🍞 Bread Production

Dividing is the first place where a good dough becomes an uneven product. Everything before it is chemistry; from here on it is handling, and handling faults are the ones customers see.

Start with the arithmetic, because the scale is where money leaks. Suppose your formula is the published white pan bread formula: flour 100, water 62, yeast 3, salt 2, sugar 4, shortening 3, milk solids 2. Those percentages add to a total formula percentage of 176.0. You want 20 loaves at 900 g of dough each, so you need 18,000 g of scaled dough. Flour needed is (18,000 divided by 176.0) times 100 = 10,227 g, round to 10,230 g. Water is 10,230 times 0.62 = 6,343 g. Yeast is 10,230 times 0.03 = 307 g. Salt 205 g, sugar 409 g, shortening 307 g, milk solids 205 g. Add them up and you get about 18,006 g, which checks out.

But 18,000 g is what must reach the divider, not what leaves the mixer. Fermentation and scaling losses reduce dough yield and can be up to 5 percent and more depending on how you work. If you mix exactly 18,000 g you will finish the batch one loaf short, every time, and you will blame the divider. Mix a margin, measure your own loss over a week, and then set the margin to your measured number.

The second piece of arithmetic runs the other way. Dough loses water in the oven, so the scaled piece is always heavier than the cooled loaf. Weigh the scaled dough, weigh the cooled loaf, and use: baking loss percent = (scaled dough weight minus cooled loaf weight) divided by scaled dough weight, times 100. There is no published baking-loss percentage you can borrow. It changes with loaf size, tin or free-standing, oven, steam and bake length. Measure your own, then work backwards: scaling weight = target sold weight divided by (1 minus your measured loss as a decimal). If your target sold loaf is 500 g and you measure a 12 percent loss, scale at 500 divided by 0.88 = 568 g. If your country regulates the declared weight of a loaf, this is a legal calculation and not merely a costing one; ask your national weights-and-measures authority what tolerance applies.

Divider accuracy is measurable in ten minutes. Weigh twenty consecutive pieces. If your target is 568 g and your pieces run 540 to 600 g, your spread is 60 g, over 10 percent of target. Underscaling is listed as a cause of poor keeping quality, excess shredding and moulder rejects; overscaling is listed as a cause of hollow bottoms and shredding. Neither is a formula problem.

Rounding comes next, and it has one job: give the piece a smooth, tight outer skin so it holds gas instead of leaking it. A rounded piece with a torn surface will spread rather than rise. After rounding, dough needs an intermediate proof, also called bench rest, before final shaping. Insufficient intermediate proof appears in the troubleshooting literature as a cause of both lack of volume and holes in bread. The dough has been tightened by rounding; it must relax before you can shape it without tearing.

Shaping degasses large bubbles and organises the gluten in the direction you want the loaf to expand. Two handling faults dominate here. Excess dusting flour is a listed cause of holes and coarse open crumb, because dry flour trapped in a seam never hydrates and leaves a void. Excess divider oil does the same thing. Use the least flour that stops sticking, and prefer a scraper to a handful of flour.

A baker in Kumasi, Ghana, ran 900 g pieces by eye and lost roughly one loaf in every batch of twenty to underweight complaints and rejects. At an ingredient cost of about 0.35 USD a loaf and a selling price near 0.80 USD, twenty batches a month meant twenty lost loaves, about 16 USD of revenue and 7 USD of ingredients gone for want of a 25 USD scale. She bought the scale, set the target at a measured scaling weight, and the reject line closed within a fortnight.

A tin does two things for a baker: it carries the loaf's weight so a weaker dough can still stand up, and it fixes the shape so slicing and packing are consistent. A free-standing loaf has to hold itself, which asks more of the flour, the shaping and the proof, but it needs no capital and no tin-washing and it gives crust on every side.

Decide by dough strength first. Bread flour runs 12 to 14 percent protein; all-purpose flour runs 8 to 11 percent and is described as workable for bread but optimal for neither bread nor cake. If your only reliable local flour is in the 8 to 11 band, a tin will carry it and a free-standing loaf will spread. That is a flour decision, not a skill deficiency.

Then decide by oven. A free-standing loaf on a wood-fired deck at around 250 degrees Celsius gets bottom heat straight through the sole and a strong lift. The same loaf on a wire rack in a domestic gas oven with no bottom mass will pale underneath and flatten. Tins buffer that unevenness.

Match the scaling weight to the tin. There is no universal gram-per-litre rule you should copy from a foreign text, so measure: fill the tin to the brim with water and weigh the water to get its volume in millilitres, which for water is grams. Then run a trial. Start with a scaling weight, bake it, and look at the result. If the loaf mushrooms over the rim and shreds down the side, the piece is too heavy or the proof went too far. If the loaf sits below the rim with a tight, flat top, it is too light or under-proofed. Write the working number on a card taped inside your tin store, one card per tin size, and stop guessing.

Proof is where tin and free-standing part company. Recommended final proof conditions are 35 to 37 degrees Celsius at 85 to 95 percent relative humidity, with a standard duration of 60 to 65 minutes and a volume increase of three to four times. Short-mix doughs take up to an hour; improved and intensive-mix doughs take one to two hours; sourdough-leavened doughs take longer than two hours. Under-proofed loaves show small volume, shell tops, inadequate flow and bursting at the sides. Over-proofed loaves show pale crust, coarse grain, poor texture and an acid flavour note. A tin hides some under-proof; it hides no over-proof at all, because an over-proofed piece collapses into the corners and bakes with a hollow bottom.

The faults map cleanly. Flat top with sharp corners: overmixing, improper moulding, under-proofing, or an oven that is too hot. Hollow bottom: over-proofing, a cool oven, improper moulding, or overscaling. Loaf bursts on the side: oven too hot, overbaking, or too much sugar. Excess shredding and capping: improper moulding, improper panning, rough handling, overscaling, over-proofing. Notice how often moulding and proof appear and how rarely the flour does.

Panning matters more than most bakers think. Moisture in pan bottoms and hot pans are both listed causes of moulder rejects, and excessive pan greasing is a listed cause of a pale crust. Pans that come out of the oven and go straight back into use with fresh dough are too hot; let them fall to hand temperature.

A bakery in Arusha, Tanzania, sold a 500 g tin loaf and wanted to add a free-standing round for the weekend market at a higher price. Their flour tested as a general-purpose grade, so the first free-standing trial spread flat. Rather than buy stronger flour at once, they shortened the final proof toward the lower end of the range, shaped tighter, and baked on a preheated steel plate for bottom heat. The round held. When they later moved to a 12 percent bread flour for that line only, they were paying more per kilogram on perhaps a fifth of their volume, which their costing sheet could carry.

Small goods are where a lot of African bakeries actually make their margin. A roll needs no tin, bakes in a fraction of the time, sells for cash at the gate, and turns a single sack of flour into many small transactions instead of a few large ones. The trade-off is labour: dividing and shaping 120 rolls takes far longer than dividing 20 loaves from the same flour.

Do that comparison in numbers before you commit. Take 10 kg of flour on a lean formula whose baker's percentages total 176. Total dough is 10 kg times 1.76 = 17.6 kg. At 900 g a piece that is 19 loaves. At 60 g a piece, allowing for the same dough, that is roughly 293 rolls, though in practice you will scale rolls a little heavier and get fewer. If a loaf sells at 0.80 USD, 19 loaves is 15.20 USD. If a roll sells at 0.10 USD, 293 rolls is 29.30 USD from the same flour. The gross looks obvious until you add the shaping hours and the higher packaging and handling count per unit. Work out your own labour per hour and subtract it before you decide.

Formulation next. A plain roll can sit on the same lean dough as your bread: water 60 to 75 percent, yeast 0.5 to 2 percent, salt 1.50 to 2.25 percent with the optimum between 1.75 and 2.25, sugar 0 to 15 percent, fat 2 to 5 percent and milk powder 2 to 8 percent of flour. A bread dough needs 3 to 3.5 percent fermentable solids for the yeast to work with; below that the yeast is living only on what the flour's own enzymes release.

Enriched buns are a different animal. Sugar above 15 percent classifies the dough as a sweet dough. The yeast feels that as osmotic pressure: sugar at 5 to 10 percent of flour increases yeast activity, but above 10 percent yeast activity decreases. So a bun at 20 percent sugar ferments more slowly than a roll at 6 percent even with the same yeast dose, and you must respond either with more yeast or with an osmotolerant yeast strain sold for sweet doughs. Bakers who simply wait longer get an over-proofed, pale, acid-tasting bun.

Always state which yeast form a formula means. Published dosage bands for lean bread run 0.5 to 2 percent, while one published white pan bread formula uses 3 percent without saying which form. Three percent is a normal fresh yeast dose and a very high instant dose. A formula card that says only "yeast 2 percent" is an incomplete card. Write "instant dry yeast 1 percent" or "fresh yeast 3 percent" and the next person to bake it will get your bread.

Doneness for small goods is a temperature question, not a colour one. Rich-dough products such as dinner rolls finish at 82 to 88 degrees Celsius internal; lean-dough breads finish at 88 to 99 degrees Celsius; scones, biscuits in the American sense, muffins and quick breads finish at 93 to 96 degrees Celsius. A probe thermometer settles arguments that eyes cannot.

Be honest about what is not known. A sourced scone ratio was not available for this course, so do not treat any scone formula you are handed as a standard. What is sourced is the finishing temperature of 93 to 96 degrees Celsius, so build a scone by test bake and let the thermometer confirm it.

A baker in Kaduna, Nigeria, supplying a school gate ran 50 g rolls from a lean dough at 8 percent sugar. When she added a sweet bun at 22 percent sugar for the same run, the buns were still slack after 70 minutes of proof while the rolls were ready at 55. She raised the yeast on the bun dough only and, later, bought an osmotolerant instant yeast for that line. The two doughs now finish together, which matters because they share one oven.

Composite flour means replacing part of the imported wheat in a bread formula with a locally grown flour, most often high-quality cassava flour in West and Central Africa, and sometimes sorghum, millet, maize or plantain. The motive is usually foreign exchange and freight cost, and in several countries it has at various times been government policy rather than a baker's choice.

Here is the honest position this course must take. The published substitution limits for cassava-wheat composite bread could not be retrieved for this reference. Three separate research sources were located and none could be read. That means there is no number in this course for how much cassava flour you may put in a loaf. Anyone who gives you a confident percentage without naming a document is guessing, and in a country where the level is regulated, guessing can be an offence as well as a baking mistake.

What you must do instead is two things. First, find out the rule. Composite-flour policy is live in West and Central Africa and it changes: it has been mandatory in some periods, voluntary in others, and it interacts with import tariffs and with bread standards. Ask your national standards body and your ministry of agriculture or trade. In Nigeria that means the Standards Organisation of Nigeria; in Ghana the Ghana Standards Authority; in Kenya the Kenya Bureau of Standards; in Uganda the Uganda National Bureau of Standards; in Tanzania the Tanzania Bureau of Standards. Ask three questions: is there a mandatory or recommended substitution level for bread; is there a bread standard that specifies composition or labelling; and must a composite loaf be declared differently on the label. Get the answer in writing and keep it in your file, because an inspector will ask.

Second, find your own technical ceiling by trial, because even where a level is permitted it may not work with your flour. The mechanism is simple. Gluten comes only from wheat, rye, barley and related cereals. Cassava, maize, sorghum, millet and plantain contribute starch and flavour but no gluten. Every percentage point of substitution removes gluten from the network that has to trap the carbon dioxide your yeast makes, so volume falls, the crumb tightens, and the dough handles differently at the divider. Water behaves differently too: cassava flour absorbs water at a different rate from wheat, so your usual hydration will feel wrong.

Run the trial as a ladder. Bake five loaves from one mixing session at 0, 5, 10, 15 and 20 percent replacement of the flour weight, holding everything else constant, and record four things for each: final proof time to reach three to four times volume, loaf height in centimetres, crumb appearance when cut, and how the bread eats on day two. You will see a point where volume falls off sharply. Your working ceiling is one step below that point, and it is your number for your flour, not a universal one.

Watch the safety side as well. Maize and groundnut-containing products carry a genuine mycotoxin risk, and the maximum levels for aflatoxin, deoxynivalenol and fumonisin in cereals could not be retrieved for this course. Ochratoxin A in raw wheat, barley and rye has a Codex maximum of 5 micrograms per kilogram, which gives you a sense of the scale of these limits, but you must get the cereal limits that apply to you from your national regulator. Module 8 Lesson 5 goes into sourcing and rejection in detail.

A bakery in Ibadan, Nigeria, ran the ladder over one morning using 2 kg of dough per step. At 10 percent cassava the loaf lost a little height and the crumb was slightly tighter but customers did not comment. At 20 percent the loaf was visibly short and dense and the sales counter noticed within a day. They settled at 10 percent, saved the difference on every sack, and wrote to the standards body to confirm they were within the current national position before printing new labels.

Oven spring is the rapid expansion in the first minutes of baking. Understanding it properly turns most loaf-shape faults from mysteries into diagnoses, because oven spring has a defined beginning and a defined end, and both are temperatures.

The loaf goes into the oven full of gas cells. As it warms, three things push outward at once: the gas already trapped expands, the yeast briefly ferments faster and releases more carbon dioxide, and water and ethanol in the dough vaporise. That expansion continues until the yeast dies, which happens somewhere between 50 and 60 degrees Celsius, with a terminal death point recorded at 59 degrees. Meanwhile the structure is setting. Gluten denaturation begins around 50 degrees and gluten coagulates between 70 and 80 degrees. Starch granules begin to gelatinise between 55 and 65 degrees and the gel forms between 60 and 80. Above 85 degrees the dough has become crumb, and the crumb structure is fully set when the core reaches 91 to 93 degrees.

So oven spring is the window between roughly 50 and 80 degrees at any given point in the loaf. Once the structure sets, nothing more expands. That single sentence explains the two commonest catastrophic loaf faults. An under-proofed loaf still has a great deal of expansion left when the crust sets, so it bursts, usually at the side or under a shoulder. An over-proofed loaf has already stretched its gas cells to failure, so when the heat arrives the cells rupture rather than expand and the loaf collapses. Adding more heat fixes neither.

The core also cannot exceed 100 degrees Celsius at normal pressure, because that is where water boils. That is why every bread doneness temperature you will ever see sits below 100, and why a probe reading of 88 to 99 degrees for a lean loaf, or 82 to 88 for a rich dough, is a real endpoint rather than an approximation.

Loading is the practical half. Heat reaches the loaf by conduction from the sole, radiation from the oven walls and roof, and convection from moving air. In a wood-fired oven all three are uneven, and where you place a loaf decides what bake it gets. Rules that hold across oven types: leave a gap between pieces, because pans positioned too closely is a listed cause of a pale crust; load fast and close the door, because every second open is heat lost from the deck; load the largest or slowest pieces to the hottest zone; and, in a small oven, plan a rotation part-way through the bake rather than pretending the oven is even.

The deck temperature and the dial temperature are rarely the same thing in a small bakery, and in a wood oven there is no dial at all. Improved wood-fired ovens in Africa are documented as baking at around 250 degrees Celsius. Buy an oven thermometer you can put on the deck and learn what your fire actually gives you at the moment of loading, then twenty minutes later. That curve, not the dial, is your oven.

One more trap. A cool oven appears in the fault lists against too much volume, hollow bottom, holes in bread, poor keeping quality and a dark crust, which surprises people. A loaf that sits too long in a slow oven keeps expanding past the point where a properly hot oven would have set it, then dries and darkens over a long bake. If your loaves are pale and yet somehow over-dark on the base, suspect the loading temperature before you blame the flour.

A cooperative in Kampala, Uganda, moved from an open fire to a rocket baking oven. The published figures for these ovens are striking: preheat time cut by at least two thirds, firewood down to about one tenth of previous use, and an efficiency increase around 70 percent. The bakers' first complaint was that their loaves now burst. They were loading a dough proofed to the old, slower oven's timing into an oven that reached temperature far faster. Extending the final proof toward the published 60 to 65 minutes and checking for a three to four times volume increase before loading closed the fault out in two bakes.

A bakery that cannot judge its own bread cannot improve it. Judging means the same six checks, in the same order, on the same day of the week, written down. It takes ten minutes and it is the cheapest quality system available.

Check one, volume. Is the loaf as tall as it should be for its scaling weight? Lack of volume has a long cause list: insufficient yeast, old dough, insufficient intermediate proof, under-proofing, over-proofing, improper mixing, oven temperature too high or too low, rough handling, wrong dough temperature, wrong dough consistency, poor yeast quality, wrong pan temperature, wrong proofer conditions, improper moulding, dough too sticky or too stiff, a dirty moulder. Notice that this is a list of possible causes, not a lookup table. Diagnosis here is elimination. Also consider the flour itself: a flour with a Falling Number above 300 seconds has very little enzyme activity and gives long proof times, pale crust, low volume and a dense crumb.

Check two, shape. Flat top with sharp corners means overmixing, improper moulding, under-proofing or an oven too hot. Hollow bottom means over-proofing, cool oven, improper moulding or overscaling. Side burst means oven too hot, overbaking or too much sugar. Excess shredding or capping means improper moulding, improper panning, rough handling, overscaling or over-proofing.

Check three, crust. Pale crust is the fault most often misdiagnosed as a broken oven. Work the list in order. Is the bake long enough? Is there enough residual sugar left after fermentation, since sugar drives caramelisation from 160 degrees Celsius and Maillard browning from above 105 degrees at the crust? Is the flour's Falling Number too high? Was the loaf over-proofed so the yeast consumed all the sugar? Only then suspect the oven. Excessive pan greasing and pans too close together are also on the pale-crust list. A crust that is too dark points to old dough, a cool oven or underbaking, and, from the flour side, a Falling Number below 220 seconds where sugars from sprouted grain caramelise hard.

Check four, crumb. Cut the loaf and look at the cell structure. Large irregular holes point to excess dusting flour, excess divider oil, insufficient intermediate proof, proofer humidity problems or, from the flour, a low Falling Number. A dense tight crumb points to insufficient proof or a high Falling Number. A sticky crumb that fouls the slicer blade is a classic low Falling Number symptom and is not your fault; it is your flour's, and it is why Falling Number is the single most useful purchase specification a small baker can ask a miller for.

Check five, flavour and aroma. Note anything sweetish, fruity or unpleasant in the crumb smell, particularly 12 to 24 hours after baking in hot weather, because that is the early sign of rope. Poor flavour is also listed against wrapping above 35 degrees Celsius, unsanitary equipment, contaminated wrappers, mouldy racks and tools, dust exposure, old dough, improper mixing and underbaking.

Check six, keeping quality. Keep one loaf from every judging day, bagged and dated, and look at it on day two and day three. Poor keeping quality is listed against old dough, improper mixing, high dough temperature against an optimum of 24 to 28 degrees Celsius, underscaling, improper shortening amount, an over-hot proofer, a cool oven, overbaking and wrapping above 35 degrees Celsius.

Run the review as a table with six rows and one column per week, and force yourself to write one change and one number to prove it. A bakery in Lusaka, Zambia, did this for eight weeks. Their recurring fault was pale crust. Working the list in order, they found their bake was two minutes shorter than their own written standard because the oven was being emptied to fit the next batch. The fix cost nothing, and the pale-crust row went blank in week three and stayed blank.