Showing posts with label Recipes. Show all posts
Showing posts with label Recipes. Show all posts

11/28/2015

How many grams is a cup of flour?

A key milestone for my bread making was beginning to measure flour with a scale. It happened very shortly after I started, and I believe it was key to the early success that kept me baking. I still advise anyone attempting bread baking to use a scale for the best results.

Recently, I received another request from my mother for my bread recipe converted to volume measurement. Such requests had been made just about every time I went home. Instead of treating this one as a heretical request like those past, I decided to embrace it as a challenge. I asked myself if it would be possible to reach some degree of accuracy using measuring cups and spoons.    

My enlightenment was partly inspired by Alice Medrich’s recent book Flavor Flours. As with most good baking books, she provides mass measurements for all of the formula. She also describes how to measure flour by volume and provides a helpful appendix of volume measurements for the non-traditional flours used in the book. The key was providing a standard way to measure the flour in her discussion on measuring by volume.

Using Alice's description as a starting point, over a weekend, I conducted an experiment and collected some data. I decided to test the primary flours I use for baking bread. This had the advantage of also extending the information provided by Alice's appendix. The flours tested were:

  • Unbleached All-Purpose Wheat Flour (Gold Medal)
  • Unbleached Bread Wheat Flour (Gold Medal)
  • Whole Wheat Flour (King Arthur)
  • Medium Rye Flour (Grocery Store Bulk )
  • White Rye Flour (Grocery Store Bulk)
  • Whole Spelt Flour (Arrowhead Mills)
  • Semolina Flour (Caputo)

I experimented with three different techniques to measure the flour. The different techniques demonstrated both the strong dependence on measuring technique, and the variability that occurs when measuring flour by volume. The three measuring techniques were as follows:

  • The first technique I call the “tamped” method. This is the “take the measuring cup and just scoop the flour out of the container” method. In most cases, flour started out rather compacted in the container and got tamped into the measuring cup by the side of the container on the way out. The only refinement in this method was a kitchen knife was used to strike the excess flour off the top of the measuring cup.
  • The second technique I call the “fluffed” method. In this case, a kitchen knife was used to aerate or fluff the flour in the container before the measuring cup was carefully used to scoop flour from the center of the container. Care was taken not to compact the flour on the way out. As with the tamped method, the kitchen knife was used to strike any excess flour off the top of the measuring cup to maintain uniformity.
  • The last technique tested was to spoon flour from the container into the measuring cup until there was a heaping mound. As with the others, a kitchen knife was used to strike the excess flour off to get a uniform measure. For this last technique, I selected the extremely original name of “spooned” method.   

For the three standard wheat flours and the two rye flours, I collected data for each of the three techniques. I don’t keep large amounts of spelt and semolina flour around so I only collected data for the fluffed and spooned techniques. In both cases, I didn’t feel I had a large enough volume to consistently collect data using the tamped technique.

To collect measurements for a full-cup of flour, I measured out four cups of flour and recorded the mass using a digital kitchen scale after each cup. After measuring four cups, I returned the flour to the storage container and repeated the procedure two more times for a total of three trials. After some subtraction, this resulted in twelve measurement samples for each experimental run.

When you combine the 7 flours with the 3 measuring techniques and 3 trials for each, you end up with a total of 57 experimental trials and 228 individual data points. A Google Docs spreadsheet was used to summarize the data and compute some statistics. With 12 measurements for each combination of flour and technique, the standard deviation for most of the combinations was in the 2-3 gram range. The tamped technique showed the most variability; this is an additional finding I will talk more about in the closing. The table below gives the mass in grams for 1 standard US cup (240 ml) of flour using each of the three measuring techniques.  

Mass in Grams per US Cup
Tamped
Fluffed
Spooned
AP Flour
148
140
135
Bread Flour
150
143
135
Whole Wheat
155
143
133
White Rye
130
120
112
Med Rye
125
119
112
Whole Spelt

149
138
Semolina

170
155

I also wanted to see how reliable the measurements would be when using a half-cup and a quarter-cup measure. I did experimental trials with bread flour as before, except I switched the size of the measuring cup from a full-cup to a half-cup and then a quarter-cup. The results were compared to what would be predicted by just taking the corresponding fraction from the mass for a full-cup. With surprising consistency, a lose of about 6% is observed as you move from one measuring cup size down to the next. For instance, a half-cup (fluffed technique) of flour measured out to be 67 grams when it would have been predicted to be 71 grams when you take the measurement for the full-cup (143 grams). For simplicity, 6% was selected as a uniform correction factor to generate the following tables for half-cup and quarter-cup measurements.   

Mass in Grams per US Half-Cup
Tamped
Fluffed
Spooned
AP Flour
69
66
64
Bread Flour
70
67
63
Whole Wheat
73
67
63
White Rye
61
56
53
Med Rye
59
56
52
Whole Spelt

70
65
Semolina

80
73


Mass in Grams per US Quarter-Cup
Tamped
Fluffed
Spooned
AP Flour
33
31
30
Bread Flour
33
31
30
Whole Wheat
34
32
29
White Rye
29
26
25
Med Rye
27
26
25
Whole Spelt

33
30
Semolina

37
34

For completeness I provide the following table for tablespoons, however it is just a guide. By the time you get down to a tablespoon fraction, the error of the experimental measurements starts to be on the same scale as the amount of flour you are measuring. You would need to actually collect experimental data using tablespoon measures to resolve this. For the same reason, I only present the fluffed technique.

Mass in Grams per US Tablespoon
Tamped
Fluffed
Spooned
AP Flour

8

Bread Flour

8

Whole Wheat

8

White Rye

7

Med Rye

7

Whole Spelt

8

Semolina

9


One additional outcome is a recommendation on measuring technique. As hinted before, the tamped technique of measurement had the most variability in the experimental data. This stands to reason given the nature of the food stuff that is being measured. I would recommend sticking to either the fluffed or spooned methods. This is consistent with the advice given by Alice, and I have found the same guidance in other cookbooks as well.    

The goal of this exercise was to come up with a reasonably accurate way to measure flour using traditional volume measuring cups. The tables presented accomplish that goal, and give a series of conversions that allow someone to go from grams to US cups, half-cups and quarter-cups. Happy baking.  

6/18/2011

French County Bread in a Fancy Shape

I like a crusty loaf of creamy white bread as much if not more than most people.  However there are those, usually with letters after their name, that say we should eat more whole grains.  “I make joke!”  I am a big proponent of the use of whole grains and especially rye in breads.  Aside from the health benefits, whole grains add a depth of flavor that is not possible with a highly refined flour.  

There are many recipes for “Country French” or Pain de Campagne.  This is my version and builds on the basic French bread of a previous post by substituting whole wheat and whole rye flour for some of the white flour.  It is usually pretty easy to take any recipe that has all white flour and add up to 15% by weight of whole grains without much change.  You will usually only need to increase the amount of water to compensate for the higher absorption characteristics of the whole grain flour.  One other quick note about using whole grain flour.  Make sure you are using fresh whole grain flour.  The natural oils it contains will go rancid after a couple of months.  Unless you are going to use the flour in 3 months or so, you should freeze it.  

A country bread dough like this can be used much like a basic white bread dough and formed into a free standing bâtard or other standard shape.  However, there is a tradition in France where breads are formed into different regional shapes which can be very decorative for a table setting or dinner party.  This past holiday season, I was asked to bring bread to dinner and I made several couronne Bordelaise which translates into “Bordeaux crown”.  I used the excellent blog post at the Wild Yeast site as a starting point for the shaping procedure I present below.      

Preparing the Dough

The procedure to prepare the dough is the same as my procedure for making a basic French bread.  Follow the same instructions for making the preferment, mixing and fermentation using the following recipe. 

Poolish Preferment
4.90 oz Unbleached Bread Flour (Gold Medal Better For Bread) *
1.65 oz Whole Wheat Flour **
6.50 oz Water (room temperature)
Pinch (scant 1/16 tsp) Instant Yeast

Final Dough
4.45 oz Unbleached Bread Flour (Gold Medal Better For Bread) *
9.10 oz Unbleached All Purpose Flour (Gold Medal Unbleached All Purpose) *
1.65 oz Whole or Dark Rye Flour **
9.35 oz Water (room temperature)
⅜ Tsp Instant Yeast
0.45 oz Non-Iodized Salt
All of the Poolish Preferment

Additional Ingredients
Extra flour for dusting
Spray oil
1 cup ice cubes


* As with the basic French bread, If you use King Arthur flours, use the All Purpose type for all the flour in the recipe.
** Any whole wheat and dark rye would do but I use Hodgson Mill stone ground whole wheat graham style flour and stone ground rye which is a pumpernickel style for added texture in this recipe.  

This recipe will make enough dough for 2 free standing bâtards or 1 couronne Bordelaise.  If you want bâtards, follow the rest of the basic French bread procedure with the only modification of bake for a total of 45 to 50 minutes.  Due to the higher hydration, you need to bake a little longer.  If you are up for a little extra work, my procedure for making a couronne Bordelaise follows.  

Forming a Couronne Bordelaise

Before embarking on a couronne Bordelaise, some thought has to be given to how you are going to proof it.  As Susan of Wild Yeast points out, you could blow a bunch of cash and get a purpose made proofing banneton.  I don’t have one but I have used her other method of improvising one using a dish towel draped over a shallow bowl with a small upturned bowl in the center.  It works well but I only have one suitable shallow bowl.  However, I stumbled upon a third option while wandering around the housewares section of a local hardware store.  The store carried an inexpensive plastic party dish for holding chips with an small bowl in the center for holding dip.



Assuming you can find a similar dish, you will want to take a small dish towel or maybe even some newspaper to fill in the inner dish.  If you cannot find a similar party dish, I suggest you use the two bowl method from the Wild Yeast post.  In either case, drape a clean lint free dish towel over the whole thing and try and make it as smooth as possible in the depression.

NOTE - At this point, preheat your oven with a baking stone as if you were making regular French bread.



Using a small hand sieve, lightly dust the inside with some white flour to prevent the dough from sticking.  I like to also sprinkle some coarse graham flour on top of the white flour to give the eventual surface of the bread some texture but this is optional.

When the dough is ready to divide, weigh out one piece at 6 oz and the rest should divide up into nine 3.5 oz pieces.  Form into rough balls, cover with cling wrap, and let rest for 15 minutes. 



After the dough has rested, start by taking the 6 oz piece of dough and placing it on a floured surface.  Degas and start to shape into a flat disk which will form the “collar” to hold the crown together.  Using a rolling pin, flatten to a disk about 11” in diameter.  You can be generous with the flour to prevent sticking.  As before, I use some graham flour at this point to give the collar some exterior texture but this is optional.    


After you have rolled out the disk, dust with flour and using the end of the rolling pin or another blunt object, feather the edge all around the disk.  Later when the bread bakes, this will crisp up to give the distinctive lip all the way around.  When done, carefully roll onto the rolling pin to transfer to the proofing form.  Unroll and adjust the disk so it is centered in the form and extends well into the depression bottom evenly around.  


NOTE - Take care to keep the edge of the collar dry.  If it gets wet, it will stick and not form a lip during baking.    

At this point, take a pastry brush with a little water and wet the inner “hill” of the collar.  This will help to bond the rest of the crown to the collar.  Degas and form the 3.5oz pieces of dough into tight balls and place evenly around the collar to form the crown.  To help with spacing, I start out by evenly spacing 3 of the pieces and then filling in the rest.  For this crown, the inside of the collar ends up being a little to large so I take a kitchen shears and remove a circle of dough from the top of the "hill".  This will make it easier when you form the tabs.  With the pastry brush, dab the top of each ball of dough with a little water.  Using the kitchen shears again, cut evenly spaced tabs from the collar and fold each over onto the crown.  Cover with cling wrap and proof for 30 to 45 minutes.  



As with my French bread recipe, the bread will be proofed when it passes the dimple test.  When ready, place a piece of parchment paper on top of the proofing basket and then place a baking sheet on top of that.  Invert the stack to transfer the bread to the baking sheet.  Carefully remove the dish towel and dust off any excess flour with a dry pastry brush.  Place in preheated oven on baking stone, add pan of ice for steam and bake for a total of 45 minutes, rotating the bread half way through.  


Once out of the oven, let cool for about 1 hour before serving.  The bread should have a nice crust with a chewy texture to the crumb.  The result pictured above shows what happens if the collar gets a bit wet at the edge.  It won’t separate as it bakes, leading to an uneven lip.  However that is part of baking bread.  You share the results with friends and try again.