Bulk fermentation is the part of a sourdough bake where most outcomes are decided. Get the timing right and you get an open crumb, a clean ear, and dough that's a pleasure to shape. Get it wrong and nothing downstream can rescue it. The biggest variable in bulk timing isn't your flour or your starter. It's the temperature of the dough as it ferments. This guide is about that one lever: how warmth and cold move your bulk clock, and how to read the dough instead of guessing.
The 2.2× rule, in plain words
Yeast and lactic bacteria work faster when it's warm and slower when it's cold. For sourdough the rate increases about 2.2× for every 8°C (about 14°F) rise in dough temperature. It's the factor the calculator uses to schedule your bake. (Scientists usually express temperature sensitivity per 10°C, as "Q10"; 2.2× per 8°C works out to a Q10 of about 2.7.)
It's worth being precise about what 2.2 means, because it's easy to mangle. A plain doubling per 8°C would be a factor of 2.0. The model's 2.2 is about 10% more than a simple doubling. Warm the dough 8°C and you don't just halve the time; you cut it to roughly 45% of what it was. That extra 10% is why summer dough sneaks past you faster than people expect.
The model that follows assumes a healthy levain at peak, white bread flour, and 20% levain. Change those and the curve shifts, but its shape, 2.2× per 8°C, stays the same.
A bulk-time table you can read at a glance
Here is the 2.2× model spelled out across the temperatures a home kitchen actually sees. Each row is the model's approximate time to the end of bulk for a 20% levain white-flour dough, anchored to the reference of 5 hours at 24°C (75°F). The math is time = 5 × 2.2^((24 − T)/8).
| Dough temp | Approx. bulk (model) | vs. 24°C reference |
|---|---|---|
| 16°C (61°F) | ~11 h 00 m | a long overnight |
| 18°C (64°F) | ~9 h 00 m | most of a working day |
| 20°C (68°F) | ~7 h 25 m | classic cool-room bulk |
| 22°C (72°F) | ~6 h 05 m | typical spring kitchen |
| 24°C (75°F) | ~5 h 00 m | the reference point |
| 26°C (79°F) | ~4 h 05 m | warm kitchen / proofer |
| 28°C (82°F) | ~3 h 20 m | fast same-day bake |
| 30°C (86°F) | ~2 h 45 m | very fast, easy to overshoot |
Read down that column and the practical consequence is obvious: the dough you mixed at 9 a.m. on a 22°C (72°F) Saturday is ready around 3 p.m., while the same dough mixed at 9 a.m. in a 16°C (61°F) winter kitchen won't be ready until evening. Treating those two days as the same bake is how loaves come out underproofed in winter and overproofed in summer. If your kitchen swings hard with the seasons, the guides on winter sourdough and summer sourdough cover how to pull the temperature back toward 24°C (75°F).
Measure dough temperature, not air temperature
The table is keyed to dough temperature on purpose: dough is often warmer than the room it sits in. Friction from mixing adds heat (1–4°C by hand, more with a stand mixer), and the dough then holds that heat because flour and water are slow to cool. A 22°C (72°F) kitchen can produce dough at 23–25°C (73–77°F) straight out of the mix, depending on your water.
That gap is why people who plan off the thermostat get caught out. They look up "22°C" expecting six hours, but the dough starts out on the 24–25°C line, so the model predicts bulk finishing roughly 20 to 40 minutes sooner as the dough cools toward the room. A cheap instant-read probe in the center of the dough right after mixing settles it; enter the reading under Dough temperature after mixing in the calculator and the schedule accounts for the drift. If you want to hit a target dough temperature rather than just measure it after the fact, the DDT water-temperature formula tells you how warm or cold to make your water before you mix.
Inoculation: the second dial that bends the clock
Temperature sets the curve; inoculation, the levain weight as a percentage of the recipe's flour, slides you along it. In the model, halving the levain makes bulk about a third longer, and doubling it takes about a quarter off. This is based on published test bakes: in The Sourdough Journey's controlled series, going from 20% to 10% levain added about a third to bulk. That series covers 5–20% levain at 19–23°C; outside it the model extrapolates. The full estimate is bulk time = 5 × 2.2^((24 − T)/8) × (20 / levain%)^0.4. For how the levain is counted and what changing it does, see how much starter in sourdough.
- 10–15% levain: slower bulk, a couple of hours longer in a cool room. Useful when you want to ferment while you sleep; overnight bulk fermentation gives the levain amount for your night.
- 20% levain: the reference used in the table above. Predictable, and the default starting point.
- 25–30% levain: faster bulk, though only by ½–1 h at 20–24°C (68–75°F), and under three hours only at 30% in a 28°C (82°F) kitchen. A small push for a same-day loaf; warmth does more.
A worked example: at 20°C (68°F) the table says ~7 h 25 m at 20% levain. Drop to 10% and the model says about 9¾ h; push to 30% and it is about 6¼ h. The levain moves the day by a few hours; the temperature moves it more. If you're not sure how your feeding ratio affects the starter's own peak timing, the starter-feeding-ratio guide handles the feed side; this article assumes your levain is already ripe at mix time.
Visual cues beat the clock
The table is a planning tool, not a stopwatch. Dough finishes bulk when it decides, and your job is to read it. The reliable tells:
- About a 50–75% rise in volume from when the salt went in (toward 50% in a warm kitchen or with whole wheat). Mark the side of a straight-walled container with tape or a wax pencil so the rise is unambiguous; eyeballing a curved bowl is where many "I thought it was ready" errors come from.
- A domed top that holds its shape when you tilt the container gently.
- A jiggle: wobble the container and the surface moves like set custard, not like wet concrete.
- Bubbles at the surface and along the sides of a clear container.
When two or more of those four line up, you're at or near the end of bulk regardless of what the clock says. The numbers tell you roughly when to start watching; the dough tells you when to stop.
How to use the calculator with your kitchen
Open the schedule calculator, set levain, hydration, and salt to your recipe, then set the room-temperature input to your measured room temperature rather than the thermostat reading. It runs the same Q10 model behind the table above, then schedules everything around the bulk estimate (folds, pre-shape, retard, and bake) so the whole day lines up from a single temperature.
If your bulks consistently finish faster than predicted, your dough is probably running warmer than the room (underfloor heating, sun on the counter, an oven light left on). Nudge the room input up 1–2°C and the prediction snaps back into line. This is the one calibration worth doing once: find your kitchen's real offset between room and dough, and every future bake inherits it.
What the temperature model can't see
Two things sit outside the temperature curve, and it's worth naming them so you don't blame the math when they bite.
Flour. The model assumes white bread flour. Whole-grain flours carry more enzymes and more wild yeast on the bran, so they ferment faster. A common rule of thumb (not a model value) is about 15–20% shorter bulk at 30–50% whole wheat, and 20–30% shorter at 60% or more. That's a flour effect, not a temperature effect, and it's covered in the whole-wheat and rye guides. Hydration and autolyse choices also nudge dough behavior, but they belong to their own topics: see hydration explained and autolyse vs. fermentolyse rather than reading them into the temperature table.
Starter health. The curve assumes a levain peaking on schedule. A sluggish or over-fed starter adds latency at the front that no dough-temperature math can erase. If yours isn't reliably doubling, fix that first; the revive-a-forgotten-starter guide is the place to start.
Common questions
Is air temperature or dough temperature what I should plug into the table?
Dough temperature, measured in the center right after mixing. The room number is only a proxy, and usually a cool one: dough often runs 1–3°C warmer than the air because of mixing friction and slow heat loss. If all you have is the room reading, add 1–2°C before you look up the row.
How do I tell underproofed from overproofed at the end of bulk?
Underproofed dough is tight, barely domed, and springs back fast when poked. Overproofed dough has lost its dome, looks slack and bubbly at the edges, and tears or jiggles loosely. A 50–75% rise plus a held dome is the window you're aiming for. In a warm kitchen you can slide past it, from perfect to overproofed, in under an hour.
Does higher hydration change the bulk time?
Wetter dough feels like it ferments faster because it's slacker and shows gas more readily, but the temperature curve is what actually sets the time. Treat hydration as a handling and crumb decision, not a clock decision; the details live in hydration explained.
Can I just retard in the fridge instead of timing a room-temperature bulk?
You can retard the shaped loaf, which moves the dough to about 4°C (39°F) where it barely moves. The 2.2× rule doesn't cover the fridge, so the retard is planned by hours and the dough's look, not by this table. Whether to proof warm or retard cold is its own trade-off, laid out in cold retard vs. same-day.
My loaf is dense even though I followed the times. What now?
Times are a starting estimate, not a guarantee. Density usually traces back to ending bulk too early, a weak starter, or under-development rather than the temperature math. The fix-a-dense-sourdough guide walks the diagnosis; come back to this curve once your starter and rise are dialed in.
Why does my bulk finish faster in summer even at the "same" kitchen temperature?
Because the dough is warmer than your thermostat suggests, and the temperature curve is steep. A kitchen that reads 24°C (75°F) in summer can hold dough at 26–27°C (79–81°F) once you account for warm flour, warm water, and ambient heat. At that point you've moved a row or two down the table without realizing it.
Calibrate once, then trust the curve
Do this a single time: mix a plain 20% levain dough, probe the dough temperature, bulk it in a marked straight-walled container, and write down the clock time when it actually reaches the done signs. Compare that to the row in the table. The difference between the two is your kitchen's offset, and it rarely changes much.
From then on you read one dough-temperature number, slide to the matching row, apply your levain percentage, and start your visual checks a little before the estimate lands.
Making pizza as well? The 2.2× per 8°C rule in this guide is for sourdough at room temperature. Baker's yeast in the fridge slows far more than that rule would predict, so the pizza dough calculator uses its own temperature model for yeasted dough, explained in the pizza dough temperature guide.