The Fresh Loaf

News & Information for Amateur Bakers and Artisan Bread Enthusiasts

Does sifting whole wheat flour produce the equivalent of high extraction flour, with wheat germ?

  • Pin It
venkitac's picture
venkitac

Does sifting whole wheat flour produce the equivalent of high extraction flour, with wheat germ?

As I understand it, high extraction flour has around 10% bran, and all the germ. I want to try using this, but it's unavailable in all nearby stores, and shipping 10 pounds of flour from wherever seems like a bad idea. I decided to sift whole wheat flour and use that instead. I have KA organic whole wheat and Bob's RM organic whole wheat flours at home. I sifted the KA, which seems like a coarser grind, and weighed the flour: about 11% of the whole wheat flour was removed in sifting. With BRM, which seems like a finer grind, about 7% of the flour was removed in sifting. I'm going to bake with the sifted BRM tomorrow. Question is, does sifting remove only the bran, or does it remove the germ too? If it removes only the bran, sifting whole wheat at home seems like a good idea to approximate high extraction flour..

ericb's picture
ericb

In general, the method home bakers use to approximate "high extraction flour" is to mix in a little bread flour with the whole wheat. Peter Reinhart recommends 75% whole wheat flour. Hamelman suggests 80-95% whole wheat (I'm working from memory here, but the numbers are close).


Based on several threads about this topic, I think the consensus is that sifting does *not* accurately reproduce high extraction flour. As you suspected, you will lose both bran and germ when sifting WW flour (although I don't know exactly how much of each). I tried it once, but any improvement was imperceptible, and it was a lot more trouble than it was worth.


I have used 75%-100% whole wheat in recipes calling for high extraction, and have been very satisfied with the results. 


Eric

venkitac's picture
venkitac

Thanks! Since I sifted already, I have like 5 pounds of flour on my hands, I'm going to try bake with it and see how it turns out..and then I'll try the mixing suggested by Hamelman.

subfuscpersona's picture
subfuscpersona

...because the bran coating is the hardest part of the grain. Based on my experience milling with a hand mill and motorized rotary grain mills, the germ is softer and is mixed into the finer flour.

Proth5 would know a lot more about this. Maybe you could email her directly from the forum (?).

suave's picture
suave

High extraction flour has extraction rate of 85% and ash content of 1.10-1.15%.  Whole wheat flour is 100% extraction and ~1.7% ash. Bread flour is 72% extraction rate and 0.55% ash.  So if you do the math you'll see that you need to use 45-50% of whole wheat flour to arrive at proper numbers.  As such I find numbers suggested by PR and JH in particular, implausible.  And, FWIW, photos of high-extraction miche in JH's book shows very light-colored bread, about what I get when I use 50% of white whole wheat in the recipe.

venkitac's picture
venkitac

That math works if you go by ash content. I believe high-extaction flour has only 10% bran content, though (I could be wrong, of course). So, if you go by the "bran content logic",  Hamelman's numbers are plausible..I think it might just be hard to approximate high-extraction flour by mixing..

suave's picture
suave

It has 10% of original bran content, not 10% percent overall bran content, that is it is milled in such a fashion that 90% of bran is removed.  Considering that wheat kernel is 17% bran, removing 90% of it gives you exactly 85% extraction rate.

venkitac's picture
venkitac

Right, my logic is this (sorry if I'm being dumb here): 85 extaction flour has 2% bran content. If you want a mix of whole wheat and bread flour that has 2% bran content, you take 10% whole wheat flour (of which 17% is bran, which is approximately 2% of the tota of 100%l), and mix it with 90% bread flour, for a total sum of 100% flour - of which 2% is now  bran. Right?

suave's picture
suave

Except that Hamelman suggests the opposite -  85-90% of whole wheat flour.

venkitac's picture
venkitac

Argh! I was reading it the other way round:) You're right of course.


I baked with my sifted flour yesterday. It is definitely milder by a lot, just removing that 6-7% of the bran makes a big difference. (I suppose I can call my sifted flour 93% flour?) I want to try a miche with it, but it'll have to wait while I get my sourdough story straightened out....

rayel's picture
rayel

Hi venkitac, you could try adding back wheat germ, like the toasted variety which keeps quite a long time. How much is lost to sifting would be the tough part. If you ended up with too much wheat germ how bad would that be?    Ray

venkitac's picture
venkitac

Unclear - I'm really not sure as to how much germ is lost, according to Eric, some would be. According to subfuscpersona, probably not a lot:) I think the only real way to find out is to order high-extaction flour, bake with my sifted flour and high-extraction flour side by side, and do a test. I plan to do that some day, but not in the next few weeks. I'll definitely update when I do it...

jkandell's picture
jkandell

Or you can go the opposite way: add germ and bran to white flour.


If whole wheat flour has 2.5% germ and about 14% bran.


Then 100g of 20% "reduced bran" flour would be  2.5 g germ, 2.8 g bran, 94.7g all purpose flour.


A "high extraction" flour that left only 10 percent of the bran would be 2.5g germ, 1.4g bran, 95.1g all purpose.


Is my math right?


I've seen other formulas such as 93g white, 5g germ, 2g bran.

sdwatts's picture
sdwatts

Has anyone read this article? 


http://hamelmanchallenge.blogspot.com/2010/06/tech-note-high-extraction-flour.html


I just worked the formula on a spreadsheet to make 85% Extraction rate flour using Bread flour which should be about 72% extraction and Whole wheat 100% extraction. I'll be baking tommorrow.  I'm going make the Pain ai Levain (page 158 - Hammelmans's BREAD)


768 grams of flour needed (349g WW & 419g BF)


 

suave's picture
suave

The problem with his reasoning is that he adopts an overly simplistic way of looking at the structure of the kernel.  That is if the inners 72% of the kernel were endosperm, and outer 28% were uniform bran then his calculations would work.  But wheat kernel is anything but that, the outer layer are not uniform, instead you have aleurone layer, and then several layers of bran each slightly different, different enough to have its own name.  The end result of this is that you get a non-linear distribution of minerals and nutrients through the outer layers of the kernel.  My point is that when you mix using this simple ratio you're  adding more stuff than should be there. 


Let me illustrate.  So you have 60/40 WW/bread ratio based on extraction rate.  Let's take a look at the ash.  Ash content for bread flour is ~.5, high extraction ~1.20, ww is ~ 1.9.  The equation you need to solve then is


0.5x + 1.9(1-x)=1.2


Solving this gives you x=0.5 that is based on ash content the correct mixture should be 50/50.


But what makes the story even more complicated is that these distributiona are not the same for different ingredients.  Let's look at the fiber.  The numbers are, bf ~ 0.2, he ~ 0.8, ww ~ 2.3.  So the new equation is


0.2x + 2.3(1-x)=0.8


Here x=0.71, so based on fiber the correct ratio is 30/70, predominantly white flour.  And if you average ash and fiber calculation you get 40/60 - exactly the opposite of the original ratio.


The real proof for me lies in the photo in Hamelman's book. His miche is incredibly light colored, and I hope anyone who has baked with red whole wheat knows that to get that color you need to go way below 50%.

MichaelH's picture
MichaelH

I recently measured the moisture content of my berries. I regularly mill red, white, spelt, rye and durum. The moisture content of my grains ranged from about 5% to 7%. I tempered the grain by adding water over 4 days to reach an estimated 13-14% moisture. After milling the grain I sifted through a #20 sieve. The results were dramatic. Under magnification the residue in the sieve was almost totally bran, and the bran flakes were several times larger than those in the dry batch. They weighed about 8% of the original sample. The #30 sieve yielded another 10%, so there was obviously some germ and maybe other parts of the berry in this sieve.

I intend to continue this experiment until I can get an 85%-90% result.

clazar123's picture
clazar123

I must not be using the correct key words because I haven't been able to locate any sieves numbered for fineness.Anybody have links,sources or just correct key words?

thomaschacon's picture
thomaschacon (not verified)

The keyword is "mesh". 

Mesh material is often used in determining the particle size distribution of a granular material.

Here's the wiki article: http://en.wikipedia.org/wiki/Mesh_(scale)#Sieve_sizing_and_conversion_charts

Here's a 16-mesh flour sifter, which translates to ~1.18 mm per "screen hole": http://www.webstaurantstore.com/stainless-steel-flour-sifter/4071260.html 

Here are several 20-mesh sieves, so screen holes are about .88 mm: http://www.bigtray.com/searchresults.asp?s=sieve&cmdGo.x=0&cmdGo.y=0&RN=1

More here when I searched for "mesh" on bigtray.com: http://www.bigtray.com/searchresults.asp?s=mesh&cmdGo.x=0&cmdGo.y=0&RN=1

You'll see some that are described as single or double mesh. That just means the number of screens per sieve/strainer/sifter. A double sieve, for example, will have two meshes, usually with different mesh sizes. 

proth5's picture
proth5

that I used.  Be warned that these are not actually intended for use as flour sifters, but they have done yeoman's worrk for me for creating non-whole grain flours.

www.lmine.com

Look for soil classifiers - they come in plastic (which some people fear) and brass (which cost a pretty penny) models.

Happy Sifting!

charbono's picture
charbono

There are lots of numbered test sieves on ebay.

clazar123's picture
clazar123

Good to have the correct terminology!

Miller1's picture
Miller1

Whole wheat flour by definition is as has been pointed out a 100% extraction product with the flour containing all of the portions of the grain in the same percentages as would be found in the whole grain.  We produce a flour which is called "Unbleached Wheat Flour" Others do as well.  I can't speak for them but I can tell you that our unbleached wheat flour is made by sifting off a portion of the bran.  We have a sifter with 14 screens in it. When we produce whole wheat flour the final screen is a US 40 mesh when we produce the unbleached wheat flour the final screen is a US 50 mesh.  Remember the higher the number is US Sieves the finer the granulation. By putting in a finer screen for unbleached wheat flour we are able to get the bran off the top of the screen.  So we end up sifting out about 80% of the bran from the whole wheat flour.   This produces a flour lighter in color then whole wheat with excellent wheat flavor characteristics.  This is an excellent flour for Miche. Of course as has also been pointed out this reduces the ash content of the flour as well. The problem you run into when trying to sift out the bran on your own (not that it cant be done) is that the particle size of the flour once in the bag is generally fairly consistent and therefore hard to separate, when done in the mill the bran is flaky and easier to remove.

Hope this helps

ppschaffer's picture
ppschaffer

Discussion by Miller1 is VERY helpful to me.  Thanks to him/her and all others who have contributed to what I consider to be a "thorny" topic.