mirror of
https://github.com/hendricius/the-sourdough-framework
synced 2025-11-19 17:41:12 -06:00
1678 lines
85 KiB
TeX
1678 lines
85 KiB
TeX
\begin{quoting}
|
||
In this chapter, you will learn how to make
|
||
freestanding wheat sourdough bread.
|
||
\end{quoting}
|
||
|
||
\begin{figure}[!htb]
|
||
\includegraphics[width=\textwidth]{loaf-pan-free-standing.jpg}
|
||
\caption[Freestanding and loaf pan bread]{A freestanding sourdough bread
|
||
next to bread made in a loaf pan. Freestanding sourdough is considered
|
||
the supreme discipline of sourdough bread by many bakers.}
|
||
\end{figure}
|
||
|
||
Freestanding sourdough bread is my favorite
|
||
type of bread. It combines a great crunchy crust, superb
|
||
flavor, and a soft fluffy crumb. This is the type of bread
|
||
that is being inhaled by my friends and family. Unfortunately,
|
||
making this type of bread requires a lot more effort, patience,
|
||
and technique than other types of bread. You have to perfectly
|
||
balance the fermentation process. You cannot ferment for too
|
||
short and also not for too long. The techniques you need to
|
||
learn also require a bit more skill. It took me several attempts
|
||
to get this right. I faced several challenges: I~had the wrong flour.
|
||
I~didn't properly know how to use my oven.
|
||
When should I~stop the fermentation? There is a lot of information
|
||
out there. I~dug through most of it and have tried almost everything.
|
||
In many cases the information was wrong; in other cases, I~found another
|
||
valuable puzzle piece. Aggregating all this
|
||
information was one of my main motivations to start \texttt{The Bread Code}.
|
||
My key learning was that there is no recipe that
|
||
you can blindly follow. You will always have to adapt the recipe
|
||
to your locally available tools and environment.
|
||
|
||
But do not worry. After reading this chapter you will know
|
||
all the signs to look out for. You will be able to read your dough.
|
||
You will turn into a confident hobby baker who can bake bread
|
||
at home, at high altitudes, at low altitudes, in summer, in winter,
|
||
at your friend's place, and even on vacation. Furthermore,
|
||
you will know how to scale your production from 1 loaf to 100 loaves of bread.
|
||
If you ever wanted to open up a bakery, consider this knowledge to
|
||
be your foundation.
|
||
|
||
Mastering this process will enable you to make amazing bread
|
||
that tastes much better than any store-bought bread.
|
||
|
||
\section{The process}
|
||
|
||
\begin{flowchart}[!htb]
|
||
\centering
|
||
\input{figures/fig-wheat-sourdough-process.tex}
|
||
\caption{The typical process of making a wheat-based sourdough bread.}%
|
||
\label{fig:wheat-sourdough-process}
|
||
\end{flowchart}
|
||
|
||
The whole process of making great sourdough bread starts with
|
||
readying your sourdough starter. The key to mastering
|
||
this process is to manage the fermentation process properly.
|
||
For this, the basis is to have an active and healthy
|
||
sourdough starter.
|
||
|
||
Once your starter is ready, you proceed to mix all the ingredients.
|
||
You want to homogenize your sourdough starter properly. This
|
||
way you ensure an even fermentation across your whole dough.
|
||
|
||
After a short break, you will proceed and create dough strength.
|
||
Kneading will create a strong gluten network. This is essential
|
||
to properly trap the \ch{CO2} created during the fermentation.
|
||
|
||
Once you've kneaded, the bulk fermentation starts. It is called bulk fermentation
|
||
because you typically ferment multiple loaves together in one bulk.
|
||
Understanding when to stop this step will take some practice.
|
||
But nothing to worry about, you will learn the exact signs to look out for.
|
||
|
||
Once this is completed you need to divide your large blob of
|
||
dough into smaller pieces and pre-shape each piece. This allows
|
||
you to apply more dough strength and shape more uniform loaves.
|
||
|
||
The proofing stage follows where you finish the fermentation process.
|
||
Depending on your time you can proof it at room temperature or in the fridge.
|
||
Mastering proofing will turn your good loaf into a great loaf.
|
||
|
||
Lastly, you will finish the whole process by baking. You will learn different
|
||
options on how to properly steam your dough. This way your
|
||
dough will have a beautiful oven spring. During the second
|
||
stage of the baking process, you will finish building your crust.
|
||
|
||
All the steps rely on each other. You will need to get each of
|
||
the steps right to make the perfect bread.
|
||
|
||
\section{Readying your starter}%
|
||
\label{section:readying-starter}
|
||
|
||
The most crucial part of the bread-making process is your starter.
|
||
The starter is what starts the fermentation in your main dough.
|
||
If your starter is off, then your main dough is also going
|
||
to cause trouble during the fermentation. Your starter's
|
||
properties are passed on to your main dough. If your starter
|
||
doesn't have a good balance of yeast to bacteria, so will your
|
||
main dough.
|
||
|
||
\begin{flowchart}[!htb]
|
||
\centering
|
||
\input{figures/fig-starter-readiness.tex}
|
||
\caption[Process to prepare your starter before baking]{The process to check
|
||
your sourdough starter when making wheat-based doughs. In practice
|
||
I~frequently use a stiff sourdough starter. The stiff starter features
|
||
enhanced yeast activity. In that case, you can use the same ratios as
|
||
shown in the chart except for the water quantity. The stiff starter has
|
||
a hydration of \qtyrange{50}{60}{\percent}. So you would have half the
|
||
shown water quantities, i.e., if the chart shows \qty{100}{\gram} of
|
||
water, use \qtyrange{50}{60}{\gram} of water for your stiff starter.}%
|
||
\label{fig:process-starter-wheat-sourdough}
|
||
\end{flowchart}
|
||
|
||
Generally, think of the dough you are mixing as a big starter with salt.
|
||
After mixing all the ingredients, you have a green field environment again.
|
||
The yeast and bacteria start to fight again to outcompete each other.
|
||
There is plenty of food available, and they all do their best to win.
|
||
Depending on the starter you mix into your dough, some of the microorganisms
|
||
might have an advantage over others.
|
||
|
||
The first option to achieve a good balance is to apply feedings.
|
||
If your starter hasn't been fed in a long period, the
|
||
bacteria dominate. This happens if your starter has been
|
||
sitting unused in the fridge, for instance. As more and more
|
||
acidity piles up, the environment is becoming more and more hostile
|
||
to the yeast. The lactic acid bacteria tolerate this environment
|
||
better. Your dough fermentation would be more on the
|
||
bacterial side with this starter. By applying a couple of
|
||
feedings, the yeast becomes more active. The older your
|
||
starter, the more acid resistant the yeast becomes. Initially,
|
||
I~had to feed my starter 2--3 times to fix the balance. With my
|
||
more mature starter, one feeding seems to be enough to balance
|
||
the microorganisms.
|
||
|
||
Some people use a 1:1:1 ratio to refresh the starter. This would
|
||
be one part of the old starter (\qty{10}{\gram} for instance), 1 part of flour,
|
||
and one part of water. I~think this is utter rubbish. As mentioned
|
||
your starter is a miniature dough. You would never opt for a 1:1:1 ratio to
|
||
make dough. You might use a maximum of \qty{20}{\percent} starter to
|
||
make dough. That's why I~advocate using a 1:5:5 ratio or a
|
||
1:10:10 ratio depending on how ripe your starter is. As I~almost
|
||
always use a stiffer sourdough starter due to its enhanced
|
||
yeast fermentation advantages (see Section~\ref{section:stiff-starter})
|
||
my ratio is never 1:5:5. My ratio would be 1:5:2.5 (1 part old starter,
|
||
5 parts flour, 2.5 parts water). If it is very warm where you live
|
||
you could opt for the aforementioned 1:10:5 or 1:20:10. This
|
||
way you slow down the ripening of your starter. You can also use this
|
||
trick to make starter feeding work with your schedule.
|
||
If your starter is typically ready in 6~hours but today you need it
|
||
ready later, simply increase how much flour/water you feed your starter.
|
||
These are all values that you need to experiment with on your own.
|
||
Every starter is unique and might behave slightly differently.
|
||
|
||
The second option at your disposal is the starter quantity that
|
||
you use to make the dough. As previously stated your starter
|
||
regrows inside of your main dough. While I~would normally use
|
||
\qtyrange{10}{20}{\percent} of starter based on the flour, sometimes I~go
|
||
as low as \qty{1}{\percent} starter. This way the microorganisms have
|
||
more room to balance out while fermenting the dough. If my sourdough
|
||
starter has not been fed in a day, I~might use \qty{5}{\percent} of sourdough
|
||
to make a dough. If I~push this to 2 days without feedings,
|
||
I~lower the starter amount even further. I~would opt for the
|
||
previously mentioned \qty{1}{\percent} starter. If the food is very scarce,
|
||
your microorganisms will sporulate. They need to regrow again
|
||
from the spores they created. In this hibernation state, it takes
|
||
longer for them to become fully active again. I~have tried
|
||
several times to make dough directly out of a dry starter.
|
||
I~wasn't successful because the fermentation took too long.
|
||
The microorganisms had to regrow from spores and then begin
|
||
the fermentation. As explained earlier there is a limit to
|
||
fermentation times as your dough naturally breaks down.
|
||
Furthermore, you want your microorganisms to outcompete
|
||
other pathogens contained in the flour. The less starter
|
||
you use, the easier it is for them to reproduce. A strong
|
||
starter will outcompete other germs. While the method of
|
||
reducing the starter works, I~recommend Option 1 more.
|
||
It will reliably create better bread. Option 2 is typically
|
||
what I~use when I~fed my starter in the morning but didn't
|
||
manage to make a dough in the evening. I~don't want to feed
|
||
my starter again the next morning. I~would like to make a dough
|
||
directly without waiting and thus use less of the very ripe starter.
|
||
|
||
Over time you will become more accustomed to your starter
|
||
and how it behaves. You will be able to read the signs of its
|
||
activity and judge its state.
|
||
|
||
\section{Ingredients}
|
||
|
||
All you need to make great sourdough bread is flour, water, and salt. You
|
||
can of course add additional things to your dough such as seeds. I~personally
|
||
enjoy the hearty taste of whole-wheat. Thus I~like to add around
|
||
\qtyrange{20}{30}{\percent} of whole-wheat flour to the mix. You could also
|
||
make this recipe with \qty{100}{\percent}
|
||
whole-wheat flour directly. In this case, look out for strong whole-wheat
|
||
flour that is made from flour with higher protein. If you don't like whole-wheat
|
||
you can omit the flour from the recipe. Simply replace the listed
|
||
quantity with bread flour. One thing to consider about whole-wheat
|
||
flour is its increased enzymatic activity. By adding some whole-wheat
|
||
flour you will speed up the whole fermentation process.
|
||
|
||
Especially when getting started I~recommend using bread flour which
|
||
contains more gluten than all-purpose or cake flour. This is essential
|
||
when trying to bake a freestanding loaf with sourdough.
|
||
|
||
Find below an example recipe for 1 loaf including baker's math calculation:
|
||
|
||
\begin{itemize}
|
||
\item \qty{400}{\gram} of bread flour
|
||
\item \qty{100}{\gram} of whole-wheat flour
|
||
% Manual unit so we can use emphasis
|
||
\item \emph{Total: 500~g of flour}
|
||
\item \qtyrange{300}{450}{\gram} of room temperature water (\qty{60}{\percent} up to \qty{90}{\percent}). More on
|
||
this topic in the next chapter.
|
||
\item \qty{50}{\gram} of stiff sourdough starter (\qty{10}{\percent})
|
||
\item \qty{10}{\gram} of salt (\qty{2}{\percent})
|
||
\end{itemize}
|
||
|
||
In case you want to make more bread simply increase the quantities based on
|
||
how much flour you have. Let's say you have \qty{2000}{\gram} of flour available. The
|
||
recipe would look like this:
|
||
|
||
\begin{itemize}
|
||
\item \qty{1600}{\gram} of bread flour
|
||
\item \qty{400}{\gram} of whole-wheat flour
|
||
% Manual unit so we can use emphasis again
|
||
\item \emph{Total: 2000~g of flour}, equaling 4 loaves
|
||
\item \qty{1200}{\gram} up to \qty{1800}{\gram} of room temperature water (60 to \qty{90}{\percent})
|
||
\item \qty{200}{\gram} of stiff sourdough starter (\qty{10}{\percent})
|
||
\item \qty{40}{\gram} of salt (\qty{2}{\percent})
|
||
\end{itemize}
|
||
|
||
This is the beauty of baker's math. Simply recalculate the percentages, and you
|
||
are good to go. If you are unsure about how this works, please check out the
|
||
full Section~\ref{section:bakers-math} which looks at the topic in detail.
|
||
|
||
\section{Hydration}
|
||
|
||
Hydration refers to how much water you use for your flour. When
|
||
beginning to make bread, I~always got this wrong. I~followed a recipe from the
|
||
internet, and my dough never looked like the dough shown in the recipe.
|
||
The amount of water your flour requires is not fixed. It depends on the flour
|
||
you have.
|
||
|
||
When a seed gets into contact initially, the outer layers soak up the water.
|
||
That's why when using whole-wheat (still containing these layers) you have to
|
||
use a little bit more water.
|
||
|
||
By forming gluten strands, water is absorbed into your dough's gluten matrix. The higher the
|
||
protein value, the more water can be used.
|
||
|
||
Some bakers like to use highly hydrated doughs to create fluffier
|
||
bread\footnote{Sometimes it almost feels like a comparison of skill value
|
||
between bakers. The more water they can handle, the more skillful the baker.}.
|
||
The reason for this
|
||
is the dough's improved extensibility. The wetter the dough, the easier it is
|
||
for the dough to be stretched. When you pull it, the dough will hold its
|
||
shape. In comparison, a very stiff (low hydration) dough will maintain its
|
||
shape for a longer period. To visualize this, think of your extensible
|
||
dough as a balloon. The stiff dough is like a car tire.
|
||
The yeast has a much harder time inflating the car tire compared to the balloon.
|
||
That’s because the rubber of the car tire is much less extensible.
|
||
It requires much more force to inflate the tire. For this reason,
|
||
an extensible dough will inflate more in the oven. The loaf will
|
||
be visually bigger and offer an airier more open crumb structure.
|
||
|
||
While this might sound great, the high hydration causes several side effects.
|
||
|
||
\begin{enumerate}
|
||
\item Your dough becomes more difficult to handle. Your dough will be stickier.
|
||
\item Your dough has to be kneaded for longer to build a proper gluten
|
||
network.
|
||
\item During the fermentation your dough might become too extensible and lose
|
||
some of the dough strength. To circumvent this, stretch and folds are applied
|
||
compared to regular dough,
|
||
requiring you to invest a lot more work.
|
||
\item Shaping becomes much more of a hassle as the dough is very sticky.
|
||
\item The dough can stick to the banneton a lot easier while proofing.
|
||
\item If you wait too long during proofing, the dough won't have enough strength
|
||
left to pull upwards and will stay flat.
|
||
\item Generally, the higher the water content, the more bacterial fermentation you
|
||
have. Thus a wetter dough will reduce gluten faster than a stiffer dough.
|
||
This is why you have to start the fermentation with a sourdough starter in
|
||
perfect shape. Bakers use a process called autolysis to shorten the main
|
||
fermentation time to circumvent this.
|
||
\item The crumb, in the end, might be perceived as somewhat sticky. It still
|
||
contains a lot of water. I~love this crumb, but this comes down to personal
|
||
taste.
|
||
\end{enumerate}
|
||
|
||
To achieve a high-hydration dough, it is best to slowly add water to
|
||
your dough. Start with \qty{60}{\percent} hydration, then slowly add a bit more water. Knead
|
||
again until the water is absorbed. Repeat and add more water. As your dough
|
||
has already formed a gluten network, new water can be absorbed much easier.
|
||
You will be surprised by how much water your dough can soak up. This
|
||
method is commonly known as the bassinage method. More on that later.
|
||
By opting for this technique, I~was easily able to push a low-gluten flour
|
||
to a hydration of \qty{80}{\percent}. This
|
||
is also my method of choice when making dough now. I~keep adding water until
|
||
I~can feel that the dough has the right consistency. As you bake more bread,
|
||
you will develop a better look and feel for your dough. When mixing
|
||
by hand this can be quite cumbersome. It is a lot easier when using a stand
|
||
mixer.
|
||
|
||
All in all, increasing hydration requires a lot of trial and error. There
|
||
is however one option that makes things easier and causes fewer headaches:
|
||
Slow fermentation. You get the same extensibility advantages the high hydration
|
||
offers by simply letting your dough ferment for a longer period.
|
||
Slowing the fermentation process is easy. Use less
|
||
sourdough starter or ferment in a cooler environment.
|
||
|
||
There are two reasons for the slow fermentation advantages.
|
||
As explained earlier, both the protease enzyme and bacteria break down your
|
||
gluten network. So as fermentation progresses, your dough will automatically
|
||
become more extensible. This is because the rubber layers of your car tire
|
||
are slowly converted and eaten. Ultimately your car tire turns into a balloon
|
||
that can very easily be inflated. When waiting too long, the
|
||
balloon will burst. You will have no gluten left anymore, and your dough
|
||
becomes very sticky. Finding the sweet spot of enough rubber eating and not
|
||
too much is what the perfect wheat sourdough bread is about. But don't worry --- after reading
|
||
this chapter you will have the right tools at your disposal.
|
||
|
||
The advantages of slow fermentation can be nicely observed when experimenting
|
||
with a fast-fermenting yeast dough (\qty{1}{\percent} dry yeast based on flour). The
|
||
crumb of such a dough is never as
|
||
open as a dough made with sourdough. Furthermore, the protease enzyme
|
||
cannot do its job within such a short fermentation period.
|
||
Large industrial bakeries add active malt which contains a
|
||
lot more enzymes. This way the time required to make the dough is shortened. You
|
||
will most likely find malt as an ingredient in supermarket bread. It is a
|
||
great hack. The baked turbo fermentation bread will feature a relatively dense
|
||
and not fluffy crumb. That is because only very little gluten is broken down when
|
||
finishing the fermentation period in 1~hour. If you were to slow things down,
|
||
the dough would look completely different.
|
||
Try this again and use much less yeast. This is the
|
||
secret of Neapolitan pizza. Only a tiny bit of yeast is used to make the
|
||
dough. My default pizza recipe calls for around \qty{150}{\mg} of dry
|
||
yeast per \unit{\kg} of flour. Give it a shot yourself the next time you
|
||
make a yeast-based dough. Try to push the fermentation to at least 8~hours.
|
||
The difference is incredible. You will have made bread with a much more
|
||
fluffy and open crumb. The flavor of the dough is drastically improved. Your
|
||
crust becomes crisper and features a better taste. This is because amylases have
|
||
converted your starches into simpler sugars which brown better during baking.
|
||
If you only learn one thing from this book, it is that slow fermentation is
|
||
the key to making great bread.
|
||
|
||
For this reason, my default hydration is much lower than the hydration of other
|
||
bakers. I~prefer slower fermentation for my recipes.
|
||
The sweet spot for my default flour is at around \qty{70}{\percent} hydration.
|
||
Again, this is a highly subjective value that works for my flour.
|
||
|
||
If you are just getting started with a new batch of flour,
|
||
I~recommend conducting the following test. This will help you to
|
||
identify the sweet spot of your flour's hydration capabilities.
|
||
|
||
Make 5 bowls with each \qty{100}{\gram} of flour. Add different slightly increasing
|
||
water amounts to each of the bowls.
|
||
|
||
\begin{itemize}
|
||
\item \qty{100}{\gram} of flour, \qty{55}{\gram} of water
|
||
\item \qty{100}{\gram} of flour, \qty{60}{\gram} of water
|
||
\item \qty{100}{\gram} of flour, \qty{65}{\gram} of water
|
||
\item \qty{100}{\gram} of flour, \qty{70}{\gram} of water
|
||
\item \qty{100}{\gram} of flour, \qty{75}{\gram} of water
|
||
\end{itemize}
|
||
|
||
Proceed and mix the flour and water mixture until you see that there
|
||
are no chunks of flour left. Wait 15~minutes and return to your dough.
|
||
Carefully pull the dough apart with your hands. Your dough should be elastic, holding
|
||
together very well. Stretch your dough until very thin. Then hold it against a light.
|
||
You should be able to see through it. The flour-water mixture that breaks without
|
||
seeing the windowpane is your no-go zone. Opt for a dough with
|
||
less hydration than this value. You will know that your flour mix can go up to
|
||
\qty{65}{\percent} hydration, for instance. Use the leftovers of this experiment
|
||
to feed your starter.
|
||
|
||
|
||
\begin{figure}[!htb]
|
||
\includegraphics[width=\textwidth]{window-pane-effect}
|
||
\caption[The window pane test]{The window pane test allows you to see if you
|
||
developed your gluten well enough.}
|
||
\end{figure}
|
||
|
||
|
||
From an economic perspective, water is the cheapest component in your bread
|
||
dough. When running a bakery, a higher hydrated dough will weigh more and have
|
||
lower production costs. The profit will be higher. This comes at the price
|
||
of increasing labor costs and more potential failures due to the enhanced
|
||
difficulty.
|
||
|
||
\section{How much starter?}
|
||
|
||
Most bakers use around \qty{20}{\percent} sourdough starter based on the
|
||
flour weight. I~recommend going much lower, to around
|
||
\qtyrange{5}{10}{\percent}.
|
||
|
||
By adjusting the amount of pre-ferment you can influence the time your dough
|
||
requires in the bulk fermentation stage. The more starter you use, the faster
|
||
this process is. The smaller the starter quantity, the slower. With a higher
|
||
quantity of starter, you are introducing more microorganisms to your main
|
||
dough. The higher this quantity, the faster the rate of fermentation in your
|
||
dough is.
|
||
|
||
The other factor influencing the rate of fermentation is the temperature of
|
||
your dough. The warmer the temperature, the faster the process; the colder, the
|
||
slower the process.
|
||
|
||
While food is available, the microorganisms will reproduce and increase in
|
||
quantity. The process is self-limiting: it stops when there is no
|
||
more food available. This can be compared to wine making where
|
||
the yeast ultimately sporulates and dies as ethanol levels increase. The ethanol creates an
|
||
environment that makes it impossible for other
|
||
microorganisms to join the feast. The same thing happens with the acidity
|
||
created by the bacteria. The high acidity slows the fermentation process and
|
||
prevents new microorganisms from entering the system.
|
||
|
||
Initially, your starter's properties are carried over to the main dough. Then,
|
||
as time progresses, the microorganisms adapt to the new environment. If your
|
||
starter is very bacterial then your main dough's fermentation will also be. You
|
||
end up with a dough that is not as fluffy as it could be. It will taste quite
|
||
sour, too sour for most people.
|
||
|
||
If you were to use an extreme value of around \qty{90}{\percent} starter based on your flour, there
|
||
would be very little room for the microorganisms to adjust in the main dough.
|
||
If you were to just use \qty{1}{\percent}, your microorganisms can regrow into a
|
||
desirable balance in the dough. Furthermore, you need to consider that a high value
|
||
of starter means a high inoculation with already fermented flour. As
|
||
mentioned earlier, enzymes break down the dough. This means the higher this
|
||
value, the more broken-down fermented flour you have. A too-long fermentation
|
||
always results in a very sticky dough that cannot be handled. The more
|
||
starter you use, the faster you will get to this point. If you were to use a
|
||
very little amount of starter, your flour might have naturally broken down
|
||
before the fermentation has reached the desired stage. You can observe this
|
||
when using a small quantity of around \qty{1}{\percent} sourdough starter. The small
|
||
amount of added microorganisms will not be able to reproduce fast enough
|
||
before the protease has broken down your dough completely.
|
||
|
||
As explained earlier the key to making great bread is a slow but not too slow
|
||
fermentation. Enzymes require time to break down your dough. Taking all this
|
||
into consideration, I~try to aim for a fermentation time of around 8 to 12~hours. This seems to be
|
||
the sweet spot for most of the flours that I~have worked with. To achieve this,
|
||
I~use around \qty{5}{\percent} of sourdough starter in summer times
|
||
(temperatures around \qty{25}{\degreeCelsius} (\qty{77}{\degF}) in the
|
||
kitchen). In winter times I~opt for around \qty{10}{\percent} up to
|
||
\qty{20}{\percent} sourdough starter (kitchen temperature around
|
||
\qty{20}{\degreeCelsius} (\qty{68}{\degF})). This
|
||
allows me to use a sourdough starter that's not in perfect condition. As
|
||
explained earlier, your
|
||
bread dough is essentially a gigantic starter. The low inoculation rate allows
|
||
the starter to regrow inside your main dough into a desirable balance.
|
||
Furthermore, the enzymes have enough time to break down the flour. This also
|
||
allows me to skip the so-called autolysis step completely (more in the next section).
|
||
This greatly simplifies the whole process.
|
||
|
||
\section{Autolysis}%
|
||
\label{section:autolysis}
|
||
|
||
Autolysis describes the process of just mixing flour and water and letting
|
||
this sit for a period of around 30~minutes up to several hours. After this
|
||
process is completed, the sourdough starter and salt are added to the
|
||
dough\footnote{I~have tested adding the salt at the start and end of the
|
||
autolysis process and could not notice a difference. Based on my current
|
||
understanding, the importance of adding salt later seems to be a myth.}.
|
||
|
||
The overall time that flour and water are in contact is extended. Thus you get the
|
||
beneficial enzymatic reactions that improve the taste and characteristics of the
|
||
dough. I~do not recommend autolysis as it adds an unnecessary step to the
|
||
process. Instead, I~recommend the fermentolysis technique which will be covered in the
|
||
next section of this book.
|
||
|
||
The effects of autolysis are very interesting. Try to mix just flour and
|
||
water and let that sit for a day. During the day, check the consistency of
|
||
your dough. Try and stretch the dough. If you dare, you can also taste the
|
||
dough throughout the day. With each hour, your dough will become
|
||
more extensible. It will be easier to stretch the dough. At the same time, your
|
||
dough will start to taste sweeter and sweeter. The protease and amylase enzymes
|
||
are doing their job. The same process is used when making oat milk. By letting
|
||
the mixture sit for some time, enzymes work on the oats. The taste is perceived as
|
||
sweeter and more appreciated. This process is further accelerated the more
|
||
whole-grain your flour is. The hull contains more enzymes. The gluten network
|
||
will ultimately tear, and your dough flattens out. For wheat sourdough, this is
|
||
your worst enemy. When this happens, your dough will become leaky and release
|
||
all that precious gas created during the fermentation. You need to find the
|
||
right balance of your dough breaking down just enough and not too much.
|
||
|
||
When you use a high inoculation rate of around \qty{20}{\percent} sourdough starter
|
||
your fermentation can be very quick. At \qty{25}{\degreeCelsius} it could be finished in as little as 5~hours.
|
||
If you ferment longer, your dough becomes leaky. At the same time, in
|
||
these 5~hours, the enzymes have not broken down the flour enough. This means
|
||
the dough might not be as elastic as it should be. Furthermore, not enough
|
||
sugars have been released and thus the flavor after baking is not good
|
||
enough\footnote{I~have not seen studies yet looking at enzymatic speeds depending on
|
||
the temperature. But I~assume the higher the temperature, the faster these
|
||
reactions. This goes up until a point when the enzymes break down under
|
||
heat.}. That's why bakers opt for autolysis. The autolysis starts the enzymatic
|
||
reactions before the microorganism fermentation begins. This way after 2~hours
|
||
of autolysis (an example) and 5~hours of fermentation the dough is in the
|
||
perfect state before beginning proofing.
|
||
|
||
When you try to mix your salt and starter into the flour/water dough you will
|
||
notice how cumbersome this is. It feels like you have to knead again from scratch
|
||
one more time. You will spend more time mixing dough.
|
||
|
||
For that reason, I~am strongly advocating utilizing the fermentolysis approach
|
||
which greatly simplifies the mixing and kneading process.
|
||
|
||
\section{Fermentolysis}%
|
||
\label{section:fermentolysis}
|
||
|
||
The fermentolysis creates the same advantageous dough properties the
|
||
autolysis creates without the headache of mixing your dough twice. You do this
|
||
by extending the fermentation time of your dough. Rather than doing a 2-hour
|
||
autolysis and 5-hour bulk fermentation you opt for an overall 7-hour
|
||
fermentation period.
|
||
|
||
To do this, you use less sourdough starter. A conventional recipe including the
|
||
autolysis step might call for \qty{20}{\percent} sourdough starter. Simply reduce this
|
||
value to \qtyrange{5}{10}{\percent}. The other option could be to place the dough in a colder
|
||
environment and thus reduce the speed at which your microorganisms replicate.
|
||
|
||
\begin{table}[!htb]
|
||
\centering
|
||
\input{tables/table-starter-usage-activity.tex}
|
||
\caption[Quantity of sourdough]{A table visualizing how much sourdough
|
||
starter to use depending on temperature and the starter's activity
|
||
level.}
|
||
\end{table}
|
||
|
||
Based on my experience and my sourdough, my ideal bread always takes around 8
|
||
to 12~hours during bulk fermentation. Based on my availability throughout
|
||
the day, I~use a higher or lower starter quantity. If I~wanted to achieve a completed
|
||
fermentation in 8~hours, I~would opt for a \qty{10}{\percent} sourdough starter. If
|
||
I~wanted it to be ready in 12~hours, I~would opt for less starter, around \qty{5}{\percent}.
|
||
Simply mix all the ingredients and your fermentation begins. The
|
||
enzymes and microorganisms commence their work. On a very warm summer day, the
|
||
mentioned quantities no longer work. With a \qty{10}{\percent} starter, the same dough
|
||
would be ready in 5~hours up to a point of no return. Another additional hour
|
||
would cause the dough to break down too much. In this case, I~would opt for 5
|
||
percent sourdough starter to slow the whole process down to reach the 8 to 12
|
||
hour window again. If it is very hot, I~might use as little as \qty{1}{\percent}
|
||
sourdough starter\footnote{Please take these values with a grain of salt as
|
||
they depend on your flour and your sourdough starter. These are values that
|
||
you have to experiment with. After baking a couple of breads you will be able
|
||
to read your dough much better.}. You have to play with the timings on your own.
|
||
Rather than relying on timing though, I~will show you a much better and more precise approach
|
||
by using a fermentation sample. This will be covered later in this chapter.
|
||
|
||
Even for yeasted doughs, I~no longer use autolysis. I~just reduce the amount
|
||
of yeast that I~am using. Opting for the fermentolysis will
|
||
save you time and simplify your bread-making process. As mentioned in previous chapters,
|
||
the secret to making great bread is a slow but not too slow fermentation.
|
||
|
||
\section{Dough strength}
|
||
|
||
Dough strength is a fancy way to describe the bread-kneading process. As you wait and
|
||
knead, the gluten bonds in your dough become stronger. The dough
|
||
becomes more elastic and holds together better. This is the basis for trapping
|
||
all the gases during the fermentation process. Without the gluten network,
|
||
the gases would just diffuse out of your dough.
|
||
|
||
\begin{flowchart}[!htb]
|
||
\centering
|
||
\input{figures/fig-kneading-process.tex}
|
||
\caption{The gluten development process for a wheat-based dough.}%
|
||
\label{fig:wheat-sourdough-kneading-process}
|
||
\end{flowchart}
|
||
|
||
It might sound odd, but the most important part of kneading is waiting. By
|
||
waiting you are allowing your flour to soak up water. This way the gluten
|
||
bonds of your dough form automatically and your dough becomes more elastic.
|
||
So you could be kneading for 10~minutes initially just to be surprised
|
||
that kneading 5~minutes and waiting 15~minutes has the same effect.
|
||
|
||
The gluten proteins glutenin and gliadin virtually instantly bond after being
|
||
hydrated. Disulfide bonds enable the longer portions of
|
||
glutenin to join with one another and form sturdy, extensible molecules.
|
||
Glutenins add strength, whilst the more compact gliadin proteins allow
|
||
the dough to flow like a fluid. Ultimately, the longer you wait, the more
|
||
your gluten network transforms into a web-like structure. This is what
|
||
traps the gases during the fermentation process~\cite{how+does+gluten+work}.
|
||
|
||
\begin{figure}[!htb]
|
||
\centering
|
||
\input{plots/fig-yeast-sourdough-strength.tex}
|
||
\caption[Dough strength over time without kneading]{A schematic
|
||
visualization of automatic gluten development. The doughs are not
|
||
kneaded, just initially mixed. Note how dough strength deteriorates
|
||
over time as enzymes break down the flour. The effect is accelerated for
|
||
sourdough due to the bacteria's gluten proteolysis.}%
|
||
\label{fig:wheat-yeast-sourdough-degradation}
|
||
\end{figure}
|
||
|
||
The soaking process has to be extended the more whole-wheat flour is used.
|
||
The purpose of the wheat kernel's outer bran is to soak up water as fast
|
||
as possible. The enzymes become activated and start the sprouting process.
|
||
Because of this, less water is available for the gluten bonds to develop.
|
||
Either wait a bit longer or proceed and use slightly more water for
|
||
the dough.
|
||
|
||
This is the same principle that popular no-knead recipes follow. By making a less
|
||
hydrated dough and waiting your gluten network automatically forms. You still
|
||
have to mix and homogenize the ingredients. You wait a few minutes just to
|
||
find your dough having developed incredible dough strength with no additional
|
||
kneading\footnote{Give it a shot yourself. The automatic formation of gluten
|
||
networks is an amazing phenomenon that still fascinates me every time I~am
|
||
making dough.}.
|
||
|
||
If you over-hydrate your dough at the beginning it becomes more difficult
|
||
for the gluten chains to form. The molecules are not as close together in
|
||
a wetter dough compared to a stiffer dough. It is harder for the molecules
|
||
to align and form the web structure. For this reason, it is always easier
|
||
to start with lower hydration and then increase the water quantity if needed.
|
||
This is also commonly known as the \emph{Bassinage method}. The gluten
|
||
bonds have formed at the lower hydration and can then be made more extensible
|
||
by adding water and kneading again. This is a great trick to make
|
||
a more extensible dough with lower-gluten flour~\cite{bassinage+technique}.
|
||
|
||
When machine kneading a dough, opt for the same technique shown in
|
||
flowchart~\ref{fig:wheat-sourdough-kneading-process}. Initially opt for a low
|
||
speed. This helps the homogenization process.
|
||
After waiting to allow the flour to soak up the water, proceed on a higher speed
|
||
setting. A good sign of a well-developed gluten network is
|
||
that your dough lets go of the container. This is because of the gluten's elasticity.
|
||
The elasticity is higher than the desire of the
|
||
dough to stick to the container.
|
||
|
||
\begin{figure}[!htb]
|
||
\includegraphics[width=\textwidth]{dough-strength-sourdough}
|
||
\caption[Dough strength over time with kneading]{A schematic visualization
|
||
of gluten development in sourdoughs with different kneading techniques.
|
||
A combination of techniques can be utilized to achieve maximum dough
|
||
strength.}%
|
||
\label{fig:dough-strength-sourdough}
|
||
\end{figure}
|
||
% See https://www.figma.com/file/wTUVe6Nm2INOvT82mJhQur/Dough-strength-visualisation?node-id=0%3A1&t=fjdPvXYuJpsdQfWN-1 for
|
||
% the source of this visualization
|
||
|
||
Generally, the more dough strength you create, the less sticky your dough is going to
|
||
feel. As the dough holds together, it will no longer stick to your hands as
|
||
much. This is a common problem beginners face. Sticky dough is frequently
|
||
the sign of a not well enough developed gluten network.
|
||
|
||
\begin{figure}[!htb]
|
||
\includegraphics[width=\textwidth]{dough-surface-touchpoints}
|
||
\caption[Touching the dough surface]{A schematic visualization of how a rough
|
||
dough surface creates more touch points compared to a smooth dough
|
||
surface. By touching the rough surface the dough will swell and get into
|
||
contact with more areas of your hand.}%
|
||
\label{fig:dough-touch-points}
|
||
\end{figure}
|
||
|
||
Kneading more is generally beneficial in almost all cases, as it results in a
|
||
stronger gluten network. However, when making soft milk breads, you might prefer
|
||
a more extensible dough from the start. In this scenario, excessive kneading
|
||
could lead to a chewier final bread, which is not desirable if you aim for a
|
||
fluffier texture. Achieving this fluffier dough can be accomplished by kneading
|
||
less. While this is an exception, properly kneading your wheat-based doughs
|
||
is generally advised.
|
||
|
||
When you use a stand mixer, you can run into the issue of kneading too much. This
|
||
is almost impossible in practice though. Even after kneading for 30~minutes on medium
|
||
speed, my doughs hardly ever were over-kneaded. The moment you knead
|
||
too much, the color of the dough can begin to change. You mostly
|
||
notice this, though, during baking. The resulting loaf looks very
|
||
pale and white. This is because mixing dough causes oxidation,
|
||
which is necessary for the development of gluten.
|
||
However, if the dough is mixed too much, the compounds that contribute
|
||
to the bread's flavor, aroma, and color may be destroyed, negatively
|
||
affecting the quality of the bread~\cite{oxidization+dough}.
|
||
|
||
The last step before beginning bulk fermentation is to
|
||
create a smooth dough ball. By making sure your dough's surface is
|
||
smooth, you will have fewer touch points when touching the dough.
|
||
See figure~\ref{fig:dough-touch-points} for a schematic visualization
|
||
of how your hand touches a rugged and smooth dough.
|
||
With the smooth surface, your dough is going to stick less on your hands. Applying
|
||
later stretches and folds will be a lot easier. Without a smooth
|
||
surface, the dough becomes almost unworkable. Folding the dough later
|
||
becomes an impossible task. This is a frequent mistake I~see many
|
||
new bakers commit.
|
||
|
||
\begin{figure}[!htb]
|
||
\includegraphics[width=\textwidth]{dough-ball-steps}
|
||
\caption[Creating a smooth surface]{The transformation of a sticky dough
|
||
blob to a dough with a smooth surface. The goal is to reduce surface
|
||
touchpoints with your hands to make the dough less sticky when working
|
||
it.}%
|
||
\label{fig:dough-ball-steps}
|
||
\end{figure}
|
||
|
||
To make the dough's surface smooth, place your dough on a wooden board or
|
||
on your kitchen's countertop. Drag the dough with your palm over the surface.
|
||
A dough scraper could be used here for assistance.
|
||
Drag the dough towards you while making sure the top center of the dough stays in place.
|
||
It can help to gently place your second hand on top of the dough so that
|
||
the dough mass moves while retaining its orientation. Once the whole dough
|
||
is too close to the edge of the container/countertop, gently move it back
|
||
with two hands. By doing so, you are stretching the outer surrounding gluten layer.
|
||
For this reason, it is important to not use any flour during this process.
|
||
By using flour, you can no longer drag the dough over the surface and thus
|
||
you can't stretch the gluten. Always imagine you are touching something utterly sticky.
|
||
By doing so you will automatically try to touch the dough as little
|
||
as possible. Keep repeating the process until you see that the dough
|
||
has a nice smooth surface. The final dough should look like the dough
|
||
shown in~\ref{fig:dough-ball-steps}.
|
||
|
||
If your outer gluten layer tears, you have overstretched your dough. In
|
||
that case, take a 10-minute break, leaving your dough on the kitchen countertop.
|
||
This allows the gluten to re-bond and heal. Repeat the same process
|
||
and the damaged rugged areas should disappear.
|
||
|
||
The same dough-rounding technique is used later during
|
||
the pre-shaping process. After creating dough strength you
|
||
have all the time you need to practice rounding. Round the dough
|
||
as much as possible until it tears. Then wait the aforementioned 10~minutes and repeat.
|
||
Later, you don't have any room for error. Your technique has to be on point.
|
||
An over-pre-shaped dough can potentially not recover.
|
||
|
||
\section{Bulk fermentation}%
|
||
\label{section:bulk-fermentation}
|
||
|
||
After mixing the starter into your dough, the next stage of
|
||
the process known as bulk fermentation begins. The term
|
||
\emph{bulk} is used because in bakeries, multiple loaves are fermented
|
||
together in bulk. If you are a home baker, you might bulk
|
||
ferment a single loaf. The bulk fermentation ends when you
|
||
divide and pre-shape, or directly shape your final loaves or loaf.
|
||
|
||
The hardest part when making sourdough bread is controlling
|
||
the fermentation process. Bulking long enough but not too
|
||
long is the deciding factor for making great bread at home.
|
||
Even with poor shaping and baking techniques, you'll be able
|
||
to make excellent bread, solely by mastering the bulk
|
||
fermentation process.
|
||
|
||
With a too-short bulk, your crumb will be
|
||
perceived as gummy. Your crumb will feature large pockets of
|
||
air commonly referred to as \emph{craters}. A too-long fermentation
|
||
results in the dough breaking down too much. The resulting
|
||
dough will stick to your banneton and spread while baking
|
||
into a pancake-like structure.
|
||
|
||
The key is to find the sweet spot between not too little
|
||
and not too much bulk fermentation. I'd always recommend pushing
|
||
the dough more toward a longer fermentation. The
|
||
flavor of the resulting bread is better compared to a pale
|
||
underfermented dough.
|
||
|
||
\begin{table}[!htb]
|
||
\centering
|
||
\input{tables/table-fermentation-effects.tex}
|
||
\caption[Stages of sourdough fermentation]{The different stages of
|
||
sourdough fermentation and the effects on crumb, alveoli, texture,
|
||
and overall taste.}
|
||
\end{table}
|
||
|
||
The worst thing you can do when fermenting sourdough
|
||
is to rely on a recipe's timing suggestions. In \qty{99}{\percent}
|
||
of the cases, the timing will not work for you. The writer
|
||
of the recipe probably has different flour and a different
|
||
sourdough starter with different levels of activity. Furthermore,
|
||
the temperature of the fermentation environment might be
|
||
different. Just small changes in one parameter result
|
||
in a completely different timing schedule. One or two~hours'
|
||
difference results in the dough not fermenting long enough, or
|
||
turning it into a gigantic sticky fermented pancake. This
|
||
is one of the reasons why the current baking industry prefers
|
||
to make solely yeast-based doughs. By removing the bacteria
|
||
from the fermentation, the whole process becomes a lot more
|
||
predictable. The room for error (as shown in figure~\ref{fig:wheat-yeast-sourdough-degradation})
|
||
is much larger. The doughs are perfect to be made in a
|
||
machine.
|
||
|
||
\begin{flowchart}[!htb]
|
||
\centering
|
||
\input{figures/fig-bulk-fermentation.tex}
|
||
\caption[Process to check the bulk fermentation]{During the bulk
|
||
fermentation, multiple doughs are fermented together in bulk. A
|
||
challenging aspect of homemade sourdough bread is to determine when this
|
||
stage of fermentation is completed. This chart shows multiple available
|
||
options to check on the bulk fermentation progress.}%
|
||
\label{fig:bulk-fermentation}
|
||
\end{flowchart}
|
||
|
||
Experienced bakers will tell you to go by the look and feel of
|
||
the dough. While this works if you have made hundreds of loaves,
|
||
this is not an option for an inexperienced baker. As
|
||
you make more and more dough, you will be able to judge
|
||
the dough's state by touching it.
|
||
|
||
My go-to method for beginners is to use an \emph{Aliquot jar}.
|
||
The aliquot is a sample that you extract from your dough. The
|
||
sample is extracted after creating the initial dough strength.
|
||
You monitor the aliquot's size increase to judge the
|
||
level of fermentation of your main dough. As your
|
||
dough ferments, so does the content of your aliquot jar. The moment your
|
||
sample reached a certain size, your main dough is ready
|
||
to be shaped and proofed. The size increase you should
|
||
aim for depends on the flour you have at hand. A flour
|
||
with a higher gluten content can be fermented for a
|
||
longer period. Generally, around \qty{80}{\percent}
|
||
of your wheat flour's protein is gluten. Check your flour's
|
||
packaging to see the protein percentage. The actual size increase
|
||
value is highly variable depending on your flour composition.
|
||
I~recommend beginning with a size increase of \qty{25}{\percent} and testing
|
||
up to \qty{100}{\percent} with subsequent bakes. Then identify a value
|
||
that you are happy with.
|
||
|
||
\begin{table}[!htb]
|
||
\centering
|
||
\input{tables/table-dough-size-increase.tex}
|
||
\caption[Increase of size versus protein content]{Reference values for
|
||
how much size increase to aim for with an aliquot jar depending on
|
||
the dough's protein content.}
|
||
\end{table}
|
||
|
||
The beauty of the aliquot is that no matter the surrounding
|
||
temperature, you will always know when your dough is ready.
|
||
While the dough might be ready in 8~hours in summer, it could
|
||
easily be 12~hours in winter. You will always ferment your
|
||
dough exactly on point.
|
||
|
||
|
||
\begin{figure}[!htb]
|
||
\includegraphics[width=\textwidth]{aliquot-before-after}
|
||
\caption[Aliquot Jar]{An aliquot jar to monitor the dough's fermentation
|
||
progress. It took 10~hours for the dough to reach a \qty{50}{\percent}
|
||
size increase.}
|
||
\end{figure}
|
||
|
||
While the aliquot sample has enabled me to consistently bake
|
||
great loaves, there are limitations to consider. It's crucial
|
||
to use a cylindrical-shaped container to properly judge
|
||
the dough's size increase. Furthermore, it is essential
|
||
to use room-temperature water when making your dough. If the
|
||
water is hotter, your aliquot, due to its smaller size,
|
||
will cool down faster. The aliquot will ferment more slowly
|
||
than your dough. Similarly, when you use too cold water,
|
||
your sample will heat up faster than the large dough mass.
|
||
In that case, your aliquot is ahead of your main dough. You
|
||
would probably stop the fermentation too early. Make sure
|
||
to keep the dough and aliquot close together. Some people even
|
||
place the aliquot in the same container. This makes sure that
|
||
both are in the same environment temperature. The aliquot
|
||
is also less reliable if your ambient temperature changes
|
||
a lot during the day. In that case, your aliquot will adapt
|
||
faster than your main dough. The readings will always be slightly
|
||
off. If you are making a large chunk of dough with more
|
||
than \qty{10}{\kg} of flour, the jar is also less reliable. The biochemical
|
||
reactions happening inside your dough will heat it.
|
||
The fermentation itself is exothermic which means
|
||
that it produces heat.
|
||
|
||
Another more expensive option is to use a pH meter
|
||
to monitor your dough's fermentation state. As the lactic
|
||
and acetic acid bacteria ferment, more acidity is piled
|
||
up inside your dough. The acidity value (pH) can be
|
||
measured using such a meter. The more acidity, the lower the pH
|
||
value of your dough. The pH scale is logarithmic, meaning
|
||
that each digit change will have a 10x increase in acidity.
|
||
A sourdough dough might begin fermenting at \pHvalue{6.0},
|
||
then shortly before baking has roughly \pHvalue{4.0}. This means
|
||
that the dough itself is 10x times 10x (= 100x) sourer
|
||
than at the beginning. By using the meter, you can always
|
||
judge the state of your dough's acidification and then act
|
||
accordingly.
|
||
|
||
To use the pH meter successfully, you need to find pH values
|
||
that work for your dough. Depending on your starter,
|
||
water, and flour composition, the pH values to look out
|
||
for are different. A stronger flour with more gluten
|
||
can be fermented for a longer period. To find out
|
||
the pH values for your bread I~recommend taking
|
||
several measurements while making your dough.
|
||
|
||
\begin{enumerate}
|
||
\item Measure the pH value of your sourdough starter before using it
|
||
\item Check the pH after mixing all the ingredients
|
||
\item Check the pH before dividing and pre-shaping
|
||
\item Check the pH before shaping
|
||
\item Check the pH of your dough before and after proofing
|
||
\item Check the pH of your bread after baking
|
||
\end{enumerate}
|
||
|
||
If the bread you made turned out successfully with your values,
|
||
you can use them as a reference for your next batch. If the
|
||
bread didn't turn out the way you like, either shorten
|
||
the fermentation or extend it a little bit.
|
||
|
||
\begin{table}[!htb]
|
||
\centering
|
||
\input{tables/table-ph-values-dough.tex}%
|
||
\caption[Dough's pH during bread preparation]{Example pH values for
|
||
the different breakpoints of my own sourdough process.}%
|
||
\label{table:sample-ph-values}
|
||
\end{table}
|
||
|
||
The beauty of this method is its reliability. Once you have found
|
||
out your good working values, you can reproduce
|
||
the same level of fermentation with each subsequent dough.
|
||
This is especially handy for large-scale bakeries that want
|
||
to achieve consistency in each bread.
|
||
|
||
While this method is very reliable, there are also certain
|
||
limitations to consider.
|
||
|
||
First of all the pH values that work for me likely won't work for
|
||
you. Depending on your own starter's composition of lactic
|
||
and acetic acid bacteria, your pH values will be different.
|
||
You can use the values shown in Table~\ref{table:sample-ph-values}
|
||
as rough ballpark figures. Regardless, you need to find values
|
||
that work for your setup.
|
||
|
||
Another limitation is the price. You will need to purchase
|
||
a high-tech pH meter, ideally, a meter featuring a spearhead
|
||
\footnote{Not every pH meter is suitable for measuring dough.
|
||
Please refer to the manual to make sure it is certified for
|
||
measuring the pH of liquid and semi-solid media. To receive
|
||
accurate pH readings further ensure that your pH meter
|
||
is properly calibrated.}.
|
||
This way you can directly poke the meter deep into the dough.
|
||
At the same time, automated temperature adjustments are a
|
||
feature to look out for. Depending on the temperature,
|
||
the pH value varies. There are tables you can use to
|
||
do the adjustment calculations. More expensive meters
|
||
have this feature built in. The pH meter loses accuracy
|
||
over time. For this reason, you need to frequently
|
||
calibrate it. The process is cumbersome and takes time.
|
||
Lastly, you need to carefully rinse the pH meter before
|
||
using it in your dough. The liquid surrounding the
|
||
head of your pH meter is not food-safe and thus should
|
||
not be eaten. I~rinse the meter for at least one minute
|
||
before using it to measure my dough's fermentation stage.
|
||
|
||
The last method to judge the state of bulk fermentation
|
||
is to read the signs of your dough. The more bread you have
|
||
made, the more accustomed you will become to this process.
|
||
Look out for the dough's size increase. This can sometimes
|
||
be a challenge when your dough is inside a container.
|
||
You can help yourself by marking your container. Some bakers
|
||
even use a transparent rectangular bulk container. You
|
||
can use a pen to mark the initial starting point. From there
|
||
on you can nicely observe the size increase. Similar to the
|
||
mentioned aliquot sample, look out for a size increase that works
|
||
for your sourdough composition.
|
||
|
||
\begin{figure}[!htb]
|
||
\includegraphics[width=\textwidth]{bulk-finished-dough}
|
||
\caption[Dough at the end of bulk fermentation]{A dough in a good state to
|
||
finish bulk fermentation. Notice the tiny bubbles on the dough's surface.
|
||
They are a sign that the dough is inflated well enough.}
|
||
\end{figure}
|
||
|
||
Look out for bubbles on the surface of your dough. They
|
||
are a good sign that your dough is inflated with gas. The
|
||
further you push the bulk fermentation the more bubbles
|
||
will appear. If you overdo this stage, the dough becomes leaky, and
|
||
the bubbles will disappear again.
|
||
|
||
Take note of the dough's smell. It should match the same
|
||
smell of a ripe starter shortly before collapsing. As mentioned
|
||
before, your dough is nothing but a gigantic starter. You
|
||
can also proceed and taste your dough. It will taste like
|
||
pickled food. Depending on the acidity you can judge how
|
||
far the dough is in the fermentation process. The final bread
|
||
will taste less sour. That's because a lot of acidity evaporates
|
||
during baking\footnote{More on this topic later.
|
||
Just by baking longer and/or shorter, you can control
|
||
the tang of your final baked bread. The longer
|
||
you bake, the less sour the final loaf. The shorter,
|
||
the more acidity is still inside the bread. The resulting
|
||
loaf will be sourer.}.
|
||
|
||
When touching the dough, it should feel tacky
|
||
on your hands. The dough should also be less sticky
|
||
compared to earlier stages. If the dough is overly
|
||
sticky, you have pushed the fermentation too far.
|
||
|
||
If you pushed the bulk fermentation too far, you won't be able
|
||
to bake a freestanding loaf with the dough anymore. But don't
|
||
worry. You can move your dough into a loaf pan, or use parts
|
||
of the dough as the starter for your next dough. When using
|
||
a loaf pan, make sure it's properly greased. You might have
|
||
to use a spatula to transfer your dough. Allow the dough
|
||
to proof for at least 30~minutes in the loaf pan before
|
||
baking it. This makes sure that large cavities induced
|
||
by the transfer are evened out. You could push the proofing
|
||
stage to 24~hours or even 72~hours. The resulting
|
||
bread would feature an excellent, very tangy taste.
|
||
|
||
|
||
\section{Stretch and folds}
|
||
|
||
\begin{figure}[!htb]
|
||
\includegraphics[width=\textwidth]{dough-being-glued}
|
||
\caption[Gluing dough]{A dough where two sticky sides are being glued
|
||
together using a stretch and fold. This process creates excellent dough
|
||
strength.}
|
||
\end{figure}
|
||
|
||
In this section, you will learn all you need to know about stretching and
|
||
folding. You will learn when to stretch and fold and how to use this technique
|
||
to your advantage.
|
||
|
||
Stretching and folding is a set of techniques used by bakers during the bulk
|
||
fermentation stage. The process involves stretching the dough and then
|
||
folding the dough onto itself. Some recipes call for a single stretch
|
||
and fold, others for multiple.
|
||
|
||
The primary goal of this technique is to provide
|
||
additional dough strength to your dough. As shown in figure~\ref{fig:dough-strength-sourdough}
|
||
there are multiple ways to create dough strength\footnote{In fact I~have seen many no-knead
|
||
recipes calling for no initial kneading, but then applying stretch and folds
|
||
during the bulk fermentation. The time required to do all the folds probably
|
||
matches the initial kneading time required.}. If you do not knead as much at
|
||
the start, you can reach the same level of dough strength by applying stretch
|
||
and folds later. The more stretch and folds you do, the more dough strength
|
||
you add to your dough. The result will be a more aesthetic loaf that has
|
||
increased vertical oven spring.
|
||
|
||
Sometimes, if the dough is very extensible
|
||
and features very high hydration, stretching and folding is essential.
|
||
Without it, the dough itself would have too little dough strength and not
|
||
spring in the oven at all.
|
||
|
||
Another benefit of stretch and folds are their homogenization properties. By
|
||
folding the dough you are redistributing areas that are fermenting faster
|
||
than other areas. The heat in your dough is not the same in all areas.
|
||
The fermentation itself produces heat. For that reason, some of the areas in
|
||
your dough will ferment a little faster than others. This means that some
|
||
areas hold more gas and more acidity than others. Applying a stretch and fold
|
||
will redistribute heat, gas, and acidity. Some bakers also refer to this
|
||
process as crumb building. Careful folds ensure that your final dough's crumb
|
||
is not overly wild featuring large cavities. If you notice overly
|
||
large cavities in your final dough's crumb, then you might be able to fix that
|
||
by applying more stretch and folds\footnote{In many cases these cavities can
|
||
also happen when a dough does not ferment enough. The crumb is commonly called
|
||
Fool's Crumb. Refer to the later Debugging Crumb Structures chapter of this
|
||
book to learn more about it.}. Please refer to Section~\ref{section:debugging-crumb-structure}
|
||
``\nameref{section:debugging-crumb-structure}'' for more information on reading
|
||
your crumb.
|
||
|
||
\begin{figure}[!htb]
|
||
\includegraphics[width=\textwidth]{stretch-and-fold-steps}
|
||
\caption[Stretch and fold steps]{An overview of the steps involved to perform
|
||
stretch and folds for wheat-based doughs.}%
|
||
\label{figure:stretch-and-fold-steps}
|
||
\end{figure}
|
||
|
||
The reason for the technique's popularity lies in its efficiency. By stretching
|
||
the dough outwards, you increase your dough's surface area. You then fold the
|
||
dough over, essentially gluing large areas of the dough together. Imagine a
|
||
piece of paper on which you place the glue. Then you fold the paper. Large areas
|
||
of the paper now stick together. Repeat the same process with more glue until
|
||
you have created multiple layers of paper and glue. This is the same thing that
|
||
happens to your dough. With only very few movements you have applied glue to your
|
||
dough.
|
||
|
||
To apply a stretch and fold first wet your hands with cold water. Watered hands
|
||
work wonders in reducing the dough's tendency to stick to your hands. Proceed and
|
||
carefully loosen the dough from the edges of your bulk container. Do this by
|
||
carefully placing your hand at the edge of the dough and pushing your hand
|
||
downwards on the container's walls. Once you have reached the bottom, drag the dough
|
||
a little bit inwards. The dough should stay in place and not move back to the
|
||
edge of your container. Try to be as swift as possible with this motion. The
|
||
slower you are, the more dough will stick to your hands. Repeat the same process
|
||
once all around your dough until the dough is free of your container's edges.
|
||
Wet your hands one more time and then carefully lift one side of the dough with
|
||
two hands placed in the center upwards. Make a fold in the center of the dough.
|
||
The upper smooth side needs to be placed on the bottom of the container. By doing
|
||
so, you will be gluing together the two sticky bottom sides. The top smooth side should
|
||
not be sticky in your hands, while the bottom rough surface should tend
|
||
to stick to your hands. Rotate the container
|
||
and repeat the same thing from the other side. Rotate the container 90°
|
||
and then repeat the process once again. Rotate the container another 180° in
|
||
the same direction
|
||
and repeat the fold one last time. By doing so you have applied 4 folds in total. Your
|
||
dough should now stay in place and resist flowing outwards\footnote{Please
|
||
also refer to~\cite{stretch+and+fold+technique} for a video showing you how to
|
||
best perform the technique.}.
|
||
|
||
In theory, there is no limit to how often you can stretch and fold. You could
|
||
apply one every 15~minutes. If your dough has enough dough strength already,
|
||
applying additional folds is just a waste of time\footnote{You could do it
|
||
just to better understand how the dough feels in your hands at different
|
||
fermentation stages.}. If you apply a large number of consecutive folds, the
|
||
outer layer of gluten
|
||
will tear. In that case, you just have to wait for at least 5--10~minutes until
|
||
the gluten bonds heal and you can try again. When the gluten does not heal
|
||
anymore, chances are you have pushed the fermentation for too long. Likely
|
||
most of the gluten has broken down and you are already
|
||
in the decay stage shown in figure~\ref{fig:dough-strength-sourdough}.
|
||
|
||
\begin{figure}[!htb]
|
||
\includegraphics[width=\textwidth]{dough-requiring-stretch-and-fold}
|
||
\caption[A flattened out dough]{A dough during bulk fermentation that has
|
||
flattened out. To improve its dough strength, a stretch and fold should
|
||
be applied.}
|
||
\end{figure}
|
||
|
||
Now, the reasonable amount of stretch and folds you should do greatly depends on how much you
|
||
kneaded initially and how extensible your dough is. A good recommendation is
|
||
to observe your dough in your bulk container. Once you see that the dough
|
||
flattens out quite a lot and spreads towards the edges of your bulk container,
|
||
you can proceed and apply a stretch and fold. For \qty{95}{\percent} of the doughs
|
||
that I~am making, this is hardly more than once. I~like to make overnight
|
||
doughs and in that case, I~typically apply one stretch and fold directly after
|
||
waking up. Then the bulk fermentation might take another 2~hours before I~proceed
|
||
with dividing and pre-shaping or directly shaping.
|
||
|
||
\section{Optional: Dividing and Preshaping}
|
||
|
||
Dividing and pre-shaping is an optional step that is done
|
||
once your sourdough finishes with the bulk fermentation stage.
|
||
The step is required if you are making multiple loaves in one
|
||
batch. It is optional if you are making a single loaf.
|
||
|
||
\begin{flowchart}[!htb]
|
||
\centering
|
||
\input{figures/fig-dividing-preshaping.tex}
|
||
\caption[Is dividing your dough required check]{Dividing is only required when you are
|
||
making multiple loaves in a single dough batch.}%
|
||
\label{fig:dividing-decision-tree}
|
||
\end{flowchart}
|
||
|
||
The goal of dividing your dough into smaller pieces is to portion
|
||
your dough accordingly. This way you'll have multiple pieces of bread
|
||
which all weigh the same. For this reason, a scale is commonly
|
||
used to weigh the pieces of dough. If one piece of dough weighs
|
||
too little you can simply cut a bit more from your dough blob
|
||
to increase its weight.
|
||
|
||
When cutting the dough, try to be as concise as possible with your
|
||
movements. You don't want to unnecessarily damage your dough too much.
|
||
Quick movements with a knife or dough scraper help to prevent the
|
||
dough from sticking too much to your tools.
|
||
|
||
\begin{figure}[!htb]
|
||
\includegraphics[width=\textwidth]{divide-preshape}
|
||
\caption{The steps of dividing and preshaping your dough.}
|
||
\end{figure}
|
||
|
||
I~sometimes like to draw small lines with the dough scraper's edge
|
||
on the large dough mass before cutting it into smaller pieces.
|
||
This helps me to better plan where I~want to do my incisions. When
|
||
I~plan to make 8 loaves I~try to use the lines to divide the dough
|
||
into 8 equally sized portions before cutting. If this is not precise enough,
|
||
you can use the aforementioned scale.
|
||
|
||
Now that you have cut your dough, the resulting chunks are not in an equal shape.
|
||
This is problematic for the next stage when you are shaping your dough.
|
||
The resulting loaves wouldn't look nice and even. You would probably
|
||
end up with areas that tear the moment you are shaping your dough.
|
||
You wouldn't start the whole proofing process on a good foundation. For that
|
||
reason, you need to pre-shape your dough.
|
||
|
||
Pre-shaping is done for several reasons:
|
||
\begin{itemize}
|
||
\item You divided your dough and require pre-shaping
|
||
\item Your dough lacks dough strength. Pre-shaping will add more strength
|
||
\item You want to even out the final loaf's crumb structure. By pre-shaping,
|
||
the resulting crumb will look more even.
|
||
\end{itemize}
|
||
|
||
If you are making a single loaf from one dough batch the step is not required.
|
||
In that case, you can directly proceed with shaping, skipping this step.
|
||
|
||
The pre-shaping technique is the same as the process figure~\ref{fig:dough-ball-steps}.
|
||
Whereas earlier you could tear the dough's surface this could now result in a catastrophe.
|
||
For this reason, I~recommend practicing this step for as long as you need after kneading.
|
||
The gluten network might be so extensible and degraded at this point that there
|
||
is hardly any room for error. The dough wouldn't come together again. The only
|
||
way to save such dough is to use a loaf pan.
|
||
|
||
\begin{figure}[!htb]
|
||
\includegraphics[width=\textwidth]{preshape-direction}
|
||
\caption[Dragging direction]{Drag the dough in the direction of the rough
|
||
surface area. This way you minimize the movements required to complete
|
||
the step.}%
|
||
\label{fig:preshape-direction}
|
||
\end{figure}
|
||
|
||
Pre-shape the dough as much as is needed to round up the top
|
||
surface area. Try to touch the dough as little as possible
|
||
to reduce its ability to stick to your hands. Drag the dough
|
||
in the direction where you see a rough surface area. In
|
||
case you have too little space to drag the dough because it might
|
||
fall from the edge of your counter, simply lift it with a swift movement and place
|
||
it in a better position for pre-shaping. Please refer to figure~\ref{fig:preshape-direction}
|
||
for a visualization showing the pre-shaping direction.
|
||
|
||
Try to set yourself a limit of movements to finish pre-shaping
|
||
a dough. Then you will be more conscious about each movement
|
||
you are performing. At the start you can try 5 movements,
|
||
iteratively reducing this to 3. The only reason for exceeding these
|
||
numbers could be if you on purpose want to even out the crumb
|
||
structure of your final loaves further.
|
||
|
||
\begin{figure}[!htb]
|
||
\includegraphics[width=\textwidth]{preshaped-dough}
|
||
\caption{Baguette doughs resting after preshaping.}%
|
||
\label{fig:dough-after-preshaping}
|
||
\end{figure}
|
||
|
||
Once you finished pre-shaping allow the dough balls to rest
|
||
on your counter for at least 10--15~minutes. Do not
|
||
cover the pre-shaped balls. By drying out the surface,
|
||
the following shaping step will be easier. The dried-out surface
|
||
will not stick to your hands as much. As
|
||
you tightened the dough's gluten you will need to
|
||
allow it to relax. Without a resting period, you wouldn't
|
||
be able to shape your dough into, for instance, a baguette-like structure.
|
||
The dough would resist each movement
|
||
always springing back into the previous shape. You
|
||
might have noticed this before, when making pizza dough. If you
|
||
don't wait long enough after balling the pizzas, it's impossible
|
||
to stretch the pizza. By waiting a few more minutes,
|
||
stretching becomes a lot easier. The dough will not resist
|
||
being transformed into the final shape that you like.
|
||
|
||
The aforementioned 10--15~minutes bench rest time depends
|
||
on how strongly you pre-shaped your dough. The more
|
||
you pre-shape the longer you need to wait. If your dough
|
||
resists a lot during shaping, extend this period up to 30~minutes.
|
||
If you wait too long, your dough's surface area can become too dry,
|
||
resulting in the dough tearing during shaping. As always, please
|
||
take these timings with a grain of salt and experiment in
|
||
your environment.
|
||
|
||
\section{Shaping}
|
||
|
||
\begin{flowchart}[!htb]
|
||
\centering
|
||
\input{figures/fig-shaping-process.tex}
|
||
\caption[Sourdough shaping process]{A schematic visualization of the shaping process
|
||
including checks for an overfermented dough.}%
|
||
\label{fig:shaping-decision-tree}
|
||
\end{flowchart}
|
||
|
||
Shaping will give your dough the final shape before baking. After
|
||
completing shaping, your dough proceeds to the proofing stage and
|
||
will then be scored and ultimately baked.
|
||
|
||
There are countless shaping techniques. The technique to choose
|
||
depends on the type of bread you want to make. Some techniques
|
||
are gentler on the dough, making sure that the dough does not
|
||
degas. Other techniques are faster but degas the dough a little
|
||
more. The tighter you shape, the more evened out your final dough's
|
||
crumb structure will look. At the same time, a tighter shaping-technique
|
||
will improve your dough's strength. More strength will ultimately result
|
||
in more vertical oven spring.
|
||
|
||
The following instructions assume that you want to make a batard-style
|
||
bread featuring an oblong shape. Learning this technique
|
||
will provide you with a solid knowledge foundation that
|
||
can easily be extended to make bread rolls or baguettes.
|
||
|
||
Mastering the challenging shaping technique will likely take you
|
||
multiple attempts. You only have a single attempt per dough, though. If you
|
||
make a mistake, the final bread is likely not going to turn out as good
|
||
as it could. If this technique causes you a headache, I~recommend making
|
||
a larger batch of dough and dividing and preshaping it into
|
||
smaller portions. Instead of making a large batard, practice making miniature
|
||
batard bread rolls.
|
||
|
||
\subsection[Flouring the surface]{Apply flour to the dough's surface.}
|
||
|
||
\begin{figure}[!htb]
|
||
\includegraphics[width=\textwidth]{step-1-flour-applied}
|
||
\caption[Step 1 of shaping process]{A dough that has flour applied to its
|
||
surface. This is the first step of the shaping process.}%
|
||
\label{fig:shaping-flour-surface}
|
||
\end{figure}
|
||
|
||
If you are only making 1 loaf out of your dough, apply flour
|
||
generously to the top layer of your dough. Rub the flour onto your
|
||
dough with your hands. Flip over your container. Wait a little bit
|
||
to allow the dough to release itself from the container. Proceed
|
||
with step 3.
|
||
|
||
If you divided and pre-shaped, apply flour generously to the dough's
|
||
top layer as well. With gentle hands spread the flour evenly across
|
||
the dough's surface. See figure~\ref{fig:shaping-flour-surface} for a
|
||
visual representation of how your dough should look after coating
|
||
the surface.
|
||
|
||
\subsection[Flipping the dough]{Flip the dough over}
|
||
|
||
\begin{figure}[!htb]
|
||
\includegraphics[width=\textwidth]{step-2-flipped-over}
|
||
\caption[Step 2 of shaping process]{A flipped-over dough. Note how the
|
||
sticky side is facing you while the floured side is facing the
|
||
countertop. The sticky side is used as glue to hold the dough together.}
|
||
\end{figure}
|
||
|
||
With gentle hands, carefully remove the dough from the surface. If
|
||
you possess a dough scraper, carefully tuck it under the dough with
|
||
rapid movements. Flip the dough over, making sure that the floured
|
||
areas are in contact with your hands. The non-floured bottom area that was
|
||
stuck to the counter is a no-touch zone. Try to avoid touching it
|
||
as it is rough and thus will stick to your hands.
|
||
|
||
Gently proceed and place the dough with the previously top-facing side
|
||
on your counter. The floured area is now on the surface, whereas the
|
||
sticky side is facing you.
|
||
|
||
\subsection[Create rectangular shape]{Make the dough rectangular}
|
||
|
||
\begin{figure}[htb!]
|
||
\includegraphics[width=\textwidth]{step-3-rectangular}
|
||
\caption[Step 3 of shaping process]{A flipped-over dough. Note how the
|
||
sticky side is facing you while the floured side is facing the
|
||
countertop.}%
|
||
\label{fig:shaping-rectangular-dough}
|
||
\end{figure}
|
||
|
||
You should be facing the sticky side of your dough now. Note how
|
||
the dough is currently round and not rectangular. The circular
|
||
shape will not be ideal when shaping the oblong batard.
|
||
|
||
For this reason, proceed and stretch the dough a little bit until
|
||
it has a more rectangular shape. While stretching, make sure to touch
|
||
the sticky side as little as possible. Place your hands on the bottom
|
||
floured side and the edge of the sticky side. With gentle hands,
|
||
stretch the dough until the shape in front of you looks rectangular.
|
||
Refer to figure~\ref{fig:shaping-rectangular-dough} and compare
|
||
your dough with the shown dough.
|
||
|
||
\subsection[Folding]{Fold the dough together}
|
||
|
||
\begin{figure}[htb!]
|
||
\includegraphics[width=\textwidth]{step-4-folding}
|
||
\caption[Step 4 of shaping process]{The process of folding a batard. Note
|
||
how the rectangle is first glued together and then rolled inwards to
|
||
create a dough roll. Ultimately the edges are sealed to create a more
|
||
uniform dough.}%
|
||
\label{fig:shaping-folding}
|
||
\end{figure}
|
||
|
||
Now that you have created the rectangular shape, your dough
|
||
is ready to be folded together. This only works because the side
|
||
facing you is sticky. Because of the dough's stickiness,
|
||
we can effectively glue it together, creating a very
|
||
strong bond.
|
||
|
||
You can practice this step with a piece of rectangular paper.
|
||
Once you mastered folding on paper you can easily apply
|
||
this to your real-life dough.
|
||
|
||
Make sure the batard is placed in front of you. Take the side
|
||
that faces you and fold it into the middle of the dough. Carefully
|
||
tuck it down so that it glues together with the sticky side.
|
||
|
||
Take the other side and fold it over the side you just folded.
|
||
Stretch the dough as much as possible towards you. Tuck it down
|
||
on the edge, creating your first glue layer.
|
||
|
||
Rotate the dough so that it is aligned lengthwise in front of you.
|
||
Rotate the dough inwards so that the seam side
|
||
now faces you.
|
||
|
||
Start to roll the dough inwards beginning at the top of the dough.
|
||
Keep rolling the dough inwards until you have created a dough roll.
|
||
|
||
Refer to figure~\ref{fig:shaping-folding} for a full visual
|
||
representation of the process.
|
||
|
||
If your dough does not hold its shape, chances are you have pushed
|
||
the fermentation too far. Most of the gluten has been degraded
|
||
and the dough won't be able to hold its shape. In this case,
|
||
the best option is to use a loaf pan to bake your bread. The
|
||
final bread will taste amazing but not offer the same texture
|
||
a freestanding bread would offer. Please refer to
|
||
Section~\ref{section:debugging-crumb-structure} for more
|
||
details on how to properly read your dough's crumb structure.
|
||
|
||
\subsection[Sealing]{Sealing the edges}
|
||
|
||
Your dough has finished shaping now. Sealing the edges
|
||
is an optional step. I~like to do it because, in my opinion,
|
||
the final baked bread will look a little bit nicer without
|
||
any rough edges.
|
||
|
||
Gently pull together the swirl-like-looking edges of your dough
|
||
with two fingers. Rotate the dough and then repeat the same process
|
||
from the other side as well.
|
||
|
||
\subsection[Proofing preparation]{Prepare for proofing}
|
||
|
||
\begin{figure}[htb!]
|
||
\includegraphics[width=\textwidth]{step-6-prepare-proofing}
|
||
\caption[Step 5 of shaping process]{The shaped dough is ready for proofing
|
||
in the banneton. Note how the seam side is now facing you. The floured
|
||
previous top side is facing downwards.}%
|
||
\label{fig:shaping-prepare-proofing}
|
||
\end{figure}
|
||
|
||
You should have a beautifully shaped dough in front of you now.
|
||
The proofing stage is about to start. To simplify later
|
||
scoring and to make sure your dough won't stick to your banneton,
|
||
apply another flour rub to the dough's surface. This
|
||
will dry out the surface and reduce the dough's tendency
|
||
to stick to everything.
|
||
|
||
For the coating, I~recommend using the same flour you used
|
||
to make your dough. Rice flour is only recommended if you
|
||
want to apply artistic scoring patterns later. It is better
|
||
to use more flour than too little flour. Excess flour can be
|
||
brushed off later.
|
||
|
||
Once your dough has been coated, it is ready to be placed on your banneton.
|
||
If you do not have a banneton, you can use a bowl
|
||
with a kitchen towel inside.
|
||
|
||
The currently top-facing floured surface will be downwards-facing in your banneton.
|
||
By doing so the banneton can be flipped
|
||
over before baking, releasing the dough\footnote{The same
|
||
applies when making other doughs such as baguette doughs. The floured
|
||
surface will always be downwards facing. The dough is then flipped over
|
||
once for baking.}.
|
||
|
||
Proceed and lift the dough with 2 hands from the counter.
|
||
Gently rotate it once and then place the dough in your
|
||
banneton for proofing\footnote{The seam side should now be facing you.
|
||
Some bakers like to seal the seam a little more. I~did
|
||
not notice that this improves the dough's strength. As far as I~can
|
||
tell, this only improves the visual appearance of the bottom side
|
||
of the final loaf.}. If you did everything right, then your
|
||
dough should look somewhat similar to the dough shown in figure~\ref{fig:shaping-prepare-proofing}.
|
||
As the last step of shaping, place a kitchen towel over your banneton
|
||
or bowl and begin proofing.
|
||
|
||
\section{Proofing}
|
||
|
||
In bread baking, proofing refers to the final rise of dough before baking,
|
||
after it has been shaped into a loaf. The chemical reactions and processes
|
||
that occur during bulk fermentation and proofing are the same.
|
||
|
||
By shaping your dough, it has lost some of the air previously generated
|
||
throughout the bulk fermentation. The goal of proofing is to inflate
|
||
the dough again. A dough without proofing wouldn't offer the same texture
|
||
as a properly proofed dough. The proofed dough features a very fluffy
|
||
and soft crumb.
|
||
|
||
There are two proofing techniques. One strategy is to proof the dough
|
||
at room temperature whereas the other proofs the dough in the fridge.
|
||
Fridge-proofing is also commonly known as retarding.
|
||
|
||
Some bakers claim that cold-proofing improves the final flavor of the bread.
|
||
In all the loaves that I~retarded I~could not tell a difference
|
||
in terms of flavor for cold-proofed doughs. The microorganisms work
|
||
at a slower rate at colder temperatures. But I~doubt that they alter
|
||
their biochemical processes. More research is needed on the topic
|
||
of retarding and flavor development.
|
||
|
||
\begin{flowchart}[!htb]
|
||
\centering
|
||
\input{figures/fig-proofing-process.tex}
|
||
\caption[Sourdough proofing process]{A schematic overview of the different steps of
|
||
the sourdough proofing process. The proofing technique to choose depends
|
||
on your availability and schedule.}%
|
||
\label{fig:proofing-process}
|
||
\end{flowchart}
|
||
|
||
To me, the sole purpose of cold-proofing is its ability to allow you
|
||
to better manage the timing of the whole process. Assuming you finished shaping
|
||
your dough at 10 pm, chances are you wouldn't want to wait for another
|
||
2~hours to proof the dough and then another 1 hour to bake it. In this case,
|
||
you can move your dough directly to the fridge after shaping. Your
|
||
dough will be proofing overnight in the fridge. Then it can be baked at any time
|
||
the following day (there are a few exceptions; more on that later).
|
||
This is especially handy for large-scale bakeries that use fridge-proofing
|
||
extensively. Some of the doughs are proofed a day before and placed in the fridge.
|
||
Early in the morning, they can be baked directly out of the fridge. Within 2
|
||
hours they will be ready to sell the first bread to morning customers. If
|
||
throughout the day more bread is needed, they simply take some proofed dough out
|
||
of the fridge and bake it. The time frame in which you can bake retarded
|
||
dough is big. It can be as little as 6~hours later up to 24~hours later.
|
||
|
||
Assuming you made an overnight dough and your dough is ready in the morning,
|
||
the situation might be different. You potentially want to bake the dough directly
|
||
for breakfast, or at lunchtime. In this case, you wouldn't want to proof the dough for
|
||
another 6~hours in the fridge. Room temperature-proofing is your technique
|
||
of choice.
|
||
|
||
To summarize, choose the technique that works for you depending on your
|
||
schedule and availability.
|
||
|
||
\subsection{Room temperature-proofing}
|
||
|
||
The easiest and most reliable way to proof your dough is to proof the dough at
|
||
room temperature. It is my method of choice if my schedule allows it. This method
|
||
works great if you make an overnight dough and then proof it the next
|
||
morning.
|
||
|
||
\begin{figure}[htb!]
|
||
\includegraphics[width=\textwidth]{step-13-finger-poke-test}
|
||
\caption[The finger poke test]{The finger poke test is a very reliable
|
||
method to check if your dough has been properly proofed. If the induced
|
||
dent is still visible one minute later, your dough can be baked.}%
|
||
\label{fig:shaping-finger-poke}
|
||
\end{figure}
|
||
|
||
The time it takes to proof your dough can be anything between 30~minutes and
|
||
3~hours. Rather than relying on timing, most bakers use the finger poke test.
|
||
|
||
Flour your thumb and gently press around 0.5cm up to 1cm deep into the dough.
|
||
Try this directly after shaping. You will notice that the created dent will
|
||
recover quickly. It will be gone again after 1 minute.
|
||
|
||
As you proceed with proofing, your dough will fill up with more gas. At the
|
||
same time, the dough will become more extensible. Once it starts to reach the
|
||
right amount of fluffiness and extensibility, the dent will disappear more slowly.
|
||
Once the dough is ready for scoring and baking the dent should still be visible after
|
||
1 minute of waiting.
|
||
|
||
I~recommend performing the finger poke test once every 15~minutes throughout
|
||
the proofing stage. Realistically, based on my experience, proofing takes at least
|
||
one hour and can sometimes take up to 3~hours. Even at warmer temperatures proofing
|
||
has never been faster than an hour for me. As always please take my timings with
|
||
a grain of salt and experiment on your own.
|
||
|
||
Once I~see that the dough is getting close to perfect proofing, I~proceed and
|
||
preheat my oven. This way I~don't overproof the dough. You would notice an
|
||
over-proofed dough when the dough suddenly becomes very sticky. At the same
|
||
time, the dough is likely to collapse during baking and will not spring back.
|
||
Generally, it is better to end proofing too early rather than too late.
|
||
|
||
\subsection{Cold-proofing (retarding)}
|
||
|
||
The second proofing option is to place your dough inside the fridge for
|
||
proofing. This option is great if you do not want to bake the dough
|
||
within the next 3~hours.
|
||
|
||
The dough will initially proof at the same rate as the room temperature dough.
|
||
As the dough cools down the rate of fermentation slows. Ultimately at below
|
||
\qty{4}{\degreeCelsius} (\qty{40}F) the fermentation comes to a halt\footnote{The actual temperature
|
||
depends on the bacteria and yeast you cultivated in your sourdough
|
||
starter.}. The dough can rest in the fridge for up to 24~hours. In some
|
||
experiments, the dough was still good even 48~hours later. Interestingly,
|
||
there is a limit to fridge proofing. I~can only explain this with continuous
|
||
fermentation activity at low temperatures.
|
||
|
||
The hard part is to judge when the dough is finished proofing in your fridge.
|
||
The previously mentioned finger poke test does not work on cold dough. Low
|
||
temperatures change the dough's elasticity. The dent from the poke test
|
||
will never recover.
|
||
|
||
For this reason, finding the best fridge-proofing time is best done
|
||
with an iterative approach. Begin with 8~hours on your first dough,
|
||
10~hours on the second, 12~hours on the third, and so on up to 24~hours.
|
||
As the temperature in your fridge is typically constant, you have an
|
||
environment in which you can rely on timings. Find the ideal proofing
|
||
time that works for you.
|
||
|
||
One additional consideration is the dough's core temperature before
|
||
placing it inside the fridge. The warmer your dough is initially
|
||
the longer it takes for the dough to cool down. This is an additional
|
||
variable to take into consideration when choosing the retarding time.
|
||
In summer times when my kitchen is hot, I~choose a shorter fridge-proofing
|
||
time compared to winter times when the dough is colder.
|
||
|
||
A reliable way to ensure consistent proofing is to opt for using a pH
|
||
meter. By checking the amount of piled-up acidity you can ensure
|
||
each of your doughs has the right amount of acidity. Opt for an iterative
|
||
approach and check the pH for multiple proofing times. Find the pH
|
||
value that creates the best bread for you. Once you have identified
|
||
your perfect pH value you can resort to that number on all following
|
||
doughs. See Table~\ref{table:sample-ph-values} for some sample pH values
|
||
to follow.
|
||
|
||
\section{Scoring}
|
||
|
||
Once your dough is done proofing, it's time to warm up your oven
|
||
to around \qty{230}{\degreeCelsius} (\qty{446}{\degF}). The next step is then
|
||
to proceed with scoring your dough.
|
||
|
||
Scoring is done for two reasons. There is functional and decorative
|
||
scoring. Functional scoring refers to making a small incision in the dough
|
||
through which it rises while baking. If the dough is not scored,
|
||
it would likely crack open at the weakest spots where you sealed
|
||
the dough after shaping. Decorative scoring can be used to apply
|
||
artistic patterns to your dough and make it more appealing. When
|
||
you want to apply artistic scoring, it is best to rub your dough
|
||
with additional rice flour before scoring. The white rice flour
|
||
greatly boosts the contrast of the scoring incisions and thus
|
||
makes the final pattern look more visually appealing.
|
||
|
||
\begin{figure}[htb!]
|
||
\includegraphics[width=\textwidth]{the-ear}
|
||
\caption[Bread's ear]{The ear is a characteristic that can be achieved on
|
||
wheat sourdough when fermenting and scoring your dough with the perfect
|
||
technique. It offers additional flavor and great texture when eating the
|
||
bread.}%
|
||
\label{fig:the-ear}
|
||
\end{figure}
|
||
|
||
When using a banneton, the dough is flipped over and
|
||
placed on an oven rack, tray, stone, steel, or dutch oven. The pros
|
||
and cons of the different baking options are covered in the next chapter.
|
||
The dough's top side which was previously at the bottom of the
|
||
banneton should now be facing you.
|
||
|
||
\begin{figure}[htb!]
|
||
\includegraphics[width=\textwidth]{artistic-scoring}
|
||
\caption[Artistic scoring]{A loaf by Nancy~Anne featuring an artistic
|
||
scoring pattern. The high contrast was achieved by rubbing the dough's
|
||
surface with rice flour before baking. Her Instagram account
|
||
\texttt{simply.beautiful.sourdough} is specialized to showcase beautiful
|
||
artistic scoring patterns.}%
|
||
\label{fig:artistic-scoring}
|
||
\end{figure}
|
||
|
||
The scoring cut is done at a \ang{45}~angle relative to the dough's
|
||
surface slightly off the dough's center. With the \ang{45}~angle cut
|
||
the overlaying side will rise more in the oven than the other side.
|
||
This way you will achieve a so-called \emph{ear} on the final bread.
|
||
The ear is a thin crisp edge that offers intriguing texture
|
||
when eating. The thin edge is typically a bit darker after baking
|
||
and thus offers additional flavor. In my opinion, the ear turns
|
||
a good loaf into a great loaf.
|
||
|
||
\begin{figure}[htb!]
|
||
\includegraphics[width=\textwidth]{bread-scoring-angle}
|
||
\caption[Scoring angle]{The \ang{45}~angle at which you score the
|
||
dough is relative to the surface of the dough. When scoring more towards
|
||
the side, you have to adjust the angle to achieve the ear on your
|
||
bread.}%
|
||
\label{fig:scoring-angle}
|
||
\end{figure}
|
||
|
||
The actual incision is done with a very sharp knife, or better, a razor
|
||
blade. You can use the razor blade directly or attach it to a chopstick.
|
||
The razor blade offers better flexibility than the sharp knife.
|
||
Regardless, the blade should be as sharp as possible. This way when cutting,
|
||
the dough is not torn and instead features a clean, non ragged incision.
|
||
|
||
To simplify scoring, your dough's surface must be dried out a little bit.
|
||
This way it is a lot easier to make the incision.
|
||
For this reason, it is crucial to rub your dough with a bit of flour
|
||
before placing it in the banneton. The dry flour will absorb some of the
|
||
moisture of the outer layers of your dough. This is especially important
|
||
when working with room temperature-proofed doughs. A cold-proofed dough
|
||
is a lot easier to score due to the dough's low viscosity. The room-temperature
|
||
dough is a lot harder to score. The scoring incision tears a lot
|
||
easier. With a ragged incision, the dough is not as likely to properly
|
||
rise in the oven. Chances are you will not achieve the previously mentioned
|
||
ear. For this reason, drying out the surface is especially important. Scoring
|
||
will become a lot easier.
|
||
|
||
\begin{figure}[htb!]
|
||
\includegraphics[width=\textwidth]{dry-dough-surface}
|
||
\caption[Drying the dough surface]{By applying flour to your dough's surface
|
||
after shaping, the outer part of the dough dries out a little bit. This
|
||
makes scoring a lot easier as the incision is less likely to tear.}%
|
||
\label{fig:dried-out-dough-scoring}
|
||
\end{figure}
|
||
|
||
|
||
Scoring requires a lot of practice. For this reason, I~recommend
|
||
practicing making the incision after creating dough strength. The dough
|
||
is going to be very wet and sticky. You can use a sharp knife or razor
|
||
blade to practice the technique. Wait a few minutes and then round
|
||
up the dough again. You can practice this for as long as you like
|
||
until you are happy with your technique. After proofing, you only
|
||
have a single chance to practice scoring. It's either hit or miss.
|
||
|
||
An additional trick that can help you to combine the benefits
|
||
of room temperature-proofing and easy cold-proofing scoring
|
||
is to place your dough in the freezer for 30~minutes before baking.
|
||
Once you notice your dough is almost done proofing, move it to the
|
||
freezer. The freezer will dry out the dough's surface even further
|
||
while also lowering its viscosity, making scoring easier.
|
||
|
||
Another interesting trick is to bake your dough for 30 seconds without steam.
|
||
The hot air will dry out the dough's surface even further and simplify
|
||
the scoring technique. Experiment with the timing to identify your personal
|
||
sweet spot.
|