Why won't idli batter ferment in winter? Temperature fixes

Idli batter has no yeast packet to blame. When it sits flat on a cold night, the bacteria living on your rice and urad dal are simply too cold to work fast enough.

Idli batter being poured into the moulds of an idli plate
Photo: Gpkp, CC BY-SA 4.0, via Wikimedia Commons

Idli batter stops fermenting in winter because its leavening comes from wild lactic acid bacteria on the rice and urad dal, and they slow down steeply as the batter cools: studies ferment idli at about 30°C, and batter held at 4°C did not ferment at all. The fixes are to keep the batter itself near 30°C (a little warmer is fine), seed it with about 10% of an older batter, add salt after fermenting rather than before, and give it more hours.

Key takeaways
  • About 30°C is the batter temperature the main idli studies use; at that heat volume doubles in about 15 hours.
  • 4°C batter did not ferment at all in the 2019 NCCS Pune study.
  • 10% old batter restored fermentation even in batter made from sterilised grains.
  • Salt late in winter: it slows the bacteria a little, which helps in summer and hurts in January.
  • 2:1 to 2.5:1 rice to urad dal holds scarce winter gas better than 4:1.

The question shows up every December on cooking forums: same rice, same dal, same grinder, and the batter that doubled in July sits flat in January.

On The Fresh Loaf, a baker who treats idli like sourdough now holds his batter at 37°C and adds salt late when it is cold. The answer is microbiology, and it gives precise fixes.

This post explains:

  • what the bacteria need, from the lab studies;
  • what your kitchen gives them in winter, city by city;
  • which folk tricks work, and which don't.
~30°Cthe ambient temperature used in the main idli microbiology studies
4°Cbatter held here did not ferment (Mandhania et al., 2019)
10%old batter that restored fermentation in batter from sterilised grains
16.1°CBengaluru's mean January minimum, 1991-2020 (IMD)

Why does idli batter need warmth to ferment?

Idli batter needs warmth because its rise is produced by living bacteria, mainly Weissella and Leuconostoc, whose growth and gas output fall sharply as temperature drops.

These are heterofermentative lactic acid bacteria: each glucose molecule they break down yields lactic acid plus carbon dioxide, and the gas, trapped by urad dal's proteins and gums, is what lifts the batter. No yeast packet is involved, and no bacteria are added; they come in on the grain surfaces.

The story

In 2019 a team of nine microbiologists at the National Centre for Microbial Resource, part of NCCS in Pune, led by Madhvi Mandhania, ran the experiment every winter cook runs by accident. They ground a 2:1 parboiled rice and black gram batter, held it at about 30°C and sampled it every three hours. The pH fell from 6.3 to 4.6 by 12 hours, and the volume had doubled by 15 hours (Mandhania et al., 2019).

Then they changed one thing at a time. Grains washed in isopropanol for 30 minutes gave a batter that did not rise in 15 hours; its pH crept only to 5.8 (same paper). Stirring in 10% of normal fermented batter brought it back. And a sample kept at 4°C did not ferment at all, which led the team to call an ambient 26 to 30°C a requirement for the microbes to work.

Mandhania et al.: Diversity and Succession of Microbiota during Fermentation of the Traditional Indian Food Idli, Applied and Environmental Microbiology (2019)

Winter failure is the same experiment run accidentally on your kitchen counter. The full microbial succession is in the science of idli batter fermentation.

Diversity and Succession of Microbiota during Fermentation of the Traditional Indian Food IdliMandhania et al. · Applied and Environmental Microbiology · 2019

Lab batter (parboiled rice and black gram, 2:1, no starter) fermented at about 30°C was sampled every 3 hours for 24 hours and analysed by amplicon sequencing, qPCR and culturing, alongside vendor batters from Bengaluru and Pune. Weissella rose from 9% to 71% of the bacterial community between hours 6 and 9 and from about 104.3 to 107.9 gene copies per gram by 15 hours. Batter at 4°C did not ferment; batter from isopropanol-sterilised grains did not rise, but adding 10% fermented batter restored fermentation to normal levels. Limits: one lab recipe and one temperature for the time course; it does not map rise time across temperatures.

Read the paper (free)
Idli batter temperature zones 4°C: no fermentation (Mandhania 2019) 10°C: fridge storage, microbes grow slowly (Sridevi 2010) 20-26°C: uncontrolled room, rise still happens but takes most of a day (Ghosh 2010) About 30°C: the lab standard; volume +100% by 15 h 37-40°C: fast home method, watch for over-souring Aim for the batter itself, not the room, to sit near 30°C
Temperature landmarks from the idli studies cited in this post and The Fresh Loaf thread. These are batter temperatures, not air temperatures.

How cold does an Indian kitchen get in winter?

Cold enough to matter almost everywhere outside the coast. Using India Meteorological Department normals for 1991 to 2020, Bengaluru's mean January minimum is 16.1°C and Delhi's (Safdarjung) is 7.5°C, while Chennai's stays at 21.2°C. A kitchen is warmer than the outdoor minimum, but batter usually ferments overnight, exactly when the house is coldest.

January mean daily minimum and maximum, 1991-2020
See the data as a table
Mean daily minimum (°C)Mean daily maximum (°C)
Chennai21.229.3
Bengaluru16.128.4
New Delhi7.520.1

Source: India Meteorological Department 1991-2020 normals, via Wikipedia climate tables · Chennai is Nungambakkam station; New Delhi is Safdarjung. Indoor temperatures sit above the outdoor minimum but follow it.

The chart explains the regional pattern people report:

  • Chennai: even a January night stays above 21°C, so batter still rises on the counter.
  • Bengaluru: a night that falls to 16°C outside can leave a kitchen in the high teens, and a 12-hour rise stretches well past breakfast.
  • Delhi and the north: a counter-top idli in January (7.5°C mean minimum) is close to a fridge experiment.

How long does batter take to rise at lower temperatures?

Much longer, and the curve is steep: at around 20 to 26°C batter still rises but needs most of a day, while at 30°C it doubles in about 15 hours. The best room-temperature evidence comes from a study that did not control temperature at all.

Preparation of idli batter, its properties and nutritional improvement during fermentationGhosh and Chattopadhyay · Journal of Food Science and Technology · 2011

The authors made batters from parboiled rice and decorticated black gram at 2:1, 3:1 and 4:1, with 2% salt, and fermented them in steel vessels for up to 22 hours with the room between 20 and 26°C. Volume rose 1.62 to 3.2 times, fastest up to 13 hours and slower afterwards; pH fell from about 5.9 to 4.2; acidity rose from 0.44% to 0.91%; and density dropped from 0.93 to about 0.59 g/cm³ for the 2:1 batter. More black gram gave more rise: at 22 hours the 2:1 batter was lightest. Limits: temperature was not controlled, so the effect of cold cannot be separated from recipe, and some units and soaking times are reported inconsistently.

Read the paper (free)

Below 20°C: the fridge evidence

Lab work on refrigerated batter shows the slowdown below that range. Sridevi, Halami and Vijayendra at CSIR-CFTRI stored batter at 30°C and at 10°C.

StorageLactic bacteria over timeWhat it means
30°CPast 1010 per gram by day four, then spoiledFast growth, then over-souring
10°CUp about half a log unit over three daysSlow, but not stopped

Cold slows the microbes; it does not kill them, which is why batter from the fridge keeps souring slowly.

Lactic bacteria in stored idli batter at 30°C vs 10°C
See the data as a table
Stored at 30°C (log cfu/g)Stored at 10°C (log cfu/g)
Day 19.28.6
Day 29.38.7
Day 39.58.8
Day 410.1

Source: Sridevi, Halami and Vijayendra, J Food Sci Technol 2010, Table 5 (Pediococcus plus Candida starter) · Day 4 was not measured at 10°C. A difference of 1 log unit is a tenfold difference in bacteria.

Selection of starter cultures for idli batter fermentation and their effect on quality of idlisSridevi, Halami, Vijayendra · Journal of Food Science and Technology · 2010

CSIR-CFTRI researchers fermented 4:1 rice and black gram batter at 30°C with three starter pairs (a lactic bacterium plus the yeast Candida versatilis) and a natural control, then stored samples at 30°C and 10°C for up to a week. Starter batters with Enterococcus faecium rose to an index of 180 against 140 for natural fermentation by day five, pH fell from 6.73 fresh to 4.3-4.5, and 50 tasters scored the starter idlis 3.9-4.4 out of 5 against 3.6 for the control. Limits: one recipe, lab starters not available to home cooks, and some figures differ between abstract and tables.

Read the paper (free)
A steel vessel of fermented idli batter with a bubbly surface
Well-fermented batter: domed, bubbly and pale, with a clean sour smell. Flat, glossy batter with no bubbles after 12 hours is too cold.Photo: SijiR, CC BY-SA 3.0, via Wikimedia Commons

Which winter fixes actually work, and why?

Every fix that works does one of two things:

  1. it raises the batter's temperature, or
  2. it raises the number of active bacteria at the start.
Tricks that do neither are habit, not chemistry.
FixWhat it changesEvidenceVerdict
Keep batter in an oven with only the lamp onBatter temperatureRaises temperature toward the 30°C range the studies useWorks; check with a thermometer
Grind with lukewarm waterStarting temperatureGives a warm start; the room then pulls it downHelps, not enough alone in the north
Add about 10% old fermented batterStarting bacteria10% fermented batter restored fermentation in sterilised-grain batter (Mandhania 2019)Works
Add salt after fermentingRemoves a brake on bacteriaLong-time idli maker on The Fresh Loaf adds salt late in the cold, early in heatReasonable; small effect
Wrap the vessel in a towel or put it in a casseroleSlows heat lossBatter makes little heat of its own, so insulation only slows coolingHelps a warm start last
A pinch of fenugreek (methi) seedsTexture, flavourNo controlled study found that isolates its effect on riseOptional; won't fix cold
Add yeast or baking sodaAdds gas directlyLifts batter but changes the flavour and texture away from idliA shortcut, not a fix

How do you set up a warm spot for batter at home?

Use the oven, the inside of a switched-off microwave with a pot of hot water, or the top of a fridge, and confirm the spot with a thermometer before you trust it. The goal is a batter temperature of about 30°C, the heat at which the Mandhania study doubled batter in about 15 hours, or a little above it: the baker on The Fresh Loaf holds his at 37°C.

  1. Measure the room. A cheap probe thermometer in a glass of water left overnight tells you what your batter will face. Below about 24°C, plan for one of the warm spots below.
  2. Grind warm. Use lukewarm water while grinding. A mixer grinder also warms batter through friction, so check the temperature after grinding; aim for lukewarm, never hot to the touch.
  3. Seed it. Stir in two to three tablespoons of yesterday's good batter per litre, roughly 5 to 10%. This is a backslop, the same idea as a sourdough starter.
  4. Hold the heat. Put the covered vessel in an oven with only the lamp on, leaving the door slightly ajar if it runs over 40°C, or in a closed microwave beside a large bowl of just-boiled water that you refresh at bedtime.
  5. Salt at the end in winter. Add salt after the rise in cold weather, then fold gently so you don't knock out all the gas.
  6. Judge by the batter, not the clock. Look for a domed top, many small bubbles and a pleasant sour smell. Then refrigerate it to slow further souring.
Oven-with-lamp proofing box lamp only batter, lid on aim about 30°C door ajar if over 40°C: Test the empty oven with a thermometer once; every oven lamp runs differently
The lamp warms the closed cavity a few degrees at a time. Measure your own oven once before trusting it overnight.

How long will your batter take tonight?

Use this to check what you just read. It is an estimate, not a measurement: it takes the 15-hour doubling at about 30°C from Mandhania and colleagues and scales it with a Q10 factor, the rule of thumb that biological rates change by a fixed multiple for every 10°C. Move the factor to see how sensitive the answer is.

Estimated time for idli batter to double

At the defaults, a 22°C night stretches the rise to roughly 26 hours, long enough for some batches to sour before they lift. That is why the warm spot matters more than any additive.

Why does urad dal quality and the ratio matter more in winter?

Because in winter you have less gas to spare, and black gram is what holds it. Urad dal's proteins and its mucilaginous gums form films around the bubbles, so a batter with too little dal, or old dal, lets the slower winter gas escape.

Rice : black gramDensity after fermentingWinter verdict
2:1about 0.59 g/cm³ (lightest)Best gas net
4:1about 0.83 g/cm³Dense when gas is scarce

Those densities are from Ghosh and Chattopadhyay. Many home recipes use 3:1 or 4:1; if your winter idlis come out flat, moving toward 2.5:1 gives the gas a better net.

The rice side matters too. Parboiled idli rice gives a softer, more forgiving batter than raw rice, for reasons covered in the science of amylose in Indian rice. And the step-by-step recipe, with soaking times and grinder settings, is in how to make idli and dosa batter.

A heap of skinned white urad dal (black gram) grains
Skinned urad dal. Its proteins and gums trap the carbon dioxide; in a cold kitchen a little more dal in the ratio helps hold what little gas forms.Photo: Thamizhpparithi Maari, CC BY-SA 4.0, via Wikimedia Commons

Is store-bought batter a better winter option?

It solves the timing problem, because packaged batter arrives already fermented and is sold refrigerated. The category has grown fast.

BatterLaunchPriceNote
Nandini (Karnataka Milk Federation)25 December 2024₹40 for 450 g, ₹80 for 900 gAbout 3,000 kg a day within a month (ThePrint)
iD Fresh specialty rangeJanuary 2025₹110 to ₹149 for 800 gProtein, ragi, multigrain and mallige

The trade-off is that packaged batter keeps souring slowly in the fridge, just like the 10°C samples in the CFTRI study, so use it within its date.

Once the batter is right, the next variable is steaming; the same pressure and steam physics show up in how many pressure cooker whistles dal and rice need. And for the extreme version of this question, see what if idli batter fermented for a year.

Idli plates stacked inside a steamer with steam rising
Steaming sets the starch around the bubbles. Batter that rose properly gives a honeycomb crumb; cold, flat batter gives dense, rubbery idlis.Photo: SijiR, CC BY-SA 3.0, via Wikimedia Commons

Questions people ask

Can idli batter ferment in the fridge?

Barely. In Mandhania and colleagues' work, batter at 4°C did not ferment, and CFTRI's 10°C samples changed slowly over days. Ferment first in warmth, then refrigerate to slow souring.

Why does my batter ferment in summer but not in winter?

The bacteria are the same; their speed is not. At about 30°C batter doubles in roughly 15 hours. On a night in the high teens, the same rise can take more than a day.

Should I add salt before or after fermenting idli batter?

In winter, after. Experienced makers such as The Fresh Loaf poster find that salt slows the bacteria a little, which is useful in hot weather to prevent over-souring but unhelpful when the batter is already struggling.

Does adding fenugreek help idli batter ferment?

No controlled study has isolated fenugreek's effect on idli rise. Many cooks add a pinch for flavour and softness, but it will not compensate for a cold kitchen.

Can I use old batter as a starter?

Yes. Adding about 10% fermented batter restored normal fermentation even in batter made from sterilised grains in the 2019 study. Use batter that smelled clean and rose well.

More from the Taste Lab on Rasa.

Sources

  1. Mandhania et al.: Diversity and Succession of Microbiota during Fermentation of the Traditional Indian Food Idli, Appl Environ Microbiol 85(13) (2019; source for 30°C, 4°C, 15-hour doubling and 10% seeding)
  2. Ghosh and Chattopadhyay: Preparation of idli batter, its properties and nutritional improvement during fermentation, J Food Sci Technol 48(5) (2011; source for 20-26°C rise, density by ratio)
  3. Sridevi, Halami, Vijayendra: Selection of starter cultures for idli batter fermentation, J Food Sci Technol 47(5) (2010; source for 10°C vs 30°C storage counts)
  4. Wikipedia: Bengaluru climate table, IMD 1991-2020 normals (accessed 2026)
  5. Wikipedia: Climate of Delhi, Safdarjung IMD 1991-2020 normals (accessed 2026)
  6. Wikipedia: Chennai climate table, Nungambakkam IMD 1991-2020 normals (accessed 2026)
  7. The Fresh Loaf: Idli, reposted from my blog (fermentation temperature and salt timing) (accessed 2026)
  8. ThePrint: Bengaluru has a hot new favourite dosa batter (2025)
  9. FoodTechBiz: iD Fresh expands flagship batter range with four new variants (2025)
  10. FnB News: KMF storms into production and supply of idli-dosa batter (2024)
idli batterfermentationwinterlactic acid bacteriaWeissellaLeuconostocurad dalbatter temperature
Sapta

Sapta (Saptarshi Sarkar) is a mechanical engineer turned marketer in Bengaluru and a solo touring rider. He writes the seven Saptaverse blogs. Work with him or read more of Foodblog.

Keep reading in Foodblog

From the other blogs