The Fresh Loaf

A Community of Amateur Bakers and Artisan Bread Enthusiasts.

Home Milled High Extraction Sourdough Miche

bwraith's picture

Home Milled High Extraction Sourdough Miche

Retsel Mill and Brass Sieves

Home Mill High Extraction Sourdough Miche

JMonkey's many blog entries on whole wheat, as well several other TFLer's posts, helped me learn to make whole grain breads that are light and flavorful, rather than the rocks and bricks I had thought were inevitable with whole wheat. Then, Peter Reinhart's Whole Grain Breads came out, and I learned more about the reasons for soakers and mashes and gave mash bread a try. Then, I read some posts by Ron, who has posted passionately about whole grains and pushed me to think about trying home milling. Finally, Goetter mentioned stone milling at home with his Retsel, which he said was easy to clean and use, and I was off and running. I just recently received my Retsel Mil-Rite, and I'm happy to say it works just great - quite fast (5 pounds/hr for reasonably finely milled flour) and easy to clean if you need a few pounds of freshly milled whole grain flour.

However, unlike many or most of the home milling aficionados, I am as interested in traditional methods and recipes as I am in the nutritional aspects of bread. As such, I do want to be able to create "high extraction flour" or maybe even some regular bread flour from my whole grain berries to satisfy interests in country style miches and other types of bread that may call for other than pure whole wheat flour. In order to accomplish that, the flour needs to be sifted. MiniOven mentioned brass sieves, and some internet searches revealed a number of places. I finally purchased a range of brass sieves from number 18 through 120 from, hoping to experiment with them to extract more refined flours from my freshly milled whole grain flour. While conversing with the people at Legend Mine, they said it would be backbreaking to sift very much flour by hand, and that I should consider a "sieve shaker". I don't have the sieve shaker yet, but it will arrive soon.

Sieve Shaker

Meanwhile, I went through a laborious process of discovering the right coarseness of grind and which sieves to use. I found that I could get very reasonable results by setting the mill stone adjustment to be just slightly looser than finger tight. The flour coming out was fairly fine - good for a whole wheat bread flour. Yet, it had some percentage of larger particles. I then successively sifted the flour through my #20, #40, and #60 sieves. The #20 caught large particles of bran, about 5% of the weight. The #40 sieve caught smaller particles of bran and other dark parts of the kernel - probably some of the germ from the look of it, with a weight of about 15%. The #60 sieve was catching what I would call a very dark flour, probably some combination of bran, germ, and outer endosperm, about another 15% of the weight. What came out of the bottom of the #60 sieve was very nice bread flour, creamy and slightly dark colored. I'm sure that flour from # 60 would have made a delicious whitish bread. So, the sifting is nowhere near where it could be with a shaker and will never be anything close to the perfect filtering done by commercial mills. However, for my purposes, even this very spur-of-the-moment hand processing was enough to get 65% fresh, creamy, bread flour.

As for grain, I ordered 25 lbs. of Wheat MT Prairie Gold, 25 lbs. of Wheat MT Bronze Chief, 10 lbs. of their wheat berries (hard red winter wheat berries, I think), and 5 lbs. each of spelt and rye berries. I stored them in 6 gallon buckets with screw on hermetically sealed lids and placed oxygen absorbers in the buckets. A 6 gallon bucket comfortably holds 25 lbs. of grain with enough room for the screw-on lid. All the storage buckets and lids were purchased form

Although the Retsel appears to be more than adequate in retrospect, I went off the deep end ealier in the week and ordered a Meadows 8 inch mill also. This one will grind much faster and hopefully won't be too hard to clean.

To create my high extraction flour, I just took the finest 85% that came out of my sifting, which amounted to all of the bread flour (throughs from the #60), all of the throughs from the #40 (a darker semolina-like flour), and some of the throughs from the #20 sieve (very dark, very coarse), such that I had 85% of the total weight of all the flour that I sifted. I then ran the coarser flours back through the mill at a fairly fine setting, which resulted in making those coarse components much more finely milled. I mixed them in with the good bread flour coming out of the #60 sieve, and that is what I used as my "high extraction flour".

I also finely milled enough spelt and rye to make 55g of whole rye flour and 105g of whole spelt flour. I just mixed all the rye and spelt berries together and ran them through the mill once.

I then made my high extraction miche, along the lines of a Thom Leonard Country French with a spelt and rye levain. The overall recipe is 15% fermented flour in a spelt and rye levain, mixed with a soaker of the high extraction flour with 1% malt syrup, 2% flour, and 1 tbsp of diastatic barley powder.

Some photos of the process are posted. Spreadsheets are posted in xls and html format.


  • 30g firm storage starter (any starter will work - use 25% more batter starter or about 50% more liquid starter)
  • 52g whole rye flour
  • 104g whole spelt flour

I mixed this starter at 12:45AM after a night of much experimentation and exercise manually sifting about 10 cups of grain into 40 samples from the sieves trying to figure out the best settings for the mill. The levain was designed to rise by double and ferment an hour or so more by 9:00 AM.


  • 10g diastatic malt powder
  • 15g malt syrup
  • 30g salt
  • 1024g water
  • 1300g home milled and manually sifted high extraction flour

I mixed the soaker in a large bowl using a scraper until it was reasonably well mixed. The mixing was done at about 1:00AM and the soaker was refrigerated overnight.


At 9:00AM in the morning, the soaker was spread out on a wet counter like a great big pizza. The levain was chopped into marshmallow-sized pieces which spread evenly over the soaker and pressed into the dough with the palms of my wet hands. The dough was rolled up and folded a few times, squished all through with wet hands a few times, rolled a couple of times, and placed in my DLX mixer. The dough was mixed/kneaded in the DLX mixer on low to medium for 4 minutes, allowed to rest for 4 minutes, and then mixed for 4 more minutes.

Bulk Fermentation and Folding

The dough was allowed to rise at a temperature of approximately 74F in a cabinet above my coffee machine. Initially the temperature was around 70, but by the end of the bulk fermentation the temperature was up to about 76F. During the bulk fermentation, I folded the dough at 10:40AM, 11:40AM, and 12:40AM. The total bulk fermentation time was 5.25 hours at roughly 74F.

Shaping and Proofing

One large boule was formed at 2:15PM, allowed to sit for 15 minutes on the counter, and turned upside down into one of those San Francisco Baking Institute lined baskets (12" diameter). I dusted the loaf and the basket liner with some of the bran and semolina-like flour from my siftings mixed with a small amount of rice flour. In retrospect, since the dough was not that hydrated (77%), it wasn't necessary to use the rice flour. I could have just used some of my home sifted bran and nothing else.

The basket was placed in a large ZipLoc "Big Bag" with a warm bowl of water and sealed. The proofing temperature was about 75F. I slashed with cross-hatch pattern and baked at 5PM for a total mix to bake time of 8 hours, and a proofing time, starting from 2:15PM of 2:45.


The loaf was baked in my brick oven. The oven was fired earlier and allowed to cool to a hearth temperature of about 450F. I sprayed the loaves with an orchid mister, sprayed the chamber of the oven until it was full of steam (20 seconds), and sealed the door with my wooden wet towel covered door. The bread was rotated every 15 minutes for a total of about 50 minutes bake time. The oven door was left open after 20 minutes, and the hearth temperature dropped to about 400F at the end of the bake.

To do this in a kitchen oven, heat oven to 450F, create steam however you do it, and then drop the oven temperature to about 400F. If the loaf becomes too dark, cover with foil and/or drop the temperature to 350F. Allow to thoroughly bake, so the color of the crust is uniformly dark but hopefully not burnt and the internal temperature is above 205F.


Allow to completely cool before cutting - several hours at least.


The miche has a color that is darker than my usual whole wheat loaves, which may be partly because my sifting wasn't that efficient, partly because the extraction rate may be higher than for Golden Buffalo, which I would normally use for this application, and maybe just the nature of freshly milled flour, which I've never tried before. The texture is definitely lighter and softer than I expect from a whole grain, so the high extraction worked in that sense. The flavor is closer to a whole grain loaf than I expected. If I want a more mild white flour flavor, it may mean using less of the darker, larger particles, i.e. use a slightly lower extraction rate. By the way, the aroma of the fresh flour when mixed with water is most definitely better than anything I've smelled using commercial flour. Everyone in the house commented on the great aroma coming from the dough and the bread. I do believe the flavor and aroma of the bread is enhanced by the freshness of the milling, something commented on by many on the site.

The Next Phase

When I receive my Meadows mill and the sieve shaker, the next phase of the project will be to discover the right settings of the mills and sieves to gain a more efficient separation of the particles from the milling.

Meadows 8 Inch Light Commercial or Home Mill

But Why Did I Do This?

OK, part of it is just fun with gadgets. However, there are several objectives beyond that. One very significant motivation is that I haven't been very happy with the availability of other than white flour or whole wheat flour. I'd like to be able to create flours with various characteristics in the amount I need when I need it. Also, any flour other than white flour will probably have spoilage issues if kept for too long. So, rather than buy a few pounds of some specific flour, pay a lot for shipping, and then use a small amount and throw out the rest when it spoils, I can create the desired flours to order. Much of the bran can be used for dusting or added to cereal, and even the middlings may be tossed into oatmeal or toasted and used in place of wheat germ, as suggested in the Essential's Columbia recipe. If I can make the process convenient and fast, then it will be easier and cheaper in the long run to occasionally buy bulk amounts of a few different berries, as I already just did. Storage is easy for the berries, and they stay fresh for a very long time in berry form.

The result is a drastic improvement in the freshness of my flours, very little waste or spoilage, and much lower cost. I seem to spend upwards of $4/pound including shipping for small quantities to get particular flours I want over the internet. The berries, purchased mostly in 25 lb. quantities, came to less than $2/pound, even if I'm very particular and buy from Wheat MT or Heartland Mill. It could be much less if I can find sources for high quality berries locally. However, it's not a bad guess to say I lose close to half my purchased whole grain flours to spoilage. I could offset the spoilage with flour freezing strategies, but I just think this home milling approach is better. No freezing, easy to use screw-on lids on buckets of grain, and absolutely fresh flour to order. At least, that's what I'm shooting for.

It's true that the cost of the mill and sifting equipment won't be offset by the lower cost of the berries for something like 2-3 years. However, for me the home milling approach is still justified because of the freshness, flexibility of flours I can generate, and the convenience of storing berries. The fact the lower cost will allow for the recover of the cost of the equipment even if it takes a few years is just an added benefit.

Of course, the benefits above are theoretical. Maybe after the next phase, I'll conclude it's not possible to produce the desired flour characteristics with simple sieves and a small stone mill. However, the first phase was almost sufficient, other than the excessive physical effort required to manually sift the flour. If I can make the separation work a little better by discovering the right series of millings and siftings, which should be far easier to do with the sieve shaker, I'm hopeful the results will justify doing it regularly going forward.


fleur-de-liz's picture


I am impressed... and exhausted. Your process is fascinating, but, most importantly, your bread looks delicious.

A few questions: what is the gauge of the mesh in your sieves? Your geological sieves seem to have a different gauge for measuring the size of the mesh. I have a standard tamis that is used for sifting flour and the gauge is listed at .24mm. I found that at that gauge most of the larger particles were sifted out of the whole wheat flour. I didn't weigh to see how much came out during the sifting, but I assume most of the larger particles of bran were removed. My understanding of Golden Buffalo is that it is whole wheat with just the coarsest bran removed.

Perhaps you could explain the rationale behind your process. From what I understand, the flour was refined almost to the point of white flour, and then you added back in some of the coarser elements and milled them more finely?

I emailed Heartland a while back with some questions about Golden Buffalo and received an email back from Thom Leonard. He wrote: "Extraction rates of stone milled flours is really not a good indicator of flour refinement as is ash level. Extraction rate is simply how much flour the miller gets out of 100 pounds of wheat."

So, do we need to start charring our flour to determine the ash content?

More seriously -- your experiments are extremely interesting to me. I totally agree with you about the practicality of grinding your own grains. I also tired of the high shipping costs, and all those little bags of miscellaneous flours in the refrigerator, and then sometimes not having what I wanted when I was ready to bake. With my mill, I ordered organic rye berries and organic winter wheat from a local health food store. The rye is spectacular. The wheat initally tasted very grassy, but with Subfuscpersona's guidance, after milling I aged it for two weeks and it now tastes very good. As a test, I also ordered some spring wheat (Bronze Chief) from Wheat Montana which was wonderful right away and didn't need aging. While the freshly milled whole wheat tastes vastly better than store bought whole wheat, I am totally knocked out by the quality of freshly milled rye. It is exponentially better than packaged rye.

I would like to better understand the rationale behind your process as I would have thought you would have achieved something akin to Golden Buffalo between sieve #20 and #40.

Looking forward to more milling experiments!



bwraith's picture


Now that I've gone through this for the first time, I have an inkling what Thom Leonard was talking about. Obviously from the experiment (I wish I had taken photos of the resulting flour from all the different permutations I ran, but I was so tired just from hand sifting that I was too tired to take photos) you could get all kinds of different separations of the darker and lighter elements of the grain, depending on how coarse the milling was and what size sieves were used. So, it's hard to say how much dark vs. light matter went into the final product, if you eliminated the largest 10% after all that. For example, if you grind very fine, and sift out the largest 10% will you get the same stuff in the flour that you get if you grind very coarse, sift out the largest 10%, then grind the remainder very fine? I expect to try some experiments like that when I have the sieve shaker. My hope is that the color and texture of the flour will give me a clue without having to do an ash test. However, maybe I should just figure out how to do an ash test at home. I am betting there are some milling experts who could send me in the right direction as to the likely best milling strategy for various types of flour, but I haven't run into the right person so far.

As far as the sieves, there is a table of sieve sizes at the site I purchased from. There is also this particle size conversion table. I'm not sure if the gauge you mention is the same as the opening size. If so, you can see the .25 mm opening is a number 60 sieve, which is the finest sieve I used. I finer ones, but it would have been pretty tough to work with them by hand. The output of the number 40 sieve is a somewhat coarse flour, a little like semolina in coarseness.  The output of the #60 sieve is finer flour, more like typical bread flour. I'll have to try running some of the Golden Buffalo and Wheat Montana flours through the sieves to see what size particles are in them, but again that just has to wait for a sieve shaker. I was absolutely exhausted from just the 40 hand siftings I did the other night.

If you put Wheat Montana flour, which is hammer milled, through the sieves, you can see it is uniform and very fine. The stone ground whole wheat from Heartland Mill has larger and smaller particles in a much wider range. My stone mill produces flour more like the Heartland Mill flour, especially if I set it for a coarser grind. My plan is basically to find the right sequence of coarse grinding, sifting, finer grinding, and possibly more sifting, maybe even three stages, if it isn't too difficult, that would hopefully allow a successful refinement. Nonetheless, I thought the result was good just doing the very simple thing I did. The miche crumb was less dense and the dough texture was clearly more in the direction of white flour than a whole wheat flour would normally produce.

Yes, the basic process I followed was to do a slightly coarse grind, sift through #20, 40, and 60 sieves (an arbitrary choice based on reading that a #20 would catch very large bran only, #40 would output something like coarse bread flour, and number 60 would output fairly fine bread flour, which seemed about true although the output of #40 was still a little coarser than I would've expected), then choose the finest 85% of the flour from that sifting. I then ran the coarser results chosen from what was caught in the #60 sieve and a small amount of what was caught in the #40 sieve back through the mill. I didn't remill that which fell through the #60 sieve, which already seemed very fine. Also, I tried putting a small amount back through the mill, and it had a tendency to go very, very slowly, whereas what was caught in the #60 and larger seemed to run through the mill almost as fast as raw grain. The process resulted in a somewhat darker whole wheatier bread than I expected, although the final flour I used looked a lot like Golden Buffalo. It did have fairly coarse whole rye and spelt 15% of the flour, that I used to make the levain, so that may have contributed to a bit more whole grain-like bread.

Now, the rationale for the process is harder to explain, since I may not have had one, hehe. OK, so I guess the thought was the if you grind way too finely, nothing can be separated on the basis of size. I imagined that a coarse grind would result in bran particles being larger and inner endosperm particles smaller (hopefully, wishfully) so that you could separate them with a sieve. However, when I ground very coarsely I had a large amount of what looked more like cracked wheat, and so I think a lot of white material was still stuck on very large particles. On the other hand grinding very finely results in all the flour falling through even the finer sieves. So, somewhere in between maybe there is a point where you get an good separation of white from dark? That's a question, not a statement. After that, I figured that once separated, you could recombine and remill the coarser stuff to get an overall fairly fine and uniform flour. It's easy to see that you could get more or less of the darker stuff or not, in the 85% you separate out, depending on exactly the process followed. So, TL's comment to measure the ash content, which should more accurately measure the proportion of the outer layers of the berry in the flour makes sense to me.

The resulting bread was a little grassy but not heavily so. I suppose a small amount of aging might have helped. I could use your advice or subfuscpersona's or whoever else has insight into this, as to the right way to age flour. However, does aging result in a loss of the freshness and nutritional benefits? What is the trade-off there?




tattooedtonka's picture

Now I understand the Retsel mill up top, pretty simple concept.  But wow, that Shaker Sieve in the middle photo, how does that work?  I dont know, but it sure is shiny.  I could even tell the kids its a force field generator.  Its real neat looking. 

And for the mill on the bottom, Im taking by the side housing it is either belt, or chain drive.  But why such a long housing, placing the motor so far away.  If it were for speed, they could have just used different size pulleys, or gears depending on whats under the housing.  With the motor closer to the mill, the vibration it will generate would be less, not to mention it would take up alot less space.  You could take the 5-6 foot contraption and bring it down to about 2 1/2'  Im guessing.  Now I understand you didnt make it, I was wondering if you know why they made it like this, did it have some milling specific reason?  Oh well, they sure are neat looking toys either way.

And as always your Sourdough Miche looks awesome.


bwraith's picture


I don't have the sieve shaker yet, but basically it's a big motor that just shakes the "H" out of those sieves until all the stuff that's small enough gets through them. The idea is to stack them up from coarse to fine, and then get a distribution of flour from top to bottom, that hopefully also is from outer to inner kernel. Not sure that part will work with just sieves as opposed to some crazy fancy separation machines in a real mill. On the other hand, my understanding is that in the old days, silk and other cloth was used to just sift the stuff, so it wouldn't be that different from a series of sieves (maybe, I sure hope so). Who knows, but I'm giving it a try as soon as I get my shaker.

As to the shape of the mill, check out several configurations at Pleasant Hill Grain. I'm not sure why assembly "B" is offered, but it made sense for my particular space. Some of the assemblies conform better with your design philosophy. I chose assembly "B" because I want to put the darn thing in my basement workshop, which is down a narrow stairway. Assembly "C" would not fit down the stairway. Also, I can set assembly "B" against the wall next to a receptacle and place my sieve shaker next to it. They will both take up a small footprint along a wall, although they are a little tall. The airspace above them probably would not have had a use anyway, so this configuration seemed to fit down my narrow stairs and also seemed to use vertical space instead of spreading out over a wider footprint horizontally. If it turns out the unit has problems from the long drive belt, which I hadn't really considered, then maybe one of the other assemblies would be better. Assembly "A" would also have worked, for example. I guess we will get to check this out when I receive it. I haven't even heard whether it has shipped yet, but I'll report further when I get to try it out.

I do feel the miche turned out very well with the fresh milled flour, regardless of the mystery of just what resulted from the sifting. The flavor is very good, and I do think the aroma of freshness emanating from the dough and the bread was undeniably better compared to what I'm used to, even with very high quality flours from Heartland Mill or Wheat Montana. Also, the texture of the crumb clearly was lighter than I normally would accomplish with whole wheat, even with the benefit of recent soaking and mash techniques from WGB and JMonkey and others. That tells me the sifting process did accomplish a refinement, even if not quite as light as what I would get with Golden Buffalo.


Uberkermit's picture

Hi Bill,

 Thanks for your detailed notes on this project! It does seem like you might need to figure out how to do an ash test on your flour. For example, it could be that what's making it through your finest sieve is close to 90% extraction by itself (maybe some of the bran & germ is just being milled very finely). In this case you would be able to get high-extraction flour, but at less than 90% yield.

Afraid I'm not sure how you would go about doing an ash test. I know KAF conducts their tests when the flour is at 14% moisture content. The European system dessicates the flour before incinerating it (ie, 0% moisture content). Seems like at the least you would need a bunsen burner/hot plate of some sort, a crucible, and a scale accurate to the mg or below. You could skip the dessicator, and just calibrate your results by performing the same test on flour of known ash content. I imagine the lab equipment could quickly become a significant expense in itself!


bwraith's picture


I wonder if a simple light meter would work for at least a relative measure. I found one for $139 (Technika 84022, I wonder if you could use some standard illumination of the flour itself or a carefully made paste of the flour and measure the light on the sample?

It might be a way to get some objective measure at home.

Meanwhile, I have some very fine sieves, but I haven't tried them yet. I am waiting for the sieve shaker. The finest one I used was a number 60, which seemed to give quite fine flour. I was tempted to just use that flour, which was about 65% yield or so depending on the mill coarseness setting, but it was light enough that I thought maybe it was closer to a white flour than I wanted. In retrospect, maybe not, as the resulting bread had a basically whole wheat color and some whole wheat flavor, more than with Golden Buffalo. The thing is, the resulting crumb was more open than I would expect from the same bread made with whole wheat flour and the same levain.

So, I'm just not sure at this point. I probably have made a very high ash content flour, but not as high as whole wheat, notwithstanding the color, but I can't quantify the result at all at this point.

Much more to follow once I can do more and finer samples with a sieve shaker. I'm also hoping the larger mill may have a finer control over the coarseness of the grind.

That was one disappointment with the Retsel, although overall it is an excellent machine. The coarseness setting is a very large awkward screw. For my purposes, it would help a lot to have a much more finely threaded screw that would allow much more adjustment of the spacing between the stonesin the relevant range. From screwing the stones tightly together (a little abitrary in itself), I only back it off about 5-10 degrees, and the variation in the coarseness is significant over probably only 1 degree. So, it's almost impossible to be consistent that way.



zolablue's picture

Really beautiful bread - WoW!  And very impressive as usual. Boy, guy, you don't ever do anything in a small way, huh!  (heehee)  I just get such a kick out of you.  I mean, I know what it is like not to be able to shut down the brain but you are really something else! 


This is really an interesting undertaking and I admire your ability to figure this all out.  Wow, again!  I hardly know what to say.  I was planning at some point on getting the grinder to fit on my DLX but.... 


Anyway, I envy that you are able to do this but like Fleur, it almost makes my brain tickle.  I will look forward to reading about your new venture and am sure you will never cease to amaze!

bwraith's picture

Hi ZB,

Nice to hear from you. I knew you would find this project amusing, at least, and maybe pretty interesting.

I think Fleur probably had it all boiled down to its essence long before me, though. I think she just does a hand-sifting on her finely ground flour, probably the #60 that seems to be the right one to get a high extraction result. It may be that simple if you don't want to get all crazy and technical and go way off the deep end like me.

So, you might well get some worthwhile results with the DLX and one good quality sieve, if you wanted to just dip your toes in the water.

In any event, I do think that even just for unsifted pure whole wheat flour, it's well worth trying to mill your own, based just on the freshness and flavor. I was by nature a little skeptical coming into this, but the difference seemed very clear to me, short of blind taste testing to confirm it.


charbono's picture


I'm wondering how much of the fresh flavor/aroma of home-milled flour is obscured, or lost,  by anything other than a straight, same-day dough process. 



bwraith's picture


That's a good question. My experience has been with either store-bought flour or freshly milled and straight into levains and soakers, so far. I've read some of  the discussions in the thread about aging in the forums. It sounds like aging for a few days might be beneficial in some ways. Proth's comments have motivated me to try tempering. I doubt the tempering would have very much effect on freshness, since the berry is just absorbing a small amount of extra moisture if done at room temperature for only 24 hours. Maybe proth5 can comment on that if listening.

I don't know how quickly I'll get around to trying aging, which may have a bigger trade-off against absolute freshness, even if only for a few days. Right now I'm too impatient to try the results of each milling and sifting session. Also, I felt the baking qualities of the flour freshly milled and sifted, especially the third pass, was very good. The slightly flat loaf probably would have happened to me with store-bought flour too, because it was a little too wet and was allowed to overproof a little. I didn't notice the grassy flavors being too strong, especially with passes two and three. My wheat was the M2 product from Heartland Mill by the way.