mirror of
https://github.com/hendricius/the-sourdough-framework
synced 2025-11-10 05:01:11 -06:00
Some checks are pending
This was written way before the mix-ins sections, made sense back then but not so much anymore so moving it out. formatted text and added centering, labels etc while at it.
766 lines
36 KiB
TeX
766 lines
36 KiB
TeX
\begin{quoting}
|
|
You could consider this chapter as an FAQ about most problems faced by
|
|
bakers, it should give you the debugging tools you need to analyze the
|
|
situation. You can then apply the appropriate measures and squash each
|
|
\emph{bug} one by one until you reach the perfect loaf.
|
|
\end{quoting}
|
|
|
|
\section{Starter}
|
|
\subsection{My starter does not double in size}
|
|
|
|
Some bakers call for the sourdough starter to
|
|
double in size before using it.
|
|
The idea is to use the sourdough starter at
|
|
peak performance to ensure a
|
|
balanced fermentation in the main dough.
|
|
|
|
The doubling in size metric should be
|
|
taken with a grain of salt when judging
|
|
your starter. Depending on the flour
|
|
you use to feed the starter, different levels
|
|
of its rising can be expected.
|
|
For instance, if you use rye flour then only
|
|
very little gas from the
|
|
fermentation can be retained inside the
|
|
starter. In consequence, your
|
|
sourdough starter will not rise as much. It
|
|
could still be in healthy shape. If you use wheat flour with less gluten,
|
|
the starter will not rise as
|
|
much either. The reason is that you have a weaker
|
|
gluten network resulting in
|
|
more gas dispersing out of your dough.
|
|
|
|
That being said, it is recommended that you develop
|
|
your volume increase
|
|
metric. Your starter will increase in size and then
|
|
ultimately lose structure
|
|
and collapse. Observe the point before it collapses.
|
|
This is the point when
|
|
you should use your starter. This could be a
|
|
\qty{50}{\percent} volume increase, 100
|
|
percent or \qty{200}{\percent}. It is always better to use
|
|
the starter a little bit
|
|
too early rather than too late. If you use the
|
|
starter later, reduce the
|
|
quantity that you use. If the recipe calls for a 20
|
|
percent starter quantity,
|
|
use only 10
|
|
percent starter in that case. Your starter will
|
|
regrow in your main dough.
|
|
|
|
On top of relying on the size increase, start
|
|
taking note of your starter's
|
|
smell. Over time you will be able to judge its
|
|
fermentation state based on the
|
|
smell. The stronger the smell becomes, the further
|
|
your dough has fermented.
|
|
This is a sign that you should use less starter
|
|
when making the actual dough.
|
|
|
|
Please refer to
|
|
Section~\ref{section:readying-starter}~``\nameref{section:readying-starter}''
|
|
for more information on the topic.
|
|
|
|
|
|
|
|
\subsection{What's the best starter feeding ratio?}
|
|
|
|
The best starter feeding ratio is commonly either 1:5:5 or 1:10:10.
|
|
In the case of 1:5:5 that's 1 part old starter,
|
|
5 parts flour and 5 parts water. If you are using a stiff starter,
|
|
use half the amount of water. So that's 1:5:2.5. Depending on when
|
|
you last fed your starter 1:10:10 might make more sense. If the starter
|
|
is old and hasn't been fed recently the 1:10:10 ratio is a better choice.
|
|
By reducing the starter inoculation ratio, you provide the microorganisms
|
|
with a cleaner environment. This way they can reproduce and regrow
|
|
into a more desirable balance to begin your dough fermentation.
|
|
|
|
Generally, think of your sourdough starter as a dough. Use the same
|
|
ratios you use for your bread dough for your starter. Your starter
|
|
should be trained in the same environment that you later use
|
|
for your dough. This way your starter is perfectly suited to
|
|
ferment the dough into which it is later inoculated.
|
|
|
|
The only exception to the 1:5:5 and 1:10:10 rule is the initial
|
|
starter set-up stage. For the first days during the starter-making
|
|
process there aren't enough microbes yet. So using a 1:1:1 ratio
|
|
can speed up the process.
|
|
\subsection{What's the benefit of using a stiff sourdough starter?}
|
|
|
|
A regular sourdough starter has equal parts of
|
|
flour and water (\qty{100}{\percent} hydration). A stiffer
|
|
sourdough starter features a hydration level of 50 to \qty{60}{\percent}.
|
|
|
|
The stiff sourdough starter boosts the yeast part
|
|
of your starter more. This way your gluten degrades
|
|
slower and you can ferment for a longer period. This
|
|
is especially handy when baking with lower gluten flours.
|
|
|
|
You can read more about the topic of stiff sourdough
|
|
starters in Section~\ref{section:stiff-starter}.
|
|
|
|
\subsection{What's the benefit of using a liquid sourdough starter?}
|
|
|
|
The liquid starter will boost anaerobic bacterial
|
|
fermentation in your starter. This way your starter
|
|
tends to produce more lactic acid rather than acetic
|
|
acid. Lactic acid is perceived as milder and more
|
|
yogurty. Acetic acid can sometimes taste quite
|
|
pungent. Acetic acid can be perfect when making
|
|
dark rye bread but not so much when making a fluffy
|
|
ciabatta-style loaf.
|
|
|
|
When converting your starter to a liquid starter you are
|
|
permanently altering the microbiome of your starter.
|
|
You cannot go back once you have eliminated acetic
|
|
acid-producing bacteria. So it is recommended to keep
|
|
a backup of your original starter.
|
|
|
|
A downside to the liquid starter is the overall
|
|
enhanced bacterial activity compared to yeast activity. This means the baked bread
|
|
will have more acidity (but milder). The dough will degrade
|
|
faster during fermentation. For this reason, you
|
|
will need to use strong high-gluten flour when using
|
|
this type of starter.
|
|
|
|
You can read more about the liquid starter in
|
|
Section~\ref{section:liquid-starter}
|
|
|
|
\subsection{My new starter doesn't rise at all}
|
|
|
|
Make sure that you use unchlorinated water.
|
|
In many areas of the world, tap water has
|
|
chlorine added to kill microorganisms. If that's
|
|
the case in your region, bottled spring water will
|
|
help.
|
|
You can also use a water filter with activated charcoal
|
|
which will remove the chlorine.
|
|
Alternatively, if you draw tap water into a pitcher or other
|
|
container and let it sit, loosely covered, the chlorine
|
|
should dissipate within 12--24~hours, and you have
|
|
the added advantage of automatically having
|
|
room-temperature water.
|
|
|
|
Make sure to use whole grain flour (whole-wheat, whole-rye, \etc{}).
|
|
These flours have more natural wild yeast and
|
|
bacterial contamination. Making a starter
|
|
from just white flour sometimes doesn't work.
|
|
Try to use organic unbleached flour to make
|
|
the starter. Industrial flour can sometimes
|
|
be treated with fungicides.
|
|
|
|
\subsection{I~made a starter, it rose on day 3 and now not anymore}
|
|
|
|
This is normal. As your starter is maturing, different
|
|
microorganisms are activated. Especially during
|
|
the first days of the process, bad microbes
|
|
like mold can be activated. These cause your
|
|
starter to rise a lot. With each subsequent
|
|
starter-feeding, you select the microbes that are best
|
|
at fermenting flour. For this reason, it is
|
|
recommended to discard the leftover unused starter
|
|
from the first days of the process. Later on, unneeded
|
|
starter amounts should never be thrown away. You can make
|
|
great discard bread out of it.
|
|
|
|
So just keep going and don't give up. The first big
|
|
rise is an indicator that you are doing everything
|
|
right. Based on my experience, it takes around 7
|
|
days to grow a starter. As you feed your starter
|
|
more and more, it will become even better at fermenting
|
|
flour. The first bread might not go exactly as you
|
|
planned, but you will get there eventually. Each
|
|
feeding makes your starter stronger and stronger.
|
|
|
|
\subsection{Liquid on top of my starter}
|
|
|
|
Sometimes a liquid, in many cases black liquid, gathers on top
|
|
of your sourdough starter. The liquid might have a pungent
|
|
smell to it. Many people confuse this with mold. I~have seen
|
|
bakers recommending to discard the starter because of this liquid.
|
|
The liquid is commonly known as \emph{hooch}. After a while
|
|
of no activity the heavier flour separates from the water. The flour
|
|
will sit at the bottom of your jar and the liquid will stay on top.
|
|
The liquid turns darker because some particles of the flour weigh
|
|
less than the water and float on top. Furthermore dead microorganisms
|
|
float in this liquid. This liquid is not a bad thing; it's actively
|
|
protecting your sourdough starter from aerobic mold entering through
|
|
the top.
|
|
|
|
\begin{figure}[!htb]
|
|
\centering
|
|
\includegraphics[width=0.5\textwidth]{sourdough-starter-hooch}
|
|
\caption[Hooch] {Hooch building on top of a sourdough
|
|
starter~\cite{liquid+on+starter}.}%
|
|
\label{fig:hooch}
|
|
\end{figure}
|
|
|
|
Simply stir your sourdough starter to homogenize the hooch back
|
|
into your starter. The hooch will disappear. Then use a little bit of
|
|
your sourdough starter to set up the starter for your next bread.
|
|
Once hooch appears, your starter has likely fermented for a long
|
|
period of time. It might be very sour. This state of starter
|
|
is excellent to make discard crackers or a discard bread. Don't throw
|
|
anything away. Your hooch is a sign that you have a long fermented
|
|
dough in front of you. Compare it to a 2 year ripened Parmigiano cheese.
|
|
The dough in front of you is full of delicious flavor.
|
|
|
|
\subsection{Fixing a moldy sourdough starter}
|
|
|
|
First of all, making a moldy sourdough starter is very difficult.
|
|
It's an indicator that something might be completely off in your starter.
|
|
Normally the symbiosis of yeast and bacteria does not allow external
|
|
pathogens such as mold to enter your sourdough starter.
|
|
The low pH created by the bacteria is a very hostile environment
|
|
that no other pathogens like. Generally everything below a pH
|
|
of 4.2 can be considered food safe~\cite{food+safe+ph}. This
|
|
is the concept of pickled foods. And your sourdough bread
|
|
is essentially pickled bread.
|
|
|
|
I~have seen this happening especially when the sourdough
|
|
starter is relatively young. Each flour naturally contains
|
|
mold spores. When beginning a sourdough starter, all
|
|
the microorganisms start to compete by metabolizing the
|
|
flour. Mold can sometimes win the race and outcompete
|
|
the natural wild yeast and bacteria. In that case simply
|
|
try cultivating your sourdough starter again. If mold reappears
|
|
again, it might be a very moldy batch of flour. Try a different
|
|
flour to begin your sourdough starter with.
|
|
|
|
Mature sourdough starters should not go moldy unless the conditions
|
|
of the starter change. I~have seen mold appearing when the starter is stored
|
|
in the fridge and the surface dried out. It also sometimes forms on the
|
|
edges of your starter's container, typically in areas where no active
|
|
starter microorganisms can reach. Simply try to extract an
|
|
area of your starter that has no mold. Feed it again with flour and
|
|
water. After a few feedings, your starter should be back to normal.
|
|
Take only a tiny bit of starter: \qtyrange{1}{2}{\gram} are enough. They already
|
|
contain millions of microorganisms.
|
|
|
|
Mold favors aerobic conditions. This means that air is required in order
|
|
for the mold fungus to grow. Another technique that has worked for me
|
|
was to convert my sourdough starter into a liquid starter. This successfully
|
|
shifted my starter from acetic acid production to lactic acid production.
|
|
Acetic acid, similarly to mold, requires oxygen to be produced. After
|
|
submerging the flour with water, over time the lactic acid bacteria
|
|
outcompeted the acetic acid bacteria. This is a similar concept to pickled
|
|
foods. By doing this you are essentially killing all live mold fungi. You
|
|
might only have some spores left. With each feeding the spores will become
|
|
fewer and fewer. Furthermore, it seems that lactic acid bacteria produce
|
|
metabolites that inhibit mold growth~\cite{mold+lactic+acid+bacteria}.
|
|
|
|
\begin{figure}[!htb]
|
|
\includegraphics[width=\textwidth]{fungi-lactic-acid-interactions}
|
|
\caption[The interaction of lactic acid bacteria and mold fungi]{The
|
|
interaction of lactic acid bacteria and mold fungi.
|
|
In~\cite{mold+lactic+acid+bacteria},
|
|
\citeauthor{mold+lactic+acid+bacteria} et al.\ show how bacteria are
|
|
producing metabolites that inhibit fungus growth.}%
|
|
\label{fig:fungi-lactic-acid-interactions}
|
|
\end{figure}
|
|
|
|
To pickle your starter, simply take a bit of your existing starter (\qty{5}{\gram} for
|
|
instance). Then feed the mixture with \qty{20}{\gram} of flour and \qty{100}{\gram} of water. You have
|
|
created a starter with a hydration of around \qty{500}{\percent}. Shake the mixture vigorously.
|
|
After a few hours you should start seeing most of the flour near the bottom
|
|
of your container. After a while most of the oxygen from the bottom mixture
|
|
is depleted and anaerobic lactic acid bacteria will start to thrive. Take a
|
|
note of the smell your sourdough starter. If it was previously acetic
|
|
it will now change to be a lot more dairy. Extract a bit of your mixture the
|
|
next day by shaking everything first. Take \qty{5}{\gram} of the previous mixture, feed
|
|
again with another \qty{20}{\gram} of flour and another \qty{100}{\gram} of water. After 2--3
|
|
additional feedings your starter should have adapted. When switching back
|
|
to a hydration of \qty{100}{\percent} the mold should have been eliminated. Please note that
|
|
more tests should be conducted on this topic. It would be nice to really
|
|
carefully analyze the microorganisms before the pickling and after.
|
|
|
|
\subsection{My sourdough starter is too sour}
|
|
|
|
If your sourdough starter is too sour it will cause problems during
|
|
the fermentation. Your fermentation will have more
|
|
bacterial activity than yeast activity. This means
|
|
you will likely create a more tangy loaf which isn't
|
|
as fluffy as it could be. The goal is to reach the right
|
|
balance: Fluffy consistency from the yeast and a great,
|
|
not-too-strong tang from the bacteria. This depends
|
|
of course on what you are looking for in terms of taste
|
|
in your bread. When making rye bread, I~prefer to be more
|
|
on the tangy side for instance. When the described balance
|
|
is off, the first thing to check is your sourdough starter.
|
|
|
|
Note the smell of your starter. Does it smell very sour?
|
|
Taste a bit of your starter too. How sour does it taste?
|
|
Over time, every starter becomes more and more sour the longer
|
|
you wait. But sometimes your starter becomes sour too fast.
|
|
In this case apply daily feedings to your starter. Reduce
|
|
the amount of old starter that you use to feed. A ratio
|
|
of 1:5:5 or 1:10:10 can do wonders. In this case you would
|
|
take 1 part of starter (\qty{10}{\gram}) and feed it with \qty{50}{\gram} of flour
|
|
and \qty{50}{\gram} of water. This way the microorganisms start
|
|
the fermentation in a greenfield environment. This is
|
|
similar to the \qty{10}{\percent} starter or \qty{20}{\percent} starter
|
|
ratio that you use to make a dough. These days I~almost
|
|
never use a 1:1:1 ratio. This only makes sense when you
|
|
are initially creating your starter. You want a sour
|
|
environment so that your microorganisms outcompete
|
|
potential pathogens. The acidic environment is toxic
|
|
to most pathogens that you do not want in your starter.
|
|
|
|
Another approach that can help is to convert your
|
|
sourdough starter into a stiff starter as
|
|
described in Section~\ref{section:stiff-starter}.
|
|
|
|
\subsection{Why does my starter smell like vinegar or acetone?}
|
|
|
|
Your sourdough starter has likely produced a lot of acetic acid.
|
|
Acetic acid is essential when creating vinegar. Once no additional
|
|
food is left some of your starter's bacteria will consume ethanol
|
|
and convert it into acetic acid. Acetic acid has a very pungent smell.
|
|
When tasting acetic acid, the flavor of your bread is often perceived
|
|
as quite strong.
|
|
|
|
\begin{figure}[!htb]
|
|
\centering
|
|
\input{figures/fig-ethanol-oxidation.tex}
|
|
\caption[Acetic acid creation]{Oxygen is required to create acetic
|
|
acid~\cite{acetic+acid+production}.}%
|
|
\label{fig:ethanol-oxidation}
|
|
\end{figure}
|
|
|
|
This is nothing bad. But if you would like to change
|
|
the flavor of your final bread, consider converting
|
|
your sourdough starter into a liquid starter. This will
|
|
help to prioritize lactic acid-producing bacteria.
|
|
Your flavor will change to dairy compared to vinegary.
|
|
You can't go back though. After the conversion your starter
|
|
will never go back to acetic acid production because you have
|
|
changed the tides towards primarily lactic acid fermentation.
|
|
I~like to have a separate rye starter. In my experiments
|
|
rye starters tend to feature many acetic acid bacteria.
|
|
This starter is excellent when you want to make a very hearty,
|
|
strong-tasting bread. A pure rye bread tastes excellent when
|
|
made with such a starter. The flavor when taking a bite
|
|
is incredible. It nicely plays with soups as well. Just take
|
|
a bit of this bread and dip it in your soup.
|
|
|
|
\subsection{Why does my starter not float after using the float test?}
|
|
The float test may not reliably determine your starter's readiness for dough
|
|
inoculation. While it's effective for wheat-based doughs, where ample gas gets
|
|
trapped in the gluten matrix, it's less reliable for non-wheat doughs. In non-
|
|
wheat doughs, the gas generated during fermentation tends to escape, causing
|
|
the starter to likely sink.
|
|
|
|
For more accurate assessments of your starter's readiness, watch for bubbles
|
|
at the container's edge and consider its aroma. A mature starter should emit a
|
|
mildly sour scent without being overly pungent.
|
|
|
|
\section{Dough}
|
|
\subsection{Should I~autolyse my dough?}
|
|
|
|
In \qty{95}{\percent} of all cases, an autolysis
|
|
makes no sense. Instead I~recommend
|
|
that you conduct a fermentolysis. You
|
|
can read more about the autolysis process in
|
|
Section~\ref{section:autolysis} and
|
|
more about the topic of fermentolysis
|
|
in Section~\ref{section:fermentolysis}.
|
|
|
|
The fermentolysis combines all the benefits
|
|
of the autolysis while eliminating disadvantages
|
|
such as having to knead the dough multiple times.
|
|
|
|
The autolysis only makes sense when you might
|
|
bake a fast-fermenting yeast-based dough with a
|
|
high yeast inoculation rate. But even in that
|
|
case you could just lower the amount of yeast
|
|
to fermentolyse rather than autolyse.
|
|
|
|
\subsection{My dough sample (aliquot) doesn't rise. What's wrong?}
|
|
|
|
If you see that your dough rises in size but your aliquot doesn't, chances
|
|
are that both are fermenting at different speeds. This can often
|
|
happen when the temperature in your kitchen changes. The aliquot
|
|
is more susceptible to temperature changes than the main dough.
|
|
Because the sample is smaller in size, it will heat up or cool down
|
|
faster.
|
|
|
|
For this reason, you must use room-temperature water when
|
|
making your dough. By having the same temperature in both the sample
|
|
and your dough, you make sure that both ferment at the same rate.
|
|
|
|
If the temperature in your room changes significantly during the day, your
|
|
best option is to use a see-through container. Mark the container to properly
|
|
measure your dough's size increase.
|
|
|
|
Another option could be to use a more expensive pH meter to measure your
|
|
dough's acidity buildup. You can read more about different ways of managing
|
|
bulk fermentation in Section~\ref{section:bulk-fermentation}.
|
|
|
|
\subsection{What's a good level of water (hydration) to make a dough?}
|
|
|
|
Especially when starting to make bread, use lower amounts of water. This will
|
|
greatly simplify the whole process. I~recommend using a level of around 60
|
|
percent hydration. So for every \qty{100}{\gram} of flour use around \qty{60}{\gram} of water.
|
|
This ballpark figure will work for most flours. With this hydration, you can
|
|
make bread, buns, pizzas, and even baguettes out of the same dough.
|
|
|
|
With the lower hydration, dough handling becomes easier and you have more yeast
|
|
fermentation, resulting in lower over-fermentation risk.
|
|
|
|
\subsection{My dough completely tears after a long fermentation}
|
|
|
|
Sometimes when touching your dough after a long fermentation
|
|
it completely tears apart. This could be for two reasons. It might
|
|
be that the bacteria completely consumed the gluten of your flour.
|
|
On the other hand, over time your gluten network automatically
|
|
degrades. This is the protease enzyme converting the gluten
|
|
network into smaller amino acids the seedling can use as
|
|
building blocks for its growth. This process starts to happen
|
|
the moment you mix flour and water. The longer your dough sits,
|
|
the more gluten is broken down. As the gluten holds the
|
|
wheat dough together, your dough will ultimately tear.
|
|
|
|
\begin{figure}[!htb]
|
|
\includegraphics[width=1.0\textwidth]{tearing-dough}
|
|
\caption[Dough tearing]{My dough tearing after 24~hours of no activity.}%
|
|
\label{fig:tearing-dough}
|
|
\end{figure}
|
|
|
|
In the picture~\ref{fig:tearing-dough} I~experimented with
|
|
using a starter that has not been fed for 30 days at room temperature.
|
|
I~tried to make a dough directly out of the unfed starter.
|
|
Typically after a long period
|
|
without feedings your microbes start to sporulate and go
|
|
into hibernation mode. This way they can survive for a long
|
|
period of time without extra feedings. Adding additional food
|
|
will activate them again. In this case the dough did not ferment
|
|
fast enough before the protease broke down the gluten. By activating
|
|
your microbes they will start to reproduce and increase in quantity
|
|
for as long as there is food available. But this process
|
|
in my case was not fast enough. After around 24~hours, the whole
|
|
dough just started to completely tear apart. The whole process was further
|
|
accelerated by my using whole-wheat flour. Whole-wheat
|
|
contains more enzymes than white flour.
|
|
|
|
To fix this, try to make sure that your sourdough starter is lively
|
|
and active. Simply apply a couple more feedings before
|
|
making your dough. This way your dough becomes ready to shape
|
|
before it has completely broken down.
|
|
|
|
\section{Bread}
|
|
\subsection{My bread stays flat}
|
|
|
|
A flat bread is in most cases related to your gluten
|
|
network breaking down fully. This is not bad; this
|
|
means you are eating a fully fermented food. However,
|
|
from a taste and consistency perspective, it might be
|
|
that your bread tastes too sour, or is not fluffy anymore.
|
|
Please also note that you can only make bread with
|
|
great oven spring when making wheat based doughs. When
|
|
starting with this hobby I~always wondered why my rye
|
|
breads would turn out so flat. Yes, rye has gluten, but
|
|
small particles called \emph{pentosans} (arabinoxylan and beta-glucan)~\cite{rye-defects}.
|
|
They prevent the dough from developing a gluten network it can
|
|
with wheat. Your efforts will be in vain, and your dough will
|
|
stay flat. Only spelt- and wheat-based doughs have the capability
|
|
of retaining the \ch{CO2} created by the fermentation.
|
|
|
|
In most cases something is probably off with your
|
|
sourdough starter. This very often happens when the starter
|
|
is still relatively young and isn't as capable of
|
|
fermenting flour. Over time your sourdough
|
|
starter is going to become better and better.
|
|
Keep your sourdough starter at room temperature
|
|
and then apply daily feedings with a 1:5:5 ratio.
|
|
This would be 1 part old starter, 5 parts flour,
|
|
5 parts water. This allows you to achieve a better
|
|
balance of yeast and bacteria in your sourdough.
|
|
Even better could be the use of a stiff sourdough
|
|
starter. The stiff sourdough starter boosts
|
|
the yeast part of your starter. This allows you
|
|
to have less bacterial fermentation, resulting
|
|
in a stronger gluten network toward the end
|
|
of the fermentation~\cite{stiff+starter}. Please
|
|
also refer to the Subsection~\ref{subsec:overfermented-dough} where
|
|
I~explained more about overfermented doughs. You can also
|
|
refer to Section~\ref{section:stiff-starter} with more details on
|
|
making a stiff sourdough starter.
|
|
|
|
Furthermore, a stronger flour containing more gluten
|
|
will help you to push the fermentation further. This
|
|
is because your flour contains more gluten and will
|
|
take longer to be broken down by your bacteria. Ultimately,
|
|
if fermented for too long, your dough is also going
|
|
to be broken down and will become sticky and flat.
|
|
|
|
To debug whether the excess bacterial fermentation is the issue,
|
|
simply taste your dough. Does it taste very sour? If yes,
|
|
that's a good indicator. When working the dough, does it
|
|
suddenly become very sticky after a few hours? That's a
|
|
another good indicator. Please also use your nose to note
|
|
the smell of the dough. It shouldn't be too pungent.
|
|
|
|
\subsection{I~want more tang in my bread}
|
|
|
|
To achieve more tang in your sourdough bread, you have
|
|
to ferment your dough for a longer period of time.
|
|
Over time the bacteria will metabolize most of the
|
|
ethanol created by the yeast in your dough. The bacteria
|
|
mostly produce lactic and acetic acid. Lactic acid
|
|
is chemically more acidic than acetic acid but sometimes
|
|
not perceived as sour. In most cases a longer fermentation
|
|
is what you want. You will either need to utilize a loaf
|
|
pan to make your dough or use a flour that can withstand
|
|
a long fermentation period. A flour like this is typically
|
|
called a \emph{strong flour}. Stronger flours tend
|
|
to be from wheat varieties that have be grown in more
|
|
sunny conditions. Because of that, stronger flours tend
|
|
to be more expensive. For freestanding loaves, I~recommend
|
|
using a flour that contains at least \qty{12}{\percent} protein.
|
|
Generally, the more protein, the longer you can ferment your dough.
|
|
|
|
Another option to achieve a more sour flavor could be to
|
|
use a starter that produces more acetic acid. Based on my own
|
|
experience, most of my pure rye starters produced stronger acetic
|
|
notes. Chemically, the acetic acid isn't as sour, but when tasting
|
|
it will seem more sour. Make sure to use a starter that is at
|
|
a hydration of around \qty{100}{\percent}. Acetic acid production
|
|
requires oxygen. A starter that is too liquid tends to favor lactic
|
|
acid production because the flour is submerged in water. By submerging
|
|
the dough very little oxygen can pass through the water to the fermenting flour.
|
|
Because of this, only very little acetic acid can be produced. Over
|
|
time the acetic acid-producing bacteria will perish from your starter.
|
|
|
|
\begin{figure}[!htb]
|
|
\includegraphics[width=\textwidth]{parbaked-bread.jpg}
|
|
\caption[Half-baked bread]{A half-baked bread, known as \emph{parbaked}.}%
|
|
\label{fig:parbaked-bread}
|
|
\end{figure}
|
|
|
|
Another easier option could be to bake your sourdough
|
|
twice. I~have observed this when shipping bread for my micro
|
|
bakery. The idea was to bake my bread for around 30~minutes
|
|
until it's sterilized, let it cool down and then ship it
|
|
to customers. Once you receive it, you just bake it again
|
|
for another 20--30~minutes to achieve the desired crust and
|
|
then you can eat it. Some of the customers reported a very sour
|
|
tasting bread. After investigating a bit more, it became
|
|
crystal clear. By baking the bread twice you don't boil off
|
|
as much acid during the baking process. Water
|
|
evaporates at around \qty{100}{\degreeCelsius} (\qty{212}{\degF}) while acetic
|
|
acid boils at \qty{118}{\degreeCelsius} (\qty{244}{\degF}) and lactic acid at
|
|
\qty{122}{\degreeCelsius} (\qty{252}{\degF}). After baking for 30~minutes at
|
|
around \qty{230}{\degreeCelsius} (\qty{446}{\degF}) some of the water has
|
|
started to evaporate, but not all the acid yet. If you were to continue to
|
|
bake, more and more of the acid would start to evaporate. Now if you were to
|
|
stop baking after 30~minutes, you would typically have reached a core
|
|
temperature of around \qty{95}{\degreeCelsius} (\qty{203}{\degF}). Your dough
|
|
would need
|
|
to be cooled down again to room temperature. The crust would
|
|
still be quite pale. Then a couple of hours later, you start
|
|
to bake your dough again. Your crust would become nice and
|
|
dark featuring delicious aroma. The aroma is coming from the
|
|
Maillard reaction. However, the core of your dough still won't
|
|
exceed the \qty{118}{\degreeCelsius} required to boil the acid. Overall, your
|
|
bread will be more sour. The enhanced acidity also helps
|
|
to prevent pathogens from entering your bread. The bread
|
|
will be good for a longer period of time. That's why
|
|
the concept of a delivery bakery works well with tangy sourdough bread.
|
|
In my own experiments, the bread stayed good for up to a week
|
|
in a plastic bag. This is much longer than a yeast-based dough that might
|
|
mold after just a few days\footnote{Some of my first test customers however
|
|
reported that the bread was overly sour and not pleasant to eat at all.
|
|
When this happens to you, consider toasting the bread. Toasting
|
|
will boil off additional acidity.}.
|
|
|
|
\subsection{My bread is too sour}
|
|
|
|
Some people like the bread less sour as well. This
|
|
is personal preference. To achieve a less sour bread
|
|
you need to ferment for a shorter period of time.
|
|
The yeast produces \ch{CO2} and ethanol. Both yeast and
|
|
bacteria consume the sugars released by the amylase enzyme
|
|
in your dough. When the sugar is depleted, bacteria starts to
|
|
consume the leftover ethanol by the yeast. Over time more
|
|
and more acidity is created, making a more sour loaf.
|
|
|
|
Another angle at this would be to change the yeast/bacteria
|
|
ratio of your sourdough. You can start the fermentation with
|
|
more yeast and less bacteria. This way, for the same given
|
|
volume increase of your dough, you will have less acidity.
|
|
A really good trick is to make sure that you feed your starter
|
|
once per day at room temperature. This way you shift
|
|
the tides of your starter towards a better yeast fermentation~\cite*{more+active+starter}.
|
|
|
|
To shift the tides even further, a real game changer
|
|
for me has been to create a stiff sourdough starter. The
|
|
stiff sourdough starter is at a hydration of around \qty{50}{\percent}.
|
|
By doing so your sourdough starter will favor yeast
|
|
activity a lot more. Your doughs will be more fluffy and less
|
|
sour for a given volume increase. I~tested this
|
|
by putting balloons over different glass jars. I~used
|
|
the same amount of flour for each of the samples.
|
|
I~tested a regular starter, a liquid starter and a stiff
|
|
starter. The stiff starter by far created the most \ch{CO2}
|
|
compared to the other starters. As a consequence, the stiff
|
|
starter balloon was inflated the most~\cite{stiff+starter}. You can read more
|
|
about the topic of stiff starters in Section~\ref{section:stiff-starter}.
|
|
|
|
Another unconventional approach could be to add baking
|
|
powder to your dough. The baking powder neutralizes the
|
|
lactic acid and will make a much milder
|
|
dough~\cite{baking+powder+reduce-acidity}.
|
|
|
|
\subsection{My bread flattens out when removing it from the banneton}
|
|
|
|
After removing your dough from the banneton, your dough will always
|
|
flatten out a bit. That's because over time your gluten network
|
|
relaxes and can no longer hold the shape. However, during the course
|
|
of baking, your dough is going to increase in size and inflate again.
|
|
|
|
If your dough however flattens out completely, it's a sign that
|
|
you have fermented your dough for too long. Please refer to
|
|
Subsection~\ref{subsec:overfermented-dough}
|
|
where I~explain about overfermented doughs. Your bacteria
|
|
has consumed most of your gluten network. That's why your
|
|
dough fully collapses and stays flat during the bake. The
|
|
\ch{CO2} and evaporating water will diffuse out of the dough.
|
|
A related symptom is that your dough sticks to the banneton.
|
|
When I~starting baking I~combated this with rice flour.
|
|
It worked for me but it might be a false find. Please refer to
|
|
Subsection~\ref{subsec:overfermented-dough} for more details on why
|
|
rice flour is not a good idea to manage sticky doughs.
|
|
|
|
These days I~gently rub my
|
|
dough with a bit of non-rice flour before placing it in
|
|
the banneton. Now if the dough starts to stick to the banneton
|
|
while I~remove it I~resort to a drastic measure. I~immediately
|
|
grease a loaf pan and directly place the dough inside. The loaf
|
|
pan provides a barrier and the dough can't flatten out as much.
|
|
The dough won't be as fluffy but it will be super delicious if you love tangy bread.
|
|
|
|
If you own a pH meter, take a note of your dough's pH before baking.
|
|
This will allow you to better judge your dough throughout
|
|
the fermentation process.
|
|
|
|
\subsection{My bread flattens out during shaping}
|
|
|
|
Similarly to a dough flattening out after removing it from the banneton,
|
|
a flattened dough after shaping is also a possible sign of over-fermentation.
|
|
|
|
When you try to shape the dough, can you easily tear pieces from the dough?
|
|
If yes, you have definitely overfermented your dough. If not, it might just
|
|
be a sign that you have not created enough dough strength for your dough.
|
|
A ciabatta, for instance, is a dough that tends to flatten out a bit after shaping.
|
|
|
|
If your dough is not able to be shaped at all, use a greased loaf pan
|
|
to rescue your dough. You can also cut a piece of the dough and use it
|
|
as the starter for your next dough. Your sourdough dough is essentially
|
|
just a gigantic starter.
|
|
|
|
\subsection{My crust becomes chewy}
|
|
|
|
Depending on which style of bread you are making a
|
|
thick crackly crust is sometimes desired. The crust
|
|
of your bread is created during the 2nd stage of the
|
|
baking process once the steaming source of your
|
|
oven has been removed. The dark colors are created by
|
|
the process known as \emph{Maillard reaction} and then followed
|
|
by another process known as \emph{caramelization}. Each
|
|
color of crust offers the taster a different aroma.
|
|
|
|
What happens quite often is that the crust becomes chewy after a day.
|
|
Sometimes when baking in the tropics with high humidity, the
|
|
crust only stays in this stage for a few hours. Afterwards
|
|
the crust becomes chewy. It's no longer as crisp compared
|
|
to the moment after baking. Your dough still contains moisture.
|
|
This moisture will start to homogenize in the final bread and
|
|
partially evaporate. The result is that your crust becomes chewy.
|
|
|
|
Similarly when storing your bread in a container or in a plastic
|
|
bag, your crust is going to become chewy. I~have no fix for this yet.
|
|
I~typically tend to store my breads in a plastic bag inside of my fridge.
|
|
This allows the moisture to stay inside of bread. When taking a slice
|
|
I~always toast each slice. This way some of the crispness returns.
|
|
If you know of a great way, please reach out and I~will update
|
|
this book with your findings.
|
|
|
|
\input{troubleshooting/crumb-structures}
|
|
|
|
\section{Misc}
|
|
\subsection{Baking in the tropics}
|
|
|
|
Depending on the temperature, your fermentation speed adapts.
|
|
In a warmer environment, everything is faster. In a colder
|
|
environment, everything is slower.
|
|
|
|
This includes the speed at which your sourdough ferments
|
|
the dough but also the speed of enzymatic reactions. The
|
|
amylase and protease enzymes work faster, making more
|
|
sugars available and degrading the gluten proteins.
|
|
|
|
At around \qty{22}{\degreeCelsius} (\qty{72}{\degF}) in my kitchen my bulk fermentation is ready
|
|
after around 10~hours. I~use around \qty{20}{\percent} of sourdough
|
|
starter based on the flour. In summertime the temperatures
|
|
in my kitchen sometimes increase to
|
|
\qty{25}{\degreeCelsius} (\qty{77}{\degF}). In that case
|
|
I~reduce the sourdough starter to around \qty{10}{\percent}.
|
|
|
|
If I~didn't do that, my fermentation would be done after
|
|
around 4--7~hours. The problem is that the dough is quite
|
|
unstable when fermenting at this high speed. This means
|
|
that you easily run into issues of over-fermentation.
|
|
Finding the perfect sweet spot between fermenting enough
|
|
and not too much becomes much harder. Normally you might
|
|
have a time window of 1~hour. But at the rapid speed it
|
|
might be reduced to a time window of 20~minutes. Now at
|
|
\qty{30}{\degreeCelsius} (\qty{86}{\degF}), everything moves much faster. Your
|
|
bulk fermentation might be complete in 2--4~hours when using
|
|
\qtyrange{10}{20}{\percent} starter. Proofing your dough in the fridge becomes
|
|
almost impossible. As your dough cools down in the fridge the fermentation
|
|
also slows down. However cooling the dough down from \qty{30}{\degreeCelsius}
|
|
to \qtyrange{4}{6}{\degreeCelsius} in your fridge takes much longer. Your
|
|
dough is much more active compared to a dough that starts at a temperature of
|
|
\qtyrange{20}{25}{\degreeCelsius}. You might end up overproofing your dough if
|
|
you leave it overnight in the fridge.
|
|
|
|
That's why I~recommend that you reduce the amount of starter
|
|
that you use in the tropics to around \qtyrange{1}{5}{\percent}
|
|
based on the flour. This will slow down the fermentation
|
|
process significantly and provides you a bigger window
|
|
of time. Try to aim for an overall bulk fermentation of at
|
|
least 8--10~hours. Reduce the amount of starter to get there.
|
|
|
|
When making dough, try to use the same water temperature
|
|
as your ambient temperature. Assuming that the temperature
|
|
will climb to \qty{30}{\degreeCelsius} try to start your dough
|
|
with \qty{30}{\degreeCelsius} water. This means that you can carefully rely on
|
|
a small fermentation sample (aliquot jar) that visualizes your fermentation
|
|
progress. To read more about this technique refer
|
|
to Section~\ref{section:bulk-fermentation}.
|
|
|
|
The sample only works reliably if your dough temperature
|
|
is equal to your ambient temperature. Else the sample heats
|
|
up or cools down faster. So tread carefully when using
|
|
the sample in this case. It's always better to stop
|
|
the fermentation a little too early rather than too late.
|
|
Stretch and folds during the bulk fermentation
|
|
will help you to develop a better feel for
|
|
the dough. An expensive but possibly useful tool
|
|
could be a pH meter that allows you to perfectly
|
|
measure how much acidity has been created by the
|
|
lactic and acetic acid bacteria. In this case measure
|
|
the pH repeatedly and figure out a value that works
|
|
for your sourdough. In my case I~tend to end bulk
|
|
fermentation at a pH of around 4.1. Please don't just
|
|
follow my pH value; it's very individual. Keep measuring
|
|
with different doughs to find out a value that works for you.
|
|
|
|
\subsection{My flour has low gluten content --- what should I~do?}
|
|
|
|
You can always mix in a little bit of vital wheat gluten. Vital wheat gluten
|
|
is concentrated extracted gluten from wheat flour.
|
|
|
|
I~recommend that you add around \qty{5}{\gram} of wheat gluten for every
|
|
\qty{100}{\gram} of flour that you are using.
|