kneading entirely.

The duration depends on temperature and starter, not on the clock. The dough is ready when it has gained 50 to 75 % in volume, looks domed and shows bubbles under the surface. Dough that has doubled has already gone too far and will tear during shaping.

Shaping. Turn out onto a lightly floured surface, fold the edges into the centre to form a ball, flip it over and drag it across the surface to tighten the skin. Place it seam-side up in a banneton or a bowl lined with a well-floured cloth.

Final proof. One to two hours at room temperature, or eight to fourteen hours in the fridge. The cold version is easier to manage, develops more flavour and lets you score a firm dough, which is considerably easier.

Baking. Preheat a cast iron pot with its lid in the oven at 240 degrees for 45 minutes. Tip the dough into the scorching pot, score decisively with a blade in one clean motion, cover, and bake for 20 minutes. Remove the lid, drop to 220 degrees and continue for 20 to 25 minutes.

The lid is not a detail: it traps the steam the dough releases, which keeps the crust supple long enough for the loaf to expand, and removing it then allows browning. Without a cast iron pot, a gratin dish inverted over the loaf does the same job. The same steam and floor-heat principles apply to wood-fired cooking and to flatbreads.

Cooling. On a rack, at least two hours. Slicing a hot loaf gives a gummy crumb, because the starch has not finished retrograding and steam is still trapped inside.

Troubleshooting

The dough does not rise. In the vast majority of cases the starter was not at its peak when it went in. Check with a float test before starting, not after.

The loaf is flat and dense. Bulk fermentation too short, or a starter used straight from the fridge without two or three waking feeds.

The loaf spreads when it goes in the oven. Bulk fermentation too long. The dough went past its point and the gluten network gave way. Cut an hour next time.

The crumb is too sour. Feed at a more dilute ratio, 1:5:5 instead of 1:1:1, use the starter earlier in its cycle, and lower the proofing temperature.

The holes are tiny and uniform. Hydration too low or not enough folds. Raise the water by 20 g and add a round of folds.

The crust is pale and soft. Oven not hot enough, or pot insufficiently preheated. Forty-five minutes of preheating with the pot inside is not negotiable.

Common mistakes

Giving up during the lull. Between the false start and the real return, there are often four to six days with nothing visible. That is the normal phase, not a failure.

Adding sugar, honey, fruit or yeast to “help”. These additions favour yeasts and bacteria other than the ones you want, and delay the stable culture settling in.

Measuring by volume. A spoonful of flour and a spoonful of water do not weigh the same, and the ratio drifts with every feed. A scale settles this permanently.

Sealing the jar airtight. The gas produced needs somewhere to go. A lid resting on top is enough.

Judging a starter by its height. A very wet starter rises less but can be perfectly active. Consistency of the cycle is the criterion, not amplitude.

By profile

Beginner. Start with wholemeal rye, the fastest and most forgiving, and move to wheat later if you want. Keep a notebook with feed times and how long doubling takes; within two weeks you will know your starter better than any tutorial does.

Anyone who bakes rarely. Keep 30 g of starter in the fridge, fed once a week. Allow two days of waking before a bake. Batch cooking pairs well with this rhythm, since dough in the cold can wait for you.

Anyone who digests bread poorly. Fructans are the most likely explanation, and a long fermentation reduces them further. Extend the cold proof rather than speeding things up with commercial yeast.

Anyone with coeliac disease. Wheat sourdough bread is out, without exception. Fermentation does not make gluten safe.

A very warm kitchen in summer. Drop to 1:10:10 and use cool water, or keep the starter in the fridge through the hot months. Above 30 degrees the population drifts.

Limits of this page

The fructan reductions cited come from in vitro trials with selected strains. A home starter hosts an unknown, uncharacterised population: the order of magnitude does not transfer directly to your jar.

The iron absorption figures come from two isotope trials with small numbers, eight and seventeen participants, in healthy women and on wholegrain rye bread. They measure absorption from a meal, not iron status, and the available systematic review does not reach a conclusion on that second point.

The times and temperatures given assume a kitchen between 20 and 24 degrees, tap water and retail flour. Heavily chlorinated water can slow a start without preventing it; letting it stand in the open for an hour is enough.

Finally, no marker on this page replaces watching your own starter across several cycles. The numbers give a framework, not a verdict.

Frequently asked questions

How long until a genuinely usable starter? Ten to fifteen days in a kitchen at 20 to 24 degrees. Tutorials promising five days are describing the false start, not a stable culture.

Do I really have to discard starter at every feed? Yes, otherwise the volume doubles each time and the amount of food per micro-organism collapses. The discard is usable as is in pancakes, waffles or flatbreads, with no need for leavening power. Other uses of the same habit appear on the page about easy homemade ferments.

Which flour should I choose? Wholemeal rye to start, it is the most reliable. For maintenance, T80 wheat gives a balanced starter. White T65 works but starts more slowly, since the bran has been removed.

My starter smells of acetone or nail polish. Is it ruined? No. That is a hunger smell: the micro-organisms have used up their food and are producing ethanol. Feed more often or at a more dilute ratio and the smell goes within two or three cycles.

A dark liquid has formed on top. What should I do? Pour it off or stir it in, then feed. Same hunger signal, entirely harmless.

Can I use tap water? Yes in almost all cases. If your water is heavily chlorinated, let it stand in the open for an hour.

Does sourdough bread contain less gluten? Fermentation partially degrades gluten proteins, but not reliably or sufficiently to change anything in coeliac disease. Wheat sourdough bread remains a gluten-containing bread.

Can a starter be frozen? It is possible but unreliable, since freezing kills some of the yeasts. Drying into flakes is a considerably more dependable backup method.

Sources