This week something very exciting happened: I got a grain mill!! Specifically, a Mockmill Lino 200.

I couldn’t wait to use it, so I skipped ahead a few recipes in the King Arthur Baking School book and made whole wheat sandwich bread. But because I’m me I couldn’t just make the recipe. I had to do an experiment. I was curious how the freshly-milled wheat would compare to store-bought flour, so I setup a head-to-head comparison.
I keep my whole wheat flour and my grain in the freezer, so I put them both out on the counter and weighed out the amount needed for my recipe the morning of the bake so they could come to room temperature. I milled my flour, then realized it was significantly warmer than its store-bought cousin and let both flours sit for a while longer so their temperatures could equalize. I was surprised by the difference in colour between these flours, but that is likely due to the type of wheat used. My flour is milled from Hard White Winter Wheat. The store bought flour is Lidl brand. I can’t find a definitive answer on what kind of wheat is used in Lidl brand whole wheat flour, but whole wheat flour is typically milled from Hard Red Spring or Winter Wheat.

For all steps in the baking process I worked with the fresh-milled flour first, and pictures of the fresh-milled flour will be on the left where there is a comparison.
When I couldn’t wait any longer I laid out my mise en place for both recipes. At this point the freshly-milled flour was 76 degrees F, and the store-bought flour was 61 degrees F. I used water from my fridge dispenser and was surprised to see a 10 degree (F) difference between the first and second cups of water (58 vs 67 degrees F), but that ended up working in my favor to equalize the temperature between the two batches of bread.
First I mixed up the dough with fresh-milled flour.




Immediately after, I mixed up the sister dough with store bought flour.




The doughs both felt quite sandy at this point. I left them to rest and hydrate for 15 minutes.

After the doughs had rested I incorporated the All Purpose flour into the dough with fresh flour and then kneaded the dough for 5 minutes. I didn’t stir all of the AP flour into the dough, but whatever was remaining (plus a little extra) got used during kneading.


I followed the same process for the dough with store-bought flour.


After both doughs had been kneaded I took their temperatures (the freshly-milled dough was 75 degrees and the store-bought dough was 76 degrees F), covered them, and let them rise. My kitchen was at a balmy 74 degrees F. After about an hour they passed the doorbell test and were ready to be divided and shaped. Before dividing I weighed both doughs: the dough with freshly-milled flour weighed 1357g and the dough with store-bought flour weighed 1367g.

I watched the video showing how to shape this bread and followed the recommended shaping method for all 4 loaves. After shaping the dough rested for another hour.


Once the final rise was complete the bread went into the oven.


The colour differences between the loaves continued into the baked bread. The bread made with store-bought flour rose higher than the bread made with fresh-milled flour, which I half-expected after reading this article. However, the bread with fresh-milled flour had much better flavor. It had a sweetness and freshness to it, whereas the bread made with store-bought flour (while still tasting very good) had a bit of a bitter flavor.


Both breads had a buttery flavor and a very tender crumb. In my opinion the crumb is too tender to be used for sandwiches, but the bread is excellent toasted.
Until the next time: may your yeast always rise and your flour never run out.






































































