Monday, June 28, 2010


I love a good bagel and have been baking them for several years using recipes from Peter Reinhart. So I was very interested to see what twists Hamelman might come up with. Is there something else I could be doing to bake an even better bagel? There are several key differences in the recipes, including the use of a bagel board. It all seemed very interesting.

Using Hamelman's recipe, I've now made bagels three times. The first try, I follow the recipe very closely using King Arthur's high gluten flour. I faked the bagel boards using a fresh cedar plank I had bought for the grill along with a linen dish towel. (The approach worked well except the kitchen smelled of pencil shavings.) The bagels were very nice but a bit gnarly looking. When boiled, they were slow to float and didn't rise or round out very nicely. Flavor, however, was not a problem. They tasted great.

The first batch exhausted my supply of high gluten flour. For the second and third batches, I had to adjust. For the second batch, I created my own high gluten flour by adding vital wheat gluten to bread flour. (See my previous post, Tech. Note: High-gluten Flour & Vital Wheat Gluten, for the detail on using vital wheat gluten.) For my third batch, which I cooked at the same time as the second batch, I used straight bread flour without adding any vital wheat gluten. I wanted to see if it really made a difference. Also, after shaping, I let the later batches sit out on the counter for another 20 to 30 minutes, that is, until that passed Reinhart's bagel floating test, before I put them into the refrigerator overnight. I was hoping the bagels would be a little less gnarly. Finally, I baked these directly on a baking stone rather than bothering with the bagel board, something I found to be a real nuisance.

The picture above shows my later efforts. I didn't taste bagels made with high gluten flour side-by-side with those made with bread flour with added vital wheat gluten, but as memory serves me, I don't think there was much difference. There was, however, a striking difference between those made with just bread flour and those that had the added gluten. It is definitely worth the very minor effort need to add the vital wheat gluten. The bagel without the addition were more bread-like lacking the chewiness I've come to expect of bagels. This was not a problem with the bagels with the added vital wheat gluten. I won't be in any hurry to order more high gluten flour.

Also, waiting until the bagel passed the float test proved to be a good idea. I got fuller looking bagels that quickly floated when boiled. And as to the bagel board, once was enough. Cooking directly on a baking stone worked just fine. Overall, these were fine bagels and not that different from Reinhart's. Still, in the future, I probably go back to Reinhart's recipe, if only because it is what I'm used to.

Friday, June 18, 2010

Tech. Note: High Extraction Flour

The recipe for Miche calls for high-extraction flour. From looking around the Internet and in some of my bread books, there appears to be some confusion about just what high-extraction flour is and how you can come by it (or fake it). So it seems worthwhile to describe what I've been able to piece together.

First, recall that a kernel is composed of three parts: indigestible outer coating or bran; the embryo or germ, the part that will develop into a plant; and the food for the developing plant or endosperm. With bread flour the germ and bran are removed leaving just the endosperm. With whole wheat flour, the germ and bran are also included in the flour.

Extraction is a measure of how much flour is extracted or milled from a given amount of wheat, i.e., how much flour is left after remove bran and germ. If you start with 100 pounds of wheat and end up with 70 pounds of flour, then you have an extraction rate of 70%. With whole-wheat flour, the entire wheat kernel is used so 100 pounds of wheat yield 100 pounds of flour and you have an extraction rate of 100%. With typical bread flour, the extraction rate is much lower, in the neighborhood of 70% to 75%. High-extraction flour lies somewhere between bread flour and whole-wheat flour. That is, more (but not all) of the bran and germ included in the flour. You can think of whole-wheat flour as the highest-extraction flour. Since high-extraction flour lies between whole-wheat and bread flour, its baking properties lie between the baking properties of whole-wheat and bread flour.

Since high extraction flour can be difficult to locate, home bakers typically simulate it. There are two approaches. First, you can take coarse or medium ground whole-wheat flour and put it through a sieve to remove some to the larger pieces of bran in the flour. If you start with 100 grams of flour and you "extract " or sieve out 10 grams of bran, then you are left with flour with a 90% extraction. It may seem odd that you are removing or "extracting" bran to get high-extraction flour, but you can think of it as going from "highest-extraction" flour to "high-extraction" flour. (I suspect this double use of "extraction" is a major part of the confusion.) You'll need to weight the flour before and after to determine the extraction rate. The extraction fraction is just the weight of the flour left divided by the weight of the flour you stared with.

Unfortunately, unless you have a whole range of different sieves, you won't have a lot of control over the extraction rate. Also, you'll need to use coarse or medium ground whole wheat flour, or everything will likely pass straight through your sieve.
Working with King Arthur Organic Whole Wheat flour and using two different sieves, I was only able to remove about 4% of the bran.

A second, simpler approach is to simply combine bread flour and whole-wheat flour. This is much easier to do, but, unfortunately, calculating the quantities needed can be more complicated. Using the definition of extraction, I worked out the following formulas to calculate flour quantities. If

Q = quantity of flour desired
R1 = desired extraction rate (average high extraction rate)
R2 = given extraction rate (for bread flour)
W = amount of whole-wheat flour to use (100% extraction)
B = amount of bread flour to use (R2 extraction)


B = Q (R2 -R1R2)/(R1 - R1R2), and,
W = Q (R1-R2)/(R1-R1R2)

For example, if you want 100 grams at 85% extraction and you have 70% extraction bread flour, then

B = 100 * (0.7 -0.85*0.7) / (0.85 -0.85*0 .7) = 41 grams, and,
W = 100 * (0.85-0.7) / (0.85*0.7-0.85) = 59 grams.

This is the approach I used with my second go at the miche and it worked great!

(There remainder of this post can be safely skipped by most readers. At this point I explain how I arrived at these equations. But, if you are so inclined, please read on and let me know if you see any problems.)

So, if you are interested, here is my reasoning:

First, a little background. Lets assume you want high-extraction flour with a 90% extraction weight and you are starting whole wheat flour and bread flour with a 70% extraction rate. If you have 100 grams of bread flour, that was what was left when 100/0.7 = 142.86 grams of flour was extracted. To get a 90% extraction rate, you will want to add enough whole-wheat flour so that the ratio of the total flour you have is 90% of the whole wheat flour plus the weight of the original flour that was extracted. If x represent the added whole-wheat flour, you'll need to satisfy the equation:

(100 + x) / (142.86 + x) = 0.90

That is, you'll need to add 285.71 grams of flour given a total of 385.71. In general, you can take the amount of flour you'll want and multiply it by 100/385.71 = 0.259 to get the amount of bread flour you'll need and multiply it by 285.71/385.71 = 0.741 to get the amount of whole-wheat flour you'll need.

Generalizing, let Q be the quantity of flour you desire, R1 is the extraction rate you desire, R2 is the extraction rate you have, W the amount of whole-wheat flour you will use, and B the amount of bread flour you will use. First, clearly the total amount of flour is just the sum of the whole-wheat and bread flour: Q = W + B

Next, the total amount of flour I would have started with before removing the bran/germ would be Q/R1, or, W + (B/R2). In the first case, we are averaging the removal over both flours, i.e., finding the effective extraction. In the second case, we are calculating the removal based on the actual bread flour used. Think of it this way---starting with a fixed amount of whole-wheat flour, we could sieve all the flour, case 1, or we could separate out part of the wheat flour, sieve the remaining wheat flour, and then add the removed flour back, case 2. Since these are equivalent, we can write: Q/R1 = W + (B/R2).

Its all downhill from here. Substituting (Q-B) for W and solving for B gives the first formula. Substituting (Q-W) for B and solving for W gives the second formula.

Thursday, June 17, 2010

Miche, Point-a-Calliere

It has been my stated intention to try every recipe at least twice, but after doing this loaf the first time I was full prepared to renig on that promise. This recipe does makes a large, spectacular loaf with the dough weighing in at over three and a half pounds. But my first result was a heavy, chewy, extremely sour loaf that cooked unevenly. Admittedly, most of these are flaws that I should have been able to address and overcome. But I felt the sourness was characteristic of the bread, i.e., something that I'd gotten right, that was intended by the recipe, and that I really didn't like.

Having made the decision to stop at one try, I began writing up my results. One reoccurring annoyance with Hamelman is that he calls for unusual ingredients but doesn't provide adquate guidance in obtaining these ingredients. The descriptions can be sketchy and there is no list of sources in the book. In particular, this recipe calls for high-extraction whole-wheat flour.

I looked back over Hamelman to see what I could say about high-extraction flour and realized that I really didn't have a handle on this. After turning to my library and to the Internet, I began to realize that this was part of my problem. Hamelman states that if high-extraction whole-wheat flour isn't available, use a blend of 85 to 90% whole-wheat flour and bread flour for the remainder. For the first loaf, I used a ratio of 12 grams white flour to 88 grams of whole-wheat flour, i.e., 88% whole-wheat and 12% bread flour. Once I had a better understanding, I was able to calculate the actual extraction-rate for my loaf. Assuming bread flour is 75% extraction, that actually works out to be 96% extraction. This is virtually a whole-wheat loaf. This brings me to another of my personal preferences. I've quite enjoyed the transitional whole-wheat breads I've made, but I'm not that big on 100% whole-wheat.

Hamelman doesn't say what extraction he uses or recommends but does mention that he uses a high-extraction flour with a 0.92% ash content. Pyler gives a table of ash content vs. extraction and from this I extrapolated an extraction-rate of 84% for Hamelman's flour.
With this in mind, I went back and recalculated my flour ratios. Assuming 75% extraction for my bread flour, I worked out conversion factors for an 85% high-extraction flour: 46.4% whole-wheat and 53.6% bread flour. I also decided to increase the initial mixing time to three minutes and use three folds with a goal of shortening the fermentation time to reduce the acidity. Overall, I cut the final fermentation by around 45 minute to an hour.

The results were a total reversal. I produced a wonderful loaf that I love. Not at all dense, it was only mildly acidic. The loaf wasn't perfect, but it was greatly improved and is now one of my favorites. It is definitely worth going back to and trying to perfect.

Tuesday, June 15, 2010

Light Rye Bread

In general, rye bread is not one of my favorites. It can turn out heavy or gummy. It is fairly easy to overwork. Still, a light, well made rye is a joy. I particularly like rye toast spread with crunchy peanut butter. So Hamelman's promise of a "Light" Rye Bread was tempting.

The first hurdle, however was flour. The recipe calls for high-gluten flour and I'd used up the last that I had. So, as described in the previous post, I mixed my own. Aside from this, I stuck very close to the recipe sprinkling the bread with caraway seeds after misting the top. I was rewarded with a wonderful light rye bread.

The second time around, I decided to try using a loaf pan. I also omitted the caraway seeds. The results were acceptable, but not spectacular. The loaf did not rise as well an was a little dense. Apart from the pan, it doesn't seem that I did anything different. Perhaps this is just a temperamental loaf. But I certainly won't use a loaf pan again.

Tech. Note: High-gluten Flour & Vital Wheat Gluten

Thus far, two of the recipes have called for high-gluten flour---the Light Rye Bread and the Bagels. Neither is a surprise. Since rye has less gluten, a high-gluten flour often added to "make up the difference", so to speak. Bagels need the added gluten to help provide the added chewiness.

At this point, I've only made the Hamelman bagels once. I'll add a post once I've had an opportunity to revisit them. I've made the rye twice and will post on it in the next day or so. With the bagels, I used up the last of the high-gluten flour I had. This is something I can't find locally and have yet ordered any more. So with the rye bread, I've had to "mix" my own high-gluten flour using bread flour and vital wheat gluten, both items readily available locally. This seems to have worked well, so I thought I'd describe the calculation I made. (I plan to use this technique when I revisit the bagels. I'll post my results when I do.)

For those of you who might not be familiar with it, Baking Science & Technology by the late E.J.Pyler is often cited as the definitive book on the science of bread. A couple of years ago Sosland Publishing enlisted L.A.Gorton to bring out a fourth edition of the book. This is a massive two volume work. I only have the first volume (772 pages) which set me back $165.

Pyler provides a table for raising the gluten level in flour through the addition of vital wheat gluten. The table has a row for starting protein levels from 6% to 16%. It has a column for target levels from 7% to 17%. Each entry in the table give the amount by weight of vital wheat gluten that needs to be added to raise the protein content to the target level. I used this table to determine how much vital wheat gluten to add. (I'm tacitly assuming that all vital wheat gluten is the same, or, at least, that the vital wheat gluten I have is typical.)

For my starting point, I'm using King Arthur Bread Flour with a protein content of 12.7%. For my target, I'm using King Arthur High-gluten Flour which has a protein content of 14.2%, or a net increase of 1.5%. The nearest table entry in Pyler says to add 1.49 to increase from 13% to 14%. So, extrapolating, an addition of 1.5 x 1.49 = 2.235 is needed. (Pyler uses pounds but I see no reason why units matter as long as measurements are done consistently.) In other words, to go from bread flour to high-gluten flour, add 2.235 grams of vital wheat gluten to ever 100 grams of bread flour you are using.

Extrapolating a bit further, since 100 of every 102.235 grams of the mix is bread flour, you'll multiple the amount of high-gluten flour called for by 0.978 to determine the amount of bread flour to use. Since 2.235 grams of every 102.235 grams of the mix is vital wheat gluten, multiple the amount of high-gluten flour called for by 0.0219 to determine the amount of vital wheat gluten to use.

For example, the Light Rye Bread calls for 1 lb, 11.2 oz or 772 grams of high-gluten flour. I would use 772 x 0.978 = 755 grams of bread flour and 772 x 0.0219 = 17 grams of vital wheat gluten. These are the weights I used when making the rye bread and they seemed to work fine.

Monday, June 14, 2010

Vermont Sourdough Breads

"One" of the June breads in the Hamelman Challenge is Vermont Sourdough. Actually, there are three recipes: Vermont Sourdough, Vermont Sourdough with Whole Wheat, and Vermont Sourdough with Increased Whole Grain. Basically these are the same recipe except the first is 10% rye, the second is 10% whole wheat, and the third is 15% rye. So, is this three recipes or one?

When listed in the Challenge, they were given separate number as though separate recipes. But if separate, then we have five new breads for June rather than the usual three. In the spirit of the Mellow Bakers, it is up to the baker to decide how to count these.

In fact, these recipes immediately suggest even more variations. For example, per the recipe, the final fermentation can be 2, 8, or 18 hours---it's the baker choice. There is also some ambiguity about what flour to use. The recipes specify bread flour. The text preceding the recipes specifically call for 11.5% to 12% bread flour. King Arthur's All Purpose Flour is 11.7%. King Arthur's Bread Flour is 12.7%. It would seem AP flour would be the logical choice based on percentage of protein. (Of course, this raise a whole lot of questions about the other recipes in the book.)

For this project, I elected to bake five types of bread. I made each of the three recipes using KA AP Flour and the overnight refrigerator retard. Additional, I made the first recipe with KA AP Four with a 2 hour fermentation and with KA Bread Flour with an 2 hour fermentation. With the overnight retard, I went directly from the refrigerator to the oven as suggested by Hamelman. Apart from these variations, I pretty much stuck to the book.

In general, there really wasn't a lot of difference among the three recipes. I had a slight preference for the wholewheat, but I would be happy with any of the three. As promised by Hamelman, the overnight retard help to develop the sour taste more. But, in the interest of full disclosure, I'm not really a sourdough fan. So, based on taste, I preferred the shorter fermentation. Moreover, I found that the dough baked directly from the refrigerator did not produce the save volume of bread, as the dough with the that had the shorter fermentation, i.e., did not have the same oven spring. So I preferred the texture of the bread with the shorter fermentation time as well. And since it takes a day less to make, for me, it won hands down.

As to bread flour vs. all purpose flour, either seemed to work well. I was careful to weight flour and water without making any adjustments. Not surprisingly, the dough from the AP flour seemed a little wetter, but not by much. And the crumb seemed a bit more open. In the photo, the slice at the bottom of the picture is 24 hour retard AP four, the middle slice is 2 hour retard AP flour, and the bread at the top of the picture is 2 hour retard bread flour.

While I don't lament having made this/these recipe/recipes, I think I'll stick to Hamelman's Rustic Bread recipe.

Tuesday, June 8, 2010

Rustic Bread

Hot Cross Buns was the first recipe selected in the Hamelman Challenge and was largely a trial run. The next official recipe was the Rustic Bread. It along with Bagels and Rye filled out the first full month of recipes.

As a place to start, the Rustic Bread was an inspired choice. This is definitely a solid recipe and a practical place to begin. The inclusion of 10% rye and 10% whole wheat gives a bread that is neither a wimpy white bread nor an heavy, stogy whole grain bread. The use of a pre-ferment, rye, and whole wheat build a satisfyingly complex flavor. 60% hydration gave a loaf that had a more open crumb, but that will still relatively tight and useful as a sandwich bread. While I love ciabatta, it is not the most practical bread for a sandwich with a leaky filling. This bread would work fine. The only down side to the bread is the boules I’ve made have been a little too flat. This is something I don’t fully understand, but need to work on.

I’ve made the recipe three times now and definitely plan to continue making it. The first time through, I stuck closely to the original recipe. I used 10 rather than 9.6 oz of water and still had a shaggy pre-ferment. The only other problem I encountered was the pre-ferment did not completely mix into the dough on my first try. This gave a bread with a slightly marbled appearance. Otherwise, the bread was fine and the problem was easy to correct in subsequent loaves. I simply added the pre-ferment to the mixer in smaller pieces and a bit earlier.

The second and third tries I played around with the cooking methods. The second time around I made one boule using a cloche lid over the bread. (I soaked the lid and then preheated it along with the pizza stone starting in a cold oven.) With a 1½ pound loaf, the boule had too much oven spring. It expanded all the way to the cloche and stuck to it. Still, I got a great crust this way and a really good oven spring. With the third try, I tried a 1 pound loaf with the cloche and this worked nicely. Still, I was right to the limit of the capacity of my cloche. I’ve also used some of the dough to make smaller rolls decreasing the cooking time. This worked very nicely.


Cornbread, for me at least, is something of a misnomer. This is not the chemically leavened batter bread that contains more cornmeal than flour, what I normally associate with the term. Rather, it is a yeasted bread that is only 25% cornmeal. It produces a lovely golden crumb when made with yellow cornmeal and an attractive crust.

The recipe calls for fine cornmeal. On both occasions that I’ve made the bread, I’ve used Bob’s Red Mill Medium Cornmeal. This is what I happened to have on hand, and it is generally what I prefer when making traditional cornbread. Still, I really should go back and make the bread again with a fine cornmeal just to see what difference it makes. This won’t be a problem. This is a bread that I particularly like, and it is not difficult to make. This bread is definitely a favorite.

I used cornmeal directly from the freezer, so the dough was a little cooler than desired, around 72 degrees, but this didn’t seem to create any problems. Perhaps it took a bit longer to rise, but I didn’t do a comparison. The bread uses a poolish, but the poolish is made without the cornmeal. The cornmeal is soaked briefly, about 15 minutes, to soften it. It would be interesting to repeat the bread adding the cornmeal to the poolish. Peter Reinhart’s Struan, of which this bread is vaguely reminiscent, soaks the corn meal overnight (along with other grains) claiming it produces a better flavor.

I’ve made this both as a boule (shown) and in a loaf pan. In many ways the loaf bread is more practical. This is a hardy bread that works very well with sandwiches, particularly egg salad, chicken salad, or pimento cheese. The boule gave a rather flat loaf (something else to work on) that wasn’t a good shape for a sandwich. It also makes a very nice toast. I prefer a jam, blackberry for example, rather than a jelly because the textures marry well.

Yet to do? I want to repeat this using a fine cornmeal, and altering the poolish to include the cornmeal. It would also be interesting to compare the recipe to Struan.

Update (6/28/10): This weekend I repeated the recipe replacing the corn meal with corn flour making a loaf bread. This worked very nicely. The result of a very light, soft bread without the graininess of the earlier loaves, but the loaves still had a nice corn meal flavor and color. The results are show to the right.