The Fresh Loaf

A Community of Amateur Bakers and Artisan Bread Enthusiasts.

Question on bulk fermenting with cold room temperatures

DutchSourdough's picture

Question on bulk fermenting with cold room temperatures

Hi all!

So winter is coming to Europe and with the energy costs my house nowadays is typically 18C/65F.This means my bulk ferment slows down drastically and I need to reconfigure the proofing time. 

I've seen a lot of posts and blogs decribing optimal proofing temp to be around 75F. Also, I've seen posts describing that bulk ferment only slows down with lower room temp. 


My question is as follows: Will I get the same results with the same recipe, just by increasing the bulk fermentation time? Or do the bacteria and wild yeast act different in such a way that my sourdough breads will always come out different?


p.s. I dont have enoug space for proofing boxes but have been fiddling around with proofing in oven with creative solutions


clazar123's picture

A desk lamp with an incandescent bulb,a heating pad on low, a light bulb in a cooler, the oven light (if it is incandescent and not LED), the microwave light (again-not LED), on top of the refrigerator (heat rises from the coils usually in the back), on top of the hot water heater. You should be able to find a warm enough place somewhere or easily create one. 


Abe's picture

You might wish to pre-ferment a higher percentage of flour which will help with the final dough. Mix the pre-ferment earlier too. Use warm water and wrap it towels. On top of the fridge, if there's space, is a good idea. 

mariana's picture


You are asking good questions and the answers for them are in the two charts below

The first one is about growth of microorganisms, for example, in sourdough (starter, preferment, dough). And the second one is about fermentation rates at different temperatures, for example, at 18C vs 24C.

You can see that at 18C yeast and bacteria will multiply at the same speed whereas at 24C yeast will multiply faster than bacteria. Colder bulk fermentation gives sharper and more acidic tasting breads because there will be relatively more bacteria than yeast.

And from here you can see that at 18C bacteria produce more acid that yeast will produce gas. And at 24C yeast produces two times more gas than at 18C, so if you are waiting for your dough to double or triple it will take 2-3 times more time at 18C giving those more numerous bacteria tome to produce a lot more acid than at 24C

Will I get the same results with the same recipe, just by increasing the bulk fermentation time?

No. 6C difference between 18C and 24C is very significant in that case. It will be a distinctly different bread. You might like it better or not, try it.

Or do the bacteria and wild yeast act different in such a way that my sourdough breads will always come out different?

Yes, they grow differently and they produce acids and gasses differently at different temperatures.

DutchSourdough's picture

Thank you for the very elaborate anwser! If I understand correctly, the difference you touch on above mostly comes out in flavor notes.

How about openess of the crumb. Can I achieve the same at 17C vs 25C by increasing the bulk time? My understanding was that the activity of the bacteria is the eventual cause of overproofing to happen (correct me if im wrong :) ) So by that logic, if you want a very open crumb, you'd want a higher proofing temp to encourage that difference in activity?

mariana's picture

Yes, the difference is in flavor. Open crumb has nothing to do with the temperature or length of fermentation, it has to do with gluten development and shaping.

Overproofing can happen at any temperature and the culprit is the baker (or yeasts), not bacteria :)

Yes, you can achieve the same crumb at 17C vs 25C, or 10C or 30C. Simply watch your dough volume as it ferments and stretch and fold it as is necessary. If you have previous experience with making your bread at 25C, you will make it at 17C looking pretty much the same, but it will be quite different to taste.

You are not alone in Europe making bread at home at colder room temperatures. Many people have done it before in unheated kitchens and chilly bakeries overnight and are making it now all over the world. You can find recipes that describe how to make bread at 17-18C. 

Ilya Flyamer's picture
Ilya Flyamer

mariana, I always had a question about the first graph above, that you've posted in the past too... it looks like each curve is actually normalized to its peak height here, i.e. the "relative growth rate" on the Y axis is just relative to the optimal rate for this particular organism. This would mean that one can't compare growth rates between the two curves to each other, and hence even when one curve is above the other it still doesn't mean that e.g. yeast are growing faster than LABs. Am I reading it right?

mariana's picture

Yes, Ilya. Each microorganism has its own rate. For example, at its peak (@32C) the population of Lb san-franciscensis doubles  every 45min whereas the population of its symbiont, C milleri (sourdough yeast), at its peak at 28C doubles every 60 min. 

At 15C doubling of the population of that bacteria happens every 3 hours and of that yeast species - every 4 hours. They both slow down but one slows down more than another (4.3times slower vs 4 times). This affects their relative proportions to each other inside that environment.

We never know which specific species of yeast and bacteria we have in our own wild starters, almost certainly none of us who keep our starters in the fridge unfed for days have those two microorganisms, they perish at 4-8C, and which species are dominant while others are just minor players. So those charts and models are just models for one of the most famous and widespread starters in commercial/artisanal baking (San-Fransisco sourdough).

We know from experience that most other starters behave similarly, their communities of yeasts and LABs behave similarly in relation to each other even if numerically their peaks might be at different temperatures (@40C or even at 50C or 60C for some LABs used in breadmaking) and their curves will have slightly different shapes.

For practical breadmaking, especially at home, they are pretty much meaningless, but they explain some things that go behind the scenes: the fact that yeasts and bacteria are different and their behaviors change differently as temperatures change which affects acidity and gassing power of starters and levains..

ejspritz's picture

Hi Mariana,

You're just the right person to ask this:

I am having a product made and I'm in the prototype stage. I was having issues with my starter being really sluggish in the winter. My ambient temperature hovers around 67 degrees. So I'm copying the warming device used for making Kumbucha and sizing it down for jars normally used for sourdough starter. I asked the engineers to make the sleeve go to 80 degrees and turn off/on according to the ambient room temp. Before I blow all my savings on this venture, do you think this is the approach I should take or should the temperature control be adjustable? eg. low/med/high. And tp anyone out there reading this: is this a product you find useful?

mariana's picture

Hi ejspritz, 

of course, you can do both, whichever you think you need in your kitchen. It must be something that will work for you, first and foremost. Otherwise, it will be difficult for you to market it, if you don't see it working for you personally. 

To me personally, 80F/27C is not useful, because it is the naturally maintained temperature in our kitchen. I have it for free, so to speak.

80F/27C setting does not favor any particular microorganism in the sourdough community and doesn't produce any special flavors, it is mostly used for white all-purpose flour (T55, 10-11% protein) starters to protect their gluten from excessive deterioration during long stretches of fermentation. 

I use our wine fridge if I need my starter to be cooler, because a lot of recipes request either 10-15C or 20-22C range, or a variety of heating devices (my programmable bread machine and heating pads) to keep it warmer, if I need it.

Normally, it is something like

33-35C for yeasted rye starters,

40C for yeastless bacterial starters with mesophilic bacteria,

up to 50C for thermophilic lactic bacteria starters (for bread or for yogurt),

and around 60-70C for small batches of scalds which I later ferment with sourdough starters at different temperatures.

So, yes, variable temperature control (and adjustable width of the band/sleeve, because my starters are never fed and fermented in jars, but in 1qt measuring cups or in covered 1Qt bowls), would be useful for me.

Some people like microstarters these days, working with a few grams of flour and water, sleeves won't work for their tiny jars, it must be something like a pad or a pot in a pot setup for them, their tiny jar inside a bigger one filled with warm and kept warm water, for example. Low, medium, and high settings are not informative enough, you have to make sure it keeps the product at certain temperature plus or minus one degree.

I have low, medium and high settings on my heating pad, for example, and have to work around that, to place bowls either directly on top or slightly above, on a rack above the heating pad, to regulate the temperature in a simple way, without the aid of an additional temperature controller. 

To do market research, please, post your question about everyone else's needs or preferences as a separate post on the forum. People will give you their feedback. 

Best wishes, 


tpassin's picture

I wouldn't recommend cycling the temperature with an on-off control as you speculated.  You want an always-on proportional controller that will maintain the target temperature for any room temperature (that is lower than the target).

Proportional controllers are very cheap these days, and you can get your engineers to experiment with an aquarium heater.  For production use, I'd be thinking about wrapping a silicon seating pad around the container and controlling it with an industrial proportional controller (both, as I said, can be found fairly cheaply).

Or buy a proof box that controls the temperature.  It may cost several hundred dollars US but that might be cheaper than paying engineers to prototype something.

Debra Wink's picture
Debra Wink

You can see that at 18C yeast and bacteria will multiply at the same speed whereas at 24C yeast will multiply faster than bacteria. Colder bulk fermentation gives sharper and more acidic tasting breads because there will be relatively more bacteria than yeast.

Apologies for coming back to this again. I was curious about the graph, so I did an image search to find out where it came from thinking it might lead me to an article that could be useful. Instead, I found its origins on a blog site called It was created in 2009 [Sourdough work in progress (part II)] from data found here: What factors affect microbial growth in sourdough?

It looks like a computer-generated type of graph, but the problem is that only the optimum growth and zero growth data points were used and the points in between were not. That changes quite a lot! Not just the shapes of the curves and rate of ascending or descending, but the very relationship between them below 20C. The lines kiss at 20C, but they do not cross until they go to 10-11ish. Here is my crude approximation including the middle data points.

(Please note that I probably should have just whited out the bit about stronger and milder flavor, but I replaced it with FQ instead. FQ is the molar ratio of Lactic acid to acetic acid which is not the same as concentration/sourness of either. The ratio increases with rising temperature within the normal range of breadmaking but is not confined to the upper/lower lines indicated on the graph.)

These kinds of graphs are often misunderstood. This one is not about multiplication speeds per se, at least not absolute rates of yeast and bacteria or their being the same rate at any point (even at 18 or 20C). Nor does it show yeast multiplying faster than bacteria at 24C although I get why it appears that way.

What it shows is that optimum growth of (this) yeast happens at 28C and optimum growth of (this) bacterium at 32-33C. Please note that "optimum growth" doesn't represent any actual growth rates. Or the same rate even though the peaks are placed at the same height -- bacterial rate at its optimum is faster than yeast at its optimum. It just is; it's the nature of these organisms. If these were based on absolute rates, the green curve would be taller than the red, and the red would be completely inside the green. But this combined graph shrinks or expands each to 1 (100% of the max achieved) to show a different kind of relationship.

What it's saying is that at 20C where the red and green lines touch, both populations slow to half (0.5 on the y-axis) of what their max rates are at their respective optimum temperatures. But since bacteria multiply faster at max than yeast, they will also be multiplying faster than yeast when both have slowed to half their individual max rates. Bacteria are generally increasing faster than yeast when only looking at absolute rates. But rates have rates too (i.e., the rate of change in growth rate) and that's different for bacteria and yeast at different temperatures. This is what the lines are showing --- the relative changes in growth rates (relative to each other) relative to temperature. It's one factor that influences the ratio of LAB:yeast.

As far as stronger vs milder, temperature alone doesn't determine that, and so it won't always follow the label on the original graph. Nor does any of this account for lag times in real starters, levains and doughs.

My best,

There isn't enough information on the second graph to translate meaningfully to baking. What are the units of activity on the y-axis and what exactly was measured? Some context is always helpful, especially the purpose and design of the study.

tpassin's picture

According the the labels, these two normalized curves are about growth rate, meaning, I'm pretty sure, the numbers of organisms.  That's quite different from the rates of gas production.  In building a starter or levain, we're probably most interested in the numbers, but to get a good rise we're interested in gas production rates too.  Not the same things.


The Roadside Pie King's picture
The Roadside Pi...

The crumb openness or lack thereof is largely determined by the amount of fermentation allowed in the bulk ferment. Secondary elements would be dough handling and shaping. That being said, an open crumb is not always what We should be aspiring to. Wild yeast fermented baguette. I want to say about 30% expansion from out of the mixer. Nothing scientific all of my proofing containers and other baking gear are already packed up.  

rondayvous's picture

If you do your bulk in an oven warmed with a pot of boiling water in the afternoon, shape and leave it out overnight you can have bread baking in the morning. If it proofs too fast you can always put it in the fridge overnight.

CrustyJohn's picture

I can't comment on the scientific relation between temperature, yeast, and bacteria, but in my personal experience, I very often do my bulk fermentation at whatever room temperature happens to be- anywhere from high 50s up to mid 80s.  I have found that I can get very similar openness of crumb doing the bulk fermentation in the low-mid 60s as at warmer temperatures as long as I give it enough time for the dough to ferment fully.  In fact I just baked a loaf yesterday with a room temp. probably right around 65-67 degrees for most of the bulk fermentation time, and I think it was actually one of my best loaves I've made.  

Sometimes it can even be nicer schedule-wise to have that longer fermentation window.  I would also assume that doing the bulk fermentation at cooler temperature expands the window of optimum fermentation, making it easier to catch the dough at the right stage to shape and bake.